issue_owner_repo listlengths 2 2 | issue_body stringlengths 0 261k ⌀ | issue_title stringlengths 1 925 | issue_comments_url stringlengths 56 81 | issue_comments_count int64 0 2.5k | issue_created_at stringlengths 20 20 | issue_updated_at stringlengths 20 20 | issue_html_url stringlengths 37 62 | issue_github_id int64 387k 2.46B | issue_number int64 1 127k |
|---|---|---|---|---|---|---|---|---|---|
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-15 15:30:42 +0100
From: baidongli
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @yzchang
Last updated: 2019-01-21 16:34:19 +0100
## Comment 16888
Date: 2012-02-15 15:30:42 +0100
From: baidongli
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0.1) Gecko/20100101 Firefox/10.0.1
Build Identifier: MonetDB 5 server v11.7.5 "Dec2011" (64-bit, 64-bit oids)
You're correct, MonetDB does not currently support non-ASCII names in identifiers. The syntax using double quotes (select source as "我" from
session;) would be the correct one if we did support it, but we don't.
Please file a bug report for this (set Severity to Enhancement).
On 2012-02-15 10:10, Dongli wrote:
> Hi folks,
>
>
>
> Please take a look at following issue that I got today:
>
>
>
> The following queries I executed aim to test if Monetdb supports GBK
> characters(Chinese), since I have to use Chinese characters in my
> system to display the reports.
>
>
>
> I’m not sure if I missed some setting such as ‘set characters=’utf-8’?
> The following alias after the ‘as’ is a Chinese character ‘I’.
>
>
>
> sql>select source as '我' from session;
>
> 42000!syntax error, unexpected STRING in: "select source as '我'"
>
>
>
> sql>select source as 我from session;
>
> unexpected character (U+6211)
>
>
>
> sql>select source as "我" from session;
>
> Invalid identifier '我'
>
>
>
>
>
> Thanks in advance,
>
> Tony
>
Reproducible: Always
## Comment 17694
Date: 2012-08-24 14:56:06 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18131
Date: 2012-11-27 15:47:49 +0100
From: @yzchang
No test needed for feature request
## Comment 26822
Date: 2019-01-21 16:34:19 +0100
From: @sjoerdmullender
We support just about any name, as long as it is in double quotes.
| Does not currently support non-ASCII names in identifiers | https://api.github.com/repos/MonetDB/MonetDB/issues/3010/comments | 0 | 2020-11-30T12:07:22Z | 2024-06-27T11:55:15Z | https://github.com/MonetDB/MonetDB/issues/3010 | 753,423,125 | 3,010 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-15 15:28:46 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @drstmane
Last updated: 2012-02-29 10:34:22 +0100
## Comment 16887
Date: 2012-02-15 15:28:46 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.19 (KHTML, like Gecko) Chrome/18.0.1025.11 Safari/535.19
Build Identifier:
I reduced the query to:
sql>select 1 as y, (select x from (select 1) as test where x = y);
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffebaa7700 (LWP 30558)]
0x00007fffec41da1a in sql_ref_inc (r=0x0) at sql_mem.c:41
41 assert(r->refcnt > 0);
(gdb) bt
0 0x00007fffec41da1a in sql_ref_inc (r=0x0) at sql_mem.c:41
1 0x00007fffec385703 in rel_dup (r=0x0) at rel_select.c:40
2 0x00007fffec397c41 in rel_select_exp (sql=0x13dfd90, rel=0x15071f0, outer=0x1506f20, sn=0x1506c40, ek=...) at rel_select.c:4621
3 0x00007fffec3995cb in rel_query (sql=0x13dfd90, rel=0x1506f20, sq=0x1506c40, toplevel=0, ek=...) at rel_select.c:4972
4 0x00007fffec39ad33 in rel_subquery (sql=0x13dfd90, rel=0x1506f20, sq=0x1506c40, ek=...) at rel_select.c:5328
5 0x00007fffec396bab in rel_value_exp2 (sql=0x13dfd90, rel=0x7fffebaa6a30, se=0x1506c40, f=3, ek=..., is_last=0x7fffebaa6934) at rel_select.c:4321
6 0x00007fffec396ecb in rel_value_exp (sql=0x13dfd90, rel=0x7fffebaa6a30, se=0x1506c40, f=3, ek=...) at rel_select.c:4385
7 0x00007fffec396fb9 in column_exp (sql=0x13dfd90, rel=0x7fffebaa6a30, column_e=0x1506d40, f=3) at rel_select.c:4398
8 0x00007fffec39722b in rel_column_exp (sql=0x13dfd90, rel=0x7fffebaa6a30, column_e=0x1506d40, f=3) at rel_select.c:4442
9 0x00007fffec39736b in rel_simple_select (sql=0x13dfd90, rel=0x1506f20, where=0x0, selection=0x1506370, distinct=0) at rel_select.c:4468
10 0x00007fffec3995a3 in rel_query (sql=0x13dfd90, rel=0x0, sq=0x1506e70, toplevel=1, ek=...) at rel_select.c:4969
11 0x00007fffec39ad33 in rel_subquery (sql=0x13dfd90, rel=0x0, sq=0x1506e70, ek=...) at rel_select.c:5328
12 0x00007fffec39add0 in rel_selects (sql=0x13dfd90, s=0x1506e70) at rel_select.c:5339
13 0x00007fffec38569a in rel_semantic (sql=0x13dfd90, s=0x1506e70) at rel_semantic.c:297
14 0x00007fffec2f2a1a in sql_symbol2relation (c=0x13dfd90, sym=0x1506e70) at sql.mx:1616
15 0x00007fffec2e1402 in SQLparser (c=0x668328) at sql_scenario.c:1386
16 0x00007ffff6a7f154 in runPhase (c=0x668328, phase=1) at mal_scenario.c:522
17 0x00007ffff6a7f2af in runScenarioBody (c=0x668328) at mal_scenario.c:564
18 0x00007ffff6a7f559 in runScenario (c=0x668328) at mal_scenario.c:601
19 0x00007ffff6a805e7 in MSserveClient (dummy=0x668328) at mal_session.c:438
20 0x00007ffff2ea9d0c in start_thread () from /lib64/libpthread.so.0
21 0x00007ffff2befd7d in clone () from /lib64/libc.so.6
Reproducible: Always
## Comment 16913
Date: 2012-02-16 15:02:13 +0100
From: @sjoerdmullender
Changeset [41944f467334](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=41944f467334) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=41944f467334](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=41944f467334)
Changeset description:
Added test for bug #3009.
It does indeed cause a crash.
## Comment 16941
Date: 2012-02-19 18:20:52 +0100
From: @drstmane
Appears to be fixed in the default branch (only) after changesets
http://dev.monetdb.org/hg/MonetDB/rev/8ee024048e42
http://dev.monetdb.org/hg/MonetDB/rev/e2d3ecc7237c
http://dev.monetdb.org/hg/MonetDB/rev/03df27701cf7
http://dev.monetdb.org/hg/MonetDB/rev/70dc90e1de2f
Remains open/failing in the Dec2011 branch.
## Comment 16944
Date: 2012-02-19 18:32:56 +0100
From: @skinkie
http://dev.monetdb.org/hg/MonetDB/rev/e2d3ecc7237c
Probably needed to backport just that one :)
## Comment 16945
Date: 2012-02-19 18:46:54 +0100
From: @drstmane
Fixed (also) in Dec2011 by transplanting changeset [e2d3ecc7237c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e2d3ecc7237c) from default to Dec2011:
http://dev.monetdb.org/hg/MonetDB/rev/98fe009c5e33
| Segmentation fault on simple query | https://api.github.com/repos/MonetDB/MonetDB/issues/3009/comments | 0 | 2020-11-30T12:07:19Z | 2024-06-27T11:55:14Z | https://github.com/MonetDB/MonetDB/issues/3009 | 753,423,102 | 3,009 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-14 18:54:46 +0100
From: Simon Brodt <<simon.brodt>>
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-21 10:36:09 +0100
## Comment 16885
Date: 2012-02-14 18:54:46 +0100
From: Simon Brodt <<simon.brodt>>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.2.18) Gecko/20110614 Firefox/3.6.18
Build Identifier: MonetDB 5 server v11.7.5 "Dec2011"
Using a boolean attribute as a condition in the WHERE part of a query results in a incorrect outcome.
Reproducible: Always
### Steps to Reproduce:
1. SELECT * FROM (SELECT CAST(false AS BOOLEAN) AS b) AS t1 WHERE b;
2. SELECT * FROM (SELECT CAST(false AS BOOLEAN) AS b) AS t1 WHERE false;
3. SELECT * FROM (SELECT CAST(false AS BOOLEAN) AS b) AS t1 WHERE b = false;
4. SELECT * FROM (SELECT CAST(false AS BOOLEAN) AS b) AS t1 WHERE b = true;
### Actual Results:
+-------+
| b |
+=======+
| false |
+-------+
+---+
| b |
+===+
+---+
+-------+
| b |
+=======+
| false |
+-------+
+---+
| b |
+===+
+---+
### Expected Results:
+---+
| b |
+===+
+---+
+---+
| b |
+===+
+---+
+-------+
| b |
+=======+
| false |
+-------+
+---+
| b |
+===+
+---+
## Comment 16962
Date: 2012-02-21 10:36:09 +0100
From: @grobian
What does "WHERE b" mean? SQL is unlike C-semantics not auto-booleaning or something. So your WHERE condition is very vague, if not unclear. Would it mean true/false? Would it mean NULL/not-NULL?
I think the problem here is that you have a certain idea on what you want the condition to be, but you don't make it explicit to the engine. The engine won't choose just an approach. It always sees a value (even NULL) so it always evaluates to a matching tuple.
| Incorrect evaluation of boolean terms | https://api.github.com/repos/MonetDB/MonetDB/issues/3008/comments | 0 | 2020-11-30T12:07:16Z | 2024-06-28T07:23:13Z | https://github.com/MonetDB/MonetDB/issues/3008 | 753,423,065 | 3,008 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-14 18:34:04 +0100
From: Simon Brodt <<simon.brodt>>
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-21 11:04:46 +0100
## Comment 16884
Date: 2012-02-14 18:34:04 +0100
From: Simon Brodt <<simon.brodt>>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.2.18) Gecko/20110614 Firefox/3.6.18
Build Identifier: MonetDB 5 server v11.7.5 "Dec2011" Compiled for x86_64-pc-winnt/64bit with 64bit OIDs dynamically linked
OS: Windows 7 Professional 64 bit SP1
When using the EXCEPT operator then the columns of the two input tables are matched by position and not by name.
Reproducible: Always
### Steps to Reproduce:
1) SELECT * FROM ((SELECT 0 AS a, 1 AS b) EXCEPT (SELECT 0 AS b, 1 AS a)) as t;
2) SELECT * FROM ((SELECT 0 AS a, '1' AS b) EXCEPT (SELECT '0' AS b, 1 AS a)) as t;
### Actual Results:
+---+---+
| a | b |
+===+===+
+---+---+
types char(1,0) and tinyint(8,0) are not equal for column 'b'
### Expected Results:
| a | b |
+===+===+
| 0 | 1 |
+---+---+
| a | b |
+===+===+
| 0 |'1'|
+---+---+
## Comment 16961
Date: 2012-02-21 10:23:09 +0100
From: @grobian
From the SQL99 specification:
... each input table must have the same number of columns and each corresponding pair of columns must have comparable data types. Usually, your DBMS decides which column to merge with which column by looking at ordinal position; it merges the first columns from each input table, then the second columns, and so on.
The standard does not restrict ordinal position matching only, but it certainly suggests it is the minimum to do, since column names don't have to match (only their types).
There is a CORRESPONDING keyword that is optional for each set operation (union, intersect, except) that means "select columns with the same names".
That said, I think you could request a feature for implementing CORRESPONDING (file a new bug for it), but this bug is invalid, since the standard does not guarantee you in any way that corresponding columns are matched.
## Comment 16963
Date: 2012-02-21 11:04:46 +0100
From: @grobian
PS. CORRESPONDING is already implemented, so just use that instead.
sql>select a, b from a except corresponding select b, c as a from b;
+------+------+
| a | b |
+======+======+
| 1 | 5 |
| 2 | 5 |
| 3 | null |
| 4 | null |
+------+------+
4 tuples (48.539ms)
sql>
| Columns are matched by position not by name in sql EXCEPT | https://api.github.com/repos/MonetDB/MonetDB/issues/3007/comments | 0 | 2020-11-30T12:07:13Z | 2024-06-28T07:23:13Z | https://github.com/MonetDB/MonetDB/issues/3007 | 753,423,030 | 3,007 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-14 17:20:14 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @yzchang
Last updated: 2016-04-11 11:46:19 +0200
## Comment 16883
Date: 2012-02-14 17:20:14 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.77 Safari/535.7
Build Identifier:
Currently, the MAL plan of a sql UNION clause is translated using appends.
When the input tables are sorted on the same attribute it makes sense to use a merge-union approach, to maintain sortedness in the result.
The expected impact is large for certain applications. For example, this would make it viable to build sorted dictionaries incrementally: appending deltas and then sorting explicitly is no option with large data. This is one of the few show-stoppers I still see to use MonetDB (more) efficiently for IR.
(hint: I know the same core functionality was used within MonetDB/XQuery)
Reproducible: Always
## Comment 17692
Date: 2012-08-24 14:56:05 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18130
Date: 2012-11-27 15:46:05 +0100
From: @yzchang
No test needed for feature request
| Keep result of UNION sorted when inputs are sorted | https://api.github.com/repos/MonetDB/MonetDB/issues/3006/comments | 0 | 2020-11-30T12:07:10Z | 2024-06-28T13:40:50Z | https://github.com/MonetDB/MonetDB/issues/3006 | 753,422,982 | 3,006 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-14 13:49:01 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @yzchang
Last updated: 2016-04-11 11:45:14 +0200
## Comment 16881
Date: 2012-02-14 13:49:01 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux ppc; rv:5.0) Gecko/20100101 Firefox/5.0
Build Identifier:
In bigger datasets it seems to have a noticeable difference if using IN or using =, for an operation with one item in the IN list. The IN operation could be optimised if the list contains one element.
sql>explain select * from (select 1 as y) as x where y = 1;
+---------------------------------------------------+
| mal |
+===================================================+
| function user.s35_6{autoCommit=true}():void; |
| X_3 := sql.single(1:bte); |
| X_4 := algebra.uselect(X_3,1:bte); |
| X_5 := algebra.markT(X_4,0@0:oid); |
| X_6 := bat.reverse(X_5); |
| X_7 := algebra.leftjoin(X_6,X_3); |
| X_8 := sql.resultSet(1,1,X_7); |
| sql.rsColumn(X_8,".x","y","tinyint",8,0,X_7); |
| X_15 := io.stdout(); |
| sql.exportResult(X_15,X_8); |
| end s35_6; |
+---------------------------------------------------+
sql>explain select * from (select 1 as y) as x where y in (1);
+------------------------------------------------------+
| mal |
+======================================================+
| function user.s36_6{autoCommit=true}():void; |
| X_3 := sql.single(1:bte); |
| X_9 := bat.reverse(X_3); |
| X_4:bat[:oid,:bte] := bat.new(nil:oid,nil:bte); |
| X_7 := bat.append(X_4,1:bte,true); |
| X_8 := bat.reverse(X_7); |
| X_10 := algebra.semijoin(X_9,X_8); |
| X_11 := bat.reverse(X_10); |
| X_12 := algebra.markT(X_11,0@0:oid); |
| X_13 := algebra.project(X_12); |
| X_14 := algebra.markT(X_13,0@0:oid); |
| X_15 := bat.reverse(X_14); |
| X_16 := algebra.leftjoin(X_15,X_3); |
| X_17 := sql.resultSet(1,1,X_16); |
| sql.rsColumn(X_17,".x","y","tinyint",8,0,X_16); |
| X_24 := io.stdout(); |
| sql.exportResult(X_24,X_17); |
| end s36_6; |
+------------------------------------------------------+
Using IN seems to trigger also a completely different plan for more extensive queries. The trivial example again;
sql>explain select * from (select 1 as y) as x where y in (select 1);
+--------------------------------------------------------+
| mal |
+========================================================+
| function user.s38_6{autoCommit=true}():void; |
| X_2 := sql.single(1:bte); |
| X_3 := bat.reverse(X_2); |
| X_4 := algebra.join(X_2,X_3); |
| X_5 := algebra.markT(X_4,0@0:oid); |
| X_6 := algebra.semijoin(X_2,X_5); |
| X_7 := algebra.markT(X_6,0@0:oid); |
| X_8 := bat.reverse(X_7); |
| X_9 := algebra.leftjoin(X_8,X_2); |
| X_10 := sql.resultSet(1,1,X_9); |
| sql.rsColumn(X_10,".x","y":str,"tinyint",8,0,X_9); |
| X_18 := io.stdout(); |
| sql.exportResult(X_18,X_10); |
| end s38_6; |
+--------------------------------------------------------+
sql>explain select * from (select 1 as y) as x where y = (select 1);
+---------------------------------------------------+
| mal |
+===================================================+
| function user.s39_6{autoCommit=true}():void; |
| X_3 := sql.single(1:bte); |
| X_4 := algebra.uselect(X_3,1:bte); |
| X_5 := algebra.markT(X_4,0@0:oid); |
| X_6 := bat.reverse(X_5); |
| X_7 := algebra.leftjoin(X_6,X_3); |
| X_8 := sql.resultSet(1,1,X_7); |
| sql.rsColumn(X_8,".x","y","tinyint",8,0,X_7); |
| X_15 := io.stdout(); |
| sql.exportResult(X_15,X_8); |
| end s39_6; |
+---------------------------------------------------+
Reproducible: Always
## Comment 17551
Date: 2012-07-21 19:02:30 +0200
From: @grobian
I guess the problem is more likely to be an issue when the argument of IN is a dynamic set.
## Comment 18128
Date: 2012-11-27 15:45:16 +0100
From: @yzchang
No need to add test for feature request
| When IN is presented with a single element, optimise to = | https://api.github.com/repos/MonetDB/MonetDB/issues/3005/comments | 0 | 2020-11-30T12:07:07Z | 2024-06-28T13:40:50Z | https://github.com/MonetDB/MonetDB/issues/3005 | 753,422,949 | 3,005 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-13 15:10:17 +0100
From: Simon Brodt <<simon.brodt>>
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @njnes
Last updated: 2012-02-29 10:34:25 +0100
## Comment 16879
Date: 2012-02-13 15:10:17 +0100
From: Simon Brodt <<simon.brodt>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:10.0.1) Gecko/20100101 Firefox/10.0.1
Build Identifier: MonetDB 5 server v11.7.5 "Dec2011" Compiled for x86_64-pc-winnt/64bit with 64bit OIDs dynamically linked
OS: Windows 7 64bit Professional SP1
Reproducible: Always
### Steps to Reproduce:
CREATE TABLE e (
"a" CLOB,
"r" BIGINT
);
CREATE TABLE s (
"a" CLOB,
"v" BIGINT
);
INSERT INTO e VALUES ('Simon', '1');
INSERT INTO s VALUES ('Simon', '0');
SELECT *
FROM
(
SELECT
"a" AS "ea",
"r" AS "er"
FROM
"e"
) AS "e"
,
(
SELECT
"a" AS "sa",
"v" AS "sv"
FROM
"s"
) AS "s"
WHERE "sv" <= "er"
AND "ea" = "sa"
;
SELECT *
FROM
(
SELECT
"a" AS "ea",
"r" AS "er"
FROM
"e"
) AS "e"
,
(
SELECT
"a" AS "sa",
"v" AS "sv"
FROM
"s"
) AS "s"
WHERE "ea" = "sa"
AND "sv" <= "er"
;
SELECT *
FROM
(
SELECT
"a" AS "ea",
"r" AS "er"
FROM
"e"
) AS "e",
(
SELECT
"a" AS "sa",
"v" AS "sv"
FROM
"s"
) AS "s"
WHERE ("sv" <= "er") = true
AND ("ea" = "sa") = true
;
DROP TABLE "e";
DROP TABLE "s";
### Actual Results:
+----+----+----+----+
| ea | er | sa | sv |
+====+====+====+====+
+----+----+----+----+
0 tuples
+-------+------+-------+------+
| ea | er | sa | sv |
+=======+======+=======+======+
| Simon | 1 | Simon | 0 |
+-------+------+-------+------+
1 tuple
+-------+------+-------+------+
| ea | er | sa | sv |
+=======+======+=======+======+
| Simon | 1 | Simon | 0 |
+-------+------+-------+------+
1 tuple
### Expected Results:
+-------+------+-------+------+
| ea | er | sa | sv |
+=======+======+=======+======+
| Simon | 1 | Simon | 0 |
+-------+------+-------+------+
1 tuple
+-------+------+-------+------+
| ea | er | sa | sv |
+=======+======+=======+======+
| Simon | 1 | Simon | 0 |
+-------+------+-------+------+
1 tuple
+-------+------+-------+------+
| ea | er | sa | sv |
+=======+======+=======+======+
| Simon | 1 | Simon | 0 |
+-------+------+-------+------+
1 tuple
## Comment 16880
Date: 2012-02-13 15:11:37 +0100
From: Simon Brodt <<simon.brodt>>
When reporting the bug the selections of platform: PC and operating system: Windows 7 were not accepted
## Comment 16959
Date: 2012-02-20 22:57:38 +0100
From: @njnes
Changeset [3ab6c6414080](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3ab6c6414080) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3ab6c6414080](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3ab6c6414080)
Changeset description:
fixed bug #3004.
Combinations of complex and normal equijoins were handled incorrectly.
Added a test for Bug #3004 including correct output
## Comment 16960
Date: 2012-02-20 22:58:32 +0100
From: @njnes
fixed by handling all equijoins before thetajoins.
| Incorrect evaluation of conditions in WHERE part of SQL statement | https://api.github.com/repos/MonetDB/MonetDB/issues/3004/comments | 0 | 2020-11-30T12:07:04Z | 2024-06-27T11:55:09Z | https://github.com/MonetDB/MonetDB/issues/3004 | 753,422,901 | 3,004 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-11 10:56:36 +0100
From: simon.xu <<xubochin>>
To: clients devs <<bugs-clients>>
Version: -- development
CC: sellam
Last updated: 2012-11-27 16:15:59 +0100
## Comment 16874
Date: 2012-02-11 10:56:36 +0100
From: simon.xu <<xubochin>>
## Comment 16875
Date: 2012-02-13 14:23:08 +0100
From: @sjoerdmullender
Can you elaborate on this?
What is the problem, and why do you think it is a blocking issue?
## Comment 16876
Date: 2012-02-13 14:31:56 +0100
From: @grobian
22018 = data exception, invalid character value for cast
## Comment 17026
Date: 2012-02-28 17:23:12 +0100
From: @sjoerdmullender
I still have no idea what this bug is about. Please elaborate.
| ODBC system error '22018' | https://api.github.com/repos/MonetDB/MonetDB/issues/3003/comments | 0 | 2020-11-30T12:07:01Z | 2024-06-27T11:55:08Z | https://github.com/MonetDB/MonetDB/issues/3003 | 753,422,873 | 3,003 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-09 00:53:35 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-03-16 14:56:56 +0100
## Comment 16873
Date: 2012-02-09 00:53:35 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
select '1970-01-01 11:00' - interval '600' second;
wrong format (1970-01-01 11:00)
select cast('1970-01-01 11:00' as timestamp) = '1970-01-01 11:00' - interval '600' second;
TypeException:user.s31_1[9]:'calc.timestamp' undefined in: _17:any := calc.timestamp(_16:lng, _10:int)
39000!program contains errors
Reproducible: Always
## Comment 16877
Date: 2012-02-13 14:47:14 +0100
From: @sjoerdmullender
It would be helpful if you didn't just copy-and-paste the query and the error, but also wrote what it was you expected and why you think something is wrong.
## Comment 16878
Date: 2012-02-13 15:02:40 +0100
From: @skinkie
In my opinion anything is wrong when a user is informed that a program contains errors, especially because it is not a message that a human can understand and is out of protocol.
## Comment 17008
Date: 2012-02-27 15:57:13 +0100
From: @njnes
we now disallow converts from interval to timestamps
## Comment 17105
Date: 2012-03-16 14:56:56 +0100
From: @grobian
Released in Dec2011-SP2
| String plus interval result in program errors upon comparison | https://api.github.com/repos/MonetDB/MonetDB/issues/3002/comments | 0 | 2020-11-30T12:06:58Z | 2024-06-27T11:55:07Z | https://github.com/MonetDB/MonetDB/issues/3002 | 753,422,837 | 3,002 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-09 00:49:31 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @yzchang
Last updated: 2016-04-11 11:45:45 +0200
## Comment 16872
Date: 2012-02-09 00:49:31 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
A SQL BETWEEN statement can be written in the form:
Y between X-10 and X+10
Typically column X will require recalculation, while Y is a constant argument. As example which takes 349ms (hot).
select count(*) from pujo where '1970-01-01 11:00' BETWEEN pujo.departuretime - interval '600' second and pujo.departuretime + interval '600' second;
The same operation could be rewritten:
Y between X-10 and X+10 = X between Y-10 and Y+10
The performance of the following query is 2ms (hot), while 4ms cold. It doesn't matter if the intervals are presubstracted manually, performance is similar.
select count(*) from pujo where pujo.departuretime between cast('1970-01-01 11:00' as timestamp) - interval '600' second and cast('1970-01-01 11:00' as timestamp) + interval '600' second;
select count(*) from pujo where pujo.departuretime between cast('1970-01-01 10:50' as timestamp) and cast('1970-01-01 11:10' as timestamp);
Reproducible: Always
## Comment 18127
Date: 2012-11-27 15:44:32 +0100
From: @yzchang
No test for feature request
| Rewrite BETWEEN in column first notation, with constant arguments last | https://api.github.com/repos/MonetDB/MonetDB/issues/3001/comments | 0 | 2020-11-30T12:06:56Z | 2024-06-28T13:40:49Z | https://github.com/MonetDB/MonetDB/issues/3001 | 753,422,819 | 3,001 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-08 16:25:30 +0100
From: Yagiz <<y.kargin>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-05-25 12:58:43 +0200
## Comment 16871
Date: 2012-02-08 16:25:30 +0100
From: Yagiz <<y.kargin>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.46 Safari/535.11
Build Identifier: MonetDB-default
Mclient is used. After creating a schema and tables under it in read-write mode, if you switch to read-only, everything works fine. Then if you switch back to read-write and drop the schema, you are not able to see the tables under it in read-write mode, however you are still able to see them in read-only mode. So the tables of the schema are removed in read-write but not in read-only.
Then if you recreate the schema and tables in read-write mode -- no complains -- the tables in read-write mode become corrupt.
Reproducible: Always
### Steps to Reproduce:
1. Connect with mclient in read-write mode.
2. sql>CREATE SCHEMA test;
3. sql>CREATE TABLE test.files (file_id INT, location STRING);
4. sql>CREATE TABLE test.catalog (file_id INT, seq_no INT, sample_count INT);
5. sql>DROP SCHEMA test;
6. sql>\d test.files
table test.files does not exist (<-- that's good)
7. go into the read-only mode.
8. sql>\d test.files (<-- returns the table definition!)
sql>SELECT * FROM test.catalog;
3F000!SELECT: no such schema 'test'
BUG is already there in step 8.
9. Now repeat steps 1, 2, 3 and 4.
10. go into the read-only mode.
11. sql>\d test.files
table test.files is not unique, corrupt catalog?
sql>SELECT * FROM test.files;
+---------+----------+
| file_id | location |
+=========+==========+
+---------+----------+
### Actual Results:
Inconsistency between read-only and read-write modes.
### Expected Results:
Should have removed the tables from read-only mode too.
MonetDB 5 server v11.8.0 (64-bit, 64-bit oids)
This is an unreleased version
Found 15.6GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.12 2011-01-15 (compiled with 8.10)
openssl: OpenSSL 1.0.0e 6 Sep 2011 (compiled with OpenSSL 1.0.0e-fips 6 Sep 2011)
libxml2: 2.7.7 (compiled with 2.7.7)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 17216
Date: 2012-05-02 21:44:31 +0200
From: @njnes
fixed, readonly mode now correctly handles the deleted rows.
Could you add a test to the system?
| Inconsistency between read-only and read-write modes after dropping a schema | https://api.github.com/repos/MonetDB/MonetDB/issues/3000/comments | 0 | 2020-11-30T11:51:25Z | 2024-06-27T11:55:05Z | https://github.com/MonetDB/MonetDB/issues/3000 | 753,412,174 | 3,000 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-07 16:24:20 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.11.7 (Jul2012-SP1)
CC: @njnes, @drstmane, @yzchang
Last updated: 2013-01-22 09:29:17 +0100
## Comment 16868
Date: 2012-02-07 16:24:20 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.77 Safari/535.7
Build Identifier:
Take the following SQL:
-- ----------------
create table a(i int);
insert into a values (1),(2),(3);
explain
select cast(i as bigint) from a;
-- ----------------
and it's explain:
+----------------------------------------------------------------+
| mal |
+================================================================+
| function user.s4_1():void; |
| X_1 := sql.mvc(); |
| X_8:bat[:oid,:int] := sql.bind(X_1,"spinque","a","i",0); |
| X_4:bat[:oid,:oid] := sql.bind_dbat(X_1,"spinque","a",1); |
| X_7 := bat.reverse(X_4); |
| X_10 := algebra.kdifference(X_8,X_7); |
| X_11 := batcalc.lng(X_10); |
| X_12 := sql.resultSet(1,1,X_11); |
| sql.rsColumn(X_12,"spinque.","i","bigint",64,0,X_11); |
| X_17 := io.stdout(); |
| sql.exportResult(X_17,X_12); |
| end s4_1; |
+----------------------------------------------------------------+
When debugging this query from SQL, I found out that X_10 is sorted on tail, as expected. However, X_11 is not sorted on tail.
Of course I would expect an up-cast not to modify the order (while a down-cast would).
What I find strange is also that trying to repeat the same in mal doesn't seem to change the order:
mal>a := bat.new(:oid,:int);
mal>bat.append(a,1);
mal>bat.append(a,2);
mal>bat.append(a,3);
mal>la := batcalc.lng(a);
mal>
mal>ra := bat.reverse(a);
mal>rla := bat.reverse(la);
mal>
mal>i_ra := bat.isSorted(ra);
mal>i_rla := bat.isSorted(rla);
mal>
mal>io.print(ra);
-----------------
t h name
int void type
-----------------
[ 1, 0@0 ]
[ 2, 1@0 ]
[ 3, 2@0 ]
mal>io.print("Ordered on tail",i_ra);
[ "Ordered on tail", true]
mal>
mal>io.print(rla);
-----------------
t h name
lng void type
-----------------
[ 1, 0@0 ]
[ 2, 1@0 ]
[ 3, 2@0 ]
mal>io.print("Ordered on tail",i_rla);
[ "Ordered on tail", true]
Reproducible: Always
### Steps to Reproduce:
1. Debug the SQL query above
2. Inspect X_10 (original int values) and X_11 (lng values)
3.
### Actual Results:
X_11 has tsorded=0
### Expected Results:
X_11 has tsorded=65
$ mserver5 --version
MonetDB 5 server v11.7.6 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2012 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 35.5GiB available memory, 8 available cpu cores
Libraries:
libpcre: 7.8 2008-09-05 (compiled with 7.8)
openssl: OpenSSL 1.0.0d 8 Feb 2011 (compiled with OpenSSL 1.0.0d-fips 8 Feb 2011)
libxml2: 2.7.7 (compiled with 2.7.7)
Compiled by: roberto@spinque01.ins.cwi.nl (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wmissing-include-dirs -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 16869
Date: 2012-02-07 16:43:38 +0100
From: @drstmane
Thanks for notifying and reporting.
We'll look into this, although it is technically not a BUG: it is technically correct to have a sorted BAT that is not marked as being sorted.
However, given that MonetDB can and does exploit (and often rely on) known properties like sortedness for performance, we should indeed be more eager on propagating such properties wherever possible...
Did you run your MAL example in the very mserver5 that also ran the SQL example?
(I.e., when running the MAL example, you did not by any chance enable property checking?)
## Comment 16870
Date: 2012-02-07 16:48:56 +0100
From: @swingbit
I agree, not technically a bug. But indeed, given how much faster many operators can be on sorted inputs, this information should not be lost whenever possible.
I ran sql and mal examples on the very same mserver5 (even on the same db), simply changing from -sql to -mal in my mclient command. Without stopping mserver5 in between.
## Comment 17218
Date: 2012-05-03 22:04:31 +0200
From: @sjoerdmullender
Changeset [b51e0d4dae3f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b51e0d4dae3f) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b51e0d4dae3f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b51e0d4dae3f)
Changeset description:
Use gdk_calc code for type coercion in MAL.
This should fix bug #2999.
## Comment 17689
Date: 2012-08-24 14:56:03 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18124
Date: 2012-11-27 15:41:33 +0100
From: @yzchang
No need, and also not possible to add a test for this bug
## Comment 18382
Date: 2013-01-22 09:29:17 +0100
From: @sjoerdmullender
Oct2012-SP3 has been released.
| Up-cast from SQL drops order information? | https://api.github.com/repos/MonetDB/MonetDB/issues/2999/comments | 0 | 2020-11-30T11:51:22Z | 2024-06-27T11:55:04Z | https://github.com/MonetDB/MonetDB/issues/2999 | 753,412,139 | 2,999 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-06 16:58:41 +0100
From: @swingbit
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @njnes
Last updated: 2012-03-16 14:56:52 +0100
## Comment 16865
Date: 2012-02-06 16:58:41 +0100
From: @swingbit
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.77 Safari/535.7
Build Identifier:
-- Example data
CREATE TABLE "treeitems" (
"tree" CHARACTER LARGE OBJECT,
"pre" BIGINT
);
INSERT INTO treeitems VALUES('documentStructure', 0);
INSERT INTO treeitems VALUES('documentStructure', 1);
-- Queries 1 and 2 are semantically equivalent
-- Query 1 gives CORRECT results, Query 2 gives WRONG results
-- Query 1
SELECT *
FROM treeitems AS a,
treeitems AS b
WHERE a.pre > b.pre
AND a.tree = 'documentStructure'
AND b.tree = 'documentStructure';
-- Query 2
SELECT *
FROM treeitems AS a,
treeitems AS b
WHERE a.pre > b.pre
AND a.tree = 'documentStructure'
AND b.tree = a.tree;
Reproducible: Always
### Steps to Reproduce:
1.execute SQL code in details
2.
3.
### Actual Results:
sql>CREATE TABLE "treeitems" (
more> "tree" CHARACTER LARGE OBJECT,
more> "pre" BIGINT
more>);
operation successful (1.147ms)
sql>INSERT INTO treeitems VALUES('documentStructure', 0);
1 affected row (0.649ms)
sql>INSERT INTO treeitems VALUES('documentStructure', 1);
1 affected row (0.230ms)
sql>
sql>SELECT *
more>FROM treeitems AS a,
more> treeitems AS b
more>WHERE a.pre > b.pre
more>AND a.tree = 'documentStructure'
more>AND b.tree = 'documentStructure';
+-------------------+------+-------------------+------+
| tree | pre | tree | pre |
+===================+======+===================+======+
| documentStructure | 1 | documentStructure | 0 |
+-------------------+------+-------------------+------+
1 tuple (2.518ms)
sql>
sql>SELECT *
more>FROM treeitems AS a,
more> treeitems AS b
more>WHERE a.pre > b.pre
more>AND a.tree = 'documentStructure'
more>AND b.tree = a.tree;
+-------------------+------+-------------------+------+
| tree | pre | tree | pre |
+===================+======+===================+======+
| documentStructure | 0 | documentStructure | 0 |
| documentStructure | 1 | documentStructure | 1 |
+-------------------+------+-------------------+------+
2 tuples (3.984ms)
sql>
### Expected Results:
sql>CREATE TABLE "treeitems" (
more> "tree" CHARACTER LARGE OBJECT,
more> "pre" BIGINT
more>);
operation successful (1.147ms)
sql>INSERT INTO treeitems VALUES('documentStructure', 0);
1 affected row (0.649ms)
sql>INSERT INTO treeitems VALUES('documentStructure', 1);
1 affected row (0.230ms)
sql>
sql>SELECT *
more>FROM treeitems AS a,
more> treeitems AS b
more>WHERE a.pre > b.pre
more>AND a.tree = 'documentStructure'
more>AND b.tree = 'documentStructure';
+-------------------+------+-------------------+------+
| tree | pre | tree | pre |
+===================+======+===================+======+
| documentStructure | 1 | documentStructure | 0 |
+-------------------+------+-------------------+------+
1 tuple (2.518ms)
sql>
sql>SELECT *
more>FROM treeitems AS a,
more> treeitems AS b
more>WHERE a.pre > b.pre
more>AND a.tree = 'documentStructure'
more>AND b.tree = a.tree;
+-------------------+------+-------------------+------+
| tree | pre | tree | pre |
+===================+======+===================+======+
| documentStructure | 1 | documentStructure | 0 |
+-------------------+------+-------------------+------+
1 tuple (2.518ms)
sql>
$ mserver5 --version
MonetDB 5 server v11.7.6 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2012 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 35.5GiB available memory, 8 available cpu cores
Libraries:
libpcre: 7.8 2008-09-05 (compiled with 7.8)
openssl: OpenSSL 1.0.0d 8 Feb 2011 (compiled with OpenSSL 1.0.0d-fips 8 Feb 2011)
libxml2: 2.7.7 (compiled with 2.7.7)
Compiled by: roberto@spinque01.ins.cwi.nl (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wmissing-include-dirs -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/bin/ld -m elf_x86_64
## Comment 16999
Date: 2012-02-26 21:03:07 +0100
From: @njnes
is fixed by the changes to the rel2bin_join some days back
## Comment 17000
Date: 2012-02-26 21:05:41 +0100
From: @njnes
Changeset [b6dd338c7c98](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b6dd338c7c98) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b6dd338c7c98](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b6dd338c7c98)
Changeset description:
added test for bug #2998
## Comment 17094
Date: 2012-03-16 14:56:52 +0100
From: @grobian
Released in Dec2011-SP2
| Equivalent multi-way join queries give different results | https://api.github.com/repos/MonetDB/MonetDB/issues/2998/comments | 0 | 2020-11-30T11:51:19Z | 2024-06-27T11:55:03Z | https://github.com/MonetDB/MonetDB/issues/2998 | 753,412,114 | 2,998 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-03 17:23:23 +0100
From: @bartscheers
To: clients devs <<bugs-clients>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-29 10:34:28 +0100
## Comment 16861
Date: 2012-02-03 17:23:23 +0100
From: @bartscheers
User-Agent: Mozilla/5.0 (X11; Linux i686 on x86_64; rv:8.0.1) Gecko/20100101 Firefox/8.0.1
Build Identifier: MonetDB-Aug2011-SP3
Ubuntu keeps python packages in /usr/lib/python2.6/dist-packages, not in site-packages. The latter is typically for user-build/installed packages (i.e., built from source code), and those should end up in /usr/local/lib/pythonx.x or a cumstom directory anyway.
Reproducible: Always
### Steps to Reproduce:
1. Follow install instructions:
2. We only want to install on the client nodes:
3. sudo apt-get install monetdb-client python-monetdb
### Actual Results:
It is installed in /usr/lib/python2.6/site-packages
### Expected Results:
installed in /usr/lib/python2.6/dist-packages
## Comment 16866
Date: 2012-02-07 16:02:53 +0100
From: @sjoerdmullender
Changeset [a9ee53420f14](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a9ee53420f14) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a9ee53420f14](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a9ee53420f14)
Changeset description:
On Debian and Ubuntu, install Python modules in dist-packages.
This fixed bug #2997.
## Comment 16867
Date: 2012-02-07 16:19:31 +0100
From: @sjoerdmullender
This is hopefully fixed now.
No test since this cannot be tested in our test setup.
| monetdb-python packages installed in wrong dir on Ubuntu 10.04 LTS (lucid) | https://api.github.com/repos/MonetDB/MonetDB/issues/2997/comments | 0 | 2020-11-30T11:51:16Z | 2024-06-27T11:55:02Z | https://github.com/MonetDB/MonetDB/issues/2997 | 753,412,078 | 2,997 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-03 16:53:24 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
To: clients devs <<bugs-clients>>
Version: -- development
Last updated: 2012-02-06 16:21:28 +0100
## Comment 16860
Date: 2012-02-03 16:53:24 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0.1) Gecko/20100101 Firefox/8.0.1
Build Identifier: Fedora 16 - MonetDB Ver 11.7.5 from 20-Jan-2012
I am using a Fedora 16 on Virtual Box, and the COPY command didn't work for a specific table, which have four integer columns. I tried to create a sample table with the same column types and the error was reproduced. With other column types the error didn't happened.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE TEST1 ( cl1 INTEGER, cl2 INTEGER, cl3 INTEGER, cl4 INTEGER);
2. Create a text file with random information well formated, like:
'1|1|1|1
'
3. COPY 1 RECORDS INTO TEST1 FROM '[PATH_TO_FILE]';
### Actual Results:
' from line 1 field 4 not inserted, expecting type int
failed to import table
### Expected Results:
operation successful.
If I change the text file to put an aditional pipe, like this:
'1|1|1|1|
'
The command works and insert that line's information, but no matter the size of the file, if you do that the COPY will only insert the first line, ignoring the rest of the file.
## Comment 16862
Date: 2012-02-05 10:20:23 +0100
From: @sjoerdmullender
I can reproduce this if I use a text file with CR-LF (a.k.a. \r\n) line endings as is used on Windows. You can tell the COPY INTO command that you have those line endings by using
copy into x from 'path-to-file' using delimiters '|','\r\n';
You can check whether you have those line endings by doing
head -1 file | od -c
and look for \r and \n at the end of the output.
## Comment 16863
Date: 2012-02-06 12:23:55 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
Thank you Sjoerd,
I was trying to find the using part command, but I couldn't, even with documentation or forums. Guess the documentation should be optmized, even having bugs, features that the DBMS offers are not well presented. But it's just an user recommendation.
Thank you again.
## Comment 16864
Date: 2012-02-06 16:21:28 +0100
From: @sjoerdmullender
Just for reference, the documentation can be found in
www.monetdb.org -> Documentation -> SQL Reference -> Data manipulation -> Bulk input/output. The URI of the page is http://www.monetdb.org/Documentation/Manuals/SQLreference/CopyInto
I'm closing this under the assumption that your answer implied that you were able to import the data without fault now. If not, please reopen.
| COPY doesn't work for tables with four integer columns, giving file formatation error | https://api.github.com/repos/MonetDB/MonetDB/issues/2996/comments | 0 | 2020-11-30T11:51:13Z | 2024-06-28T13:18:27Z | https://github.com/MonetDB/MonetDB/issues/2996 | 753,412,041 | 2,996 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-03 11:24:45 +0100
From: @drstmane
To: SQL devs <<bugs-sql>>
Version: -- development
Last updated: 2012-02-29 10:34:24 +0100
## Comment 16858
Date: 2012-02-03 11:24:45 +0100
From: @drstmane
Test geom/sql/Tests/example.sql crashes (segfault, bus error) or hangs (timeout) since changeset [6399077ff8a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6399077ff8a4) (http://dev.monetdb.org/hg/MonetDB/rev/6399077ff8a4); cf.
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=43018:6e534d939e36&module=geom
http://monetdb.cwi.nl/testweb/tests/43018:6e534d939e36/GNU-Fedora-x86_64-propcheck-assert/geom/mTests/sql/example.err.diff.html
## Comment 16886
Date: 2012-02-15 14:53:54 +0100
From: @sjoerdmullender
Fixed with changesets [be05fff18a93](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=be05fff18a93) and [0b6018784f6b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0b6018784f6b)
| Test geom/sql/Tests/example.sql crashes since changeset 6399077ff8a4 | https://api.github.com/repos/MonetDB/MonetDB/issues/2995/comments | 0 | 2020-11-30T11:51:10Z | 2024-06-27T11:55:00Z | https://github.com/MonetDB/MonetDB/issues/2995 | 753,412,007 | 2,995 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-02-03 11:04:22 +0100
From: @drstmane
To: MonetDB5 devs <<bugs-monetdb5>>
Version: -- development
CC: @yzchang
Last updated: 2012-11-27 14:14:52 +0100
## Comment 16857
Date: 2012-02-03 11:04:22 +0100
From: @drstmane
The following tests fail since changeset [ada288cef37c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ada288cef37c)
http://dev.monetdb.org/hg/MonetDB/rev/ada288cef37c
monetdb5/mal/Tests/tst654.mal
http://monetdb.cwi.nl/testweb/tests/43018:6e534d939e36/GNU-Fedora-x86_64-propcheck-assert/monetdb5/mTests/mal/tst654.out.diff.html
monetdb5/mal/Tests/tst655.mal
http://monetdb.cwi.nl/testweb/tests/43018:6e534d939e36/GNU-Fedora-x86_64-propcheck-assert/monetdb5/mTests/mal/tst655.out.diff.html
IMHO the key property must not be set in both cases, since there is a NULL value (rendered as "unknown")
monetdb5/modules/kernel/Tests/vacuum.mal
http://monetdb.cwi.nl/testweb/tests/43018:6e534d939e36/GNU-Fedora-x86_64-propcheck-assert/monetdb5/mTests/modules/kernel/vacuum.err.diff.html
times out after 100s
## Comment 16859
Date: 2012-02-03 12:29:20 +0100
From: @drstmane
Correction for tests monetdb5/mal/Tests/tst654.mal & monetdb5/mal/Tests/tst655.mal
Since there is only one NULL values, the "key" property does hold after all (at least for the first/left of the two BATs is was already approved as correct.
Hence, I approved the new output for tests monetdb5/mal/Tests/tst654.mal & monetdb5/mal/Tests/tst655.mal and changed the subject of this bug report to refer only to test monetdb5/modules/kernel/Tests/vacuum.mal
## Comment 17174
Date: 2012-04-27 01:15:50 +0200
From: @drstmane
This appears to be fixed in Apr2012_1, Apr2012 (latest) and default:
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=43870:4702495ed780&module=monetdb5
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=43950:702b55cbad26&module=monetdb5
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=43935:bd7bcdd865f7&module=monetdb5
## Comment 18075
Date: 2012-11-27 14:14:52 +0100
From: @yzchang
This is about existing tests.
| test monetdb5/modules/kernel/Tests/vacuum.mal fails since changeset ada288cef37c | https://api.github.com/repos/MonetDB/MonetDB/issues/2994/comments | 0 | 2020-11-30T11:51:07Z | 2024-06-27T11:54:59Z | https://github.com/MonetDB/MonetDB/issues/2994 | 753,411,981 | 2,994 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-31 22:35:05 +0100
From: Kostis Kyzirakos <<kkyzir>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @romulogoncalves, @njnes
Last updated: 2012-11-27 16:41:23 +0100
## Comment 16856
Date: 2012-01-31 22:35:05 +0100
From: Kostis Kyzirakos <<kkyzir>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.77 Safari/535.7
Build Identifier: Aug2011, Dec2011
I created a synthetic dataset that contains a relation with a spatial attribute. Geometries in this dataset are placed uniformly on a grid. I have a query template that selects (among others) the geometries that are located inside a bounding box. For specific values of the bounding box the results are wrong. If I use inner joins instead of left joins, the results are correct.
Reproducible: Always
### Steps to Reproduce:
1. Download and extract the archive from http://dev.strabon.di.uoa.gr/~kkyzir/MonetDB-bugreport-2011-01-31.tgz
2. Run load.sh to create and load the database.
3. Evaluate the following query: SELECT u_node.id, u_node.value, u_tag.id, u_tag.value FROM hastag_3 h0 INNER JOIN hasgeography1_6 h1 ON (h1.subj = h0.subj) INNER JOIN key_4 k2 ON (k2.obj = 805306370 AND k2.subj = h0.obj) INNER JOIN geo_values l_geo ON (l_geo.id = h1.obj) LEFT JOIN uri_values u_node ON (u_node.id = h0.subj) LEFT JOIN uri_values u_tag ON (u_tag.id = h0.obj) WHERE (Within(l_geo.strdfgeo, GeomFromText('POLYGON ((-1 -1, 5.12 -1, 5.12 5.12, -1 5.12, -1 -1))',4326))) ORDER BY u_node.id, u_tag.id LIMIT 5;
4. Run query.sh that executes the same query with different bounding boxes.
### Actual Results:
+------+-------+------+-------+
| id | value | id | value |
+======+=======+======+=======+
| 10 | | 11 | |
| 13 | | 14 | |
| 15 | | 16 | |
| 19 | | 20 | |
| 21 | | 22 | |
+------+-------+------+-------+
### Expected Results:
+------+-----------------------------+------+----------------------------+
| id | value | id | value |
+======+=============================+======+============================+
| 10 | http://www.wktrdf.org/Node1 | 11 | http://www.wktrdf.org/Tag1 |
| 13 | http://www.wktrdf.org/Node2 | 14 | http://www.wktrdf.org/Tag3 |
| 15 | http://www.wktrdf.org/Node3 | 16 | http://www.wktrdf.org/Tag4 |
| 19 | http://www.wktrdf.org/Node4 | 20 | http://www.wktrdf.org/Tag7 |
| 21 | http://www.wktrdf.org/Node5 | 22 | http://www.wktrdf.org/Tag8 |
+------+-----------------------------+------+----------------------------+
I am using the query template described earlier with the following bounding boxes:
A=POLYGON ((-1 -1, 4.5 -1, 4.5 4.5, -1 4.5, -1 -1)),
B=POLYGON ((-1 -1, 5.12 -1, 5.12 5.12, -1 5.12, -1 -1)),
C=POLYGON ((-1 -1, 6 -1, 6 6, -1 6, -1 -1)) and
D=POLYGON ((-1 -1, 7 -1, 7 6, -1 7, -1 -1)).
The results are correct when I use A and D but wrong when I use B and C. The problem is that 1400 tuples have empty values for the u_node.value and u_tag.value attributes.
If I replace the left joins with inner joins, the results are correct.
## Comment 17001
Date: 2012-02-26 21:07:03 +0100
From: @njnes
could you try this with the latest version. Many fixes were done in this area. I cannot download your query/data to test myself.
## Comment 17007
Date: 2012-02-27 14:18:57 +0100
From: Kostis Kyzirakos <<kkyzir>>
You should be able to download the file now (wrong permissions).
I followed your suggestion and I re-run the tests using the latest version of the Dec2011 branch and it seems to be working. I will run some more tests and let you know if I have any other issues.
## Comment 18154
Date: 2012-11-27 16:41:23 +0100
From: @romulogoncalves
Any progress in testing the latest version?
We would like to know the current status to close the bug.
| Left joins are not evaluated correctly (in some cases) | https://api.github.com/repos/MonetDB/MonetDB/issues/2993/comments | 0 | 2020-11-30T11:51:04Z | 2024-06-27T11:54:58Z | https://github.com/MonetDB/MonetDB/issues/2993 | 753,411,941 | 2,993 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-31 15:05:08 +0100
From: sellam
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-05-25 12:58:49 +0200
## Comment 16854
Date: 2012-01-31 15:05:08 +0100
From: sellam
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.11 (KHTML, like Gecko) Chrome/17.0.963.44 Safari/535.11
Build Identifier: Dec2011 - changeset 42883:baf7d5fc1f6f (Thu Jan 19)
Consider the following UDF:
create function truncode(lon decimal(9,6), lat decimal(9,6))
returns int external name truncatenate.encode;
The SQL module should pass the argument to MAL as integers. That works fine except when an argument has more than 2 digits after the comma
EXAMPLES:
sql>explain select truncode(12,12); *******ALL GOOD
+-------------------------------------------------------------------------+
| mal |
+=========================================================================+
| function user.s8_1{autoCommit=true}():void; |
| X_4 := truncatenate.encode(12000000:int,12000000:int); |
| sql.exportValue(1,".","truncode_single_value","int",32,0,6,X_4,""); |
| end s8_1; |
+-------------------------------------------------------------------------+
4 tuples (0.770ms)
sql>explain select truncode(12.0,12);******** ALL GOOD
+-------------------------------------------------------------------------+
| mal |
+=========================================================================+
| function user.s9_1{autoCommit=true}():void; |
| X_5 := truncatenate.encode(12000000:int,12000000:int); |
| sql.exportValue(1,".","truncode_single_value","int",32,0,6,X_5,""); |
| end s9_1; |
+-------------------------------------------------------------------------+
4 tuples (0.680ms)
sql>explain select truncode(12.00,12); ******* NOT GOOD !
+-------------------------------------------------------------------------+
| mal |
+=========================================================================+
| function user.s10_1{autoCommit=true}():void; |
| X_4 := truncatenate.encode(1200:int,1200:int); |<-!!
| sql.exportValue(1,".","truncode_single_value","int",32,2,6,X_4,""); |
| end s10_1; |
+-------------------------------------------------------------------------+
4 tuples (0.626ms)
* SQL:
create function truncode(lon decimal(9,6), lat decimal(9,6))
returns int external name truncatenate.encode;
* MAL:
module truncatenate;
command encode(lon:int, lat:int):int
address Truncatenate
comment "Pastes the two leftmost bytes of two integers";
* C:
str Truncatenate(int *res, int *lon, int *lat) {
*res = (*lon & 0xFFFF0000) | (*lat >> 16);
return MAL_SUCCEED;
Reproducible: Always
## Comment 16855
Date: 2012-01-31 15:25:18 +0100
From: sellam
There seems to be a workaround: putting arguments in variables instead of putting the function:
* WORKS:
sql>declare x decimal(9,6);
sql>set x=12.00;
sql>select truncode(x,0);
+-----------------------+
| truncode_single_value |
+=======================+
| 11993088 |
+-----------------------+
1 tuple (0.571ms)
* DOES NOT WORK:
sql>select truncode(12.00,0);
+-----------------------+
| truncode_single_value |
+=======================+
| 0 |
+-----------------------+
1 tuple (0.556ms)
* EXPLAINS
sql>explain select truncode(12.00,0);
+-------------------------------------------------------------------------+
| mal |
+=========================================================================+
| function user.s11_1{autoCommit=true}():void; |
| X_4 := truncatenate.encode(1200:int,0:int); |->!!!!! PROBLEM
| sql.exportValue(1,".","truncode_single_value","int",32,2,6,X_4,""); |
| end s11_1; |
+-------------------------------------------------------------------------+
4 tuples (0.677ms)
sql>explain select truncode(x,0);
+-------------------------------------------------------------------------+
| mal |
+=========================================================================+
| function user.s12_1{autoCommit=true}():void; |
| X_2 := sql.mvc(); |
| X_5:int := sql.getVariable(X_2,"x"); |
| X_7 := truncatenate.encode(X_5,0:int); |
| sql.exportValue(1,".","truncode_single_value","int",32,6,6,X_7,""); |
| end s12_1; |
+-------------------------------------------------------------------------+
6 tuples (0.724ms)
## Comment 17222
Date: 2012-05-07 08:34:59 +0200
From: @njnes
Changeset [a8c386663dfe](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a8c386663dfe) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a8c386663dfe](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=a8c386663dfe)
Changeset description:
fixed bug in handeling decimals in user defined functions (bug #2992)
## Comment 17223
Date: 2012-05-07 08:36:35 +0200
From: @njnes
fixed, user defined functions need matching scales with the argument types. No need for the dynamic scales as needed for internal functions such as 'sql_sub' etc.
| SQL module truncates DECIMALS when passing integers to MAL module | https://api.github.com/repos/MonetDB/MonetDB/issues/2992/comments | 0 | 2020-11-30T11:51:01Z | 2024-06-27T11:54:57Z | https://github.com/MonetDB/MonetDB/issues/2992 | 753,411,905 | 2,992 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-31 13:29:29 +0100
From: Wouter Alink <<alink>>
To: clients devs <<bugs-clients>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-29 10:36:50 +0100
## Comment 16853
Date: 2012-01-31 13:29:29 +0100
From: Wouter Alink <<alink>>
Using the JDBC Control class for managing a dbfarm.
For slow connections the server sometimes closes the connection before the client has read the final prompt. This causes a SQL exception to be thrown at the client.
For me (connecting remotely over SSH using a Fedora Core machine) it happens in roughly 20% of the attempts (I tried a 1000 times), which makes it too unreliable to be useful. (therefore the severity increase)
It can be reproduced with various Java versions (Sun 6, Sun 7, OpenJDK 6, OpenJDK 7). I couldn't reproduce a single failure when using a Windows OS (tried 1000 attempts).
Note: I could not reproduce on the Aug2011 release.
## Comment 17030
Date: 2012-02-29 10:36:50 +0100
From: @grobian
I think this is fixed in Dec2011-SP1
| merovingian: protocol incorrect | https://api.github.com/repos/MonetDB/MonetDB/issues/2991/comments | 0 | 2020-11-30T11:50:59Z | 2024-06-27T11:54:56Z | https://github.com/MonetDB/MonetDB/issues/2991 | 753,411,872 | 2,991 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-29 00:53:39 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-05-09 17:11:34 +0200
## Comment 16842
Date: 2012-01-29 00:53:39 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.159 Safari/535.1
Build Identifier:
create table test (id integer, id2 integer, primary key(id, id2));
create table foreign_test (id integer references test (id));
CONSTRAINT FOREIGN KEY: could not find referenced PRIMARY KEY in table test
PostgreSQL gave me a more informative error that could directly be applied to MonetDB;
ERROR: there is no unique constraint matching given keys for referenced table "test"
alter table test add unique(id);
create table foreign_test (id integer references test (id));
operation successful (20.015ms)
None the less; MonetDB was right and I shouldn't have tried this.
Reproducible: Always
## Comment 16843
Date: 2012-01-29 09:37:17 +0100
From: @grobian
Hmmm, isn't MonetDB's message a little bit more precise compared to PostgreSQL's one?
There is no PRIMARY KEY(id), as you reference it. So what's wrong with the error?
## Comment 16844
Date: 2012-01-29 13:24:21 +0100
From: @skinkie
Actually getting the hint 'you can do this if you make sure you have a unique column' would be valuable information.
But it seems to be that the ERD I implemented is just wrong.
| Wrong error for foreign key that matches a part of a primary key | https://api.github.com/repos/MonetDB/MonetDB/issues/2990/comments | 0 | 2020-11-30T11:50:55Z | 2024-06-28T13:40:49Z | https://github.com/MonetDB/MonetDB/issues/2990 | 753,411,831 | 2,990 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-29 00:05:14 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
Duplicates: #3019
Last updated: 2012-02-16 16:56:52 +0100
## Comment 16840
Date: 2012-01-29 00:05:14 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.159 Safari/535.1
Build Identifier:
sql>alter table pujo add FOREIGN KEY (DataOwnerCode, LinePlanningNumber, TimeDemandGroupCode, JourneyPatternCode) REFERENCES timdemgrp;
M0M29!UPDATE: FOREIGN KEY constraint 'pujo.pujo_dataownercode_lineplanningnumber_timedemandgroupcode_journeypatterncode_fkey' violated
sql>select distinct DataOwnerCode, LinePlanningNumber, TimeDemandGroupCode, JourneyPatternCode from sys.pujo except select distinct DataOwnerCode, LinePlanningNumber, TimeDemandGroupCode, JourneyPatternCode from timdemgrp;
+---------------+--------------------+---------------------+--------------------+
| dataownercode | lineplanningnumber | timedemandgroupcode | journeypatterncode |
+===============+====================+=====================+====================+
+---------------+--------------------+---------------------+--------------------+
0 tuples (87.470ms)
sql>SELECT DISTINCT pujo.DataOwnerCode, pujo.LinePlanningNumber, pujo.TimeDemandGroupCode, pujo.JourneyPatternCode from pujo LEFT JOIN timdemgrp ON pujo.DataOwnerCode = timdemgrp.DataOwnerCode and pujo.LinePlanningNumber = timdemgrp.LinePlanningNumber and pujo.TimeDemandGroupCode = timdemgrp.TimeDemandGroupCode and pujo.JourneyPatternCode = timdemgrp.JourneyPatternCode WHERE timdemgrp.dataownercode is null;
+---------------+--------------------+---------------------+--------------------+
| dataownercode | lineplanningnumber | timedemandgroupcode | journeypatterncode |
+===============+====================+=====================+====================+
+---------------+--------------------+---------------------+--------------------+
0 tuples (193.979ms)
Reproducible: Always
Latest head.
## Comment 16917
Date: 2012-02-16 16:56:52 +0100
From: @skinkie
More reasoning in the new bug.
*** This bug has been marked as a duplicate of bug #3019 ***
| Foreign key constraints are not correctly checked | https://api.github.com/repos/MonetDB/MonetDB/issues/2989/comments | 0 | 2020-11-30T11:50:53Z | 2024-06-28T13:18:26Z | https://github.com/MonetDB/MonetDB/issues/2989 | 753,411,809 | 2,989 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-28 16:00:48 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-03-16 14:56:54 +0100
## Comment 16838
Date: 2012-01-28 16:00:48 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
sql>select 1, (select 1) as x;
+------+------+------+
| L12 | L11 | x |
+======+======+======+
| 1 | 1 | 1 |
+------+------+------+
select 1, (select 1 as y, 2 as z) as x;
+------+------+------+------+
| L21 | y | L20 | x |
+======+======+======+======+
| 1 | 1 | 2 | 2 |
+------+------+------+------+
Reproducible: Always
### Expected Results:
PostgreSQL says;
select 1, (select 1 as y) as x;
?column? | x
----------+---
1 | 1
select 1, (select 1 as y, 2 as z) as x;
ERROR: subquery must return only one column
(actually trying to find the crash that caused my group by bug)
## Comment 17002
Date: 2012-02-26 23:07:46 +0100
From: @njnes
corrected bug in simple_select, need alway to project full list of outer projection.
## Comment 17003
Date: 2012-02-26 23:09:17 +0100
From: @njnes
Changeset [85e9bea7a75f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=85e9bea7a75f) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=85e9bea7a75f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=85e9bea7a75f)
Changeset description:
added test for bug #2988
fixed bug in rel_simple_select outer projection handling.
## Comment 17010
Date: 2012-02-27 19:26:47 +0100
From: @njnes
*** Bug #2985 has been marked as a duplicate of this bug. ***
## Comment 17099
Date: 2012-03-16 14:56:54 +0100
From: @grobian
Released in Dec2011-SP2
| Subquery in select part goes wrong | https://api.github.com/repos/MonetDB/MonetDB/issues/2988/comments | 0 | 2020-11-30T11:50:49Z | 2024-06-27T11:54:53Z | https://github.com/MonetDB/MonetDB/issues/2988 | 753,411,779 | 2,988 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-27 14:56:26 +0100
From: Holger <<holger>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-05-25 12:58:49 +0200
## Comment 16837
Date: 2012-01-27 14:56:26 +0100
From: Holger <<holger>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/534.52.7 (KHTML, like Gecko) Version/5.1.2 Safari/534.52.7
Build Identifier: v11.7.1
sql>\f raw
sql>explain select round(750,1);
% .explain table_name
% mal name
% clob type
% 71 length
function user.s15_1{autoCommit=true}():void;
X_4 := calc.bte(750:sht,2,0);
X_8 := sql.round(X_4,2,0,1:bte);
sql.exportValue(1,".","round_single_value","decimal",2,0,8,X_8,"");
end s15_1;
sql>select round(750,1);
MAPI = (monetdb) /tmp/.s.monetdb.50000
QUERY = select round(750,1);
ERROR = !too many digits (3 > 2)
Reproducible: Always
### Steps to Reproduce:
1. run sql query: select round(750,1);
### Actual Results:
22003!too many digits (3 > 2)
### Expected Results:
750
## Comment 17225
Date: 2012-05-07 13:48:39 +0200
From: @njnes
fixed by adding another check for functions with decimals as input (when the input expressions are numbers)
| Broken type inference when using sql round | https://api.github.com/repos/MonetDB/MonetDB/issues/2987/comments | 0 | 2020-11-30T11:50:47Z | 2024-06-27T11:54:52Z | https://github.com/MonetDB/MonetDB/issues/2987 | 753,411,755 | 2,987 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-26 11:53:26 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2012-02-29 23:55:38 +0100
## Comment 16812
Date: 2012-01-26 11:53:26 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0.1) Gecko/20100101 Firefox/8.0.1
Build Identifier: mclient.bat
I have a query that calls DENSE_RANK with a SUM inside a OVER condition, this way:
SELECT
DENSE_RANK() OVER(ORDER BY SUM('COL-INTEGER') DESC) AS RANK;
It works fine on PostgreSQL, but not for MonetDB
Reproducible: Always
### Steps to Reproduce:
1.CREATE TABLE tab(col integer);
2.INSERT INTO tab(col) VALUES (5);
2.SELECT DENSE_RANK() OVER(ORDER BY col DESC) AS RANK FROM tab;
### Actual Results:
Answer for PostgreSQL:
1
Answer for MonetDB (mclient):
cannot use non GROUP BY column 'col' in query results without an aggregate function
### Expected Results:
The same result as Pg was expected, which is right if I take off the SUM() clause.
## Comment 16813
Date: 2012-01-26 13:14:28 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
Sorry, the query is wrong, this is the right one:
SELECT DENSE_RANK() OVER(ORDER BY SUM(col) DESC) AS RANK FROM tab;
## Comment 17033
Date: 2012-02-29 23:55:38 +0100
From: @njnes
not a bug. The order by only allows for column expressions not aggregates. At best the error message could/should be improved
| Cannot use SUM() inside OVER() condition | https://api.github.com/repos/MonetDB/MonetDB/issues/2986/comments | 0 | 2020-11-30T11:50:44Z | 2024-06-28T13:18:26Z | https://github.com/MonetDB/MonetDB/issues/2986 | 753,411,731 | 2,986 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-25 16:54:53 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Duplicates: #2988
Last updated: 2012-02-27 19:26:47 +0100
## Comment 16811
Date: 2012-01-25 16:54:53 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
CREATE TABLE "sys"."timdemrnt" (
"version" DECIMAL(2) NOT NULL,
"implicit" BOOLEAN NOT NULL,
"dataownercode" VARCHAR(10) NOT NULL,
"lineplanningnumber" VARCHAR(10) NOT NULL,
"journeypatterncode" VARCHAR(10) NOT NULL,
"timedemandgroupcode" VARCHAR(10) NOT NULL,
"timinglinkorder" DECIMAL(3) NOT NULL,
"userstopcodebegin" VARCHAR(10) NOT NULL,
"userstopcodeend" VARCHAR(10) NOT NULL,
"totaldrivetime" DECIMAL(5) NOT NULL,
"drivetime" DECIMAL(5) NOT NULL,
"expecteddelay" DECIMAL(5),
"layovertime" DECIMAL(5),
"stopwaittime" DECIMAL(5) NOT NULL,
"minimumstoptime" DECIMAL(5)
);
CREATE TABLE "sys"."pujo" (
"version" DECIMAL(2) NOT NULL,
"implicit" BOOLEAN NOT NULL,
"dataownercode" VARCHAR(10) NOT NULL,
"timetableversioncode" VARCHAR(10) NOT NULL,
"organizationalunitcode" VARCHAR(10) NOT NULL,
"periodgroupcode" VARCHAR(10) NOT NULL,
"specificdaycode" VARCHAR(10) NOT NULL,
"daytypemonday" BOOLEAN NOT NULL,
"daytypetuesday" BOOLEAN NOT NULL,
"daytypewednesday" BOOLEAN NOT NULL,
"daytypethursday" BOOLEAN NOT NULL,
"daytypefriday" BOOLEAN NOT NULL,
"daytypesaturday" BOOLEAN NOT NULL,
"daytypesunday" BOOLEAN NOT NULL,
"lineplanningnumber" VARCHAR(10) NOT NULL,
"journeynumber" DECIMAL(6) NOT NULL,
"timedemandgroupcode" VARCHAR(10) NOT NULL,
"journeypatterncode" VARCHAR(10) NOT NULL,
"departuretime" TIMESTAMP NOT NULL,
"wheelchairaccessible" BOOLEAN NOT NULL,
"dataownerisoperator" BOOLEAN NOT NULL
);
CREATE TABLE "sys"."pegrval" (
"version" DECIMAL(2) NOT NULL,
"implicit" BOOLEAN NOT NULL,
"dataownercode" VARCHAR(10) NOT NULL,
"organizationalunitcode" VARCHAR(10) NOT NULL,
"periodgroupcode" VARCHAR(10) NOT NULL,
"validfrom" DATE NOT NULL,
"validthru" DATE NOT NULL
);
CREATE TABLE "sys"."usrstop" (
"version" DECIMAL(2) NOT NULL,
"implicit" BOOLEAN NOT NULL,
"dataownercode" VARCHAR(10) NOT NULL,
"userstopcode" VARCHAR(10) NOT NULL,
"timingpointcode" VARCHAR(10),
"getin" BOOLEAN NOT NULL,
"getout" BOOLEAN NOT NULL,
"name" VARCHAR(50) NOT NULL,
"town" VARCHAR(50) NOT NULL,
"userstopareacode" VARCHAR(10),
"stopsidecode" VARCHAR(10),
"roadsideeqdataownercode" VARCHAR(10),
"roadsideequnitnumber" DECIMAL(5),
"minimalstoptime" DECIMAL(5) NOT NULL,
"stopsidelength" DECIMAL(3),
"description" VARCHAR(255),
"userstoptype" VARCHAR(10) NOT NULL
);
CREATE TABLE "sys"."line" (
"version" DECIMAL(2) NOT NULL,
"implicit" BOOLEAN NOT NULL,
"dataownercode" VARCHAR(10) NOT NULL,
"lineplanningnumber" VARCHAR(10) NOT NULL,
"linepublicnumber" VARCHAR(4) NOT NULL,
"linename" VARCHAR(50) NOT NULL,
"linevetagnumber" DECIMAL(3) NOT NULL,
"description" VARCHAR(255),
"transporttype" VARCHAR(5) NOT NULL
);
select linepublicnumber as lpn, timinglinkorder as tlo, name, totaldrivetime,
(select sum(totaldrivetime) from timdemrnt where
huge.dataownercode = timdemrnt.dataownercode and
huge.lineplanningnumber = timdemrnt.lineplanningnumber and
huge.journeypatterncode = timdemrnt.journeypatterncode and
huge.timedemandgroupcode = timdemrnt.timedemandgroupcode and
huge.timinglinkorder >= timdemrnt.timinglinkorder group by totaldrivetime)
from
(select timdemrnt.*, pujo.dataownercode, pujo.lineplanningnumber, pujo.timetableversioncode, pujo.journeypatterncode, pujo.timedemandgroupcode from pujo, timdemrnt, pegrval where
pujo.dataownercode = timdemrnt.dataownercode and
pujo.lineplanningnumber = timdemrnt.lineplanningnumber and
pujo.journeypatterncode = timdemrnt.journeypatterncode and
pujo.timedemandgroupcode = timdemrnt.timedemandgroupcode and
pujo.daytypemonday = True and
pujo.departuretime = '1970-01-01 05:35:00' and
pujo.timetableversioncode = pegrval.periodgroupcode and
'2012-01-18' between pegrval.validfrom and pegrval.validthru and
pujo.lineplanningnumber = 'A060') as huge,
usrstop, line
where
huge.userstopcodebegin = usrstop.userstopcode and
huge.lineplanningnumber = line.lineplanningnumber
order by
huge.dataownercode,
huge.lineplanningnumber,
huge.timetableversioncode,
huge.journeypatterncode,
huge.timedemandgroupcode,
huge.timinglinkorder
limit 40;
And all this magic results in:
>mserver5: rel_select.c:603: rel_project: Assertion `exps_card(rel->exps) <= rel->card' failed.
I'll try to make this query smaller.
Reproducible: Always
MonetDB 5 server v11.8.0 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2012 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 3.9GiB available memory, 1 available cpu core
Libraries:
libpcre: 8.20 2011-10-21 (compiled with 8.20)
openssl: OpenSSL 1.0.0e 6 Sep 2011 (compiled with OpenSSL 1.0.0e 6 Sep 2011)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: root@openov_cxx (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 16839
Date: 2012-01-28 16:10:15 +0100
From: @skinkie
Or the trivial case:
select x, (select 1 from (select 1 as y, 2 as z) as y where x.x = y.y group by y.z) from (select 1 as x) as x;
Without the group by it seems to work. But I would love to use a group by here, that would hopefully allow me to do incremental summations.
from w to x takes 60s
from x to y takes 30s
from y to z takes 50s
transformed to:
from w to x takes 60s
from w to y takes 90s
from w to z takes 140s
## Comment 17009
Date: 2012-02-27 19:26:47 +0100
From: @njnes
*** This bug has been marked as a duplicate of bug #2988 ***
| >mserver5: rel_select.c:603: rel_project: Assertion `exps_card(rel->exps) <= rel->card' failed. | https://api.github.com/repos/MonetDB/MonetDB/issues/2985/comments | 0 | 2020-11-30T11:50:40Z | 2024-06-28T13:18:25Z | https://github.com/MonetDB/MonetDB/issues/2985 | 753,411,693 | 2,985 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-23 10:34:18 +0100
From: @grobian
To: Merovingian devs <<bugs-merovingian>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-29 10:34:25 +0100
## Comment 16808
Date: 2012-01-23 10:34:18 +0100
From: @grobian
(vomac:/) fabian% /usr/local/monetdb/bin/monetdb -hlocalhost -p50000 -Ptest create mytest2
create: OK
(vomac:/) fabian% /usr/local/monetdb/bin/monetdb -hlocalhost -p50000 -Ptest release mytest2
release: OK
(vomac:/) fabian% /usr/local/monetdb/bin/monetdb -hlocalhost -p50000 -Ptest start mytest2
starting database 'mytest2'... FAILED
start: OK
(vomac:/) fabian% /usr/local/monetdb/bin/monetdb -hlocalhost -p50000 -Ptest stop mytest2
stopping database 'mytest2'... FAILED
stop: OK
(vomac:/) fabian% /usr/local/monetdb/bin/monetdb -hlocalhost -p50000 -Ptest status
name state uptime health
mynewdb stopped 100%, 34s
mytest2 stopped 100%, 13s
(vomac:/) fabian% /usr/local/monetdb/bin/monetdb -hlocalhost -p50000 -Ptest discover
location
mapi:monetdb://vomac.cwi.nl:50000/mynewdb
mapi:monetdb://vomac.cwi.nl:50000/mytest2
## Comment 16809
Date: 2012-01-23 10:37:06 +0100
From: @grobian
incidently, the merovingian.log says:
2012-01-23 10:33:06 ERR control[55208]: localhost:54694: error reading from cont
rol channel: Unknown error: 0
## Comment 16845
Date: 2012-01-30 16:28:11 +0100
From: @grobian
Changeset [12897126f239](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=12897126f239) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=12897126f239](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=12897126f239)
Changeset description:
ctl_handle_client: don't send double trailing newlines
Fix for bug #2984. Due to sending too many newlines, the client code
interpreted the answer from the server as not being the exact string
"OK\n". (Correct, there was another '\n').
## Comment 16846
Date: 2012-01-30 16:56:59 +0100
From: @grobian
Changeset [da50d8af4828](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=da50d8af4828) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=da50d8af4828](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=da50d8af4828)
Changeset description:
recvWithTimeout: tweak logic involving mnstr_errnr return
Return something more useful, avoiding "error reading from control
channel" error messages in the log, as observed in bug #2984.
## Comment 16847
Date: 2012-01-30 16:58:20 +0100
From: @grobian
Fixed with previous two commits.
| monetdb create/release/start/stop returns error "OK" for remote connections | https://api.github.com/repos/MonetDB/MonetDB/issues/2984/comments | 0 | 2020-11-30T11:50:37Z | 2024-06-27T11:54:49Z | https://github.com/MonetDB/MonetDB/issues/2984 | 753,411,662 | 2,984 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-23 10:17:23 +0100
From: @grobian
To: Merovingian devs <<bugs-merovingian>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-29 10:34:21 +0100
## Comment 16807
Date: 2012-01-23 10:17:23 +0100
From: @grobian
Upon shutdown sequence, monetdbd logs the following error:
2012-01-23 10:14:53 ERR merovingian[54777]: strange, process 54777 serves database 'control' which does not exist
It shouldn't try to stop itself (or its pseudo databases).
## Comment 16848
Date: 2012-01-30 17:25:08 +0100
From: @grobian
Changeset [39f63e96af65](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=39f63e96af65) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=39f63e96af65](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=39f63e96af65)
Changeset description:
merovingian: don't try to shutdown internal database 'control'
Skip shutting down the pseudo 'control' database that was introduced for
allowing remote control connections to be made over the same (mapi)
connection served by monetdbd. This fixes bug #2983.
## Comment 16849
Date: 2012-01-30 17:25:28 +0100
From: @grobian
per comment 1
| monetdbd attempts to stop non-existing database 'control' | https://api.github.com/repos/MonetDB/MonetDB/issues/2983/comments | 0 | 2020-11-30T11:50:34Z | 2024-06-27T11:54:48Z | https://github.com/MonetDB/MonetDB/issues/2983 | 753,411,624 | 2,983 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-23 10:11:40 +0100
From: @grobian
To: Merovingian devs <<bugs-merovingian>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-02-29 10:34:29 +0100
## Comment 16806
Date: 2012-01-23 10:11:40 +0100
From: @grobian
continuously running monetdbd get all dbfarm with a running monetdbd returns random values for status and mserver like so:
status monetdbd[54777] (unknown version) is serving this dbfarm
mserver /usr/local/monetdb/bin/mserver5
status monetdbd[54777] 1.6 (Dec2011) is serving this dbfarm
mserver /usr/local/monetdb/bin/mserver5
status monetdbd[54777] 1.6 (Dec2011) is serving this dbfarm
mserver unknown (unsupported monetdbd)
This is odd, merovingian.log doesn't report any problems.
## Comment 16851
Date: 2012-01-30 18:24:17 +0100
From: @grobian
Changeset [fd13abca7fc6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fd13abca7fc6) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fd13abca7fc6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fd13abca7fc6)
Changeset description:
monetdbd properties: mserver and version are multi-line responses
make sure we expect more output to follow after requesting mserver or
version properties
make sure we terminate the connection after sending the answer to
mserver and version properties
this solves a problem where the answer might not arrive, and monetdbd
eventually thinking more should have come, bug #2982
## Comment 16852
Date: 2012-01-30 18:24:39 +0100
From: @grobian
Fixed per comment 1
| monetdbd status and mserver properties have random values | https://api.github.com/repos/MonetDB/MonetDB/issues/2982/comments | 0 | 2020-11-30T11:50:32Z | 2024-06-27T11:54:47Z | https://github.com/MonetDB/MonetDB/issues/2982 | 753,411,605 | 2,982 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-19 17:36:03 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
To: clients devs <<bugs-clients>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-26 15:32:05 +0100
## Comment 16793
Date: 2012-01-19 17:36:03 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0.1) Gecko/20100101 Firefox/8.0.1
Build Identifier: Embarcadero RAD Studio XE Version 15.0.3953.35171
I got some queries that return different results in my application using Delphi and ODBC, compared to the mclient and from a copy of the DB in PostgreSQL. Apparently I can have the correct answer if I put a NULL column in SELECT, like
"SELECT Col01, NULL FROM Table01;"
But still my BI application have a different behaviour in the result of the queries compared to PostgreSQL. I ran separately the queries in mclient and the Postgres's PgAdmin, and there the results were the same.
Reproducible: Always
### Steps to Reproduce:
1.Running the following query in an application with Delphi
2.SELECT 2001 as Ano;
### Actual Results:
2001
1991
1992
1993
1994
1995
(NOTE: My application can't retrieve the data, giving a lot of errors)
### Expected Results:
2001
3. now run the following query
4. SELECT 2001 as Ano, NULL;
5. You'll get the following result, which is:
2001 | NULL
## Comment 16794
Date: 2012-01-19 17:38:00 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
Created attachment 97
log from my application run, but apparently no errors are found from the queries.
> Attached file: [odbc.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2981_odbc.log_97) (application/octet-stream, 131090 bytes)
> Description: log from my application run, but apparently no errors are found from the queries.
## Comment 16796
Date: 2012-01-19 18:32:02 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
I ran over the bug. I put a NULL column in all the queries and the application worked.
## Comment 16798
Date: 2012-01-20 10:20:26 +0100
From: @sjoerdmullender
I don't see a correlation in the log with the queries that you mention in your report. Can you create a log where you try the queries you mention?
## Comment 16799
Date: 2012-01-20 11:03:30 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
I'm attaching the log with the exact queries I mentioned on the report.
## Comment 16800
Date: 2012-01-20 11:03:53 +0100
From: Luiz Gustavo de Souza <<luiz.souza>>
Created attachment 98
Log with the report queries
> Attached file: [odbc.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2981_odbc.log_98) (application/octet-stream, 4021 bytes)
> Description: Log with the report queries
## Comment 16801
Date: 2012-01-20 13:47:54 +0100
From: @sjoerdmullender
Changeset [df6d0b68a429](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=df6d0b68a429) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=df6d0b68a429](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=df6d0b68a429)
Changeset description:
ODBC: Use correct row descriptor when returning values in SQLFetch.
SQLFetch can return information about the fetched data to the
application through the SQL_ATTR_ROW_STATUS_PTR and
SQL_ATTR_ROWS_FETCHED_PTR statement attributes. These values are
stored in the Implementation Row Descriptor, not in the Application
Row Descriptor.
This should fix bug #2981.
## Comment 16802
Date: 2012-01-20 14:05:22 +0100
From: @sjoerdmullender
Hopefully fixed.
The fix is in the Dec2011 branch and will be in the next release.
## Comment 16829
Date: 2012-01-26 15:32:05 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| Different answer from querys in the ODBC compared to mcliente or other DBMSs | https://api.github.com/repos/MonetDB/MonetDB/issues/2981/comments | 0 | 2020-11-30T11:50:30Z | 2024-06-27T11:54:46Z | https://github.com/MonetDB/MonetDB/issues/2981 | 753,411,588 | 2,981 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-19 14:40:53 +0100
From: sellam
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2012-01-19 15:43:08 +0100
## Comment 16790
Date: 2012-01-19 14:40:53 +0100
From: sellam
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.63 Safari/535.7
Build Identifier: Dec2011 42876:bc492bfbc040 / Aug2011 42389:9e3ad41dfa21
The Website provides an example of UDF:
http://www.monetdb.org/Documentation/Cookbooks/SQLrecipies/Extensions
However, I did not manage to run the example reverse(ra1:str):str.
Same issue was observed on another workstation.
Reproducible: Always
### Steps to Reproduce:
sql>select reverse('this is a test');
TypeException:user.s2_1[4]:object code for command udf.reverse missing
TypeException:user.s2_1[4]:'udf.reverse' undefined in: _7:str := udf.reverse(_6:str)
39000!program contains errors
mal>udf.reverse("lalala");
MAPI = (monetdb) /tmp/.s.monetdb.50000
QUERY = udf.reverse("lalala");
ERROR = !TypeException:user.main[1]:object code for command udf.reverse missing
!TypeException:user.main[1]:'udf.reverse' undefined in: _1:str := udf.reverse(_2:str)
mal>
Tested with latest Aug2011 and Dec2011
## Comment 16791
Date: 2012-01-19 15:06:30 +0100
From: @grobian
Changeset [baf7d5fc1f6f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=baf7d5fc1f6f) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=baf7d5fc1f6f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=baf7d5fc1f6f)
Changeset description:
UDF: get module including right
For bug #2980, don't declare signatures twice, and actually include the
module, such that the library is also loaded.
## Comment 16792
Date: 2012-01-19 15:43:08 +0100
From: sellam
Problem solved - website and source updated.
Thanks for your help Fabian and Sjoerd
| UDF template "reverse(ra1:str):str" seems broken | https://api.github.com/repos/MonetDB/MonetDB/issues/2980/comments | 0 | 2020-11-30T11:50:27Z | 2024-06-27T11:54:46Z | https://github.com/MonetDB/MonetDB/issues/2980 | 753,411,565 | 2,980 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-19 12:26:19 +0100
From: Wouter Alink <<alink>>
To: Merovingian devs <<bugs-merovingian>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @drstmane
Last updated: 2012-02-29 10:34:22 +0100
## Comment 16787
Date: 2012-01-19 12:26:19 +0100
From: Wouter Alink <<alink>>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.24) Gecko/20111108 Fedora/3.6.24-1.fc14 Firefox/3.6.24
Build Identifier: latest
Various bugs in merovingian:
1. Not all advertised hashes are supported (SHA256)
2. java Control doesn't understand new connection methods
3. errors aren't returned properly according to the protocol
4. response should not expect trailing new-line.
The issue numnber 4 is a show-stopper for the release.
Reproducible: Always
## Comment 16788
Date: 2012-01-19 12:33:45 +0100
From: @drstmane
"Version: unspecified"
Since you're talking about release show stoppers, know the version you're using would be very useful for us.
Thanks!
## Comment 16789
Date: 2012-01-19 13:07:59 +0100
From: Wouter Alink <<alink>>
sorry, i know it is already Jan, but it's the Dec release we are talking about.
## Comment 16795
Date: 2012-01-19 18:18:22 +0100
From: @grobian
Changeset [468d83f26d17](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=468d83f26d17) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=468d83f26d17](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=468d83f26d17)
Changeset description:
monetdb TCP control: properly use block-based protocol
Instead of a hacked up blockwise login sequence followed by a
non-blockwise communication sequence, do complete communication
blockwise for TCP connections, and without for UNIX socket connections.
This allows other clients (like merocontrol Java library) not to have to
hack to the max, but use existing connection code for the blockwise
protocol to communicate to monetdbd.
This commit is part of a fix for bug #2979.
## Comment 16850
Date: 2012-01-30 17:31:36 +0100
From: @grobian
Closing this bug, we should file new bugs for the issues we're sometimes seeing with the merocontrol java library, ok?
| misc merovingian control problems | https://api.github.com/repos/MonetDB/MonetDB/issues/2979/comments | 0 | 2020-11-30T11:50:24Z | 2024-06-27T11:54:44Z | https://github.com/MonetDB/MonetDB/issues/2979 | 753,411,536 | 2,979 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-18 17:20:50 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: 11.13.3 (Oct2012)
CC: @njnes, @elpetrak
Last updated: 2013-01-22 09:29:19 +0100
## Comment 16782
Date: 2012-01-18 17:20:50 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
sql>select now() + 1;
IllegalArgumentException:calc.second_interval:illegal argument
(yeah it should be done with interval)
Reproducible: Always
## Comment 17552
Date: 2012-07-21 19:03:59 +0200
From: @grobian
sql>select now() + interval '1' day;
+----------------------------------+
| current_timestamp |
+==================================+
| 2012-07-22 18:03:33.000000+01:00 |
+----------------------------------+
1 tuple (0.613ms)
I guess this bug is then just about making the error message a bit more clear?
## Comment 17666
Date: 2012-08-24 14:54:32 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18047
Date: 2012-11-27 13:03:25 +0100
From: @elpetrak
To make a test for this case, I should have a substraction (now()-now()) in order to have a stable result. However, this substraction returns an error. Moreover, the query returns an error when the interval unit is smaller than second (e.g., millisecond).
## Comment 18207
Date: 2012-11-28 13:45:54 +0100
From: @elpetrak
Changeset [973503cfeacd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=973503cfeacd) made by Eleni Petraki <petraki@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=973503cfeacd](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=973503cfeacd)
Changeset description:
Add test for bug #2978 (now() + 1 results in illegal argument).
## Comment 18217
Date: 2012-11-28 19:33:37 +0100
From: @njnes
addition/substraction of timestamps isn't supported. We now give a proper error.
## Comment 18386
Date: 2013-01-22 09:29:19 +0100
From: @sjoerdmullender
Oct2012-SP3 has been released.
| now() + 1 results in illegal argument | https://api.github.com/repos/MonetDB/MonetDB/issues/2978/comments | 0 | 2020-11-30T11:50:22Z | 2024-06-27T11:54:43Z | https://github.com/MonetDB/MonetDB/issues/2978 | 753,411,516 | 2,978 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-18 16:27:35 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: 11.11.7 (Jul2012-SP1)
CC: @njnes, sellam
Last updated: 2013-01-22 09:29:13 +0100
## Comment 16781
Date: 2012-01-18 16:27:35 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
select (cast('1970-01-01 0:00' as timestamp) - cast('1970-01-01' as date));
TypeException:user.s9_3[7]:'calc.second_interval' undefined in: _12:any := calc.second_interval(_11:date, _13:int, _14:int)
39000!program contains errors
sql>select (cast('1970-01-01 0:00' as timestamp) - cast('1970-01-01' as timestamp));
+----------------------+
| sql_sub_single_value |
+======================+
| 0 |
+----------------------+
1 tuple (1.363ms)
Reproducible: Always
/opt/monetdb/bin/mserver5 --version
MonetDB 5 server v11.8.0 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2012 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 3.9GiB available memory, 1 available cpu core
Libraries:
libpcre: 8.20 2011-10-21 (compiled with 8.20)
openssl: OpenSSL 1.0.0e 6 Sep 2011 (compiled with OpenSSL 1.0.0e 6 Sep 2011)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: root@openov_cxx (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 17553
Date: 2012-07-21 19:04:55 +0200
From: @grobian
problem still exists in current Jul2012
## Comment 17554
Date: 2012-07-21 19:05:15 +0200
From: @grobian
sql bug
## Comment 17680
Date: 2012-08-24 14:55:57 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18029
Date: 2012-11-27 11:25:01 +0100
From: sellam
This combination is not allowed in SQL-99.
You can perform arithmetic on
- Date and Interval
- Timestamp and Interval
- Date and Date
- Timestamp and timestamp
but not timestamp/interval
(In reply to comment 3)
> Jul2012-SP1 has been released.
## Comment 18031
Date: 2012-11-27 11:27:40 +0100
From: sellam
oops typo, I meant:
but not timestamp and date
(In reply to comment 4)
> This combination is not allowed in SQL-99.
> You can perform arithmetic on
> - Date and Interval
> - Timestamp and Interval
> - Date and Date
> - Timestamp and timestamp
> but not timestamp/interval
>
> (In reply to comment 3)
> > Jul2012-SP1 has been released.
## Comment 18032
Date: 2012-11-27 11:30:06 +0100
From: @skinkie
(In reply to comment 4)
> This combination is not allowed in SQL-99.
> You can perform arithmetic on
> - Date and Interval
> - Timestamp and Interval
> - Date and Date
> - Timestamp and timestamp
> but not timestamp/interval
It works as a charm on PostgreSQL.
select (cast('1970-01-01 0:00' as timestamp) - cast('1970-01-01' as date));
?column?
----------
00:00:00
(1 row)
Date can be converted to timestamp by the parser.
## Comment 18197
Date: 2012-11-28 13:45:27 +0100
From: sellam
Changeset [389c9f8e78f4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=389c9f8e78f4) made by Thibault Sellam <sellam@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=389c9f8e78f4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=389c9f8e78f4)
Changeset description:
Added test for bug #2977
## Comment 18212
Date: 2012-11-28 16:23:36 +0100
From: @njnes
Changeset [1ee4a27c1126](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1ee4a27c1126) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1ee4a27c1126](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1ee4a27c1126)
Changeset description:
fixed bugs 2977 and 3143
return errors in psm when we don't expect select statements (solves 3143)
Give proper error when a function isn't found and don't allow converts from
date into second intervals. (solves 2977)
## Comment 18214
Date: 2012-11-28 16:24:00 +0100
From: @njnes
give proper error now
## Comment 18373
Date: 2013-01-22 09:29:13 +0100
From: @sjoerdmullender
Oct2012-SP3 has been released.
| timestamp minus date results in program error | https://api.github.com/repos/MonetDB/MonetDB/issues/2977/comments | 0 | 2020-11-30T11:50:20Z | 2024-06-27T11:54:42Z | https://github.com/MonetDB/MonetDB/issues/2977 | 753,411,497 | 2,977 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-18 16:13:00 +0100
From: sco <<scoulibaly>>
To: GDK devs <<bugs-common>>
Version: -- development
CC: @drstmane
Last updated: 2012-01-19 10:01:31 +0100
## Comment 16780
Date: 2012-01-18 16:13:00 +0100
From: sco <<scoulibaly>>
User-Agent: Mozilla/5.0 (Windows NT 6.1) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.75 Safari/535.7
Build Identifier: v11.5.7 "Aug2011-SP2"
When trying to import a 4 millions lines (940 MB) long CSV file, the following occurs.
PC, Windows7, 4GB RAM
c:\Program Files\MonetDB\MonetDB5\bin>mclient -u monetdb -lsql --database=demo -
s "COPY INTO cr_epoch FROM 'e:\synthesis\01_cr_all.csv' USING DELIMITERS ';','\n
';"
password:
invalid/unknown response from server, ignoring output
? GDKmmap(100663296) fails, try to free up space [memory in use=13054864,virtua
l memory in use=1046478848]
? GDKmmap(100663296) result [mem=12521960,vm=1015021568]
? GDKmmap(100663296) fails, try to free up space [memory in use=12513728,virtua
l memory in use=1015021568]
? GDKmmap(100663296) result [mem=12495136,vm=1015021568]
MALException:!ERROR: HEAPextend: failed to extend to 100663296 for 25\2557theap
Reproducible: Always
### Steps to Reproduce:
1. Create a 4 millions lines (940 MB) long CSV file
2. Create a suitable table
3. Import with c:\Program Files\MonetDB\MonetDB5\bin>mclient -u monetdb -lsql --database=demo -
s "COPY INTO cr_epoch FROM 'e:\synthesis\01_cr_all.csv' USING DELIMITERS ';','\n
';"
### Actual Results:
The import fails with mclient displaying :
invalid/unknown response from server, ignoring output
? GDKmmap(100663296) fails, try to free up space [memory in use=13054864,virtua
l memory in use=1046478848]
? GDKmmap(100663296) result [mem=12521960,vm=1015021568]
? GDKmmap(100663296) fails, try to free up space [memory in use=12513728,virtua
l memory in use=1015021568]
? GDKmmap(100663296) result [mem=12495136,vm=1015021568]
MALException:!ERROR: HEAPextend: failed to extend to 100663296 for 25\2557theap
### Expected Results:
The import is performed properly, all data is ingested in the table.
Big files not fitting in memory shall be able to be imported.
If to be imported data is greater than available RAM, the import shall finish properly.
Major severity since this make MonetDB unusable for many use cases.
## Comment 16783
Date: 2012-01-19 08:40:14 +0100
From: @grobian
Could you paste the contents of the mserver dos box in here, please?
## Comment 16784
Date: 2012-01-19 09:48:25 +0100
From: sco <<scoulibaly>>
Sure Fabian, here it is. Let me know if you need anything else.
MonetDB 5 server v11.5.7 "Aug2011-SP2"
Serving database 'demo', using 4 threads
Compiled for i686-pc-winnt/32bit with 32bit OIDs dynamically linked
Found 3.165 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
MonetDB/SQL module loaded
## Comment 16785
Date: 2012-01-19 09:54:26 +0100
From: @drstmane
What does the schema of your table look like?
While MonetDB is in pricinple very well able to load data that is larger than
the physically available memory --- on 8 GB 64-bit Linux machines, we eaasily
load 100 GB data ---, on 32-bit systems, in particular 32-bit Windows, with
large data MonetDB can hit the 32-bit address space limitation. For details,
see also
http://www.monetdb.org/Documentation/UserGuide/resources
## Comment 16786
Date: 2012-01-19 10:01:31 +0100
From: @grobian
(In reply to comment 2)
> Compiled for i686-pc-winnt/32bit with 32bit OIDs dynamically linked
You're on a 32-bits server/platform. By design of MonetDB, we can't support this. Please switch to 64-bits server/platform.
| MALException when importing a 1GB (4Million lines) CSV file | https://api.github.com/repos/MonetDB/MonetDB/issues/2976/comments | 0 | 2020-11-30T11:50:16Z | 2024-06-28T13:18:24Z | https://github.com/MonetDB/MonetDB/issues/2976 | 753,411,457 | 2,976 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-18 15:23:28 +0100
From: sco <<scoulibaly>>
To: clients devs <<bugs-clients>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-26 15:31:57 +0100
## Comment 16779
Date: 2012-01-18 15:23:28 +0100
From: sco <<scoulibaly>>
User-Agent: Mozilla/5.0 (Windows NT 6.1) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.75 Safari/535.7
Build Identifier: v11.5.7 "Aug2011-SP2"
After having tried to import a CSV file into the database, the file can not be written from any text editor. I guess a reboot or session closing is necessary to be able to write to the CSV file again.
Command line used :
c:\Program Files\MonetDB\MonetDB5\bin>mclient -u monetdb -lsql --database=demo -
s "COPY INTO cr_epoch FROM 'e:\synthesis\cr_all_yyyy_mm_dd.csv' USING DELIMITERS
';','\n';"
password:
SQLException:importTable:value 'Timestamp' from line 1 field 1 not inserted, exp
ecting type timestamp(7)
failed to import table
Reproducible: Always
### Steps to Reproduce:
1. Create a table with a timestamp element.
2. Create CSV file with an invalid timestamp at first and single column
3. Import with c:\Program Files\MonetDB\MonetDB5\bin>mclient -u monetdb -lsql --database=demo -
s "COPY INTO cr_epoch FROM 'e:\synthesis\cr_all_yyyy_mm_dd.csv' USING DELIMITERS
';','\n';"
### Actual Results:
The file cr_all_yyyy_mm_dd.csv can not be written (although acecss rights are set to all from write rights)
### Expected Results:
File is saved normally
Since it can be huge files, using temporary copies that can be modified and imported again can be difficult, hence major severity.
Platform : PC
OS : Windows 7 64 bits (sorry, the bigzilla refuses these items to be set...)
## Comment 16797
Date: 2012-01-20 09:49:39 +0100
From: @sjoerdmullender
I can see that the server in the Aug2011 releases keep the file open, but this has been fixed in the upcoming Dec2011 release.
Can you try the release candidate at http://dev.monetdb.org/downloads/testings/Windows/ and report back?
## Comment 16803
Date: 2012-01-20 14:35:42 +0100
From: @sjoerdmullender
I believe this is fixed, but please reopen if it wasn't.
## Comment 16816
Date: 2012-01-26 15:31:57 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| CSV file is kept write-locked after mclient failed to COPY | https://api.github.com/repos/MonetDB/MonetDB/issues/2975/comments | 0 | 2020-11-30T11:50:14Z | 2024-06-27T11:54:40Z | https://github.com/MonetDB/MonetDB/issues/2975 | 753,411,427 | 2,975 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-14 13:43:49 +0100
From: @grobian
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2012-03-16 14:56:56 +0100
## Comment 16773
Date: 2012-01-14 13:43:49 +0100
From: @grobian
The following snippet demonstrates a problem with the result of a thetajoin.
(it is not blocking me at the moment, because the thetajoin doesn't do what I need, which I'm affraid is by design of the thetajoin, similar to antijoin problems)
l := bat.new(:oid, :int);
r := bat.new(:oid, :int);
l := bat.insert(l, 2@0, 1);
l := bat.insert(l, 6@0, 2);
r := bat.insert(r, 3@0, 1);
r := bat.insert(r, 7@0, 2);
rr := bat.reverse(r);
t := algebra.thetajoin(l, rr, 1); gt
io.print(t); returns 6@0 3@0 (2 is greater than 1? w/e)
b := bat.new(:oid, :oid);
b := bat.insert(b, t); BUG: ERROR: BATkey: dense column must be unique.
## Comment 17023
Date: 2012-02-28 15:49:22 +0100
From: @sjoerdmullender
(In reply to comment 0)
> The following snippet demonstrates a problem with the result of a thetajoin.
> (it is not blocking me at the moment, because the thetajoin doesn't do what I
> need, which I'm affraid is by design of the thetajoin, similar to antijoin
> problems)
>
>
> l := bat.new(:oid, :int);
> r := bat.new(:oid, :int);
>
> l := bat.insert(l, 2@0, 1);
> l := bat.insert(l, 6@0, 2);
>
> r := bat.insert(r, 3@0, 1);
> r := bat.insert(r, 7@0, 2);
>
> rr := bat.reverse(r);
>
> t := algebra.thetajoin(l, rr, 1); gt
>
> io.print(t); returns 6@0 3@0 (2 is greater than 1? w/e)
Yes, 2 *is* greater than 1.
> b := bat.new(:oid, :oid);
> b := bat.insert(b, t); BUG: ERROR: BATkey: dense column must be unique.
This is a matter of incorrect property setting. It can be solved by calling BATsettrivprop() at the end of BATnlthetajoin.
It is worth checking for similar situations (all functions that return a new BAT?).
## Comment 17024
Date: 2012-02-28 16:53:01 +0100
From: @sjoerdmullender
Changeset [84a308d197ae](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=84a308d197ae) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=84a308d197ae](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=84a308d197ae)
Changeset description:
When returning a BAT to the intrepreter, set the trivial properties.
BBPkeepref is called whenever a new BAT is returned to the
interpreter, so it is a good place to set trivial properties
(properties which can be inferred trivially).
This fixes bug #2974.
Also added test for the bug.
## Comment 17025
Date: 2012-02-28 16:54:29 +0100
From: @sjoerdmullender
Fixed.
## Comment 17104
Date: 2012-03-16 14:56:56 +0100
From: @grobian
Released in Dec2011-SP2
| thetajoin produces invalid/inconsistent bat | https://api.github.com/repos/MonetDB/MonetDB/issues/2974/comments | 0 | 2020-11-30T11:50:11Z | 2024-06-27T11:54:40Z | https://github.com/MonetDB/MonetDB/issues/2974 | 753,411,397 | 2,974 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-14 13:39:42 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @drstmane
Last updated: 2012-02-29 10:34:27 +0100
## Comment 16772
Date: 2012-01-14 13:39:42 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier: Dec2011
The test file date_script_test.sql should be completed and tested.
Reproducible: Always
## Comment 16775
Date: 2012-01-14 14:12:15 +0100
From: @drstmane
Changeset [fdc3b8b379f8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fdc3b8b379f8) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fdc3b8b379f8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fdc3b8b379f8)
Changeset description:
renamed new tests to include bug report id in test name:
url_script_test.sql -> url_script_test.Bug-2972.sql
date_script_test.sql -> date_script_test.Bug-2973.sql
this simplies linking tests to bug reports and vice versa
## Comment 16777
Date: 2012-01-14 14:16:14 +0100
From: @drstmane
Changeset [49f931f6754f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=49f931f6754f) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=49f931f6754f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=49f931f6754f)
Changeset description:
added skeleton stable output for tests url_script_test.Bug-2972 & date_script_test.Bug-2973
correct / expected output still needs to be verified and approved
## Comment 16804
Date: 2012-01-20 14:59:38 +0100
From: @sjoerdmullender
Looking at the test and the code, I wonder what we think should happen.
The first test is
select str_to_date('2012','%y');
When given a format %y, the conversion function (strptime) expects a 2 digit year and returns a value for a year in the range 1969..2068. This means that in this case, the year returned is 2020 and the input string is not fully consumed.
This raises the question:
-> Should an error be returned if the input is not fully consumed? (My answer: yes.)
Furthermore, the conversion function will set the day-of-month and month both to 0. Further processing in mtime will then refuse this date as invalid (month has to be in range 1..12, day-of-month in the appropriate range for the month), and will return a date_nil value.
This raises another couple of question:
-> When given an invalid date, should the mtime module raise an error or just return nil? (My answer: it should raise an error.)
-> When the parser returns 0 for day-of-month and/or month, should we set the value to 1? (My answer: yes.)
Any opnions?
## Comment 16942
Date: 2012-02-19 18:20:56 +0100
From: @drstmane
date_script_test.Bug-2973.sql:
Appears to be fixed in the default branch (only) after changesets
http://dev.monetdb.org/hg/MonetDB/rev/8ee024048e42
http://dev.monetdb.org/hg/MonetDB/rev/e2d3ecc7237c
http://dev.monetdb.org/hg/MonetDB/rev/03df27701cf7
http://dev.monetdb.org/hg/MonetDB/rev/70dc90e1de2f
Remains open/failing in the Dec2011 branch.
## Comment 16947
Date: 2012-02-19 18:47:07 +0100
From: @drstmane
Fixed (also) in Dec2011 by transplanting changeset [70dc90e1de2f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=70dc90e1de2f) from default to Dec2011:
http://dev.monetdb.org/hg/MonetDB/rev/ec5692ecaf8c
| Date coercion functionality tests | https://api.github.com/repos/MonetDB/MonetDB/issues/2973/comments | 0 | 2020-11-30T11:50:08Z | 2024-06-27T11:54:39Z | https://github.com/MonetDB/MonetDB/issues/2973 | 753,411,356 | 2,973 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-14 13:23:06 +0100
From: @mlkersten
To: SQL devs <<bugs-sql>>
Version: 11.21.11 (Jul2015-SP1)
CC: joerg.strebel, @njnes
Last updated: 2016-03-25 09:59:37 +0100
## Comment 16771
Date: 2012-01-14 13:23:06 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier: Dec2011
The URL data type and operations are incomplete.
In particular, when the script url_script_test is ran completely
as
mserver5 --dbinit()="sql.start();" url_script_test.sql
...
% newurl_single_value name
% url type
% 28 length
[ "://www.monetdb.org:8080/blah" ]
which should have produced
[ "https://www.monetdb.org:8080/blah" ]
Reproducible: Always
Test file has been added, pending further investigation
## Comment 16774
Date: 2012-01-14 14:12:15 +0100
From: @drstmane
Changeset [fdc3b8b379f8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fdc3b8b379f8) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fdc3b8b379f8](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fdc3b8b379f8)
Changeset description:
renamed new tests to include bug report id in test name:
url_script_test.sql -> url_script_test.Bug-2972.sql
date_script_test.sql -> date_script_test.Bug-2973.sql
this simplies linking tests to bug reports and vice versa
## Comment 16776
Date: 2012-01-14 14:16:14 +0100
From: @drstmane
Changeset [49f931f6754f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=49f931f6754f) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=49f931f6754f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=49f931f6754f)
Changeset description:
added skeleton stable output for tests url_script_test.Bug-2972 & date_script_test.Bug-2973
correct / expected output still needs to be verified and approved
## Comment 16805
Date: 2012-01-20 15:53:36 +0100
From: @sjoerdmullender
When I try the test, I do get the expected result (https://www.monetdb.org:8080/blah). I also see that this result on the testweb.
## Comment 17134
Date: 2012-04-18 15:13:55 +0200
From: @grobian
Changeset [adc577be86a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=adc577be86a4) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=adc577be86a4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=adc577be86a4)
Changeset description:
open_urlstream: provide file:// implementation without cURL
For Windows (and other systems), where we don't have cURL, check if the
URL happens to be a file:// URL, and fall back to a regular ASCII stream
in that case. This should allow url_script_test.Bug-2972 test to come a
little bit further on Windows.
## Comment 17226
Date: 2012-05-07 13:52:54 +0200
From: @njnes
seems okay on linux, still broken on windows
## Comment 17244
Date: 2012-05-09 17:02:13 +0200
From: @njnes
output wasn't stable because of incorrect NIL checking, this is fixed now.
## Comment 17401
Date: 2012-07-03 12:56:41 +0200
From: @sjoerdmullender
Changeset [5aebbb6f96b0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5aebbb6f96b0) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5aebbb6f96b0](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5aebbb6f96b0)
Changeset description:
Make that sql test BugTracker-2012/url_script_test.Bug-2972 works on Windows.
This involves creating a special environment variable within Mtest
called UTSTSRCDIR which is the file: URL version of TSTSRCDIR, and
fixing the non-cURL implementation of open_urlstream to skip all
slashes that are part of a file: url on Windows.
## Comment 21720
Date: 2016-01-08 18:48:48 +0100
From: Jörg Strebel <<joerg.strebel>>
The test url_script_test.Bug-2972.sql.in contains the line:
insert into t2972 values ('$UTSTSRCDIR/url_script_test.Bug-2972.txt');
The Mtest script creates the following SQL statement:
insert into t2972 values ('file:/home/jstrebel/devel/MonetDB-Jul2015/MonetDB/sql/test/BugTracker-2012/Tests/url_script_test.Bug-2972.txt');
This table entry then makes the following "select getcontent(u) from t2972" throw an error (ERROR = !failed to open urlstream), as the double slashes after the colon are missing.
With the double slashes in the url (file:///home...), getcontent() works nicely.
## Comment 21924
Date: 2016-03-16 18:49:22 +0100
From: @njnes
added the additional //.
## Comment 21970
Date: 2016-03-25 09:59:37 +0100
From: @sjoerdmullender
Jul2015-SP3 has been released.
| SQL URL functionality contains errors | https://api.github.com/repos/MonetDB/MonetDB/issues/2972/comments | 0 | 2020-11-30T11:50:05Z | 2024-06-27T11:54:38Z | https://github.com/MonetDB/MonetDB/issues/2972 | 753,411,325 | 2,972 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-13 13:35:32 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes, sellam
Last updated: 2012-11-27 16:37:39 +0100
## Comment 16769
Date: 2012-01-13 13:35:32 +0100
From: @skinkie
sql>\d sys.kv6_current
CREATE TABLE "sys"."kv6_current" (
"messagetype" VARCHAR(10) NOT NULL,
"dataownercode" VARCHAR(10) NOT NULL,
"lineplanningnumber" VARCHAR(10) NOT NULL,
"operatingday" DATE NOT NULL,
"journeynumber" DECIMAL(6) NOT NULL,
"reinforcementnumber" DECIMAL(2) NOT NULL,
"userstopcode" VARCHAR(10),
"passagesequencenumber" DECIMAL(4),
"timestamp" TIMESTAMP NOT NULL,
"source" VARCHAR(10) NOT NULL,
"vehiclenumber" DECIMAL(6),
"punctuality" DECIMAL(4),
"blockcode" DECIMAL(8),
"wheelchairaccessible" BOOLEAN,
"numberofcoaches" DECIMAL(2),
"distancesincelastuserstop" DECIMAL(5),
"rd_x" DECIMAL(6),
"rd_y" DECIMAL(6),
"terminated" BOOLEAN
);
sql>select dataownercode, lineplanningnumber, operatingday, journeynumber, reinforcementnumber from kv6_current group by dataownercode, lineplanningnumber, operatingday, journeynumber, reinforcementnumber having count(*) > 1;
+---------------+--------------------+--------------+---------------+---------------------+
| dataownercode | lineplanningnumber | operatingday | journeynumber | reinforcementnumber |
+===============+====================+==============+===============+=====================+
+---------------+--------------------+--------------+---------------+---------------------+
0 tuples (362.126ms)
sql>alter table kv6_current add constraint unique(dataownercode, lineplanningnumber, operatingday, journeynumber, reinforcementnumber);
42000!syntax error, unexpected UNIQUE in: "alter table kv6_current add constraint unique"
sql>alter table kv6_current add constraint semi_key unique(dataownercode, lineplanningnumber, operatingday, journeynumber, reinforcementnumber);
MALException:mkey.bulk_rotate_xor_hash:operation failed Operands not synced on head
MonetDB 5 server v11.7.0 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 3.9GiB available memory, 1 available cpu core
Libraries:
libpcre: 8.20 2011-10-21 (compiled with 8.20)
openssl: OpenSSL 1.0.0e 6 Sep 2011 (compiled with OpenSSL 1.0.0e 6 Sep 2011)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: root@localhost (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 16770
Date: 2012-01-13 13:45:59 +0100
From: @skinkie
A restart of MonetDB did not resolve this.
## Comment 17015
Date: 2012-02-27 22:43:39 +0100
From: @njnes
we need data to test this, with out the data it runs now.
## Comment 17016
Date: 2012-02-27 23:09:42 +0100
From: @skinkie
Having the same situation at the moment.
| MALException:mkey.bulk_rotate_xor_hash:operation failed Operands not synced on head (on altertable add unique) | https://api.github.com/repos/MonetDB/MonetDB/issues/2971/comments | 0 | 2020-11-30T11:50:02Z | 2024-06-27T11:54:37Z | https://github.com/MonetDB/MonetDB/issues/2971 | 753,411,286 | 2,971 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-12 22:02:01 +0100
From: @grobian
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @yzchang
Last updated: 2012-11-27 14:17:23 +0100
## Comment 16768
Date: 2012-01-12 22:02:01 +0100
From: @grobian
Consider the following snippet. `a' contains the cross-product of l and r, while I expected all tuples that do not match. The example is modeled after the wikipedia explanation of the antijoin: http://en.wikipedia.org/wiki/Relational_algebraAntijoin_.28.E2.96.B7.29
l := bat.new(:oid, :int);
r := bat.new(:oid, :int);
l := bat.insert(l, 0@0, 1);
l := bat.insert(l, 1@0, 2);
l := bat.insert(l, 2@0, 1);
l := bat.insert(l, 3@0, 3);
r := bat.insert(r, 4@0, 2);
r := bat.insert(r, 5@0, 3);
rr := bat.reverse(r);
a := algebra.antijoin(l, rr);
io.print(a);
## Comment 16778
Date: 2012-01-16 14:36:25 +0100
From: @grobian
this is likely intended behaviour
## Comment 18076
Date: 2012-11-27 14:17:23 +0100
From: @yzchang
No test for invalid bug
| antijoin returns join results | https://api.github.com/repos/MonetDB/MonetDB/issues/2970/comments | 0 | 2020-11-30T11:49:59Z | 2024-06-28T13:18:24Z | https://github.com/MonetDB/MonetDB/issues/2970 | 753,411,250 | 2,970 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-10 13:00:39 +0100
From: sellam
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @drstmane
Last updated: 2012-02-29 10:34:27 +0100
## Comment 16762
Date: 2012-01-10 13:00:39 +0100
From: sellam
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.63 Safari/535.7
Build Identifier: [9e3ad41dfa21](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9e3ad41dfa21) (Aug2011)
Consider the following function:
create function bug (x decimal(7,5), y decimal(8,5)) returns decimal(15,0) begin
return power(10,7);
end;
We have the following:
select bug(12,12);
+------------------+
| bug_single_value |
+==================+
| 10000000 |
+------------------+
That's fine. However:
sql>select bug(12.1111,12.1111);
+-------------------+
| bug_single_value |
+===================+
| 1000.0000 |
+-------------------+
Reproducible: Always
hg id
[9e3ad41dfa21](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=9e3ad41dfa21) (Aug2011)
## Comment 16763
Date: 2012-01-10 13:36:20 +0100
From: @sjoerdmullender
Changeset [475e900c55dc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=475e900c55dc) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=475e900c55dc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=475e900c55dc)
Changeset description:
Added test for bug #2969.
## Comment 16943
Date: 2012-02-19 18:21:00 +0100
From: @drstmane
Appears to be fixed in the default branch (only) after changesets
http://dev.monetdb.org/hg/MonetDB/rev/8ee024048e42
http://dev.monetdb.org/hg/MonetDB/rev/e2d3ecc7237c
http://dev.monetdb.org/hg/MonetDB/rev/03df27701cf7
http://dev.monetdb.org/hg/MonetDB/rev/70dc90e1de2f
Remains open/failing in the Dec2011 branch.
## Comment 16946
Date: 2012-02-19 18:47:01 +0100
From: @drstmane
Fixed (also) in Dec2011 by transplanting changeset [70dc90e1de2f](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=70dc90e1de2f) from default to Dec2011:
http://dev.monetdb.org/hg/MonetDB/rev/ec5692ecaf8c
| Result precision issue with Decimal data type | https://api.github.com/repos/MonetDB/MonetDB/issues/2969/comments | 0 | 2020-11-30T11:49:56Z | 2024-06-27T11:54:34Z | https://github.com/MonetDB/MonetDB/issues/2969 | 753,411,213 | 2,969 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-09 07:09:19 +0100
From: james.laign
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @mlkersten, @njnes
Last updated: 2012-03-16 14:56:52 +0100
## Comment 16758
Date: 2012-01-09 07:09:19 +0100
From: james.laign
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier: 11.5.9
mserver5 slowly consumes memory when processing queries.
Reproducible: Always
### Steps to Reproduce:
1. Send lots of queries
### Actual Results:
Memory usage grows approx 20MB per 2000 queries.
I compiled MonetDB from the 11.5.9 sources, using:
./configure --enable-debug=yes --enable-optimize=no
--enable-testing=yes --enable-developer=yes --enable-valgrind=yes
`uname -a` outputs:
Linux 3.0-ARCH 1 SMP PREEMPT Wed Aug 17 21:55:57 CEST 2011 x86_64
Intel(R) Core(TM)2 CPU 6700 @ 2.66GHz GenuineIntel GNU/Linux
## Comment 16759
Date: 2012-01-09 07:11:17 +0100
From: james.laign
Created attachment 95
valgrind log
How i produced the log file:
1. starting mserver5 with the command:
valgrind --log-file=valgrind_mserver.log --show-possibly-lost=no --leak-check=full --tool=memcheck /usr/local/bin/mserver5 --dbname=test_db --dbfarm=/mnt/160a/ --dbinit="include sql;" --set mapi_open=true --daemon=yes
2. running the attached python script, it does the following:
a) Creates a new table 'test_tbl' with 4 int columns.
b) Generates 1000 rows, and loads them with COPY INTO...
c) sends 2000 simple SELECT queries with mclient.
3) stopping valgrind.
> Attached file: [valgrind_mserver.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2968_valgrind_mserver.log_95) (application/octet-stream, 13906 bytes)
> Description: valgrind log
## Comment 16760
Date: 2012-01-09 07:12:52 +0100
From: james.laign
Created attachment 96
python script used to generate data and send queries.
> Attached file: [test_monetdb.py](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2968_test_monetdb.py_96) (text/x-python, 992 bytes)
> Description: python script used to generate data and send queries.
## Comment 16761
Date: 2012-01-09 08:06:52 +0100
From: @mlkersten
Dear James,
Thank you for the report and test script. We will look into it a.s.a.p.
## Comment 17011
Date: 2012-02-27 19:34:25 +0100
From: @njnes
leaks have been fixed
## Comment 17095
Date: 2012-03-16 14:56:52 +0100
From: @grobian
Released in Dec2011-SP2
| Memory Leaks | https://api.github.com/repos/MonetDB/MonetDB/issues/2968/comments | 0 | 2020-11-30T11:49:54Z | 2024-06-27T11:54:34Z | https://github.com/MonetDB/MonetDB/issues/2968 | 753,411,189 | 2,968 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-07 00:43:06 +0100
From: David Pennell <<dpennell>>
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2012-01-12 18:15:38 +0100
## Comment 16754
Date: 2012-01-07 00:43:06 +0100
From: David Pennell <<dpennell>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.63 Safari/535.7
Build Identifier:
The documentation at http://www.monetdb.org/Documentation/Manuals/SQLreference/CopyInto leaves out the int_val for the offset value. Furthermore, its unclear which int_val applies to OFFSET and which one applies to RECORDS (one is prefix, the other postfix)
An example would be handy. Let me know if you would like me to draft something.
Reproducible: Always
## Comment 16755
Date: 2012-01-07 10:36:30 +0100
From: @sjoerdmullender
There is more wrong with that page:
It says the default separator is '\t' (TAB), but in reality it is '|' (vertical bar).
The offset/count syntax is:
[ int_val [ OFFSET int_val ] RECORDS ]
where the first int_val is number of records and the second is the offset.
I've fixed both issues, but if you want to suggest better wording, then please feel free.
## Comment 16756
Date: 2012-01-07 19:59:06 +0100
From: David Pennell <<dpennell>>
Good catch on the default separator, I stumbled on that earlier. I would suggest a clarification off OFFSET vs "skip". The two concepts are somewhat conflated on the page. It took me a while to figure out that if I wanted to skip one header record, I needed to use OFFSET 2.
## Comment 16764
Date: 2012-01-11 16:05:28 +0100
From: @sjoerdmullender
I added a line to the paragraph describing offset: "The offset value specifies the record at which loading should commence, the first record having offset 1."
Does that make things clear enough?
## Comment 16767
Date: 2012-01-12 17:25:38 +0100
From: David Pennell <<dpennell>>
It looks good to me. Thanks!
| copy into documentation missing OFFSET count. | https://api.github.com/repos/MonetDB/MonetDB/issues/2967/comments | 0 | 2020-11-30T11:49:51Z | 2024-06-27T11:54:33Z | https://github.com/MonetDB/MonetDB/issues/2967 | 753,411,159 | 2,967 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-06 15:31:22 +0100
From: @grobian
To: Merovingian devs <<bugs-merovingian>>
Version: -- development
CC: @sjoerdmullender
Last updated: 2012-01-06 15:31:55 +0100
## Comment 16753
Date: 2012-01-06 15:31:22 +0100
From: @grobian
bdksksdlkfsd
| testbug | https://api.github.com/repos/MonetDB/MonetDB/issues/2966/comments | 0 | 2020-11-30T11:49:49Z | 2024-06-28T13:18:23Z | https://github.com/MonetDB/MonetDB/issues/2966 | 753,411,130 | 2,966 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-04 20:32:44 +0100
From: Babis <<charnik>>
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @mlkersten, @yzchang
Last updated: 2012-11-27 14:19:26 +0100
## Comment 16744
Date: 2012-01-04 20:32:44 +0100
From: Babis <<charnik>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/534.52.7 (KHTML, like Gecko) Version/5.1.2 Safari/534.52.7
Build Identifier:
Executing a naive algebra.joinPath on three BATs does not terminate. CPU usage is at 100%.
Reproducible: Always
### Steps to Reproduce:
1. Run mserver5
2. Paste the following series of MAL commands
a:bat[:oid, :int] := bat.new(:oid, :int);
b:bat[:oid, :int] := bat.new(:oid, :int);
c:bat[:oid, :int] := bat.new(:oid, :int);
bat.append(a, 1);
bat.append(a, 2);
bat.append(a, 3);
bat.append(b, 1);
bat.append(b, 2);
bat.append(b, 0);
bat.append(c, 0);
bat.append(c, 3);
bat.append(c, 1);
algebra.joinPath(a, b, c);
Run on Ubuntu Linux and Mac OS X both using a 64bit architecture CPU.
## Comment 16745
Date: 2012-01-04 21:58:27 +0100
From: @mlkersten
The joinpath optimizer should have complained about a typing error.
## Comment 16746
Date: 2012-01-04 22:30:55 +0100
From: @mlkersten
Changeset [fa4034576f02](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fa4034576f02) made by Martin Kersten <mk@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa4034576f02](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa4034576f02)
Changeset description:
Fix for bug #2965
Typechecking should be included as extra defense line.
Furthermore, we need to remove reference counts during
error handling.
## Comment 16747
Date: 2012-01-04 22:31:42 +0100
From: @mlkersten
Code has been extended with type checking in ALGjoinPathZZ
## Comment 16748
Date: 2012-01-05 13:14:35 +0100
From: @mlkersten
Changeset [3216c175f19c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3216c175f19c) made by Martin Kersten <mk@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3216c175f19c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3216c175f19c)
Changeset description:
Fix for bug #2965
Typechecking should be included as extra defense line.
Furthermore, we need to remove reference counts during
error handling.
(transplanted from fa4034576f02bf1a1aa3a755852b021d8da1540b)
## Comment 16749
Date: 2012-01-05 16:25:48 +0100
From: @sjoerdmullender
Changeset [7a5d896c4c2a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7a5d896c4c2a) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7a5d896c4c2a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7a5d896c4c2a)
Changeset description:
Added test for bug #2965.
## Comment 16757
Date: 2012-01-08 13:09:31 +0100
From: @drstmane
Changeset [3ee6c9d194f6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3ee6c9d194f6) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3ee6c9d194f6](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3ee6c9d194f6)
Changeset description:
ALGjoinPath: refined fix for bug #2965 (type check)
While algebra.join() allows only joins on equal types,
and opt_joinpath only recognizes joins on equal types,
algebra.joinpath() did allow joins on equal storage types.
With this checkin, also algebra.joinpath() allows only
joins on equal types.
This should make test monetdb5/tests/BugTracker/Tests/joinpath-loop.Bug-2965
work as intended also with 32-bit OIDs.
The test tries to trigger a type mismatch exception with
algebra.joinpath() by tyring to join int with oid.
This attempt failed, i.e., the exception was not triggered,
with 32-bit OIDs, as 32-bit OIDs have the same storage type
as (32-bit) int.
Cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42715:298eda2b30da&module=monetdb5
http://monetdb.cwi.nl/testweb/tests/42715:298eda2b30da/GNU-Darwin-i386-propcheck/monetdb5/mTests/tests/BugTracker/joinpath-loop.Bug-2965.out.diff.html
## Comment 16815
Date: 2012-01-26 15:31:56 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
## Comment 18077
Date: 2012-11-27 14:19:26 +0100
From: @yzchang
Test already added in monetdb5/tests/BugTracker/Tests/joinpath-loop.Bug-2965
| joinPath runs indeterminately | https://api.github.com/repos/MonetDB/MonetDB/issues/2965/comments | 0 | 2020-11-30T11:49:46Z | 2024-06-27T11:54:31Z | https://github.com/MonetDB/MonetDB/issues/2965 | 753,411,097 | 2,965 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-04 14:11:32 +0100
From: Alfred Nordman <<alfred.nordman>>
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @drstmane
Last updated: 2012-02-29 10:34:26 +0100
## Comment 16735
Date: 2012-01-04 14:11:32 +0100
From: Alfred Nordman <<alfred.nordman>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0.1) Gecko/20100101 Firefox/8.0.1
Build Identifier:
When a LEFT OUTER JOIN query is run, if there are constants in the condition the query returns errors.
Reproducible: Always
### Steps to Reproduce:
1.SELECT * FROM (SELECT 1 as c) a LEFT OUTER JOIN (SELECT 1 as c) b ON 1 = 1;
2.
3.
### Actual Results:
TypeException:user.s7_1[17]:'algebra.markT' undefined in: _25:any := algebra.markT(_24:bit, _21:oid)
TypeException:user.s7_1[18]:'algebra.kdifference' undefined in: _26:any := algebra.kdifference(_18:bat[:oid,:bte], _25:any)
TypeException:user.s7_1[19]:'algebra.markT' undefined in: _27:any := algebra.markT(_26:any, _19:oid)
TypeException:user.s7_1[20]:'bat.reverse' undefined in: _28:any := bat.reverse(_27:any)
TypeException:user.s7_1[21]:'algebra.join' undefined in: _29:any := algebra.join(_28:any, _18:bat[:oid,:bte])
TypeException:user.s7_1[22]:'bat.reverse' undefined in: _30:any := bat.reverse(_25:any)
TypeException:user.s7_1[23]:'algebra.leftjoin' undefined in: _31:any := algebra.leftjoin(_30:any, _18:bat[:oid,:bte])
TypeException:user.s7_1[25]:'bat.append' undefined in: _35:any := bat.append(_32:bat[:oid,:bte], _31:any, _5:bit)
TypeException:user.s7_1[26]:'bat.append' undefined in: _36:any := bat.append(_35:any, _29:any, _5:bit)
TypeException:user.s7_1[32]:'algebra.project' undefined in: _44:any := algebra.project(_28:any, _43:bte)
TypeException:user.s7_1[40]:'bat.reverse' undefined in: _52:any := bat.reverse(_24:bit)
TypeException:user.s7_1[41]:'algebra.markT' undefined in: _53:any := algebra.markT(_52:any, _51:oid)
TypeException:user.s7_1[42]:'bat.reverse' undefined in: _54:any := bat.reverse(_53:any)
TypeException:user.s7_1[43]:'algebra.leftjoin' undefined in: _55:any := algebra.leftjoin(_54:any, _50:bat[:oid,:bte])
TypeException:user.s7_1[45]:'bat.append' undefined in: _57:any := bat.append(_56:bat[:oid,:bte], _55:any, _58:bit)
TypeException:user.s7_1[46]:'bat.append' undefined in: _59:any := bat.append(_57:any, _44:any, _58:bit)
SQLException:SQLengine:Program contains errors
### Expected Results:
+------+------+
| c | c |
+======+======+
| 1 | 1 |
+------+------+
MonetDB 5 server v11.5.9 "Aug2011-SP3"
Serving database 'demo', using 8 threads
Compiled for x86_64-pc-winnt/64bit with 64bit OIDs dynamically linked
Found 15.990 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
MonetDB/SQL module loaded
## Comment 16979
Date: 2012-02-23 14:44:25 +0100
From: @njnes
Changeset [86d94f21e92c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=86d94f21e92c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=86d94f21e92c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=86d94f21e92c)
Changeset description:
fixed bugs 3034, 2964 2946 and 3017
The rel2bin_join didn't handle a bunch of corner cases
| LEFT OUTER JOIN with constants returns errors | https://api.github.com/repos/MonetDB/MonetDB/issues/2964/comments | 0 | 2020-11-30T11:49:43Z | 2024-06-27T11:54:30Z | https://github.com/MonetDB/MonetDB/issues/2964 | 753,411,054 | 2,964 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-04 09:27:23 +0100
From: @skinkie
To: clients devs <<bugs-clients>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @grobian
Last updated: 2012-01-26 15:32:02 +0100
## Comment 16719
Date: 2012-01-04 09:27:23 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux ppc; rv:5.0) Gecko/20100101 Firefox/5.0
Build Identifier:
note: to disable dropping columns and/or truncating fields use \w-1
sql>select * from kv17;
+-------------+-------+------+------------+---------+------+-------------+------+------+-------------+------+------+-------------+----------+-------+-------+----------+----------+-------------+---------+-------------+-------------+
| messagetype | datao | line | operatingd | journey | rein | timestamp | reas | subr | reasonconte | advi | suba | adviceconte | userstop | passa | lagti | targetar | targetde | journeystop | destina | destination | destination |>
: : wnerc : plan : ay : number : forc : : onty : easo : nt : cety : dvic : nt : code : geseq : me : rivaltim : parturet : type : tioncod : name50 : name16 :>
: : ode : ning : : : emen : : pe : ntyp : : pe : etyp : : : uence : : e : ime : : e : : :>
: : : numb : : : tnum : : : e : : : e : : : numbe : : : : : : : :>
: : : er : : : ber : : : : : : : : : r : : : : : : : :>
+=============+=======+======+============+=========+======+=============+======+======+=============+======+======+=============+==========+=======+=======+==========+==========+=============+=========+=============+=============+
| CANCEL | ARR | N198 | 2007-10-31 | 1021 | 0 | 2007-10-31 | 1 | 19_1 | null | 1 | 2 | null | null | null | null | 2 | 2 | null | null | null | null |
: : : : : : : 10:44:09.00 : : : : : : : : : : : : : : : :
: : : : : : : 0000 : : : : : : : : : : : : : : : :
invalid tuple received from server, got 17 columns, expected 24, ignoring
invalid tuple received from server, got 17 columns, expected 24, ignoring
+-------------+-------+------+------------+---------+------+-------------+------+------+-------------+------+------+-------------+----------+-------+-------+----------+----------+-------------+---------+-------------+-------------+
95 tuples (16.519ms) !2 columns dropped!
sql>\f raw
sql>select * from kv17;
% sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17, sys.kv17 table_name
% messagetype, dataownercode, lineplanningnumber, operatingday, journeynumber, reinforcementnumber, timestamp, reasontype, subreasontype, reasoncontent, advicetype, subadvicetype, advicecontent, userstopcode, passagesequencenumber, lagtime, targetarrivaltime, targetdeparturetime, journeystoptype, destinationcode, destinationname50, destinationname16, destinationdetail16, destinationdisplay16 name
% varchar, varchar, varchar, date, decimal, decimal, timestamp, decimal, varchar, varchar, decimal, varchar, varchar, varchar, decimal, decimal, time, time, varchar, varchar, varchar, varchar, varchar, varchar type
% 17, 5, 4, 10, 7, 3, 26, 4, 4, 68, 4, 1, 41, 8, 5, 5, 8, 8, 12, 7, 20, 12, 15, 15 length
[ "CANCEL", "ARR", "N198", 2007-10-31, 1021, 0, 2007-10-31 10:44:09.000000, 1, "19_1", NULL, 1, "2", NULL, NULL, NULL, NULL, "2", "2", NULL, NULL, NULL, NULL, NULL, NULL ]
[ "SHORTEN", "ARR", "N198", 2007-10-31, 1021, 0, 2007-10-31 10:44:09.000000, NULL, NULL, NULL, NULL, NULL, NULL, "57330090", 1, NULL, 1
[ "SHORTEN", "ARR", "N198", 2007-10-31, 1021, 0, 2007-10-31 10:44:09.000000, NULL, NULL, NULL, NULL, NULL, NULL, "57330090", 1, NULL, 1
Nothing else.
Reproducible: Always
### Steps to Reproduce:
1. query with large number of columns
2. limiting the colomns works
MonetDB 5 server v11.8.0 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 3.9GiB available memory, 2 available cpu cores
Libraries:
libpcre: 8.20 2011-10-21 (compiled with 8.20)
openssl: OpenSSL 1.0.0e 6 Sep 2011 (compiled with OpenSSL 1.0.0e 6 Sep 2011)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: skinkie@openkvk (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 16720
Date: 2012-01-04 09:47:26 +0100
From: @sjoerdmullender
What is the contents of columns 17 and 18 (targetarrivaltime and targetdeparturetime)? They are both of type TIME according to the header, but I would then expect something like a time value in the output. The first row has "2" and "2" for those values (including quotes). It should be something like 09:42:00 (without quotes).
What do you mean with "limiting the columns"? Do you mean using something like \w80 in mclient or do you mean projecting in the SELECT query?
Can you run the command
mclient -d database -X -s 'select * from kv17' > file
and attach the file?
## Comment 16721
Date: 2012-01-04 10:06:46 +0100
From: @skinkie
(In reply to comment 1)
> What is the contents of columns 17 and 18 (targetarrivaltime and
> targetdeparturetime)? They are both of type TIME according to the header, but
> I would then expect something like a time value in the output. The first row
> has "2" and "2" for those values (including quotes). It should be something
> like 09:42:00 (without quotes).
>
> What do you mean with "limiting the columns"? Do you mean using something like
> \w80 in mclient or do you mean projecting in the SELECT query?
Projecting the query.
But I just selected on the two time fields, and I'm not happy with what I see.
sql>select targetarrivaltime, targetdeparturetime from kv17;
+-------------------+---------------------+
| targetarrivaltime | targetdeparturetime |
+===================+=====================+
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| ½½½ | ½½½ |
| 10:00:00 | 10:00:00 |
| 10: | 10: |
| 10: | 10: |
| 10: | 10: |
| 19:28:00 | 19:30:00 |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 19: | 19: |
| 08:17:00 | 08:20:00 |
| 08: | 08: |
| 08: | 08: |
| 08: | 08: |
| 10:00:00 | 10:00:00 |
.....
## Comment 16722
Date: 2012-01-04 10:07:21 +0100
From: @skinkie
Created attachment 93
Output from mclient
> Attached file: [naarsjoerd.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2963_naarsjoerd.txt_93) (text/plain, 11686 bytes)
> Description: Output from mclient
## Comment 16723
Date: 2012-01-04 10:51:40 +0100
From: @grobian
The tuples actually even lack a closing ], so it seems like the reconstruction terminates for the rest of the line.
## Comment 16724
Date: 2012-01-04 10:53:58 +0100
From: @grobian
I would be interested to see what jdbcclient (sorry Java) thinks of this query.
## Comment 16725
Date: 2012-01-04 12:29:58 +0100
From: @grobian
Ok, there's control chars in the output:
[ "SHORTEN", "ARR", "N198", 2007-10-31, 1021, 0, 2007-10-31 10:44:09. 000000, NULL, NULL, NULL, NULL, NULL, NULL, "57330090", 1, NULL,
1^@7, 1^@7, NULL, NULL, NULL, NULL, NULL, NULL ]
(Vim makes it ^@), so the tuples are complete, but the clients get confused since they treat the null-byte as end of the line.
## Comment 16728
Date: 2012-01-04 12:54:03 +0100
From: @sjoerdmullender
Can you give the result of the queries
select cast(targetarrivaltime as interval hour to second) from kv17;
select cast(cast(targetarrivaltime as interval hour to second) as int) from kv17;
## Comment 16729
Date: 2012-01-04 13:00:02 +0100
From: @skinkie
(In reply to comment 7)
> Can you give the result of the queries
> select cast(targetarrivaltime as interval hour to second) from kv17;
> select cast(cast(targetarrivaltime as interval hour to second) as int) from
> kv17;
sql>select cast(targetarrivaltime as interval hour to second) from kv17;
+-------------------------------------------+
| targetarrivaltime |
+===========================================+
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| 36000.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
| 68400.000 |
| -2145600.000 |
| -2145600.000 |
| -2145600.000 |
....
Observe the width of the column above, it is full screen width, don't know if that is cast related.
+-------------------+
| targetarrivaltime |
+===================+
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| 36000000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| 68400000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
| -2145600000 |
## Comment 16730
Date: 2012-01-04 13:10:52 +0100
From: @sjoerdmullender
The problem has to do with negative times.
First of all, a negative time doesn't make sense (the time type is supposed to represent a time-of-day which goes from 00:00:00 to 23:59:59). Negative intervals do make sense, but negative times don't.
Because of that, the server code deals badly with negative times, so that is what's causing the bad behavior.
Of course, the server should do something better: either refuse to allow negative values in a time column, or somehow convert the values to something reasonable on output.
## Comment 16731
Date: 2012-01-04 13:23:42 +0100
From: @skinkie
(In reply to comment 9)
> The problem has to do with negative times.
The "funny" thing is... I am not inserting negative times as far as I can see.
<tmi8:targetarrivaltime>19:28:00</tmi8:targetarrivaltime>
<tmi8:targetarrivaltime>08:17:00</tmi8:targetarrivaltime>
<tmi8:targetarrivaltime>10:00:00</tmi8:targetarrivaltime>
<tmi8:targetarrivaltime>11:00:00</tmi8:targetarrivaltime>
<tmi8:targetarrivaltime>12:00:00</tmi8:targetarrivaltime>
<tmi8:targetdeparturetime>19:30:00</tmi8:targetdeparturetime>
<tmi8:targetdeparturetime>08:20:00</tmi8:targetdeparturetime>
<tmi8:targetdeparturetime>10:00:00</tmi8:targetdeparturetime>
<tmi8:targetdeparturetime>11:01:00</tmi8:targetdeparturetime>
<tmi8:targetdeparturetime>12:00:00</tmi8:targetdeparturetime>
Attached will give an indication what goes in...
## Comment 16732
Date: 2012-01-04 13:25:11 +0100
From: @skinkie
Created attachment 94
Indication what goes in in 'pseudo' SQL
Attached contains the SQL statements that are used to insert the data. Sadly I have no clue how I can get 'marked up queries' so the field values are not 'escaped', but as you can see those are normal times.
> Attached file: [naarsjoerd-insert.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2963_naarsjoerd-insert.txt_94) (text/plain, 8582 bytes)
> Description: Indication what goes in in 'pseudo' SQL
## Comment 16733
Date: 2012-01-04 13:39:33 +0100
From: @sjoerdmullender
I see that not all rows get values for the two times. Do you have a default value for those columns? I'm guessing you don't, and that therefore the values that are actually inserted are cast(null as time).
If that is the case, there is something wrong with null handling.
## Comment 16734
Date: 2012-01-04 14:03:22 +0100
From: @skinkie
(In reply to comment 12)
> I see that not all rows get values for the two times. Do you have a default
> value for those columns? I'm guessing you don't, and that therefore the values
> that are actually inserted are cast(null as time).
> If that is the case, there is something wrong with null handling.
No default values are in there. NULLs are allowed.
## Comment 16737
Date: 2012-01-04 14:59:58 +0100
From: @sjoerdmullender
Changeset [29e91cf9f4f1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=29e91cf9f4f1) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=29e91cf9f4f1](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=29e91cf9f4f1)
Changeset description:
Copy result to correct buffer.
This partially fixes bug #2963 in that now nil time values are now
actually reported as nil.
## Comment 16738
Date: 2012-01-04 15:01:51 +0100
From: @sjoerdmullender
I just committed a fix which makes that the query runs.
However, there is still a problem:
When a null value is inserted into a time column, the internal representation of the null value is incorrect. This is masked by the (now fixed) function that converts a time value to a string for output.
## Comment 16739
Date: 2012-01-04 15:03:19 +0100
From: @sjoerdmullender
The bug also occurs in Dec2011.
## Comment 16740
Date: 2012-01-04 16:27:29 +0100
From: @sjoerdmullender
Changeset [c0edd6cd8ecf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c0edd6cd8ecf) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c0edd6cd8ecf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c0edd6cd8ecf)
Changeset description:
Before fixing daytime fractions, check whether the value is nil.
This fixes the rest of bug #2963.
## Comment 16741
Date: 2012-01-04 16:29:39 +0100
From: @sjoerdmullender
Changeset [6e6b3637b011](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6e6b3637b011) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6e6b3637b011](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6e6b3637b011)
Changeset description:
Added test for bug #2963.
## Comment 16742
Date: 2012-01-04 16:31:52 +0100
From: @skinkie
Bedankt :)
## Comment 16743
Date: 2012-01-04 16:32:31 +0100
From: @sjoerdmullender
Fixed in Dec2011 branch.
## Comment 16825
Date: 2012-01-26 15:32:02 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| invalid tuple received from server, got 17 columns, expected 24, ignoring | https://api.github.com/repos/MonetDB/MonetDB/issues/2963/comments | 0 | 2020-11-30T11:49:39Z | 2024-06-27T11:54:29Z | https://github.com/MonetDB/MonetDB/issues/2963 | 753,411,017 | 2,963 |
[
"MonetDB",
"MonetDB"
] | Date: 2012-01-01 06:16:16 +0100
From: hsaddington <<hs.addington>>
To: Holger <<holger>>
Version: 11.15.11 (Feb2013-SP3)
CC: bugs-sql, holger.pirk, @njnes
Last updated: 2013-09-27 13:47:20 +0200
## Comment 16709
Date: 2012-01-01 06:16:16 +0100
From: hsaddington <<hs.addington>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.7 (KHTML, like Gecko) Chrome/16.0.912.63 Safari/535.7
Build Identifier:
current_date and current_timestamp are behaving in odd ways. It looks like the current_date ignores the local timezone and gives a date that is exactly 24 hours ahead. I think it is giving the correct local time, but is giving the date based on GMT. The current_timestamp gives correct result. But, if we extract the day (extract(day from current_timestamp), it also give a wrong result. On New Year's Eve (i.e. tonight), it gives wrong year, month, and day!
It is the New Year's Eve. And, the following query gives incorrect results.
"select current_date" results in 2012-01-01.
Reproducible: Always
### Steps to Reproduce:
1.select current_date
2.select current_timestamp, extract(day from current_timestamp)
3.Run the above queries before and after the day switch based on GMT
### Actual Results:
It is the New Year's Eve. And, the following query gives incorrect result.
select current_date ====> 2012-01-01
select current_time ====> 21:13:06
select current_timestamp ====> 2011-12-31 21:13:15
select extract(month from current_timestamp) ======> 1
## Comment 16713
Date: 2012-01-02 15:09:27 +0100
From: @sjoerdmullender
I'm assuming you're located somewhere West of Greenwich, UK. At the time you did your queries, it was already past midnight in Greenwich, and the server gave back the times in GMT (or UTC).
The server uses GMT internally, but clients can tell the server what timezone they are in. mclient does that.
It looks like you used a client that didn't tell the server what timezone it is in, so all the server can tell is what the current time/date is in Greenwich.
So, which client were you using for your test?
If not mclient, you could use something like
SET TIME ZONE INTERVAL '-05:00' HOUR TO MINUTE
to set the time zone for the client (this example: 5 hours West of Greenwich).
When using mclient for these queries, the results (for me) include the timezone information (after I fixed a bug in the DST determination):
sql>select current_timestamp;
+----------------------------------+
| current_timestamp |
+==================================+
| 2012-01-02 15:05:47.000000+01:00 |
+----------------------------------+
1 tuple (0.519ms)
sql>select current_time;
+-----------------------+
| current_time |
+=======================+
| 15:05:50.000000+01:00 |
+-----------------------+
1 tuple (0.533ms)
## Comment 16714
Date: 2012-01-02 22:35:19 +0100
From: hsaddington <<hs.addington>>
Hello:
Yes. I am near Los Anageles, USA.
I was using DBVisualizer and mclient. For DBVisualizer, I was using JDBC driver version 1.17. I thought I got exact same results in both clients except the mclient did not account for the DST so it was off by 1 hour compared to what I was getting in DBVisualizer.
I think DBVisualizer tells the server what timezone I am in because current_time that I get from the server is the correct local time.
Let me try again later in the evening when it is past midnight in Greenwich. I thought I was getting results that did not look right even in mclient.....
## Comment 16715
Date: 2012-01-03 02:17:24 +0100
From: hsaddington <<hs.addington>>
Ok. I ran the same queries in DBVisualizer and mclient.
from DBVisualizer around local time, 2012-01-02 17:05
select current_date, extract(year from current_date) as years, extract(month from current_date) as months, extract(day from current_date) as days;
select current_time;
select current_timestamp, extract(year from current_timestamp) as years, extract(month from current_timestamp) as months, extract(day from current_timestamp) as days;
current_date years months days
2012-01-03 2012 1 3
17:05:13
current_timestamp years months days
2012-01-02 17:05:14 2012 1 3
from mclient around local time, 2012-01-02 17:05
select current_date, extract(year from current_date) as years, extract(month from current_date) as months, extract(day from current_date) as days;
select current_time;
select current_timestamp, extract(year from current_timestamp) as years, extract(month from current_timestamp) as months, extract(day from current_timestamp) as days;
+--------------+-------------------+--------------------+------------------+
| current_date | year_current_date | month_current_date | day_current_date |
+==============+===================+====================+==================+
| 2012-01-03 | 2012 | 1 | 3 |
+--------------+-------------------+--------------------+------------------+
+-----------------------+
| current_time |
+=======================+
| 18:08:21.000000-07:00 |
+-----------------------+
+----------------------------------+---------------+----------------+----------------+
| current_timestamp | current_times | current_timest | current_timest |
: : tamp : amp : amp :
+==================================+===============+================+================+
| 2012-01-02 18:10:04.000000-07:00 | 2012 | 1 | 3 |
+----------------------------------+---------------+----------------+----------------+
As you can see above, the just the times are off by 1 hour due to DST, perhaps. But, both clients return incorrect current_date. Extracting the year, month, or day from the current_date or current_timestamp returns the results from GMT, not from local timezone....
Where does the local timezone translation occur? Is it in the JDBC driver or the client? Thanks.
## Comment 16716
Date: 2012-01-03 08:29:04 +0100
From: @grobian
JDBC sets the timezone you use on the server. Possibly it does this wrong, though.
## Comment 16717
Date: 2012-01-03 10:31:13 +0100
From: @sjoerdmullender
I now understand what goes wrong:
The server works in GMT and needs to convert times to the client's local time zone (as set by e.g. mclient).
When using the query SELECT CURRENT_TIMESTAMP this conversion happens during "export", i.e. when the value is being sent from the server to the client.
With the query SELECT EXTRACT(MONTH FROM CURRENT_TIMESTAMP), the conversion doesn't happen. The month is extracted before any timezone conversion, and then when exported, there is no need to de timezone conversion.
I expect that you get the wrong result also when extracting the day from the timestamp during the time that your date is different from the date in GMT.
This is a problem that is not so easy to fix.
## Comment 16718
Date: 2012-01-04 04:04:58 +0100
From: hsaddington <<hs.addington>>
Yes, you are right. I do get the wrong day when I extract it from the current_timestamp. I think your explanation of where/when the conversion occurs makes sense based on what we are seeing... Too bad, it is not an easy fix, though... That is not what I wanted to hear from the experts! :-)
## Comment 17695
Date: 2012-08-24 14:56:07 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18066
Date: 2012-11-27 14:02:45 +0100
From: @grobian
keep alias in the loop
## Comment 18107
Date: 2012-11-27 15:15:14 +0100
From: Holger <<holger.pirk>>
Still occurs in default and Oct2012. Added a test in revision 33598ae3797e
## Comment 18203
Date: 2012-11-28 13:45:42 +0100
From: Holger <<holger>>
Changeset [ef4c8fa0fcb3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ef4c8fa0fcb3) made by Holger Pirk <holger@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ef4c8fa0fcb3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ef4c8fa0fcb3)
Changeset description:
* added test for bug #2962
## Comment 19072
Date: 2013-08-25 15:49:24 +0200
From: @njnes
functions with timestamp with time zone arguments (not returning timestamps) are now first transformed into the local timezone.
## Comment 19074
Date: 2013-08-25 16:23:37 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [0620e9df2003](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0620e9df2003) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0620e9df2003](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0620e9df2003)
Changeset description:
fixed bug #3232, use dayofweek, starting at Monday
fixed bug #2962, convert timestamp with time zone to local time zone, when used as input to the extract functions.
fixed bug #2781, use correct time zone in mclient
| client timezone information needs to be handled also for modifications to timestamp values | https://api.github.com/repos/MonetDB/MonetDB/issues/2962/comments | 0 | 2020-11-30T11:49:36Z | 2024-06-27T11:54:28Z | https://github.com/MonetDB/MonetDB/issues/2962 | 753,410,979 | 2,962 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-30 22:27:19 +0100
From: adamb
To: Merovingian devs <<bugs-merovingian>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-02 12:47:44 +0100
## Comment 16706
Date: 2011-12-30 22:27:19 +0100
From: adamb
User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X 10.5; en-US; rv:1.9.2.25) Gecko/20111212 Firefox/3.6.25
Build Identifier:
Reproducible: Always
### Steps to Reproduce:
1. monetdbd start
2. monetdb status //everything looks fine
3. monetdb -h localhost -P anything status
4. Fails with: "status: internal error: no response from merovingian"
5. Execute `ps -A | grep monetdbd` and you'll see that monetdbd process is gone
6. log file contains no error
### Actual Results:
monetdbd process crashed without error.
### Expected Results:
No crash
Notes:
* I was able to reproduce this on an Ubuntu box as well as a CentOS box.
* Using all default monetdbd properties.
* Everything was executed as the user 'monetdb'.
* If mserver5 processes were spawned they remain orphaned after the crash.
* Unable to connect to any databases after the crash.
## Comment 16707
Date: 2011-12-31 09:18:11 +0100
From: @grobian
I've never seen this before, but it's really severe. I'll look into it ASAP.
## Comment 16708
Date: 2011-12-31 15:46:03 +0100
From: @grobian
Hmmm, with Dec2011 (upcoming release), I just get
status: internal error: access denied
(not really internal, but it doesn't crash)
## Comment 16711
Date: 2012-01-02 12:43:17 +0100
From: @grobian
Changeset [28e16671d2ae](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=28e16671d2ae) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=28e16671d2ae](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=28e16671d2ae)
Changeset description:
monetdbd: install SIGSEGV handler
For bug #2961, add SIGSEGV handler which attempts to write a message
identifying the crash to the logfile before going down. This is just a
best-effort attempt to make it visible in the log that a crash has
occurred.
## Comment 16712
Date: 2012-01-02 12:47:44 +0100
From: @grobian
I cannot reproduce this on Aug2011-SP3 either. Since this code has changed considerable in the Dec2011 branch, I think there is a good chance this has been fixed in the upcoming Dec2011 release.
Feel free to reopen if you have additional info, or when this also crashes with a Dec2011 snapshot.
| monetdbd crashes when attempting to administer remotely | https://api.github.com/repos/MonetDB/MonetDB/issues/2961/comments | 0 | 2020-11-30T11:49:33Z | 2024-06-28T13:18:22Z | https://github.com/MonetDB/MonetDB/issues/2961 | 753,410,952 | 2,961 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-28 01:31:40 +0100
From: Babis <<charnik>>
To: SQL devs <<bugs-sql>>
Version: 11.9.1 (Apr2012) [obsolete]
CC: @njnes, @drstmane
Last updated: 2012-05-25 12:58:47 +0200
## Comment 16702
Date: 2011-12-28 01:31:40 +0100
From: Babis <<charnik>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/534.52.7 (KHTML, like Gecko) Version/5.1.2 Safari/534.52.7
Build Identifier:
mclient terminates the connection to the mserver when executing some "CREATE TABLE" commands and then a large INSERT command. Sometimes the database crashes (reported as crashed from the monetdb status command).
During execution of SQL commands mclient reports the following:
operation successful
….
MAPI = (monetdb) /tmp/.s.monetdb.50000
ACTION= read_line
QUERY = …
ERROR = !Connection terminated
sql>
Reproducible: Always
### Steps to Reproduce:
1. mclient -d test -lsql
2. sql> \< <attatched_file_1_here>
3.
What's more, I observed a somewhat similar bad behavior with another big INSERT statement, which does not close the connection, but it executes correctly and no records have been stored in the table. To see this, do the above steps with the second attached file.
## Comment 16703
Date: 2011-12-28 01:36:22 +0100
From: Babis <<charnik>>
Created attachment 91
first input
> Attached file: [bug_large-insert1.sql.zip](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2960_bug_large-insert1.sql.zip_91) (application/zip, 341614 bytes)
> Description: first input
## Comment 16704
Date: 2011-12-28 01:37:07 +0100
From: Babis <<charnik>>
Created attachment 92
second input
> Attached file: [bug_large-insert2.sql.zip](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2960_bug_large-insert2.sql.zip_92) (application/zip, 24652 bytes)
> Description: second input
## Comment 16705
Date: 2011-12-28 09:16:58 +0100
From: @grobian
Ok, the 1st statement looks like this:
INSERT INTO foo VALUES
(1,1,255),
[... x100K]
(255,255,255);
IOW one very large statement. Likely the server crashes due to overallocation or something similar. Mapi at the same time could go wrong here.
The second is similar, but has much less values (~8K).
## Comment 16710
Date: 2012-01-02 10:04:51 +0100
From: @sjoerdmullender
The server crashes when running the first input due to extremely deep recursion in the function rel_properties in the file rel_optimizer.c (line 264 in the Dec2011 branch). I'm thinking it recurses into that function for each tuple being inserted, and there are some 130000.
## Comment 16972
Date: 2012-02-22 23:35:29 +0100
From: @drstmane
Most probably, this one won't be fixed soon (if at all ...).
Instead of one large insert statement for "very many" tuples, use one insert statement per tuple (slow), or better (and strongly recommended) a COPY INTO (from file or stdin) statement (fast); cf., http://www.monetdb.org/Documentation/Manuals/SQLreference/CopyInto
E.g., try
cat bug_large-insert1.sql | awk '/^INSERT INTO .+ VALUES *$/ {n=0; s=""; t=$3;} /^ *\(.+\)[,;] *$/ {n++; s=s gensub("^ *\\((.+)\\)[,;] *$","\\1",1,$0) "\n";} /^ *\(.+\); *$/ {printf "COPY %d RECORDS INTO %s FROM stdin USING DELIMITERS '\'','\'','\''\\n'\'','\''\\'\'\'';\n%s",n,t,s;} !/^(INSERT INTO .+ VALUES| *\(.+\)[,;]) *$/ {print $0;}' > bug_large-insert1x.sql
mclient bug_large-insert1x.sql
cat bug_large-insert2.sql | awk '/^INSERT INTO .+ VALUES *$/ {n=0; s=""; t=$3;} /^ *\(.+\)[,;] *$/ {n++; s=s gensub("^ *\\((.+)\\)[,;] *$","\\1",1,$0) "\n";} /^ *\(.+\); *$/ {printf "COPY %d RECORDS INTO %s FROM stdin USING DELIMITERS '\'', '\'','\''\\n'\'','\''\\'\'\'';\n%s",n,t,s;} !/^(INSERT INTO .+ VALUES| *\(.+\)[,;]) *$/ {print $0;}' > bug_large-insert2x.sql
mclient bug_large-insert2x.sql
(If necessary, set the record delimiters according to your original insert statements syntax.)
## Comment 17241
Date: 2012-05-09 16:21:41 +0200
From: @njnes
fixed the recursion problem. Instead of a recursive union we now use a list of values.
## Comment 17243
Date: 2012-05-09 16:56:55 +0200
From: @njnes
Changeset [5a8a62a9f0b5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5a8a62a9f0b5) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5a8a62a9f0b5](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5a8a62a9f0b5)
Changeset description:
fixed problem with deep recursion, when loading/inserting lots of values
using values (),() etc. This fixes bug #2960
| mclient terminates connection during insertion of a large number of records from a file | https://api.github.com/repos/MonetDB/MonetDB/issues/2960/comments | 0 | 2020-11-30T11:49:30Z | 2024-06-27T11:54:25Z | https://github.com/MonetDB/MonetDB/issues/2960 | 753,410,916 | 2,960 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-27 20:32:20 +0100
From: Alfred Nordman <<alfred.nordman>>
To: SQL devs <<bugs-sql>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
CC: @njnes
Last updated: 2012-01-26 15:32:01 +0100
## Comment 16701
Date: 2011-12-27 20:32:20 +0100
From: Alfred Nordman <<alfred.nordman>>
The following query crashes the server:
SELECT 123 AS dummy1 FROM tables n WHERE EXISTS ( SELECT 456 AS dummy2 FROM (SELECT * FROM columns) nnn WHERE n.id > 0 );
## Comment 16726
Date: 2012-01-04 12:42:19 +0100
From: @njnes
fixed in Dec2011 (test added to exists-select.Bug-2933.sql)
## Comment 16727
Date: 2012-01-04 12:51:41 +0100
From: @njnes
Changeset [f6bfa6df48ad](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f6bfa6df48ad) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f6bfa6df48ad](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f6bfa6df48ad)
Changeset description:
added test for Bug #2959
## Comment 16824
Date: 2012-01-26 15:32:01 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| Nested SELECT with EXISTS causes the server to crash | https://api.github.com/repos/MonetDB/MonetDB/issues/2959/comments | 0 | 2020-11-30T11:49:27Z | 2024-06-27T11:54:24Z | https://github.com/MonetDB/MonetDB/issues/2959 | 753,410,893 | 2,959 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-27 03:30:38 +0100
From: @skinkie
To: SQL devs <<bugs-sql>>
Version: 11.9.1 (Apr2012) [obsolete]
CC: @njnes
Last updated: 2012-05-25 12:58:43 +0200
## Comment 16689
Date: 2011-12-27 03:30:38 +0100
From: @skinkie
sql>grant select on sys.faillissementen to dbslayer;
42000!syntax error, unexpected '.', expecting TO in: "grant select on sys."
sql>grant select on faillissementen to dbslayer;
operation successful (2.880ms)
## Comment 16693
Date: 2011-12-27 09:46:04 +0100
From: @grobian
On a side-note: what client is this, the 42000! shouldn't be in the output.
## Comment 16695
Date: 2011-12-27 10:13:02 +0100
From: @skinkie
December 2011 branch mclient.
## Comment 17235
Date: 2012-05-07 20:37:40 +0200
From: @njnes
has been fixed some time back, by adding the proper qname rule to the parser.
## Comment 17236
Date: 2012-05-07 20:38:29 +0200
From: @njnes
Changeset [44068741341a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=44068741341a) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=44068741341a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=44068741341a)
Changeset description:
added test for allready fixed bug #2958
| Grant does't work with schema syntax | https://api.github.com/repos/MonetDB/MonetDB/issues/2958/comments | 0 | 2020-11-30T11:49:25Z | 2024-06-27T11:54:23Z | https://github.com/MonetDB/MonetDB/issues/2958 | 753,410,860 | 2,958 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-27 02:23:56 +0100
From: @skinkie
To: GDK devs <<bugs-common>>
Version: -- development
CC: risc
Last updated: 2012-11-07 19:44:09 +0100
## Comment 16688
Date: 2011-12-27 02:23:56 +0100
From: @skinkie
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.83 Safari/535.2
Build Identifier:
After having to terminate MonetDB because of full memory consumption. MonetDB wouldn't start anymore, resulting in a Segmentation Fault. Using GDB it was showing that this error kept another error hidden: !FATAL: GDKlockHome: Database lock '.gdk_lock' denied
I verified no other database was running in this same directory. And I removed the .gdk_lock; this didn't matter.
Using GDB I can find the following;
Program received signal SIGSEGV, Segmentation fault.
0x00007fffec019f40 in temp_create (b=0x0) at bat_utils.c:105
105 temp_dup(b->batCacheid);
(gdb) bt
0 0x00007fffec019f40 in temp_create (b=0x0) at bat_utils.c:105
1 0x00007fffec013d73 in load_dbat (bat=0x1215b10, bid=0) at bat_storage.c:736
2 0x00007fffec013e89 in create_del (tr=0x1006df0, t=0x12159c0) at bat_storage.c:755
3 0x00007fffebffd46e in load_table (tr=0x1006df0, s=0x1006e70, rid=88) at store.c:617
4 0x00007fffebffeed0 in load_schema (tr=0x1006df0, id=2151, rid=88) at store.c:910
5 0x00007fffebfff383 in load_trans (tr=0x1006df0, id=2151) at store.c:963
6 0x00007fffec001d06 in store_init (debug=0, store=store_bat, logdir=0x7fffec037a50 "sql_logs", dbname=0x63b920 "kvk-2011-sinterklaas", stk=0) at store.c:1489
7 0x00007fffebf8160e in mvc_init (dbname=0x63b920 "kvk-2011-sinterklaas", debug=0, store=store_bat, stk=0) at sql_mvc.c:51
8 0x00007fffebee8061 in SQLinit () at sql_scenario.c:237
9 0x00007fffebee7ce3 in SQLprelude () at sql_scenario.c:166
10 0x00007ffff6ae5c63 in runMALsequence (cntxt=0x645090, mb=0x676250, startpc=1, stoppc=0, stk=0x10fe380, env=0x0, pcicaller=0x0) at mal_interpreter.mx:1949
11 0x00007ffff6ad74ed in runMAL (cntxt=0x645090, mb=0x676250, startpc=1, mbcaller=0x0, env=0x0, pcicaller=0x0) at mal_interpreter.mx:336
12 0x00007ffff6ac0c47 in MALengine (c=0x645090) at mal_session.c:631
13 0x00007ffff6abf165 in malBootstrap () at mal_session.c:73
14 0x00007ffff6a9cfea in mal_init () at mal.c:247
15 0x0000000000403648 in main (argc=6, av=0x7fffffffdd18) at mserver5.c:567
At this point I am unable to access my database, while creating new databases in a different folder isn't an issue. I have started to fiddle in some dbfarm stuff to get it going again. This wasn't helping anything. If anyone would like to have the remaining dbfarm dir for investigation I could package it.
Now probably the best thing is to take a backup and start rebuilding the database.
Reproducible: Didn't try
MonetDB 5 server v11.7.0 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 3.9GiB available memory, 2 available cpu cores
Libraries:
libpcre: 8.20 2011-10-21 (compiled with 8.12)
openssl: OpenSSL 1.0.0d 8 Feb 2011 (compiled with OpenSSL 1.0.0d 8 Feb 2011)
libxml2: 2.7.7 (compiled with 2.7.7)
Compiled by: skinkie@openkvk (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wmissing-include-dirs -Wp,-D_FORTIFY_SOURCE=2
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
## Comment 16692
Date: 2011-12-27 09:44:59 +0100
From: @grobian
The lockfile is only valid if a process actually still lives and holds the lock on that file. You removed it, but that feels wrong. Is your dbfarm on a networked filesystem or something?
## Comment 16694
Date: 2011-12-27 10:12:32 +0100
From: @skinkie
(In reply to comment 1)
> The lockfile is only valid if a process actually still lives and holds the lock
> on that file. You removed it, but that feels wrong. Is your dbfarm on a
> networked filesystem or something?
No, just on a local disk. And removing that file doesn't fix the segmentation fault.
## Comment 16696
Date: 2011-12-27 10:19:28 +0100
From: @grobian
(In reply to comment 2)
> (In reply to comment 1)
> > The lockfile is only valid if a process actually still lives and holds the lock
> > on that file. You removed it, but that feels wrong. Is your dbfarm on a
> > networked filesystem or something?
>
> No, just on a local disk. And removing that file doesn't fix the segmentation
> fault.
... that's obvious, since the lockfile doesn't do anything but protect two instances accessing the same data.
## Comment 16736
Date: 2012-01-04 14:33:09 +0100
From: @skinkie
Latest development, normal exit; abnormal startup :(
Program received signal SIGSEGV, Segmentation fault.
0x00007fffeb2ac87c in temp_create (b=0x0) at bat_utils.c:105
105 temp_dup(b->batCacheid);
(gdb) bt
0 0x00007fffeb2ac87c in temp_create (b=0x0) at bat_utils.c:105
1 0x00007fffeb2a65fd in load_dbat (bat=0x12db7c0, bid=0) at bat_storage.c:737
2 0x00007fffeb2a6713 in create_del (tr=0x1212360, t=0x12db910) at bat_storage.c:756
3 0x00007fffeb28fce3 in load_table (tr=0x1212360, s=0x1099540, rid=116) at store.c:617
4 0x00007fffeb291717 in load_schema (tr=0x1212360, id=2151, rid=116) at store.c:910
5 0x00007fffeb291bca in load_trans (tr=0x1212360, id=2151) at store.c:963
6 0x00007fffeb294555 in store_init (debug=0, store=store_bat, logdir=0x7fffeb2ca780 "sql_logs",
dbname=0x668650 "openov_cxx", stk=0) at store.c:1488
7 0x00007fffeb210885 in mvc_init (dbname=0x668650 "openov_cxx", debug=0, store=store_bat, stk=0)
at sql_mvc.c:51
8 0x00007fffeb175585 in SQLinit () at sql_scenario.c:237
9 0x00007fffeb175207 in SQLprelude () at sql_scenario.c:166
10 0x00007ffff6ab6ed8 in runMALsequence (cntxt=0x671dc0, mb=0x6a30a0, startpc=1, stoppc=0,
stk=0x112ab50, env=0x0, pcicaller=0x0) at mal_interpreter.mx:1949
11 0x00007ffff6aa7b43 in runMAL (cntxt=0x671dc0, mb=0x6a30a0, startpc=1, mbcaller=0x0, env=0x0,
pcicaller=0x0) at mal_interpreter.mx:336
12 0x00007ffff6a90ec3 in MALengine (c=0x671dc0) at mal_session.c:631
13 0x00007ffff6a8f3dd in malBootstrap () at mal_session.c:73
14 0x00007ffff6a6cc56 in mal_init () at mal.c:247
15 0x0000000000403648 in main (argc=6, av=0x7fffffffdd58) at mserver5.c:564
## Comment 17753
Date: 2012-09-25 17:36:01 +0200
From: @grobian
feels like a corrupt log
## Comment 17904
Date: 2012-11-07 19:36:30 +0100
From: @grobian
perhaps we should use a virtual machine with limited resources or something to do a crash session, but for now we don't have enough info for what we need here
## Comment 17906
Date: 2012-11-07 19:39:09 +0100
From: @skinkie
In general: would you want to have the 'custom compiled' MonetDB + the dbfarm directory? If we have established these kind of crashes are annoying and should be fixed once they occur, it might be a good idea to have a procedure in place to reproduce the specific segmentation fault.
## Comment 17909
Date: 2012-11-07 19:44:09 +0100
From: @grobian
I think we should just stress test monetdb in scenarios where it runs short of resources, see for example http://bugs.monetdb.org/show_bug.cgi?id=3162c1
| Startup of MonetDB fails with segmentationfault | https://api.github.com/repos/MonetDB/MonetDB/issues/2957/comments | 0 | 2020-11-30T11:49:22Z | 2024-06-27T11:54:22Z | https://github.com/MonetDB/MonetDB/issues/2957 | 753,410,825 | 2,957 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-26 16:26:55 +0100
From: @mlkersten
To: MonetDB5 devs <<bugs-monetdb5>>
Version: -- development
CC: @yzchang
Last updated: 2012-11-27 14:21:23 +0100
## Comment 16686
Date: 2011-12-26 16:26:55 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier:
DFLOWstep under performance profiling will re-initialize the complete block. This is an overkill to be removed at some moment by inlineing DFLOWstep into DFLOWworker and proper factoring out the GDKalloc/GDKfree etc.
Reproducible: Always
## Comment 17905
Date: 2012-11-07 19:36:56 +0100
From: @grobian
status here?
## Comment 17915
Date: 2012-11-07 20:09:15 +0100
From: @mlkersten
It is a reminder for the future. It may improve MAL interpretation overhead with a few microseconds each. In light of other optimization opportunities it is not urgent.
Furthermore, the code between MALinterpreter and DFLOWinterpreter has been synchronised.
## Comment 18073
Date: 2012-11-27 14:13:43 +0100
From: @mlkersten
Code synchronization is enough.
## Comment 18078
Date: 2012-11-27 14:21:23 +0100
From: @yzchang
No test (possible) for performance issues
| DFLOW too expensive | https://api.github.com/repos/MonetDB/MonetDB/issues/2956/comments | 0 | 2020-11-30T11:49:19Z | 2024-06-28T13:40:48Z | https://github.com/MonetDB/MonetDB/issues/2956 | 753,410,800 | 2,956 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-26 14:37:33 +0100
From: @mlkersten
To: clients devs <<bugs-clients>>
Version: -- development
Last updated: 2012-11-12 22:15:14 +0100
## Comment 16685
Date: 2011-12-26 14:37:33 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier:
- stethoscope errors are confusion, in general one can not access
the log in this case. perhaps dropping them and limit to successful
connections.
mk@vienna::~/current> stethoscope -d demo -u monetdb -P monetdb
-- connection with server mapi:monetdb://vienna.ins.cwi.nl:50000/demo is [3]
-- [3] profiler.activate("event");
-- [3] profiler.activate("time");
-- [3] profiler.activate("ticks");
MAPI = monetdb@budapest.ins.cwi.nl:50000
ACTION= mapi_start_talking
ERROR = !monetdbd: internal error while starting mserver, please refer to the logs
-- connection with server mapi:monetdb://budapest.ins.cwi.nl:50000/demo closed
-- [3] profiler.activate("stmt");
MAPI = monetdb@munich.ins.cwi.nl:50000
-- connection with server mapi:monetdb://sirius.ins.cwi.nl:50000/demo is [2]
ACTION= mapi_start_talking
ERROR = !merovingian: internal error while starting mserver, please refer to the logs
-- [3] profiler.activate("thread");
-- connection with server mapi:monetdb://munich.ins.cwi.nl:50000/demo closed
-- [3] profiler.activate("start");
...
Reproducible: Always
### Steps to Reproduce:
1.stethoscope -d demo -u monetdb -P monetdb
## Comment 16691
Date: 2011-12-27 09:42:34 +0100
From: @grobian
Please clarify what's so confusing about this.
## Comment 17937
Date: 2012-11-12 22:15:14 +0100
From: @mlkersten
This has been solved by looking at localhost as default host,
instead of a broadcast to all merovingians visible.
| stethoscope errors are confusion | https://api.github.com/repos/MonetDB/MonetDB/issues/2955/comments | 0 | 2020-11-30T11:49:16Z | 2024-06-27T11:54:20Z | https://github.com/MonetDB/MonetDB/issues/2955 | 753,410,767 | 2,955 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-26 14:33:52 +0100
From: @mlkersten
To: clients devs <<bugs-clients>>
Version: -- development
Last updated: 2011-12-27 09:40:58 +0100
## Comment 16684
Date: 2011-12-26 14:33:52 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier:
- stethoscope does not obey localhost restriction
mk@vienna::~/current> stethoscope -d sf10 -u monetdb -P monetdb -h localhost
-- connection with server mapi:monetdb://vienna.ins.cwi.nl:50000/sf10 is [2]
...
-- [2] opened UDP profile stream for vienna.ins.cwi.nl:50010
-- [2] profiler.setAll();
-- [2] profiler.start();
-- connection with server mapi:monetdb://scilens10.ins.cwi.nl:50000/sf10 is [1]
...;
-- [1] opened UDP profile stream for scilens10.ins.cwi.nl:50011
Reproducible: Always
## Comment 16690
Date: 2011-12-27 09:40:58 +0100
From: @grobian
This is intended behaviour. You might want a flag/mode where it doesn't do discovery.
| stethoscope does not obey localhost restriction | https://api.github.com/repos/MonetDB/MonetDB/issues/2954/comments | 0 | 2020-11-30T11:49:13Z | 2024-06-28T13:20:03Z | https://github.com/MonetDB/MonetDB/issues/2954 | 753,410,738 | 2,954 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-26 14:31:55 +0100
From: @mlkersten
To: Merovingian devs <<bugs-merovingian>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-26 15:32:01 +0100
## Comment 16683
Date: 2011-12-26 14:31:55 +0100
From: @mlkersten
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:9.0.1) Gecko/20100101 Firefox/9.0.1
Build Identifier:
erroneous input causes crash
Reproducible: Always
### Steps to Reproduce:
monetdb get */demo
name prop source value
get: no such property: */demo
Segmentation fault
## Comment 16697
Date: 2011-12-27 10:31:54 +0100
From: @grobian
Changeset [7d7e87343c9c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=7d7e87343c9c) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7d7e87343c9c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=7d7e87343c9c)
Changeset description:
monetdb: avoid crash when using invalid properties with get command
Check properties first, such that warnings about invalid properties go
on top, then print, and silently skip invalid properties.
Fixes bug #2953.
## Comment 16822
Date: 2012-01-26 15:32:01 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| segfault for monetdb get with invalid property | https://api.github.com/repos/MonetDB/MonetDB/issues/2953/comments | 0 | 2020-11-30T11:49:11Z | 2024-06-27T11:54:19Z | https://github.com/MonetDB/MonetDB/issues/2953 | 753,410,712 | 2,953 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-21 10:59:51 +0100
From: Wouter Alink <<alink>>
To: clients devs <<bugs-clients>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-26 15:31:58 +0100
## Comment 16670
Date: 2011-12-21 10:59:51 +0100
From: Wouter Alink <<alink>>
User-Agent: Mozilla/5.0 (Windows NT 6.0; WOW64; rv:8.0.1) Gecko/20100101 Firefox/8.0.1
Build Identifier:
When inserting a long string via JDBC (using batched prepared statements) containing multi-byte UTF-8 characters, the JDBC-driver raises an exception.
Reproducible: Always
### Steps to Reproduce:
import java.nio.charset.Charset;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.SQLException;
import java.sql.Statement;
import org.junit.Test;
public class TestChar {
@Test
public void testChars() throws ClassNotFoundException, SQLException {
// 3-byte UTF-8 symbol (a dot, but I think it could be any 3-byte character):
byte[] errorBytes = new byte[] { (byte) 0xe2, (byte) 0x80, (byte) 0xa7 };
String errorStr = new String(errorBytes, Charset.forName("UTF-8"));
StringBuilder repeatedErrorStr = new StringBuilder();
for(int i = 0; i < 8170;i++) {
repeatedErrorStr.append(errorStr);
}
System.out.println("TestString: " + errorStr);
Class.forName("nl.cwi.monetdb.jdbc.MonetDriver");
String connectionURL = "jdbc:monetdb://localhost:50000/vv01?language=sql&user=monetdb&password=monetdb";
Connection conn = DriverManager.getConnection(connectionURL);
try {
Statement st = conn.createStatement();
try { st.execute("DROP TABLE x;"); } catch (SQLException e) { }
st.execute("CREATE TABLE x (c INT, a CLOB, b DOUBLE);");
PreparedStatement ps = conn.prepareStatement("INSERT INTO x VALUES (?,?,?);");
ps.setLong(1, 1);
ps.setString(2, repeatedErrorStr.toString());
ps.setDouble(3, 1.0);
ps.addBatch();
ps.executeBatch();
st.close();
} catch (SQLException e) {
e.printStackTrace();
if (e.getNextException() != null) {
e.getNextException().printStackTrace();
}
throw e;
} finally {
conn.close();
}
}
}
### Actual Results:
TestString: ‧
java.sql.BatchUpdateException: Error(s) occurred while executing the batch, see next SQLExceptions for details
at nl.cwi.monetdb.jdbc.MonetStatement.executeBatch(MonetStatement.java:199)
at com.spinque.tools.importStream.io.TestChar.testChars(TestChar.java:44)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:616)
at org.junit.internal.runners.TestMethodRunner.executeMethodBody(TestMethodRunner.java:99)
at org.junit.internal.runners.TestMethodRunner.runUnprotected(TestMethodRunner.java:81)
at org.junit.internal.runners.BeforeAndAfterRunner.runProtected(BeforeAndAfterRunner.java:34)
at org.junit.internal.runners.TestMethodRunner.runMethod(TestMethodRunner.java:75)
at org.junit.internal.runners.TestMethodRunner.run(TestMethodRunner.java:45)
at org.junit.internal.runners.TestClassMethodsRunner.invokeTestMethod(TestClassMethodsRunner.java:66)
at org.junit.internal.runners.TestClassMethodsRunner.run(TestClassMethodsRunner.java:35)
at org.junit.internal.runners.TestClassRunner$1.runUnprotected(TestClassRunner.java:42)
at org.junit.internal.runners.BeforeAndAfterRunner.runProtected(BeforeAndAfterRunner.java:34)
at org.junit.internal.runners.TestClassRunner.run(TestClassRunner.java:52)
at org.eclipse.jdt.internal.junit4.runner.JUnit4TestReference.run(JUnit4TestReference.java:50)
at org.eclipse.jdt.internal.junit.runner.TestExecution.run(TestExecution.java:38)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:467)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:683)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.run(RemoteTestRunner.java:390)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.main(RemoteTestRunner.java:197)
java.sql.SQLException: Overflow: don't use multi statements when batching (0)
at nl.cwi.monetdb.jdbc.MonetStatement.internalBatch(MonetStatement.java:250)
at nl.cwi.monetdb.jdbc.MonetStatement.executeBatch(MonetStatement.java:219)
at com.spinque.tools.importStream.io.TestChar.testChars(TestChar.java:44)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:616)
at org.junit.internal.runners.TestMethodRunner.executeMethodBody(TestMethodRunner.java:99)
at org.junit.internal.runners.TestMethodRunner.runUnprotected(TestMethodRunner.java:81)
at org.junit.internal.runners.BeforeAndAfterRunner.runProtected(BeforeAndAfterRunner.java:34)
at org.junit.internal.runners.TestMethodRunner.runMethod(TestMethodRunner.java:75)
at org.junit.internal.runners.TestMethodRunner.run(TestMethodRunner.java:45)
at org.junit.internal.runners.TestClassMethodsRunner.invokeTestMethod(TestClassMethodsRunner.java:66)
at org.junit.internal.runners.TestClassMethodsRunner.run(TestClassMethodsRunner.java:35)
at org.junit.internal.runners.TestClassRunner$1.runUnprotected(TestClassRunner.java:42)
at org.junit.internal.runners.BeforeAndAfterRunner.runProtected(BeforeAndAfterRunner.java:34)
at org.junit.internal.runners.TestClassRunner.run(TestClassRunner.java:52)
at org.eclipse.jdt.internal.junit4.runner.JUnit4TestReference.run(JUnit4TestReference.java:50)
at org.eclipse.jdt.internal.junit.runner.TestExecution.run(TestExecution.java:38)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:467)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:683)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.run(RemoteTestRunner.java:390)
at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.main(RemoteTestRunner.java:197)
### Expected Results:
no exception is expected.
note: with 8170 3-byte characters the exception is raised, with 8169 3-byte characters it is not. Might it perhaps be a buffer issue?
## Comment 16671
Date: 2011-12-21 11:17:23 +0100
From: @grobian
you fall outside of the BLOCK size, which apparently makes things go boom with a misleading exception:
java.sql.SQLException: Overflow: don't use multi statements when batching (0)
## Comment 16672
Date: 2011-12-21 11:36:30 +0100
From: @grobian
Changeset [2232a2c1f642](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2232a2c1f642) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2232a2c1f642](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2232a2c1f642)
Changeset description:
JDBC add test for bug #2952
## Comment 16673
Date: 2011-12-21 12:18:00 +0100
From: @grobian
Changeset [e16c4bb803ce](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e16c4bb803ce) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e16c4bb803ce](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e16c4bb803ce)
Changeset description:
JDBC: fix "overflow" due to very large statements
Off-by-one error that was never triggered before, because offset and max
were never equal. However, for bug #2952 the first statement is larger
than what the MAPI block can hold, and hence offset = i = 0. max is
expected to be a count, though.
Fixes bug #2952
## Comment 16674
Date: 2011-12-21 12:21:06 +0100
From: @grobian
Changeset [ef01bf784b71](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ef01bf784b71) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ef01bf784b71](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ef01bf784b71)
Changeset description:
Add test for bug #2952
## Comment 16818
Date: 2012-01-26 15:31:58 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| utf-8 characters in long string | https://api.github.com/repos/MonetDB/MonetDB/issues/2952/comments | 0 | 2020-11-30T11:49:08Z | 2024-06-27T11:54:18Z | https://github.com/MonetDB/MonetDB/issues/2952 | 753,410,672 | 2,952 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-20 22:06:43 +0100
From: Josip <<josip.djuricic>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: josip.djuricic, @drstmane
Last updated: 2012-02-14 15:49:30 +0100
## Comment 16666
Date: 2011-12-20 22:06:43 +0100
From: Josip <<josip.djuricic>>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.17) Gecko/20110717 Gentoo Firefox/3.6.17
Build Identifier:
sql>select count(originating) as count, originating from testing group by originating order by count desc limit 20;
+----------+-----------------+
| count | sms_originating |
+==========+=================+
| 23321699 | |
| 14653910 | |
| 10604551 | |
| 6938043 | |
| 5302587 | xxxx1 |
| 2757905 | xxxx2 |
| 1208261 | xxxx3 |
| 825955 | xxxx4 |
| 823496 | xxxx5 |
| 807673 | xxxx6 |
| 506432 | xxxx7 |
| 440624 | xxxx8 |
| 412879 | xxxx9 |
| 193748 | xxxx10 |
| 152259 | xxxx11 |
| 147790 | xxxx12 |
| 145981 | xxxx13 |
| 128437 | xxxx14 |
| 103344 | xxxx15 |
| 98601 | xxxx16 |
+----------+-----------------+
20 tuples (30m 52s)
Field originating is of varchar type.
Current number of records is:
sql>select count(id) from testing;
+-----------+
| L5 |
+===========+
| 169045755 |
+-----------+
1 tuple (1.273ms)
Here in output I changed real data with xxx but what is wrong is first 4 rows. They contain no data no matter there is some. I am sure it has something to do with number of records since if I use smaller subset of data I get normal results.
Since we are doing testing of product to see if it's possible to use it in commercial application (in which case I am sure we can negotiate support contract) I need to understand what happened and how we can fix it.
Log file has no error of any kind.
Reproducible: Always
### Steps to Reproduce:
1. Do a group by query on large dataset
### Actual Results:
sql>select count(originating) as count, originating from testing group by originating order by count desc limit 20;
+----------+-----------------+
| count | sms_originating |
+==========+=================+
| 23321699 | |
| 14653910 | |
| 10604551 | |
| 6938043 | |
| 5302587 | xxxx1 |
| 2757905 | xxxx2 |
| 1208261 | xxxx3 |
| 825955 | xxxx4 |
| 823496 | xxxx5 |
| 807673 | xxxx6 |
| 506432 | xxxx7 |
| 440624 | xxxx8 |
| 412879 | xxxx9 |
| 193748 | xxxx10 |
| 152259 | xxxx11 |
| 147790 | xxxx12 |
| 145981 | xxxx13 |
| 128437 | xxxx14 |
| 103344 | xxxx15 |
| 98601 | xxxx16 |
+----------+-----------------+
20 tuples (30m 52s)
### Expected Results:
sql>select count(originating) as count, originating from testing group by originating order by count desc limit 20;
+----------+-----------------+
| count | sms_originating |
+==========+=================+
| 23321699 | xxxx20 |
| 14653910 | xxxx21 |
| 10604551 | xxxx22 |
| 6938043 | xxxx23 |
| 5302587 | xxxx1 |
| 2757905 | xxxx2 |
| 1208261 | xxxx3 |
| 825955 | xxxx4 |
| 823496 | xxxx5 |
| 807673 | xxxx6 |
| 506432 | xxxx7 |
| 440624 | xxxx8 |
| 412879 | xxxx9 |
| 193748 | xxxx10 |
| 152259 | xxxx11 |
| 147790 | xxxx12 |
| 145981 | xxxx13 |
| 128437 | xxxx14 |
| 103344 | xxxx15 |
| 98601 | xxxx16 |
+----------+-----------------+
20 tuples (30m 52s)
./mserver5 --version
MonetDB 5 server v11.5.9 "Aug2011-SP3" (64-bit, 64-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 15.7GiB available memory, 4 available cpu cores
Libraries:
libpcre: 8.12 2011-01-15 (compiled with 8.12)
openssl: OpenSSL 1.0.0d 8 Feb 2011 (compiled with OpenSSL 1.0.0d 8 Feb 2011)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: root@dbn01 (x86_64-unknown-linux-gnu)
Compilation: gcc -O3 -fomit-frame-pointer -pipe
Linking : /usr/x86_64-pc-linux-gnu/bin/ld -m elf_x86_64
Also worth noticing is that testing server is dell poweredge 1950 with 2x Dual core xeons at 3.0 ghz and 16gig of memory. Database on the disk is 64GB in size.
Operating system is Gentoo 64bit with kernel 2.6.39-r3
## Comment 16667
Date: 2011-12-21 00:09:22 +0100
From: @drstmane
What do the following queries return on your case?
select min(length(originating)) from testing;
select max(length(originating)) from testing;
select l, count(l) from ( select length(originating) as l from testing ) as t group by l order by l;
?
## Comment 16668
Date: 2011-12-21 10:05:57 +0100
From: Josip <<josip.djuricic>>
(In reply to comment 1)
> What do the following queries return on your case?
>
> select min(length(originating)) from testing;
+------+
| L5 |
+======+
| 1 |
+------+
1 tuple (38.4s)
>
> select max(length(originating)) from testing;
+------+
| L6 |
+======+
| 20 |
+------+
1 tuple (37.2s)
>
> select l, count(l) from ( select length(originating) as l from testing ) as t
> group by l order by l;
+------+----------+
| l | L7 |
+======+==========+
| null | 0 |
| 1 | 4309 |
| 2 | 146190 |
| 3 | 20940041 |
| 4 | 177290 |
| 5 | 1728982 |
| 6 | 10921783 |
| 7 | 661 |
| 8 | 2241140 |
| 9 | 530681 |
| 10 | 46243 |
| 11 | 66299281 |
| 12 | 3194830 |
| 13 | 102097 |
| 14 | 5009 |
| 15 | 40653 |
| 16 | 37878 |
| 17 | 200 |
| 18 | 102 |
| 19 | 161 |
| 20 | 224 |
+------+----------+
21 tuples (41.4s)
>
> ?
## Comment 16669
Date: 2011-12-21 10:46:48 +0100
From: @drstmane
Josip,
assuming that you cannot share your data with us, here are some more questions to analyze the problem further:
Which is the largest number of records in table testing that your query works fine with? Well, the largest you successfully tested so far is sufficient for now, i.e., no need for an exhaustive search.
Likewise, which is the smallest number of records in table testing that your query "fails" as reported?
How did you load/insert your data into MonetDB? Bulkload via COPY INTO from a textual CSV file?
Are there any non-ASCII characters in your data; if so, which encoding does/did the original .CSV use?
Could you please run the following queries at least on the "large" testing table (where your original query fails as reported), possibly also on the "small" testing table (where your original query succeeds):
select count(*), count(id), count(originating) from testing;
select count(distinct id), count(distinct originating) from testing;
select count(*) from testing where originating is null;
select min(length(originating)), max(length(originating)) from testing;
select l, count(l), count(*) from ( select length(originating) as l from testing ) as t group by l order by l;
select sum(c1), sum(c0) from ( select count(l) as c1, count(*) as c0 from ( select length(originating) as l from testing ) as t group by l ) as tt;
select count(originating) as count, originating, length(originating) as len from testing group by originating order by count desc limit 20;
select count(originating) as count, originating, length(originating) as len from testing group by originating order by len asc limit 20;
Thanks!
Stefan
## Comment 16675
Date: 2011-12-21 19:51:01 +0100
From: Josip <<josip.djuricic>>
(In reply to comment 3)
> Josip,
>
> assuming that you cannot share your data with us, here are some more questions
> to analyze the problem further:
I could share it off the list, directly with you on the email.
>
> Which is the largest number of records in table testing that your query works
> fine with? Well, the largest you successfully tested so far is sufficient for
> now, i.e., no need for an exhaustive search.
> Likewise, which is the smallest number of records in table testing that your
> query "fails" as reported?
For this to answer I will need to reimport the data day by day on my local machine and try to figure it out. will take some time, especially since transferring the data over vpn can take a LOT of time since it's located off the continent and data connection is very slow, however will try to do it with data repetition and inserting same data.
>
> How did you load/insert your data into MonetDB? Bulkload via COPY INTO from a
> textual CSV file?
used bulkload, since I am not 100% sure why with jdbc driver there was no way to enter 2000 records as a single transaction -- since it's a test machine the disk just could not handle data inserts (1x SAS 10k)
>
> Are there any non-ASCII characters in your data; if so, which encoding does/did
> the original .CSV use?
There is always possibility for that, but we used UTF-8
>
> Could you please run the following queries at least on the "large" testing
> table (where your original query fails as reported), possibly also on the
> "small" testing table (where your original query succeeds):
>
> select count(*), count(id), count(originating) from testing;
large(fails):
+-----------+-----------+-----------+
| L11 | L12 | L13 |
+===========+===========+===========+
| 169045755 | 169045755 | 106417755 |
+-----------+-----------+-----------+
1 tuple (1m 8s)
> select count(distinct id), count(distinct originating) from testing;
large(fails):
+-----------+---------+
| L24 | L25 |
+===========+=========+
| 169000755 | 1552921 |
+-----------+---------+
1 tuple (4m 32s)
> select count(*) from testing where originating is null;
large(fails):
+----------+
| L26 |
+==========+
| 62628000 |
+----------+
1 tuple (21.8s)
Please note that I have 2 different types of record data in the table originating and destination, so originating has destination field and destination has originating field among others. I also tried doing group by with just selecting destination by using destination_id = 2 in where clause but it still had same results.
> select min(length(originating)), max(length(originating)) from testing;
large(fails):
+------+------+
| L30 | L31 |
+======+======+
| 1 | 20 |
+------+------+
1 tuple (38.2s)
> select l, count(l), count(*) from ( select length(originating) as l from
> testing ) as t group by l order by l;
large(fails):
+------+----------+
| l | L7 |
+======+==========+
| null | 0 |
| 1 | 4309 |
| 2 | 146190 |
| 3 | 20940041 |
| 4 | 177290 |
| 5 | 1728982 |
| 6 | 10921783 |
| 7 | 661 |
| 8 | 2241140 |
| 9 | 530681 |
| 10 | 46243 |
| 11 | 66299281 |
| 12 | 3194830 |
| 13 | 102097 |
| 14 | 5009 |
| 15 | 40653 |
| 16 | 37878 |
| 17 | 200 |
| 18 | 102 |
| 19 | 161 |
| 20 | 224 |
+------+----------+
21 tuples (41.4s)
> select sum(c1), sum(c0) from ( select count(l) as c1, count(*) as c0 from (
> select length(originating) as l from testing ) as t group by l ) as tt;
large(fails):
+-----------+-----------+
| L34 | L35 |
+===========+===========+
| 106417755 | 169045755 |
+-----------+-----------+
1 tuple (46.5s)
> select count(originating) as count, originating, length(originating) as len
> from testing group by originating order by count desc limit 20;
large(fails):
+----------+-----------------+------+
| count | sms_originating | len |
+==========+=================+======+
| 23321699 | | 0 |
| 14653910 | | 0 |
| 10604551 | | 0 |
| 6938043 | | 0 |
| 5302587 | xxxx01 | 6 |
| 2757905 | xxxx02 | 6 |
| 1208261 | xxx03 | 5 |
| 825955 | MyTunez@ | 8 |
| 823496 | xxxxxx xxxx | 11 |
| 807673 | Bxxxlysx | 8 |
| 506432 | 167 | 3 |
| 440624 | StanXXXXX | 9 |
| 412879 | 170030 | 6 |
| 193748 | 170032 | 6 |
| 152259 | 1362 | 4 |
| 147790 | 108 | 3 |
| 145981 | 71 | 2 |
| 128437 | OPTIONS@ | 8 |
| 103344 | 13366 | 5 |
| 98601 | 170042 | 6 |
+----------+-----------------+------+
20 tuples (31m 0s)
> select count(originating) as count, originating, length(originating) as len
> from testing group by originating order by len asc limit 20;
large(fails):
+----------+-----------------+------+
| count | originating | len |
+==========+=================+======+
| 2 | | 0 |
| 6 | | 0 |
| 83 | | 0 |
| 3 | | 0 |
| 6938043 | | 0 |
| 8 | | 0 |
| 6 | | 0 |
| 84 | | 0 |
| 1 | | 0 |
| 23321699 | | 0 |
| 5 | | 0 |
| 243 | | 0 |
| 8605 | | 0 |
| 2 | | 0 |
| 274 | | 0 |
| 12 | | 0 |
| 0 | | 0 |
| 14653910 | | 0 |
| 22 | | 0 |
| 7 | | 0 |
+----------+-----------------+------+
20 tuples (30m 50s)
>
> Thanks!
> Stefan
## Comment 16676
Date: 2011-12-21 19:53:45 +0100
From: Josip <<josip.djuricic>>
Please note originating is actually sms_originating if you wonder why it's different field names.
(In reply to comment 4)
> (In reply to comment 3)
> > Josip,
> >
> > assuming that you cannot share your data with us, here are some more questions
> > to analyze the problem further:
>
> I could share it off the list, directly with you on the email.
>
> >
> > Which is the largest number of records in table testing that your query works
> > fine with? Well, the largest you successfully tested so far is sufficient for
> > now, i.e., no need for an exhaustive search.
> > Likewise, which is the smallest number of records in table testing that your
> > query "fails" as reported?
>
> For this to answer I will need to reimport the data day by day on my local
> machine and try to figure it out. will take some time, especially since
> transferring the data over vpn can take a LOT of time since it's located off
> the continent and data connection is very slow, however will try to do it with
> data repetition and inserting same data.
>
> >
> > How did you load/insert your data into MonetDB? Bulkload via COPY INTO from a
> > textual CSV file?
>
> used bulkload, since I am not 100% sure why with jdbc driver there was no way
> to enter 2000 records as a single transaction -- since it's a test machine the
> disk just could not handle data inserts (1x SAS 10k)
>
> >
> > Are there any non-ASCII characters in your data; if so, which encoding does/did
> > the original .CSV use?
>
> There is always possibility for that, but we used UTF-8
>
> >
> > Could you please run the following queries at least on the "large" testing
> > table (where your original query fails as reported), possibly also on the
> > "small" testing table (where your original query succeeds):
>
> >
> > select count(*), count(id), count(originating) from testing;
> large(fails):
> +-----------+-----------+-----------+
> | L11 | L12 | L13 |
> +===========+===========+===========+
> | 169045755 | 169045755 | 106417755 |
> +-----------+-----------+-----------+
> 1 tuple (1m 8s)
>
>
> > select count(distinct id), count(distinct originating) from testing;
> large(fails):
> +-----------+---------+
> | L24 | L25 |
> +===========+=========+
> | 169000755 | 1552921 |
> +-----------+---------+
> 1 tuple (4m 32s)
>
>
> > select count(*) from testing where originating is null;
> large(fails):
> +----------+
> | L26 |
> +==========+
> | 62628000 |
> +----------+
> 1 tuple (21.8s)
>
> Please note that I have 2 different types of record data in the table
> originating and destination, so originating has destination field and
> destination has originating field among others. I also tried doing group by
> with just selecting destination by using destination_id = 2 in where clause but
> it still had same results.
>
>
> > select min(length(originating)), max(length(originating)) from testing;
> large(fails):
> +------+------+
> | L30 | L31 |
> +======+======+
> | 1 | 20 |
> +------+------+
> 1 tuple (38.2s)
>
>
> > select l, count(l), count(*) from ( select length(originating) as l from
> > testing ) as t group by l order by l;
> large(fails):
> +------+----------+
> | l | L7 |
> +======+==========+
> | null | 0 |
> | 1 | 4309 |
> | 2 | 146190 |
> | 3 | 20940041 |
> | 4 | 177290 |
> | 5 | 1728982 |
> | 6 | 10921783 |
> | 7 | 661 |
> | 8 | 2241140 |
> | 9 | 530681 |
> | 10 | 46243 |
> | 11 | 66299281 |
> | 12 | 3194830 |
> | 13 | 102097 |
> | 14 | 5009 |
> | 15 | 40653 |
> | 16 | 37878 |
> | 17 | 200 |
> | 18 | 102 |
> | 19 | 161 |
> | 20 | 224 |
> +------+----------+
> 21 tuples (41.4s)
>
> > select sum(c1), sum(c0) from ( select count(l) as c1, count(*) as c0 from (
> > select length(originating) as l from testing ) as t group by l ) as tt;
> large(fails):
> +-----------+-----------+
> | L34 | L35 |
> +===========+===========+
> | 106417755 | 169045755 |
> +-----------+-----------+
> 1 tuple (46.5s)
>
>
> > select count(originating) as count, originating, length(originating) as len
> > from testing group by originating order by count desc limit 20;
> large(fails):
> +----------+-----------------+------+
> | count | sms_originating | len |
> +==========+=================+======+
> | 23321699 | | 0 |
> | 14653910 | | 0 |
> | 10604551 | | 0 |
> | 6938043 | | 0 |
> | 5302587 | xxxx01 | 6 |
> | 2757905 | xxxx02 | 6 |
> | 1208261 | xxx03 | 5 |
> | 825955 | MyTunez@ | 8 |
> | 823496 | xxxxxx xxxx | 11 |
> | 807673 | Bxxxlysx | 8 |
> | 506432 | 167 | 3 |
> | 440624 | StanXXXXX | 9 |
> | 412879 | 170030 | 6 |
> | 193748 | 170032 | 6 |
> | 152259 | 1362 | 4 |
> | 147790 | 108 | 3 |
> | 145981 | 71 | 2 |
> | 128437 | OPTIONS@ | 8 |
> | 103344 | 13366 | 5 |
> | 98601 | 170042 | 6 |
> +----------+-----------------+------+
> 20 tuples (31m 0s)
>
>
> > select count(originating) as count, originating, length(originating) as len
> > from testing group by originating order by len asc limit 20;
> large(fails):
> +----------+-----------------+------+
> | count | originating | len |
> +==========+=================+======+
> | 2 | | 0 |
> | 6 | | 0 |
> | 83 | | 0 |
> | 3 | | 0 |
> | 6938043 | | 0 |
> | 8 | | 0 |
> | 6 | | 0 |
> | 84 | | 0 |
> | 1 | | 0 |
> | 23321699 | | 0 |
> | 5 | | 0 |
> | 243 | | 0 |
> | 8605 | | 0 |
> | 2 | | 0 |
> | 274 | | 0 |
> | 12 | | 0 |
> | 0 | | 0 |
> | 14653910 | | 0 |
> | 22 | | 0 |
> | 7 | | 0 |
> +----------+-----------------+------+
> 20 tuples (30m 50s)
>
> >
> > Thanks!
> > Stefan
## Comment 16677
Date: 2011-12-21 20:56:27 +0100
From: @drstmane
(In reply to comment 4)
> > select l, count(l), count(*) from ( select length(originating) as l from
> > testing ) as t group by l order by l;
> large(fails):
> +------+----------+
> | l | L7 |
> +======+==========+
> | null | 0 |
> | 1 | 4309 |
> | 2 | 146190 |
> | 3 | 20940041 |
> | 4 | 177290 |
> | 5 | 1728982 |
> | 6 | 10921783 |
> | 7 | 661 |
> | 8 | 2241140 |
> | 9 | 530681 |
> | 10 | 46243 |
> | 11 | 66299281 |
> | 12 | 3194830 |
> | 13 | 102097 |
> | 14 | 5009 |
> | 15 | 40653 |
> | 16 | 37878 |
> | 17 | 200 |
> | 18 | 102 |
> | 19 | 161 |
> | 20 | 224 |
> +------+----------+
> 21 tuples (41.4s)
I would have expected 3 result columns. Did you by accident omit count(*) i your query?
Could you maybe also run these queries?
select count(sms_originating) as count1, count(*) as count2, sms_originating, length(sms_originating) as len from testing group by sms_originating order by count1 desc limit 20;
select count(sms_originating) as count1, count(*) as count2, sms_originating, length(sms_originating) as len from testing group by sms_originating order by len asc limit 20;
select count(sms_originating) as count1, count(*) as count2, sms_originating, length(sms_originating) as len from testing group by sms_originating limit 20;
select count(sms_originating) as count1, count(*) as count2, sms_originating, length(sms_originating) as len from testing where length(sms_originating) = 0 group by sms_originating limit 20;
select count(*), count(sms_originating) from testing where length(sms_originating) = 0;
select sum(count1), sum(count2) from ( select count(sms_originating) as count1, count(*) as count2 from testing group by sms_originating ) as t;
Sharing your data off the list / bug tracker is fine. Whether email is the feasible way depends on how large the (compressed) .CSV file for the bulkload is.
## Comment 16678
Date: 2011-12-21 21:32:06 +0100
From: Josip <<josip.djuricic>>
(In reply to comment 6)
> (In reply to comment 4)
> > > select l, count(l), count(*) from ( select length(originating) as l from
> > > testing ) as t group by l order by l;
> > large(fails):
> > +------+----------+
> > | l | L7 |
> > +======+==========+
> > | null | 0 |
> > | 1 | 4309 |
> > | 2 | 146190 |
> > | 3 | 20940041 |
> > | 4 | 177290 |
> > | 5 | 1728982 |
> > | 6 | 10921783 |
> > | 7 | 661 |
> > | 8 | 2241140 |
> > | 9 | 530681 |
> > | 10 | 46243 |
> > | 11 | 66299281 |
> > | 12 | 3194830 |
> > | 13 | 102097 |
> > | 14 | 5009 |
> > | 15 | 40653 |
> > | 16 | 37878 |
> > | 17 | 200 |
> > | 18 | 102 |
> > | 19 | 161 |
> > | 20 | 224 |
> > +------+----------+
> > 21 tuples (41.4s)
>
> I would have expected 3 result columns. Did you by accident omit count(*) i
> your query?
Could be I pasted wrong one by accident, here is revised one:
+------+----------+----------+
| l | L40 | L41 |
+======+==========+==========+
| null | 0 | 62628000 |
| 1 | 4309 | 4309 |
| 2 | 146190 | 146190 |
| 3 | 20940041 | 20940041 |
| 4 | 177290 | 177290 |
| 5 | 1728982 | 1728982 |
| 6 | 10921783 | 10921783 |
| 7 | 661 | 661 |
| 8 | 2241140 | 2241140 |
| 9 | 530681 | 530681 |
| 10 | 46243 | 46243 |
| 11 | 66299281 | 66299281 |
| 12 | 3194830 | 3194830 |
| 13 | 102097 | 102097 |
| 14 | 5009 | 5009 |
| 15 | 40653 | 40653 |
| 16 | 37878 | 37878 |
| 17 | 200 | 200 |
| 18 | 102 | 102 |
| 19 | 161 | 161 |
| 20 | 224 | 224 |
+------+----------+----------+
21 tuples (55.9s)
>
>
> Could you maybe also run these queries?
They will take a lot of time, where group by sms_originating is used it needs 30min per query, I will let them run and once they are done will post the results back here, ok?
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating order by
> count1 desc limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating order by
> len asc limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing where length(sms_originating) = 0
> group by sms_originating limit 20;
>
> select count(*), count(sms_originating) from testing where
> length(sms_originating) = 0;
>
> select sum(count1), sum(count2) from ( select count(sms_originating) as count1,
> count(*) as count2 from testing group by sms_originating ) as t;
>
>
> Sharing your data off the list / bug tracker is fine. Whether email is the
> feasible way depends on how large the (compressed) .CSV file for the bulkload
> is.
Well over email I could share table structure, CSVs are 40gig tgz files, so I doubt I could transfer it easily.
However if there is possibility to sign NDA agreement between us I could think of a way to prepare test env to access the DB? Also we could enter commercial discussions for support contracts (in this case we would act more as an OEM partner)
## Comment 16679
Date: 2011-12-21 23:18:39 +0100
From: Josip <<josip.djuricic>>
sql>select count(sms_originating) as count1, count(*) as count2, sms_originating, length(sms_originating) as len from testing group by sms_originating order by count1 desc limit 20;
+----------+----------+-----------------+------+
| count1 | count2 | sms_originating | len |
+==========+==========+=================+======+
| 23321699 | 40041525 | | 0 |
| 14653910 | 31077151 | | 0 |
| 10604551 | 10604551 | | 0 |
| 6938043 | 6938043 | | 0 |
| 5302587 | 5302587 | Mascom | 6 |
| 2757905 | 2757905 | MySpot | 6 |
| 1208261 | 1208261 | 15728 | 5 |
| 825955 | 825955 | MyTunez@ | 8 |
| 823496 | 823496 | Mascom SMSC | 11 |
| 807673 | 807673 | Barclays | 8 |
| 506432 | 506432 | 167 | 3 |
| 440624 | 440624 | StanChart | 9 |
| 412879 | 412879 | 170030 | 6 |
| 193748 | 193748 | 170032 | 6 |
| 152259 | 152259 | 1362 | 4 |
| 147790 | 147790 | 108 | 3 |
| 145981 | 145981 | 71 | 2 |
| 128437 | 128437 | OPTIONS@ | 8 |
| 103344 | 103344 | 13366 | 5 |
| 98601 | 98601 | 170042 | 6 |
+----------+----------+-----------------+------+
20 tuples (31m 54s)
Others will follow...
## Comment 16680
Date: 2011-12-22 10:44:23 +0100
From: Josip <<josip.djuricic>>
Hi Stefan,
I've sent you results directly to your email address.
Thank you.
(In reply to comment 6)
> (In reply to comment 4)
> > > select l, count(l), count(*) from ( select length(originating) as l from
> > > testing ) as t group by l order by l;
> > large(fails):
> > +------+----------+
> > | l | L7 |
> > +======+==========+
> > | null | 0 |
> > | 1 | 4309 |
> > | 2 | 146190 |
> > | 3 | 20940041 |
> > | 4 | 177290 |
> > | 5 | 1728982 |
> > | 6 | 10921783 |
> > | 7 | 661 |
> > | 8 | 2241140 |
> > | 9 | 530681 |
> > | 10 | 46243 |
> > | 11 | 66299281 |
> > | 12 | 3194830 |
> > | 13 | 102097 |
> > | 14 | 5009 |
> > | 15 | 40653 |
> > | 16 | 37878 |
> > | 17 | 200 |
> > | 18 | 102 |
> > | 19 | 161 |
> > | 20 | 224 |
> > +------+----------+
> > 21 tuples (41.4s)
>
> I would have expected 3 result columns. Did you by accident omit count(*) i
> your query?
>
>
> Could you maybe also run these queries?
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating order by
> count1 desc limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating order by
> len asc limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing where length(sms_originating) = 0
> group by sms_originating limit 20;
>
> select count(*), count(sms_originating) from testing where
> length(sms_originating) = 0;
>
> select sum(count1), sum(count2) from ( select count(sms_originating) as count1,
> count(*) as count2 from testing group by sms_originating ) as t;
>
>
> Sharing your data off the list / bug tracker is fine. Whether email is the
> feasible way depends on how large the (compressed) .CSV file for the bulkload
> is.
## Comment 16687
Date: 2011-12-26 23:06:49 +0100
From: Josip <<josip.djuricic>>
Hi Stefan,
first of all Merry Xmas.
I have wondered if you had any more ideas?
Josip
(In reply to comment 6)
> (In reply to comment 4)
> > > select l, count(l), count(*) from ( select length(originating) as l from
> > > testing ) as t group by l order by l;
> > large(fails):
> > +------+----------+
> > | l | L7 |
> > +======+==========+
> > | null | 0 |
> > | 1 | 4309 |
> > | 2 | 146190 |
> > | 3 | 20940041 |
> > | 4 | 177290 |
> > | 5 | 1728982 |
> > | 6 | 10921783 |
> > | 7 | 661 |
> > | 8 | 2241140 |
> > | 9 | 530681 |
> > | 10 | 46243 |
> > | 11 | 66299281 |
> > | 12 | 3194830 |
> > | 13 | 102097 |
> > | 14 | 5009 |
> > | 15 | 40653 |
> > | 16 | 37878 |
> > | 17 | 200 |
> > | 18 | 102 |
> > | 19 | 161 |
> > | 20 | 224 |
> > +------+----------+
> > 21 tuples (41.4s)
>
> I would have expected 3 result columns. Did you by accident omit count(*) i
> your query?
>
>
> Could you maybe also run these queries?
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating order by
> count1 desc limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating order by
> len asc limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing group by sms_originating limit 20;
>
> select count(sms_originating) as count1, count(*) as count2, sms_originating,
> length(sms_originating) as len from testing where length(sms_originating) = 0
> group by sms_originating limit 20;
>
> select count(*), count(sms_originating) from testing where
> length(sms_originating) = 0;
>
> select sum(count1), sum(count2) from ( select count(sms_originating) as count1,
> count(*) as count2 from testing group by sms_originating ) as t;
>
>
> Sharing your data off the list / bug tracker is fine. Whether email is the
> feasible way depends on how large the (compressed) .CSV file for the bulkload
> is.
## Comment 16882
Date: 2012-02-14 15:49:30 +0100
From: Josip <<josip.djuricic>>
I can confirm that with Release 11.7.5 this seems to be solved.
Thank you
| wrong results when using group by on large dataset | https://api.github.com/repos/MonetDB/MonetDB/issues/2951/comments | 0 | 2020-11-30T11:49:04Z | 2024-06-27T11:54:16Z | https://github.com/MonetDB/MonetDB/issues/2951 | 753,410,619 | 2,951 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-19 12:18:56 +0100
From: robertminer1222
To: clients devs <<bugs-clients>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-26 15:32:08 +0100
## Comment 16658
Date: 2011-12-19 12:18:56 +0100
From: robertminer1222
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
ResultSets that need String to Date parsing, resulting from queries executed in concurrent sessions, trigger exceptions such as the one below. Possibly this is because of the use of static SimpleDateFormat variables in the MonetResultSet class, which is not thread safe.
Stack trace:
java.lang.NumberFormatException: For input string: "2000099.E2000099E44"
at sun.misc.FloatingDecimal.readJavaFormatString(FloatingDecimal.java:1222) ~[na:1.6.0_26]
at java.lang.Double.parseDouble(Double.java:510) ~[na:1.6.0_26]
at java.text.DigitList.getDouble(DigitList.java:151) ~[na:1.6.0_26]
at java.text.DecimalFormat.parse(DecimalFormat.java:1303) ~[na:1.6.0_26]
at java.text.SimpleDateFormat.subParse(SimpleDateFormat.java:1591) ~[na:1.6.0_26]
at java.text.SimpleDateFormat.parse(SimpleDateFormat.java:1312) ~[na:1.6.0_26]
at nl.cwi.monetdb.jdbc.MonetResultSet.getJavaDate(MonetResultSet.java:1666) ~[monetdb-jdbc-1.22.jar:na]
at nl.cwi.monetdb.jdbc.MonetResultSet.getDate(MonetResultSet.java:1788) ~[monetdb-jdbc-1.22.jar:na]
at nl.cwi.monetdb.jdbc.MonetResultSet.getDate(MonetResultSet.java:1769) ~[monetdb-jdbc-1.22.jar:na]
at nl.cwi.monetdb.jdbc.MonetResultSet.getObject(MonetResultSet.java:1376) ~[monetdb-jdbc-1.22.jar:na]
at nl.cwi.monetdb.jdbc.MonetResultSet.getObject(MonetResultSet.java:1338) ~[monetdb-jdbc-1.22.jar:na]
Reproducible: Sometimes
## Comment 16659
Date: 2011-12-20 10:13:05 +0100
From: @grobian
for my understanding, are you using multiple threads on a single ResultSet, or just multiple threads on each its own ResultSet?
## Comment 16662
Date: 2011-12-20 15:17:06 +0100
From: robertminer1222
Thanks for following up on this possible issue.
It's the latter: (multiple threads on each its own ResultSet: each session opens its own JDBC connection, and closes it when finished. All of the session's DB work is executed in a single thread.
## Comment 16663
Date: 2011-12-20 16:00:12 +0100
From: @grobian
Thank you. For that case the simplest solution is to drop the 'static'. I guess the performance penalty incurred by that can be neglected.
## Comment 16664
Date: 2011-12-20 16:09:28 +0100
From: @grobian
Changeset [5cc6927f19df](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5cc6927f19df) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5cc6927f19df](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5cc6927f19df)
Changeset description:
ResultSet: make date parsers instance local
It was nice for performance to share the same date parsers between all
instances of MonetResultSet, but it caused problems in concurrent use,
so better have them instance local. With some luck, Java does lazy/jit
instantiation, which means we still don't pay if we don't use
date-related getters. Solves bug #2950.
## Comment 16665
Date: 2011-12-20 16:10:18 +0100
From: @grobian
Fix will be part of the Dec2011 feature release.
## Comment 16681
Date: 2011-12-22 12:28:20 +0100
From: @grobian
If possible, could you please try the driver from
http://monetdb.cwi.nl/testweb/web/42598:ef01bf784b71/Java6/monetdb-jdbc-2.0.jar
or
http://monetdb.cwi.nl/testweb/web/42598:ef01bf784b71/Java7/monetdb-jdbc-2.0.jar
to verify if this fix is sufficient for your scenario? Thanks!
## Comment 16682
Date: 2011-12-22 18:34:31 +0100
From: robertminer1222
(In reply to comment 6)
Works like a charm. Great response time, Fabian.
Thanks a lot!
## Comment 16836
Date: 2012-01-26 15:32:08 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| MonetResultSet use of SimpleDateFormat not threadsafe | https://api.github.com/repos/MonetDB/MonetDB/issues/2950/comments | 0 | 2020-11-30T11:49:01Z | 2024-06-27T11:54:15Z | https://github.com/MonetDB/MonetDB/issues/2950 | 753,410,591 | 2,950 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-18 13:28:16 +0100
From: Babis <<charnik>>
To: SQL devs <<bugs-sql>>
Version: 11.9.1 (Apr2012) [obsolete]
CC: @njnes
Last updated: 2012-05-25 12:58:47 +0200
## Comment 16656
Date: 2011-12-18 13:28:16 +0100
From: Babis <<charnik>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/534.52.7 (KHTML, like Gecko) Version/5.1.2 Safari/534.52.7
Build Identifier:
The user defines a function, F, that returns a table. The function does nothing but returns a relational table which is constructed from a simple SQL select query, Q. Although query Q is correct, using function F in another SQL query produces an SQL exception stating that the program contains errors.
Reproducible: Always
### Steps to Reproduce:
1. mclient -d test
2. sql> create table r (u varchar(32), v varchar(32), r varchar(5));
2. sql> CREATE FUNCTION refine ()
RETURNS TABLE (u varchar(32), v varchar(32), r varchar(5))
BEGIN
RETURN TABLE (
( SELECT R.u, R.v, R.r FROM R
INTERSECT
SELECT R.u, R.v, R.r FROM R
)
);
END;
3. sql> select * from refine();
### Actual Results:
SyntaxException:user.refine[0]:RETURN missing
TypeException:user.s4_1[5]:'user.refine' undefined in: (_7:bat[:oid,:str], _8:bat[:oid,:str], _9:bat[:oid,:str]) := user.refine()
SQLException:SQLengine:Program contains errors
### Expected Results:
Expect to get an empty relation such as the following:
+---+---+---+
| u | v | r |
+===+===+===+
+---+---+---+
0 tuples (3.155ms)
Tested on Ubuntu 10.04 and Mac OS X 10.7.2 (Snow Leopard) for the Aug2011, Dec2011, and development branches. It works as expected only in the Aug2011 branch.
## Comment 16657
Date: 2011-12-18 14:51:20 +0100
From: Babis <<charnik>>
Created attachment 90
SQL exception in a while clause
I also attach a similar exception. The .sql file defines two functions, the first (refine) is the same as the one reported initially in this bug report. The second (trans) makes use of this function in a while clause. If you comment out the while statement, things are OK. If you leave the while loop and comment out the "INSERT INTO" statement, things are again OK. But if you leave them all (as it is in the attached .sql script), then you get the following exception:
sql>\< bug.sql
operation successful
operation successful
operation successful
operation successful
sql>select * from trans();
SyntaxException:user.s24_1[151]:barrier '_32' without exit in s24_1[152]
SQLException:SQLengine:Program contains errors
I have tested this in the Aug2011 branch, because the other branches fail (as reported initially in this but report) even in the case that you execute a "select * from refine()" query.
> Attached file: [bug.sql](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2949_bug.sql_90) (application/octet-stream, 707 bytes)
> Description: SQL exception in a while clause
## Comment 17090
Date: 2012-03-16 14:52:29 +0100
From: Babis <<charnik>>
(In reply to comment 1)
> Created attachment 90 [details]
> SQL exception in a while clause
>
> I also attach a similar exception. The .sql file defines two functions, the
> first (refine) is the same as the one reported initially in this bug report.
> The second (trans) makes use of this function in a while clause. If you comment
> out the while statement, things are OK. If you leave the while loop and comment
> out the "INSERT INTO" statement, things are again OK. But if you leave them all
> (as it is in the attached .sql script), then you get the following exception:
>
> sql>\< bug.sql
> operation successful
> operation successful
> operation successful
> operation successful
> sql>select * from trans();
> SyntaxException:user.s24_1[151]:barrier '_32' without exit in s24_1[152]
> SQLException:SQLengine:Program contains errors
>
> I have tested this in the Aug2011 branch, because the other branches fail (as
> reported initially in this but report) even in the case that you execute a
> "select * from refine()" query.
Just to add that the bug is still active for the latest changeset of Dec2011. Time permitting I will come back with a test.
## Comment 17233
Date: 2012-05-07 20:27:25 +0200
From: @njnes
fixed by removing incorrect chkProgram during optimizing. Should only be done at end of each optimizer (with a full program)
## Comment 17234
Date: 2012-05-07 20:34:59 +0200
From: @njnes
Changeset [6c699953279b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6c699953279b) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6c699953279b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6c699953279b)
Changeset description:
fixed bug #2949. Checking program (types/declaration/flow) with half complete
programs fails. So only check at the end of the optimizer.
| SQL exception during execution of a correct user defined function that just returns a relation table constructed by a simple SQL select query | https://api.github.com/repos/MonetDB/MonetDB/issues/2949/comments | 0 | 2020-11-30T11:48:58Z | 2024-06-27T11:54:14Z | https://github.com/MonetDB/MonetDB/issues/2949 | 753,410,543 | 2,949 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-16 13:08:57 +0100
From: @grobian
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
CC: renalkoclok, @yzchang
Last updated: 2020-06-03 09:16:04 +0200
## Comment 16655
Date: 2011-12-16 13:08:57 +0100
From: @grobian
We should document some general info about when MonetDB (GDK) creates and destroys indices.
Next we should write something about the relation of the SQL frontend and indices. Do's and don't for creating primary, unique and foreign keys, and what their possible effect on performance is.
## Comment 16750
Date: 2012-01-06 13:25:20 +0100
From: @grobian
this is a test
## Comment 16751
Date: 2012-01-06 13:29:59 +0100
From: @grobian
test 2
## Comment 16752
Date: 2012-01-06 13:30:54 +0100
From: @grobian
test 3
## Comment 18114
Date: 2012-11-27 15:23:43 +0100
From: @yzchang
No test needed for documentation issues
## Comment 18628
Date: 2013-03-20 17:28:32 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [cd3000a6ee6b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=cd3000a6ee6b) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=cd3000a6ee6b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=cd3000a6ee6b)
Changeset description:
Dummy change to test out mercurial-bugzilla interface.
This change is not relevant to bug #2948, but should trigger a message
there anyway.
## Comment 19804
Date: 2014-05-06 16:19:58 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [00e4e1d0da53](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=00e4e1d0da53) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=00e4e1d0da53](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=00e4e1d0da53)
Changeset description:
Dummy change to test out mercurial-bugzilla interface.
This change is not relevant to bug #2948, but should trigger a message
there anyway.
## Comment 27770
Date: 2020-06-03 09:16:04 +0200
From: @sjoerdmullender
The content of attachment 671 has been deleted for the following reason:
spam
| Document MonetDB's usage of indices | https://api.github.com/repos/MonetDB/MonetDB/issues/2948/comments | 0 | 2020-11-30T11:48:55Z | 2024-06-28T13:19:24Z | https://github.com/MonetDB/MonetDB/issues/2948 | 753,410,514 | 2,948 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-16 09:31:36 +0100
From: bug.monetdb.org.snap.348
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @njnes
Last updated: 2012-03-16 14:56:51 +0100
## Comment 16654
Date: 2011-12-16 09:31:36 +0100
From: bug.monetdb.org.snap.348
User-Agent: Mozilla/5.0 (Ubuntu; X11; Linux x86_64; rv:8.0) Gecko/20100101 Firefox/8.0
Build Identifier:
If a column has an indexed varchar column and a query is issued with "is null" on the indexed column the query does not return the correct result. This only happens if there are further clauses in the query.
For example, these queries work as expected:
select * from c__has_a where owner_table is null and property_table='SIMPLE';
select * from c__has_a where owner_table is null and property_id = 4;
select * from c__has_a where owner_table is null;
But this fails:
select * from c__has_a where owner_table is null and property_table='SIMPLE' and property_id = 4;
Reproducible: Always
### Steps to Reproduce:
1. Enter following into database:
START TRANSACTION;
SET SCHEMA "sys";
CREATE TABLE "sys"."c__has_a" (
"owner_table" VARCHAR(32672),
"owner_id" BIGINT,
"relation_name" VARCHAR(32672),
"property_table" VARCHAR(32672),
"property_id" BIGINT,
"property_class" VARCHAR(32672)
);
CREATE INDEX "c__has_a_property_index" ON "sys"."c__has_a" ("property_table", "property_id");
COPY 1 RECORDS INTO "sys"."c__has_a" FROM stdin USING DELIMITERS '\t','\n','"';
NULL NULL NULL "SIMPLE" 4 "simpleclass"
COMMIT;
2. Then execute this query:
select * from c__has_a where owner_table is null and property_table='SIMPLE' and property_id = 4;
### Actual Results:
No rows returned.
### Expected Results:
One row returned.
Dropping the index fixes the problem.
## Comment 16660
Date: 2011-12-20 11:32:38 +0100
From: bug.monetdb.org.snap.348
The bug manifests if the size of the indexed columns are reduced down to 8.
## Comment 16661
Date: 2011-12-20 11:38:22 +0100
From: bug.monetdb.org.snap.348
Simplifying the table does not help:
CREATE TABLE FOO (
"a" VARCHAR(8),
"b" VARCHAR(8),
"c" BIGINT);
CREATE INDEX FOO_INDEX ON FOO ("b","c");
insert into foo (b,c)values('foo',3);
select * from foo where a is null and b='foo' and c=3;
## Comment 17013
Date: 2012-02-27 21:59:44 +0100
From: @njnes
fixed, first select based on hash, then handle the other select expressions
## Comment 17014
Date: 2012-02-27 22:10:55 +0100
From: @njnes
Changeset [1a5795e15129](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=1a5795e15129) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1a5795e15129](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=1a5795e15129)
Changeset description:
fixed bug #2947
properly handle hash based selection before the other
select expressions
## Comment 17091
Date: 2012-03-16 14:56:51 +0100
From: @grobian
Released in Dec2011-SP2
| Queries where a "is null" clause is combined with two equals clauses fail to produce the expected result. | https://api.github.com/repos/MonetDB/MonetDB/issues/2947/comments | 0 | 2020-11-30T11:48:52Z | 2024-06-27T11:54:12Z | https://github.com/MonetDB/MonetDB/issues/2947 | 753,410,492 | 2,947 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-15 16:47:52 +0100
From: George Garbis <<ggarbis>>
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @drstmane
Last updated: 2012-02-29 10:34:28 +0100
## Comment 16653
Date: 2011-12-15 16:47:52 +0100
From: George Garbis <<ggarbis>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.1 (KHTML, like Gecko) Ubuntu/11.04 Chromium/14.0.835.202 Chrome/14.0.835.202 Safari/535.1
Build Identifier:
A left join which have only right table columns in join condition seems to produce invalid bat.
Assume we have the tables below:
CREATE TABLE tableA(x integer, y integer);
INSERT INTO tableA values(1,10);
INSERT INTO tableA values(2,20);
CREATE TABLE tableB(x integer, y integer);
INSERT INTO tableB values(1,-10);
INSERT INTO tableB values(2,-20);
and pose the query:
SELECT A.x, A.y, B.x, B.y
FROM tableA A
LEFT JOIN tableB B ON ( B.y < -10 ) ;
the expected results are:
+-------+------+-------+-------+
| x | y | x | y |
+=======+======+=======+=======+
| 1 | 10 | 2 | -20 |
| 2 | 20 | 2 | -20 |
+-------+------+-------+-------+
but mclient prints this message:
TypeException:user.s1_1[36]:'algebra.kdifference' undefined in: _55:any := algebra.kdifference(_54:bat[:oid,:int], _44:bat[:void,:oid])
TypeException:user.s1_1[37]:'algebra.markT' undefined in: _56:any := algebra.markT(_55:any, _21:oid)
TypeException:user.s1_1[38]:'bat.reverse' undefined in: _57:any := bat.reverse(_56:any)
TypeException:user.s1_1[39]:'algebra.join' undefined in: _58:any := algebra.join(_57:any, _20:bat[:oid,:int])
TypeException:user.s1_1[41]:'algebra.leftjoin' undefined in: _60:any := algebra.leftjoin(_59:bat[:oid,:void], _54:bat[:oid,:int])
TypeException:user.s1_1[43]:'bat.append' undefined in: _64:any := bat.append(_61:bat[:oid,:int], _60:any, _65:bit)
TypeException:user.s1_1[44]:'bat.append' undefined in: _66:any := bat.append(_64:any, _58:any, _65:bit)
TypeException:user.s1_1[55]:'algebra.join' undefined in: _80:any := algebra.join(_57:any, _79:bat[:oid,:int])
TypeException:user.s1_1[61]:'algebra.leftjoin' undefined in: _90:any := algebra.leftjoin(_59:bat[:oid,:void], _89:bat[:oid,:int])
TypeException:user.s1_1[63]:'bat.append' undefined in: _92:any := bat.append(_91:bat[:oid,:int], _90:any, _65:bit)
TypeException:user.s1_1[64]:'bat.append' undefined in: _93:any := bat.append(_92:any, _80:any, _65:bit)
TypeException:user.s1_1[70]:'algebra.project' undefined in: _102:any := algebra.project(_57:any, _100:int)
TypeException:user.s1_1[83]:'bat.append' undefined in: _119:any := bat.append(_117:bat[:oid,:int], _102:any, _118:bit)
TypeException:user.s1_1[89]:'algebra.project' undefined in: _126:any := algebra.project(_57:any, _125:int)
TypeException:user.s1_1[96]:'bat.append' undefined in: _133:any := bat.append(_132:bat[:oid,:int], _126:any, _118:bit)
SQLException:SQLengine:Program contains errors
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE tableA(x integer, y integer);
2. INSERT INTO tableA values(1,10);
3. INSERT INTO tableA values(2,20);
4. CREATE TABLE tableB(x integer, y integer);
5. INSERT INTO tableB values(1,-10);
6. INSERT INTO tableB values(2,-20);
7. SELECT A.x, A.y, B.x, B.y FROM tableA A LEFT JOIN tableB B ON ( B.y < -10 ) ;
### Actual Results:
TypeException:user.s1_1[36]:'algebra.kdifference' undefined in: _55:any := algebra.kdifference(_54:bat[:oid,:int], _44:bat[:void,:oid])
TypeException:user.s1_1[37]:'algebra.markT' undefined in: _56:any := algebra.markT(_55:any, _21:oid)
TypeException:user.s1_1[38]:'bat.reverse' undefined in: _57:any := bat.reverse(_56:any)
TypeException:user.s1_1[39]:'algebra.join' undefined in: _58:any := algebra.join(_57:any, _20:bat[:oid,:int])
TypeException:user.s1_1[41]:'algebra.leftjoin' undefined in: _60:any := algebra.leftjoin(_59:bat[:oid,:void], _54:bat[:oid,:int])
TypeException:user.s1_1[43]:'bat.append' undefined in: _64:any := bat.append(_61:bat[:oid,:int], _60:any, _65:bit)
TypeException:user.s1_1[44]:'bat.append' undefined in: _66:any := bat.append(_64:any, _58:any, _65:bit)
TypeException:user.s1_1[55]:'algebra.join' undefined in: _80:any := algebra.join(_57:any, _79:bat[:oid,:int])
TypeException:user.s1_1[61]:'algebra.leftjoin' undefined in: _90:any := algebra.leftjoin(_59:bat[:oid,:void], _89:bat[:oid,:int])
TypeException:user.s1_1[63]:'bat.append' undefined in: _92:any := bat.append(_91:bat[:oid,:int], _90:any, _65:bit)
TypeException:user.s1_1[64]:'bat.append' undefined in: _93:any := bat.append(_92:any, _80:any, _65:bit)
TypeException:user.s1_1[70]:'algebra.project' undefined in: _102:any := algebra.project(_57:any, _100:int)
TypeException:user.s1_1[83]:'bat.append' undefined in: _119:any := bat.append(_117:bat[:oid,:int], _102:any, _118:bit)
TypeException:user.s1_1[89]:'algebra.project' undefined in: _126:any := algebra.project(_57:any, _125:int)
TypeException:user.s1_1[96]:'bat.append' undefined in: _133:any := bat.append(_132:bat[:oid,:int], _126:any, _118:bit)
SQLException:SQLengine:Program contains errors
### Expected Results:
+-------+------+-------+-------+
| x | y | x | y |
+=======+======+=======+=======+
| 1 | 10 | 2 | -20 |
| 2 | 20 | 2 | -20 |
+-------+------+-------+-------+
## Comment 16973
Date: 2012-02-22 23:51:03 +0100
From: @drstmane
Since this left join is a Cartesian product, consider this work around:
SELECT A.x, A.y, B.x, B.y FROM tableA A, ( select * from tableB B where B.y < -10 ) as B;
## Comment 16978
Date: 2012-02-23 14:44:24 +0100
From: @njnes
Changeset [86d94f21e92c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=86d94f21e92c) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=86d94f21e92c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=86d94f21e92c)
Changeset description:
fixed bugs 3034, 2964 2946 and 3017
The rel2bin_join didn't handle a bunch of corner cases
| Invalid BAT if left join condition has only columns from right table | https://api.github.com/repos/MonetDB/MonetDB/issues/2946/comments | 0 | 2020-11-30T11:48:50Z | 2024-06-27T11:54:11Z | https://github.com/MonetDB/MonetDB/issues/2946 | 753,410,463 | 2,946 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-08 18:34:48 +0100
From: @drstmane
To: MonetDB5 devs <<bugs-monetdb5>>
Version: -- development
CC: @njnes, @yzchang
Last updated: 2014-10-31 14:14:17 +0100
## Comment 16640
Date: 2011-12-08 18:34:48 +0100
From: @drstmane
[For now, I file this as a feature request; it might be considered a (performance) bug, though ...]
There appear to be no bulk implementations of binary MIN(a,b) / MAX(a,b) (e.g., in module batcalc).
Consequently, the translation of SQL queries using "between SYMMETRIC" is sub-optimal, using BAT iterators in MAL, rather than an efficient bulk implementation of binary MIN(a,b) / MAX(a,b).
Cf.:
$ mclient -s 'create table T (a int, b int, c int);'
operation successful
$ mclient -s 'PLAN select * from T where a between b and c;'
+------------------------------------------------------+
| rel |
+======================================================+
| project ( |
| | select ( |
| | | table(sys.t) [ t.a, t.b, t.c, t.%TID% NOT NULL ] |
| | ) [ t.b <= t.a <= t.c ] |
| ) [ t.a, t.b, t.c ] |
+------------------------------------------------------+
5 tuples
$ mclient -s 'EXPLAIN select * from T where a between b and c;' | grep -9 -i iterator
$ mclient -s 'PLAN select * from T where a between SYMMETRIC b and c;'
+---------------------------------------------------------------+
| rel |
+===============================================================+
| project ( |
| | select ( |
| | | table(sys.t) [ t.a, t.b, t.c, t.%TID% NOT NULL ] |
| | ) [ sys.sql_min(t.b, t.c) <= t.a <= sys.sql_max(t.b, t.c) ] |
| ) [ t.a, t.b, t.c ] |
+---------------------------------------------------------------+
5 tuples
$ mclient -s 'EXPLAIN select * from T where a between SYMMETRIC b and c;' | grep -9 -i iterator
| X_36 := algebra.kunion(X_35,X_27); |
| X_37 := algebra.kdifference(X_36,X_16); |
| X_10:bat[:oid,:int] := sql.bind(X_2,"sys","t","a",0); |
| X_8:bat[:oid,:int] := sql.bind(X_2,"sys","t","a",2); |
| X_12 := algebra.kdifference(X_10,X_8); |
| X_13 := algebra.kunion(X_12,X_8); |
| X_5:bat[:oid,:int] := sql.bind(X_2,"sys","t","a",1); |
| X_14 := algebra.kunion(X_13,X_5); |
| X_156 := bat.new(nil:oid,nil:int); |
| barrier (X_159,X_160,X_161) := bat.newIterator(X_24); |
| X_163 := algebra.find(X_37,X_160); |
| X_165 := calc.max(X_161,X_163); |
| bat.insert(X_156,X_160,X_165); |
| redo (X_159,X_160,X_161) := bat.hasMoreElements(X_24); |
| exit (X_159,X_160,X_161); |
| X_38:bat[:oid,:int] := X_156; |
| X_68 := bat.mirror(X_10); |
| X_42 := bat.mirror(X_38); |
| X_41 := bat.mirror(X_5); |
--
| X_49 := algebra.join(X_44,X_38); |
| X_50:bat[:oid,:bit] := batcalc.<=(X_48,X_49); |
| X_51 := algebra.uselect(X_50,true); |
| X_53 := bat.reverse(X_51); |
| X_54 := algebra.join(X_53,X_44); |
| X_55 := bat.reverse(X_54); |
| X_83 := algebra.kunion(X_82,X_55); |
| X_84 := algebra.kdifference(X_83,X_16); |
| X_170 := bat.new(nil:oid,nil:int); |
| barrier (X_172,X_173,X_174) := bat.newIterator(X_24); |
| X_176 := algebra.find(X_37,X_173); |
| X_178 := calc.min(X_174,X_176); |
| bat.insert(X_170,X_173,X_178); |
| redo (X_172,X_173,X_174) := bat.hasMoreElements(X_24); |
| exit (X_172,X_173,X_174); |
| X_24 := nil:BAT; |
| X_37 := nil:BAT; |
| X_85:bat[:oid,:int] := X_170; |
| X_87 := bat.mirror(X_85); |
## Comment 18080
Date: 2012-11-27 14:25:10 +0100
From: @yzchang
No test for performance issues
## Comment 20363
Date: 2014-10-31 14:14:17 +0100
From: @sjoerdmullender
Oct2014 has been released.
| evaluation of SQL "between SYMMETRIC" requires MAL iterator because there is no (bulk) MIN/MAX(a,b) | https://api.github.com/repos/MonetDB/MonetDB/issues/2945/comments | 0 | 2020-11-30T11:48:47Z | 2024-06-27T11:54:10Z | https://github.com/MonetDB/MonetDB/issues/2945 | 753,410,434 | 2,945 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-07 14:06:19 +0100
From: Viktor Rosenfeld <<24hesk>>
To: clients devs <<bugs-clients>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
Last updated: 2012-01-26 15:31:58 +0100
## Comment 16612
Date: 2011-12-07 14:06:19 +0100
From: Viktor Rosenfeld <<24hesk>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_6_8) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
Connecting to database with stethoscope using the -d option results in an error message.
Reproducible: Always
### Steps to Reproduce:
1. start monetdbd process
2. run "stethoscope -d ... -u ... -P ..." with valid connection details
### Actual Results:
Stethoscope exits with the following error message:
stethoscope: no databases found for '...'
## Comment 16622
Date: 2011-12-07 17:17:26 +0100
From: @grobian
confirmed on Linux, i.e. it's not platform specific.
## Comment 16623
Date: 2011-12-07 17:36:46 +0100
From: @grobian
Changeset [ed417790216c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ed417790216c) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ed417790216c](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=ed417790216c)
Changeset description:
stethoscope: fix connecting without monetdbd
Make sure we can connect directly to mservers, and that we don't print
garbage if making a connection actually failed. Fixes bug #2944.
## Comment 16624
Date: 2011-12-07 17:37:09 +0100
From: @grobian
Thanks for reminding me. Fixed for the Dec2011 release.
## Comment 16819
Date: 2012-01-26 15:31:58 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| stethoscope on can't connect to database without monetdbd (discovery enabled) | https://api.github.com/repos/MonetDB/MonetDB/issues/2944/comments | 0 | 2020-11-30T11:48:44Z | 2024-06-27T11:54:09Z | https://github.com/MonetDB/MonetDB/issues/2944 | 753,410,406 | 2,944 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 21:43:49 +0100
From: Shantanu Kumar <<shantanu_k06>>
To: clients devs <<bugs-clients>>
Version: 11.9.7 (Apr2012-SP2) [obsolete]
Last updated: 2012-07-17 15:34:13 +0200
## Comment 16605
Date: 2011-12-06 21:43:49 +0100
From: Shantanu Kumar <<shantanu_k06>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
MonetDB being Open Source database doesn't have the the JDBC drivers uploaded to a Maven repository. This causes hassle for projects that fetch dependencies from the Internet.
Please consider uploading JDBC drivers to Maven Central (or suitable repository) as and when they are released.
Reproducible: Always
### Steps to Reproduce:
1. Try to use MonetDB JDBC driver as a Maven dependency.
2.
3.
### Actual Results:
JDBC driver not found on any Maven repository.
### Expected Results:
JDBC driver should be available on Maven Central (or an appropriate Maven repo.)
## Comment 16606
Date: 2011-12-06 21:57:37 +0100
From: @grobian
can you give an example?
## Comment 16609
Date: 2011-12-07 08:57:02 +0100
From: Shantanu Kumar <<shantanu_k06>>
(In reply to comment 1)
> can you give an example?
Examples below:
The Maven coordinates for various versions of MySQL can be found here:
http://mvnrepository.com/artifact/mysql/mysql-connector-java
Similarly, Maven coordinates for MckoiDDB v1.0.2 are here:
http://mvnrepository.com/artifact/mckoi/mckoi/1.0.2
Typically Maven Central http://search.maven.org/ is the place where all Open Source projects upload their project JARs. However, uploading artifacts to Maven Central can be fairly involved. Hence, another simpler option is to upload to Clojars http://clojars.org/ -- that is what the Open Source CUBRID database uses too:
http://www.cubrid.org/cubrid_java_programming
http://clojars.org/cubrid/cubrid-jdbc
Fabian Groffen: Does that answer your question?
## Comment 16614
Date: 2011-12-07 14:41:05 +0100
From: @grobian
(In reply to comment 2)
> Fabian Groffen: Does that answer your question?
Yes it does. Thank you. We're completely unknown with this system. I looked at clojars, but it doesn't really look like a place to upload anything to, but rather a database system of itself?
## Comment 16619
Date: 2011-12-07 15:51:04 +0100
From: Shantanu Kumar <<shantanu_k06>>
(In reply to comment 3)
> (In reply to comment 2)
> > Fabian Groffen: Does that answer your question?
>
> Yes it does. Thank you. We're completely unknown with this system. I looked
> at clojars, but it doesn't really look like a place to upload anything to, but
> rather a database system of itself?
Okay, I will be happy to help. I am explaining the steps below to upload artifacts to Clojars. I am going to assume you are using some variant of UNIX. (If you are on Windows you may like to use TortoiseGit's built-in SSH and SCP commands instead of the OpenSSH package.) The example below is for MonetDB JDBC version 1.19 -- please use appropriate version number as necessary in your case.
1. First, you should have your SSH public key created and handy (one-time activity):
http://rcsg-gsir.imsb-dsgi.nrc-cnrc.gc.ca/documents/internet/node31.html
2. Then you will need to register on Clojars (one-time activity):
http://clojars.org/register
3. You will have to create a pom.xml file for the JDBC JAR file, which should look as follows (one-time activity):
(When doing it for the first time, you may like to consider renaming the file to monetdb-jdbc-1.19.jar as a standard practice, and edit the `artifactId` as monetdb-jdbc and `version` as 1.19 instead of 1.19-jdbc. This is the style that Maven recommends though it allows other styles too.)
<?xml version="1.0" encoding="UTF-8"?>
<project xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd" xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<modelVersion>4.0.0</modelVersion>
<groupId>monetdb</groupId>
<artifactId>monetdb</artifactId>
<version>1.19-jdbc</version>
<name>monetdb</name>
<description>MonetDB JDBC driver</description>
<repositories>
<repository>
<id>central</id>
<url>http://repo1.maven.org/maven2</url>
</repository>
<repository>
<id>clojars</id>
<url>http://clojars.org/repo/</url>
</repository>
</repositories>
</project>
3a. You should update the version of the JDBC driver in pom.xml file every time you want to upload a new version of the JAR.
4. To upload the JDBC driver JAR to Clojars, run the shell command below:
$ scp pom.xml monetdb-1.19-jdbc.jar clojars@clojars.org:
(Note that there is a colon after `clojars.org` hostname.)
I hope this was clear. Should you need more details or explanation please let me know.
## Comment 16620
Date: 2011-12-07 16:55:24 +0100
From: @grobian
ok, thank you very much!
I'll look if we can make this update part of our release process.
## Comment 16632
Date: 2011-12-08 15:01:28 +0100
From: @grobian
Changeset [fa7a8c7e896b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=fa7a8c7e896b) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa7a8c7e896b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=fa7a8c7e896b)
Changeset description:
JDBC: change naming scheme to something more standard
As noted in bug #2943, we use a rather unusal naming scheme for our JDBC
driver. It was based on what the PostgreSQL folks did.
With the change from a JDBCv3 to a JDBCv4 driver, we bumped the major
version, and also take the opportunity to change the naming scheme into
something more common: monetdb-jdbc-2.0.jar.
## Comment 16633
Date: 2011-12-08 15:28:07 +0100
From: @grobian
Aparently it is as simple as it looks:
https://clojars.org/monetdb/monetdb-jdbc
## Comment 16642
Date: 2011-12-08 20:19:13 +0100
From: Shantanu Kumar <<shantanu_k06>>
(In reply to comment 7)
> Aparently it is as simple as it looks:
> https://clojars.org/monetdb/monetdb-jdbc
That's awesome - thanks! You may like to put up the Clojars link (mentioning "Maven Dependency") on the Downloads page for everybody to know:
http://www.monetdb.org/Downloads
## Comment 16643
Date: 2011-12-08 20:21:26 +0100
From: @grobian
yup, I'll keep this bug open until we sorted that out
## Comment 17493
Date: 2012-07-17 15:34:13 +0200
From: @grobian
It's now on http://dev.monetdb.org/downloads/Java/Latest/
| Maven coordinates for MonetDB JDBC driver | https://api.github.com/repos/MonetDB/MonetDB/issues/2943/comments | 0 | 2020-11-30T11:48:41Z | 2024-06-27T11:54:08Z | https://github.com/MonetDB/MonetDB/issues/2943 | 753,410,380 | 2,943 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 20:30:50 +0100
From: Babis <<charnik>>
To: buildtools devs <<bugs-buildtools>>
Version: -- development
CC: a_lurie, @drstmane, @yzchang
Last updated: 2020-04-16 19:04:13 +0200
## Comment 16602
Date: 2011-12-06 20:30:50 +0100
From: Babis <<charnik>>
Created attachment 86
Log for configuration
Configuration is done as follows:
../MonetDB/configure --prefix=/Users/charnik/apps/MonetDB_default --enable-sql --enable-monetdb5 --disable-geom --disable-jdbc
Make fails with error: "cc1: warnings being treated as errors"
Attached there are the configuration and compilation logs. Configuring with —disable-strict, make succeeds.
> Attached file: [config.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2942_config.log_86) (application/octet-stream, 285797 bytes)
> Description: Log for configuration
## Comment 16603
Date: 2011-12-06 20:31:27 +0100
From: Babis <<charnik>>
Created attachment 87
Log for compilation
> Attached file: [make.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2942_make.log_87) (application/octet-stream, 18182 bytes)
> Description: Log for compilation
## Comment 16604
Date: 2011-12-06 20:34:31 +0100
From: @grobian
Not much we can do about a host system which deprecates still often used functions.
## Comment 16607
Date: 2011-12-06 22:10:01 +0100
From: @drstmane
Apparently, this compiler
i686-apple-darwin11-llvm-gcc-4.2 (GCC) 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2336.1.00)
on this system
Darwin Kernel Version 11.2.0: Tue Aug 9 20:54:00 PDT 2011; root:xnu-1699.24.8~1/RELEASE_X86_64
says
/bin/sh ../../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I../.. -I../../../MonetDB/common/utils -I../../../MonetDB/common/utils -DLIBMCRYPT -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wmissing-include-dirs -Wp,-D_FORTIFY_SOURCE=2 -D_REENTRANT -c -o libmcrypt_la-mcrypt.lo `test -f 'mcrypt.c' || echo '../../../MonetDB/common/utils/'`mcrypt.c
libtool: compile: gcc -DHAVE_CONFIG_H -I. -I../.. -I../../../MonetDB/common/utils -I../../../MonetDB/common/utils -DLIBMCRYPT -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wmissing-include-dirs -Wp,-D_FORTIFY_SOURCE=2 -D_REENTRANT -c ../../../MonetDB/common/utils/mcrypt.c -fno-common -DPIC -o .libs/libmcrypt_la-mcrypt.o
cc1: warnings being treated as errors
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_MD5Sum':
../../../MonetDB/common/utils/mcrypt.c:60: warning: 'MD5' is deprecated (declared at /usr/include/openssl/md5.h:116)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_SHA1Sum':
../../../MonetDB/common/utils/mcrypt.c:83: warning: 'SHA1' is deprecated (declared at /usr/include/openssl/sha.h:124)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_SHA224Sum':
../../../MonetDB/common/utils/mcrypt.c:106: warning: 'SHA224' is deprecated (declared at /usr/include/openssl/sha.h:146)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_SHA256Sum':
../../../MonetDB/common/utils/mcrypt.c:133: warning: 'SHA256' is deprecated (declared at /usr/include/openssl/sha.h:150)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_SHA384Sum':
../../../MonetDB/common/utils/mcrypt.c:162: warning: 'SHA384' is deprecated (declared at /usr/include/openssl/sha.h:193)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_SHA512Sum':
../../../MonetDB/common/utils/mcrypt.c:195: warning: 'SHA512' is deprecated (declared at /usr/include/openssl/sha.h:197)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_RIPEMD160Sum':
../../../MonetDB/common/utils/mcrypt.c:233: warning: 'RIPEMD160' is deprecated (declared at /usr/include/openssl/ripemd.h:102)
../../../MonetDB/common/utils/mcrypt.c: In function 'mcrypt_hashPassword':
../../../MonetDB/common/utils/mcrypt.c:294: warning: 'RIPEMD160_Init' is deprecated (declared at /usr/include/openssl/ripemd.h:98)
../../../MonetDB/common/utils/mcrypt.c:295: warning: 'RIPEMD160_Update' is deprecated (declared at /usr/include/openssl/ripemd.h:99)
../../../MonetDB/common/utils/mcrypt.c:296: warning: 'RIPEMD160_Update' is deprecated (declared at /usr/include/openssl/ripemd.h:99)
../../../MonetDB/common/utils/mcrypt.c:297: warning: 'RIPEMD160_Final' is deprecated (declared at /usr/include/openssl/ripemd.h:100)
../../../MonetDB/common/utils/mcrypt.c:303: warning: 'SHA512_Init' is deprecated (declared at /usr/include/openssl/sha.h:194)
../../../MonetDB/common/utils/mcrypt.c:304: warning: 'SHA512_Update' is deprecated (declared at /usr/include/openssl/sha.h:195)
../../../MonetDB/common/utils/mcrypt.c:305: warning: 'SHA512_Update' is deprecated (declared at /usr/include/openssl/sha.h:195)
../../../MonetDB/common/utils/mcrypt.c:306: warning: 'SHA512_Final' is deprecated (declared at /usr/include/openssl/sha.h:196)
../../../MonetDB/common/utils/mcrypt.c:312: warning: 'SHA384_Init' is deprecated (declared at /usr/include/openssl/sha.h:190)
../../../MonetDB/common/utils/mcrypt.c:313: warning: 'SHA384_Update' is deprecated (declared at /usr/include/openssl/sha.h:191)
../../../MonetDB/common/utils/mcrypt.c:314: warning: 'SHA384_Update' is deprecated (declared at /usr/include/openssl/sha.h:191)
../../../MonetDB/common/utils/mcrypt.c:315: warning: 'SHA384_Final' is deprecated (declared at /usr/include/openssl/sha.h:192)
../../../MonetDB/common/utils/mcrypt.c:321: warning: 'SHA256_Init' is deprecated (declared at /usr/include/openssl/sha.h:147)
../../../MonetDB/common/utils/mcrypt.c:322: warning: 'SHA256_Update' is deprecated (declared at /usr/include/openssl/sha.h:148)
../../../MonetDB/common/utils/mcrypt.c:323: warning: 'SHA256_Update' is deprecated (declared at /usr/include/openssl/sha.h:148)
../../../MonetDB/common/utils/mcrypt.c:324: warning: 'SHA256_Final' is deprecated (declared at /usr/include/openssl/sha.h:149)
../../../MonetDB/common/utils/mcrypt.c:330: warning: 'SHA224_Init' is deprecated (declared at /usr/include/openssl/sha.h:143)
../../../MonetDB/common/utils/mcrypt.c:331: warning: 'SHA224_Update' is deprecated (declared at /usr/include/openssl/sha.h:144)
../../../MonetDB/common/utils/mcrypt.c:332: warning: 'SHA224_Update' is deprecated (declared at /usr/include/openssl/sha.h:144)
../../../MonetDB/common/utils/mcrypt.c:333: warning: 'SHA224_Final' is deprecated (declared at /usr/include/openssl/sha.h:145)
../../../MonetDB/common/utils/mcrypt.c:339: warning: 'SHA1_Init' is deprecated (declared at /usr/include/openssl/sha.h:121)
../../../MonetDB/common/utils/mcrypt.c:340: warning: 'SHA1_Update' is deprecated (declared at /usr/include/openssl/sha.h:122)
../../../MonetDB/common/utils/mcrypt.c:341: warning: 'SHA1_Update' is deprecated (declared at /usr/include/openssl/sha.h:122)
../../../MonetDB/common/utils/mcrypt.c:342: warning: 'SHA1_Final' is deprecated (declared at /usr/include/openssl/sha.h:123)
../../../MonetDB/common/utils/mcrypt.c:348: warning: 'MD5_Init' is deprecated (declared at /usr/include/openssl/md5.h:113)
../../../MonetDB/common/utils/mcrypt.c:349: warning: 'MD5_Update' is deprecated (declared at /usr/include/openssl/md5.h:114)
../../../MonetDB/common/utils/mcrypt.c:350: warning: 'MD5_Update' is deprecated (declared at /usr/include/openssl/md5.h:114)
../../../MonetDB/common/utils/mcrypt.c:351: warning: 'MD5_Final' is deprecated (declared at /usr/include/openssl/md5.h:115)
I'm afraid there's indeed not much we can do about that ...
## Comment 16608
Date: 2011-12-07 08:19:12 +0100
From: @grobian
the header file looks like this:
unsigned char *SHA384(const unsigned char *d, size_t n,unsigned char *md) DEPRECATED_IN_MAC_OS_X_VERSION_10_7_AND_LATER;
(That's an Apple local change.)
Perhaps passing -Wno-deprecated-declarations (in CFLAGS) helps?
-Wno-deprecated-declarations
Do not warn about uses of functions, variables, and types marked as
deprecated by using the "deprecated" attribute.
## Comment 16610
Date: 2011-12-07 09:29:24 +0100
From: Babis <<charnik>>
Created attachment 88
Log for configuration in Linux
> Attached file: [config_linux.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2942_config_linux.log_88) (application/octet-stream, 319496 bytes)
> Description: Log for configuration in Linux
## Comment 16611
Date: 2011-12-07 09:29:47 +0100
From: Babis <<charnik>>
Created attachment 89
Log for compilation
> Attached file: [make_linux.log](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2942_make_linux.log_89) (application/octet-stream, 102511 bytes)
> Description: Log for compilation
## Comment 16613
Date: 2011-12-07 14:38:36 +0100
From: @grobian
You already reported that one, and we already concluded that it's a ubuntu screwup, that we're not going to fix. Thank you.
## Comment 16615
Date: 2011-12-07 14:45:48 +0100
From: Babis <<charnik>>
(In reply to comment 7)
> You already reported that one, and we already concluded that it's a ubuntu
> screwup, that we're not going to fix. Thank you.
Nope, I haven't done this. Maybe it is another user. Which is the bug you are referring to?
## Comment 16616
Date: 2011-12-07 14:48:46 +0100
From: Babis <<charnik>>
(In reply to comment 8)
> (In reply to comment 7)
> > You already reported that one, and we already concluded that it's a ubuntu
> > screwup, that we're not going to fix. Thank you.
>
> Nope, I haven't done this. Maybe it is another user. Which is the bug you are
> referring to?
Hmm, just saw that my answer on this comment http://bugs.monetdb.org/show_bug.cgi?id=2942c4 hasn't been posted. Anyway, I was saying that I'll try your suggestion as soon as I get my hands on my mac.
## Comment 16621
Date: 2011-12-07 17:13:02 +0100
From: @grobian
my apologies, that was indeed someone else:
http://old.nabble.com/stack-protector-error-td32305603.html
## Comment 16625
Date: 2011-12-07 20:34:41 +0100
From: Babis <<charnik>>
(In reply to comment 4)
> the header file looks like this:
>
> unsigned char *SHA384(const unsigned char *d, size_t n,unsigned char *md)
> DEPRECATED_IN_MAC_OS_X_VERSION_10_7_AND_LATER;
>
> (That's an Apple local change.)
>
> Perhaps passing -Wno-deprecated-declarations (in CFLAGS) helps?
>
> -Wno-deprecated-declarations
> Do not warn about uses of functions, variables, and types marked as
> deprecated by using the "deprecated" attribute.
Congratulations Fabian, that made the trick. How do you plan to fix it? Including, maybe, the option -Wno-deprecated-declarations in case you are on Mac OS X with Snow Leopard or just by the exact compiler version (4.2.1)? Dodgy, eh? I sympathize with you about that...
## Comment 16626
Date: 2011-12-08 09:59:14 +0100
From: @grobian
We cannot/will not fix this, you'll have to do that manually if you want to use development sources and/or --enable-strict.
Release tarballs have exactly for reasons like these strictness not enabled by default.
## Comment 16627
Date: 2011-12-08 10:07:50 +0100
From: Babis <<charnik>>
(In reply to comment 12)
> We cannot/will not fix this, you'll have to do that manually if you want to use
> development sources and/or --enable-strict.
>
> Release tarballs have exactly for reasons like these strictness not enabled by
> default.
I see.
Thanks.
## Comment 27659
Date: 2020-04-14 17:58:55 +0200
From: Andrei L. <<a_lurie>>
Even though this was closed as "won't fix", I just wanted to document/update that this MacOS build issue still exists for those of us who use Mac to build MonetDB.
Hopefully this would be useful, if you encounter same errors:
Specifically, there are two errors on current Catalina 10.15.4:
1) Nullability warning results in the error:
In file included from /Library/Developer/CommandLineTools/SDKs/MacOSX10.15.sdk/usr/include/stdio.h:64:
/Library/Developer/CommandLineTools/SDKs/MacOSX10.15.sdk/usr/include/_stdio.h:93:16: error: pointer is missing a
nullability type specifier (_Nonnull, _Nullable, or _Null_unspecified) [-Werror,-Wnullability-completeness]
unsigned char *_base;
^
This can be resolved by adding -Wno-nullability-completeness to CFLAGS.
2) MD5 deprecation warning results in the error:
../../../monetdb/common/utils/mcrypt.c:80:2: error: 'CC_MD5_Init' is deprecated: first deprecated in macOS 10.15 - This
function is cryptographically broken and should not be used in security contexts. Clients should migrate to SHA256
(or stronger). [-Werror,-Wdeprecated-declarations]
This can be resolved by adding -Wno-deprecated-declarations to CFLAGS.
Because there are many Makefiles generated by MonetDB configuration, these changes should ideally be made inside the configuration script (e.g. I've added the changes in configuration script right before line: CFLAGS="$CFLAGS ${X_CFLAGS}" ), and then re-run the config which will regenerate Makefiles with these -W settings.
## Comment 27668
Date: 2020-04-16 16:57:35 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [e509350361fc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e509350361fc) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e509350361fc](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=e509350361fc)
Changeset description:
On newer Darwin, don't check for MD5: it gives a deprecation warning.
This should solve at least part of the latest problem mentioned in bug #2942.
## Comment 27670
Date: 2020-04-16 18:10:12 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [53fa1a67b5b3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=53fa1a67b5b3) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [https//devmonetdborg/hg/MonetDB?cmd=changeset;node=53fa1a67b5b3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=https//devmonetdborg/hg/MonetDB?cmd=changeset;node=53fa1a67b5b3)
Changeset description:
Don't warn about nullability incompleteness.
See bug #2942.
## Comment 27671
Date: 2020-04-16 18:20:51 +0200
From: @yzchang
Hai Andrei,
Thanks for bringing these problems back into attention. Can you please check if Sjoerd's fixes also work for you? I also use Catalina, but I only had problem with MD5, not Nullability.
## Comment 27672
Date: 2020-04-16 19:04:13 +0200
From: Andrei L. <<a_lurie>>
Hi Ying, and Sjoerd,
I tried and both MD5 and Nullability errors are solved.
Thank you very much for such quick response and fix!!!
| make fails with —enabled-strict (default behavior) | https://api.github.com/repos/MonetDB/MonetDB/issues/2942/comments | 0 | 2020-11-30T11:48:38Z | 2024-06-28T13:19:23Z | https://github.com/MonetDB/MonetDB/issues/2942 | 753,410,351 | 2,942 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 12:32:15 +0100
From: manuela <<m.sagona>>
To: clients devs <<bugs-clients>>
Version: -- development
CC: y.kargin, @yzchang
Last updated: 2016-04-11 11:41:43 +0200
## Comment 16591
Date: 2011-12-06 12:32:15 +0100
From: manuela <<m.sagona>>
User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
there is no support for transactions, tryng to
Reproducible: Always
### Steps to Reproduce:
1. execute
monetdb_query($db, 'START TRANSACTION');
### Actual Results:
causes the driver to interpret the empty server response as an error
(i.e. returns FALSE);
### Expected Results:
we would expect being able to execute a next 'COMMIT' or 'ROLLBACK' statement too
## Comment 16599
Date: 2011-12-06 13:35:38 +0100
From: @grobian
This would require addition of the functions monetdb_set_autocommit(boolean), monetdb_commit() and monetdb_rollback(), IMO.
Executing the statement "START TRANSACTION" is never going to work correctly.
## Comment 17656
Date: 2012-08-24 14:50:24 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
| please add transaction functions to php driver | https://api.github.com/repos/MonetDB/MonetDB/issues/2941/comments | 0 | 2020-11-30T11:48:36Z | 2024-06-28T13:40:47Z | https://github.com/MonetDB/MonetDB/issues/2941 | 753,410,325 | 2,941 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 12:29:35 +0100
From: manuela <<m.sagona>>
To: clients devs <<bugs-clients>>
Version: 11.3.3 (Apr2011-SP1) [obsolete]
Last updated: 2012-01-26 15:32:08 +0100
## Comment 16590
Date: 2011-12-06 12:29:35 +0100
From: manuela <<m.sagona>>
User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
PHP split() function is deprecated, but php_mapi.inc does a large use of it
Reproducible: Always
## Comment 16598
Date: 2011-12-06 13:33:26 +0100
From: @grobian
All our split usage is non-pattern wise, so we can use explode, like the patch below. Would you mind testing if that works correctly?
diff --git a/clients/php/lib/php_mapi.inc b/clients/php/lib/php_mapi.inc
--- a/clients/php/lib/php_mapi.inc
+++ b/clients/php/lib/php_mapi.inc
@@ -163,12 +163,12 @@
if ($row[1] == Q_TABLE) {
$operation = Q_TABLE;
// save info about the query
- $fields = split(" ", $row);
+ $fields = explode(" ", $row);
$handle["query"] = array("id" => $fields[1], "rows" => $fields[2], "cols" => $fields[3], "index" => $fields[4]);
} else if ($row[1] == Q_UPDATE) {
$operation = Q_UPDATE;
- $fields = split(" ", $row);
+ $fields = explode(" ", $row);
$handle["query"] = array("affected" => $fields[1]);
} else if ($row[1] == Q_CREATE) {
$operation = Q_CREATE;
@@ -222,13 +222,13 @@
//$parsed_rows = "";
$rows = rtrim($rows, "\t]");
- return split("\t]", $rows);
+ return explode("\t]", $rows);
// print_r($parsed_rows);
/*
foreach ($rows as &$row) {
$row = ltrim($row, "[ ");
- $row = split(",\t", $row);
+ $row = explode(",\t", $row);
foreach ($row as &$field) {
$field = stripslashes($field);
@@ -244,7 +244,7 @@
function php_parse_row($row) {
$row = ltrim($row, "[ ");
- $row = split(",\t", $row);
+ $row = explode(",\t", $row);
foreach ($row as &$field) {
if ($field == "NULL") {
$field = NULL;
@@ -269,12 +269,12 @@
/* Field names */
$header[1] = ltrim($header[1], "% ");
- $header[1] = split("", $header[1]);
+ $header[1] = explode("", $header[1]);
$header_array["fields"] = explode(",\t", $header[1][0]);
/* Field types */
$header[2] = ltrim($header[2], "% ");
- $header[2] = split("", $header[2]);
+ $header[2] = explode("", $header[2]);
$header_array["types"] = explode(",\t", $header[2][0]);
return $header_array;
@@ -328,7 +328,7 @@
)
*/
- $credentials = split(":", $challenge);
+ $credentials = explode(":", $challenge);
$algos = explode(',', $credentials[3]);
@@ -352,7 +352,7 @@
if ($response != MSG_PROMPT) {
// not ready to authenticate yet
if ($response[0] == MSG_REDIRECT) {
- $redirects = split("\n", $response);
+ $redirects = explode("\n", $response);
/* Follow the first redirect */
if ( ($redirects[0] == "") || (substr($redirects[0], 0, 6) != "^mapi:") ) {
## Comment 16628
Date: 2011-12-08 14:16:05 +0100
From: @grobian
Changeset [2af65b58ba4b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=2af65b58ba4b) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2af65b58ba4b](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=2af65b58ba4b)
Changeset description:
php_mapi: use explode instead of deprecated split
Bug #2940: drop deprecated split() usage.
## Comment 16629
Date: 2011-12-08 14:16:35 +0100
From: @grobian
local testing showed no problems, committed.
## Comment 16835
Date: 2012-01-26 15:32:08 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| PHP split() function deprecated | https://api.github.com/repos/MonetDB/MonetDB/issues/2940/comments | 0 | 2020-11-30T11:48:33Z | 2024-06-27T11:54:05Z | https://github.com/MonetDB/MonetDB/issues/2940 | 753,410,291 | 2,940 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 12:26:21 +0100
From: manuela <<m.sagona>>
To: clients devs <<bugs-clients>>
Version: 11.3.3 (Apr2011-SP1) [obsolete]
Last updated: 2012-01-26 15:31:57 +0100
## Comment 16589
Date: 2011-12-06 12:26:21 +0100
From: manuela <<m.sagona>>
User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
we notice that a "PHP Notice" was rised every time the mapi_execute() returns a FALSE
Reproducible: Always
Only variable references should be returned by reference in .../php_mapi.inc on line 96) due to the by reference return behaviour of the function;
we suggest to remove that behaviour since it does not increase performance (see http://php.net/manual/en/language.references.return.php).
Finally note that pass-by-reference has been deprecated (Deprecated: Call-time pass-by-reference has been deprecated in php_monetdb.php on line 302)
## Comment 16597
Date: 2011-12-06 13:26:06 +0100
From: @grobian
Do you mean something like this?
diff --git a/clients/php/lib/php_mapi.inc b/clients/php/lib/php_mapi.inc
--- a/clients/php/lib/php_mapi.inc
+++ b/clients/php/lib/php_mapi.inc
@@ -64,7 +64,7 @@
/**
* Execute an SQL query and return the resulting handle by reference.
*/
- function &mapi_execute($conn=NULL, $query) {
+ function mapi_execute($conn=NULL, $query) {
global $connection_pool;
global $rows;
/**
## Comment 16630
Date: 2011-12-08 14:35:51 +0100
From: @grobian
Changeset [6c59acf6021e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6c59acf6021e) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6c59acf6021e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6c59acf6021e)
Changeset description:
mapi_execute: don't return a reference, since we can return FALSE
Bug #2939, PHP spits a notice mapi_execute tries to return FALSE. Since
there is no performance improvement by actually returning a reference,
just don't do it.
## Comment 16631
Date: 2011-12-08 14:36:15 +0100
From: @grobian
Change seems to work fine, committed.
## Comment 16817
Date: 2012-01-26 15:31:57 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| "PHP Notice" in mapi_execute() | https://api.github.com/repos/MonetDB/MonetDB/issues/2939/comments | 0 | 2020-11-30T11:48:31Z | 2024-06-27T11:54:04Z | https://github.com/MonetDB/MonetDB/issues/2939 | 753,410,262 | 2,939 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 12:20:34 +0100
From: manuela <<m.sagona>>
To: clients devs <<bugs-clients>>
Version: 11.3.3 (Apr2011-SP1) [obsolete]
Last updated: 2012-01-26 15:32:07 +0100
## Comment 16587
Date: 2011-12-06 12:20:34 +0100
From: manuela <<m.sagona>>
User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
there was an error retrieving the socket for the mapi_write() call
Reproducible: Always
## Comment 16588
Date: 2011-12-06 12:21:01 +0100
From: manuela <<m.sagona>>
Created attachment 85
proposed patch
> Attached file: [php_monetdb.php_patch_auto_commit.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2938_php_monetdb.php_patch_auto_commit.txt_85) (text/plain, 390 bytes)
> Description: proposed patch
## Comment 16595
Date: 2011-12-06 13:22:10 +0100
From: manuela <<m.sagona>>
Changeset [554fd39f52ab](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=554fd39f52ab) made by Manuela Sagona <m.sagona@acsys.it> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=554fd39f52ab](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=554fd39f52ab)
Changeset description:
auto_commit: reference existing variable
socket is a member for the $conn array. Fixes bug #2938.
## Comment 16596
Date: 2011-12-06 13:22:26 +0100
From: @grobian
Thanks, pushed
## Comment 16832
Date: 2012-01-26 15:32:07 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| minor bug into auto_commit() (php_monetdb.php) | https://api.github.com/repos/MonetDB/MonetDB/issues/2938/comments | 0 | 2020-11-30T11:48:28Z | 2024-06-27T11:54:03Z | https://github.com/MonetDB/MonetDB/issues/2938 | 753,410,237 | 2,938 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 12:16:11 +0100
From: manuela <<m.sagona>>
To: clients devs <<bugs-clients>>
Version: 11.3.3 (Apr2011-SP1) [obsolete]
Last updated: 2012-01-26 15:32:03 +0100
## Comment 16585
Date: 2011-12-06 12:16:11 +0100
From: manuela <<m.sagona>>
User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
there is no method db_quote_ident()
Reproducible: Always
use monetdb_quote_ident() instead
## Comment 16586
Date: 2011-12-06 12:17:22 +0100
From: manuela <<m.sagona>>
Created attachment 84
proposed patch
> Attached file: [php_monetdb.php_patch_quote_ident.txt](https://github.com/MonetDB/monetdb-issues-attachments/blob/master/issue_2937_php_monetdb.php_patch_quote_ident.txt_84) (text/plain, 433 bytes)
> Description: proposed patch
## Comment 16593
Date: 2011-12-06 13:19:44 +0100
From: manuela <<m.sagona>>
Changeset [cff56fc02d72](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=cff56fc02d72) made by Manuela Sagona <m.sagona@acsys.it> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=cff56fc02d72](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=cff56fc02d72)
Changeset description:
monetdb_insert_id: use existing monetdb_quote_ident
Replace db_quote_ident with monetdb_quote_ident. Fixes bug #2937.
## Comment 16594
Date: 2011-12-06 13:20:07 +0100
From: @grobian
thanks for your patch, pushed
## Comment 16826
Date: 2012-01-26 15:32:03 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| minor bug into monetdb_insert_id() (php_monetdb.php) | https://api.github.com/repos/MonetDB/MonetDB/issues/2937/comments | 0 | 2020-11-30T11:48:26Z | 2024-06-27T11:54:02Z | https://github.com/MonetDB/MonetDB/issues/2937 | 753,410,206 | 2,937 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 12:04:52 +0100
From: manuela <<m.sagona>>
To: clients devs <<bugs-clients>>
Version: 11.3.3 (Apr2011-SP1) [obsolete]
Last updated: 2012-01-26 15:32:04 +0100
## Comment 16584
Date: 2011-12-06 12:04:52 +0100
From: manuela <<m.sagona>>
User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.121 Safari/535.2
Build Identifier:
monetdb_connected() in php_monetdb.php does not work due to a bug into mapi_ping()
Reproducible: Always
### Steps to Reproduce:
1.execute "select true;"
### Actual Results:
the resulting handle has not an 'id' element to use in mapi_free_result() call
### Expected Results:
an 'id' element to use in mapi_free_result() call
to fix the bug:
@@ -637,6 +637,7 @@
break;
}
if ($res != NULL) {
+ if (! array_key_exists('id', $res)) return TRUE;
if (mapi_free_result($conn['id'], $res['id'])) {
return TRUE;
}
## Comment 16592
Date: 2011-12-06 13:15:15 +0100
From: @grobian
Changeset [0c5bc1adc7b3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0c5bc1adc7b3) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0c5bc1adc7b3](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0c5bc1adc7b3)
Changeset description:
mapi_ping: Fix call to mapi_free_result
The result id is actually stored in the query array of the handle, not
the handle itself. Since not all results have such thing, check for it
before calling mapi_free_result, to avoid needless work (and some
warnings from PHP).
Inspired by patch from manuela in bug #2936.
## Comment 16827
Date: 2012-01-26 15:32:04 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| bug into mapi_ping() | https://api.github.com/repos/MonetDB/MonetDB/issues/2936/comments | 0 | 2020-11-30T11:48:23Z | 2024-06-27T11:54:01Z | https://github.com/MonetDB/MonetDB/issues/2936 | 753,410,178 | 2,936 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-06 08:50:08 +0100
From: @grobian
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
CC: @yzchang
Last updated: 2012-11-27 14:27:41 +0100
## Comment 16583
Date: 2011-12-06 08:50:08 +0100
From: @grobian
Aparently, vuurbal has 0.0GiB of memory. In reality, it has 16GiB.
## Comment 16600
Date: 2011-12-06 14:16:46 +0100
From: @grobian
The Solaris manpage reportes under the NOTES section:
Multiplying sysconf(_SC_PHYS_PAGES) or sysconf(_SC_AVPHYS_PAGES) by
sysconf(_SC_PAGESIZE) to determine memory amount in bytes can exceed
the maximum values representable in a 32-bit signed or unsigned inte‐
ger.
Aparently this has changed between the Solaris version on the T1000, since there it reported nicely 4.0GiB available memory. (Out of the 8 it had.)
## Comment 16601
Date: 2011-12-06 17:25:55 +0100
From: @grobian
Changeset [312c7c400304](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=312c7c400304) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=312c7c400304](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=312c7c400304)
Changeset description:
MT_init: avoid overflow for memory on Solaris 32-bits
Fix for bug #2935. Cap the memorysize to something which can be
addressed in 32-bits, when it happens to be larger than that. Fixes
memory detection on 32-bits targets for Solaris systems that have lots
of memory installed.
## Comment 16823
Date: 2012-01-26 15:32:01 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
## Comment 18081
Date: 2012-11-27 14:27:41 +0100
From: @yzchang
Not possible to add test for this bug, unless someone else knows how...
| mserver5 doesn't detect memory on Solaris/Sparc | https://api.github.com/repos/MonetDB/MonetDB/issues/2935/comments | 0 | 2020-11-30T11:48:21Z | 2024-06-27T11:54:00Z | https://github.com/MonetDB/MonetDB/issues/2935 | 753,410,145 | 2,935 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-12-05 16:28:28 +0100
From: @sjoerdmullender
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
Last updated: 2012-01-26 15:32:06 +0100
## Comment 16582
Date: 2011-12-05 16:28:28 +0100
From: @sjoerdmullender
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.24) Gecko/20111108 Fedora/3.6.24-1.fc14 Firefox/3.6.24
Build Identifier:
Yet another blocker for a release.
Reproducible: Always
### Steps to Reproduce:
1.create database using Aug2011 release (e.g. SP3)
2.(maybe insert some data)
3.stop server, upgrade to Dec2011, start server
4.create new user with new schema, set schema to be default for user
5.select * from sys.tables;
### Actual Results:
SELECT: access denied for testuser to table 'sys.tables'
### Expected Results:
Contents of sys.tables.
I used my "testdb.sql" script *after* the upgrade. It creates a user with a default schema.
Without an upgrade, things work as expected. In both branches.
## Comment 16634
Date: 2011-12-08 16:03:18 +0100
From: @sjoerdmullender
Changeset [8f99f0ff2208](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8f99f0ff2208) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8f99f0ff2208](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8f99f0ff2208)
Changeset description:
After destroying and recreating sys.columns and sys.tables, add privileges.
This fixes bug #2934.
The problem: during an catalog upgrade, sys.columns and sys.tables are
dropped and recreated (in case they changed). The new versions of
these views were not represented in the sys.privileges table. We now
add entries for the new views, solving the bug.
## Comment 16635
Date: 2011-12-08 16:06:17 +0100
From: @sjoerdmullender
Bug has been fixed.
No test added since we don't yet have a way to test upgrades.
The summary is changed since this bug actually occurred in many previous releases but was never noticed.
## Comment 16830
Date: 2012-01-26 15:32:06 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
| Database upgrade with SQL catalog upgrade messes up privileges | https://api.github.com/repos/MonetDB/MonetDB/issues/2934/comments | 0 | 2020-11-30T11:48:18Z | 2024-06-27T11:53:59Z | https://github.com/MonetDB/MonetDB/issues/2934 | 753,410,116 | 2,934 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-16 15:03:01 +0100
From: Viktor Rosenfeld <<24hesk>>
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @njnes
Last updated: 2013-06-09 14:17:39 +0200
## Comment 16560
Date: 2011-11-16 15:03:01 +0100
From: Viktor Rosenfeld <<24hesk>>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_6_8) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.120 Safari/535.2
Build Identifier:
The following query results in an error message:
SELECT
count(*)
FROM
_rank AS _rank1,
_rank AS _rank2
WHERE
NOT _rank1.pre = _rank2.pre AND
EXISTS (SELECT pre FROM _rank AS ancestor WHERE
ancestor.pre < _rank1.pre AND _rank1.pre < ancestor.post AND
ancestor.pre < _rank2.pre AND _rank2.pre < ancestor.post);
Reproducible: Always
### Steps to Reproduce:
The following table can be used to reproduce the error:
CREATE TABLE _rank (pre INTEGER, post INTEGER);
INSERT INTO _rank VALUES (22, 37);
INSERT INTO _rank VALUES (23, 24);
INSERT INTO _rank VALUES (33, 34);
### Actual Results:
The following error message:
TypeException:user.s0_1[77]:'algebra.join' undefined in: _100:any := algebra.join(_99:bat[:oid,:oid], _80:bat[:int,:oid])
SQLException:SQLengine:Program contains errors
### Expected Results:
+------+
| L147 |
+======+
| 2 |
+------+
1 tuple
If actual data is used inside the EXISTS clause then the query works, e.g.
SELECT
count(*)
FROM
_rank AS _rank1,
_rank AS _rank2
WHERE
_rank2.pre = 33 AND
NOT _rank1.pre = _rank2.pre AND
EXISTS (SELECT pre FROM _rank AS ancestor WHERE
ancestor.pre < _rank1.pre AND _rank1.pre < ancestor.post AND
ancestor.pre < 33 AND 33 < ancestor.post);
MonetDB is installed from the OS X download package.
mserver5 --version follows:
MonetDB 5 server v11.5.7 "Aug2011-SP2" (64-bit, 64-bit oids)
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 4.0GiB available memory, 2 available cpu cores
Libraries:
libpcre: 8.11 2010-12-10 (compiled with 8.11)
openssl: OpenSSL 1.0.0d 8 Feb 2011 (compiled with OpenSSL 1.0.0d 8 Feb 2011)
libxml2: 2.7.8 (compiled with 2.7.8)
Compiled by: monetdb-bin-builder (x86_64-apple-darwin9)
Compilation: x86_64-apple-darwin9-gcc -fast -pipe
Linking : /Volumes/scratch/gentoo/x86_64/usr/bin/x86_64-apple-darwin9-ld -Wl,-dead_strip_dylibs
## Comment 16561
Date: 2011-11-16 15:06:42 +0100
From: Viktor Rosenfeld <<24hesk>>
Not sure, if this is related, but if I try to rewrite the query as follows:
SELECT
count(*)
FROM
_rank AS _rank1,
_rank AS _rank2
WHERE
NOT _rank1.pre = _rank2.pre AND
EXISTS (SELECT pre FROM _rank AS ancestor WHERE
ancestor.pre < _rank1.pre AND _rank1.pre < ancestor.post
INTERSECT SELECT pre FROM _rank AS ancestor WHERE
ancestor.pre < _rank2.pre AND _rank2.pre < ancestor.post);
I get another error message: SELECT: no such column '_rank1.pre'
## Comment 16638
Date: 2011-12-08 17:17:43 +0100
From: @sjoerdmullender
Changeset [0244e96a2c5e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0244e96a2c5e) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0244e96a2c5e](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0244e96a2c5e)
Changeset description:
Added test for bug #2933.
## Comment 16700
Date: 2011-12-27 15:28:34 +0100
From: @njnes
Changeset [57700c6efacb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=57700c6efacb) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=57700c6efacb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=57700c6efacb)
Changeset description:
fix bug with semijoins. This partialy fixes bug #2933 and bug #2915.
## Comment 16810
Date: 2012-01-25 10:43:41 +0100
From: Viktor Rosenfeld <<24hesk>>
The original query now works on the latest Dec2011 release. However the version with INTERSECT now throws a new error message: 21000!zero_or_one: cardinality violation (2>1).
I'm using the following query.
SELECT count(*)
FROM rank AS rank1, rank AS rank2
WHERE
rank1.pre IN (1, 3, 5, 6, 8, 13) AND
rank2.pre IN (1, 3, 5, 6, 8, 13) AND
NOT rank1.pre = rank2.pre AND
EXISTS (
SELECT ancestor.pre
FROM rank AS ancestor
WHERE ancestor.pre < rank1.pre AND rank1.pre < ancestor.post
INTERSECT SELECT ancestor.pre
FROM rank AS ancestor
WHERE ancestor.pre < rank2.pre AND rank2.pre < ancestor.post
);
## Comment 18799
Date: 2013-06-09 14:17:39 +0200
From: MonetDB Mercurial Repository <<hg>>
Changeset [054504135056](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=054504135056) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=054504135056](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=054504135056)
Changeset description:
fixed (again) bug exists-select.Bug-2933.stable.out
| "TypeException: algebra.join undefined" when evaluating EXISTS (SELECT ...) predicate | https://api.github.com/repos/MonetDB/MonetDB/issues/2933/comments | 0 | 2020-11-30T11:48:16Z | 2024-06-27T11:53:58Z | https://github.com/MonetDB/MonetDB/issues/2933 | 753,410,093 | 2,933 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-14 20:53:14 +0100
From: Alfred Nordman <<alfred.nordman>>
To: SQL devs <<bugs-sql>>
Version: 11.13.5 (Oct2012-SP1)
CC: bugs-monetdb5, holger.pirk, @mlkersten, @yzchang
Last updated: 2012-12-08 22:40:54 +0100
## Comment 16559
Date: 2011-11-14 20:53:14 +0100
From: Alfred Nordman <<alfred.nordman>>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.2 (KHTML, like Gecko) Chrome/15.0.874.106 Safari/535.2
Build Identifier:
Using the DISTINCT statement seems to cause the utilization of just one machine-core per query, while an equivalent query without DISTINCT seems to be utilizing more cores and returns results 3 times faster.
Reproducible: Always
### Steps to Reproduce:
1. CREATE TABLE t (c1 INT, c2 INT);
2. INSERT INTO t VALUES (3, 5), (4,7);
3. INSERT INTO t (SELECT * FROM t);
4. Repeat step 3 several times to populate t with millions of rows.
5. SELECT c1, SUM(DISTINCT c2) FROM t GROUP BY c1;
6. SELECT c1, SUM(c2) FROM (SELECT c1, c2 FROM t GROUP BY c1, c2) abc GROUP BY c1;
### Actual Results:
For the query in step 5:
+------+------+
| c1 | L62 |
+======+======+
| 3 | 5 |
| 4 | 7 |
+------+------+
2 tuples (1.0s)
For the query in step 6:
+------+------+
| c1 | L64 |
+======+======+
| 3 | 5 |
| 4 | 7 |
+------+------+
2 tuples (319.504ms)
The first query runs about 3 times slower than the second query.
### Expected Results:
We'd expect both queries to run for about the same time.
MonetDB 5 server v11.5.7 "Aug2011-SP2"
Serving database 'demo', using 8 threads
Compiled for x86_64-pc-winnt/64bit with 64bit OIDs dynamically linked
Found 7.990 GiB available main-memory.
Copyright (c) 1993-July 2008 CWI.
Copyright (c) August 2008-2011 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
## Comment 17665
Date: 2012-08-24 14:54:31 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 18064
Date: 2012-11-27 14:01:52 +0100
From: @grobian
keep alias in the loop
## Comment 18082
Date: 2012-11-27 14:30:46 +0100
From: @yzchang
No test (possible) for performance issues
## Comment 18243
Date: 2012-12-08 22:40:54 +0100
From: @mlkersten
A SQL code generation issue followed by optimizers.
No real chance to make these two queries to produce the same plan.
| Using distinct returns results slower than when using an equivalent complex query with group by | https://api.github.com/repos/MonetDB/MonetDB/issues/2932/comments | 0 | 2020-11-30T11:48:12Z | 2024-06-28T13:19:23Z | https://github.com/MonetDB/MonetDB/issues/2932 | 753,410,052 | 2,932 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-14 11:00:08 +0100
From: @drstmane
To: clients devs <<bugs-clients>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: rchibois
Duplicates: #2891
Last updated: 2011-11-26 16:23:33 +0100
## Comment 16556
Date: 2011-11-14 11:00:08 +0100
From: @drstmane
The fix for bug #2819, i.e., changeset [f584fadb4847](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f584fadb4847) (cf., http://dev.monetdb.org/hg/MonetDB/rev/f584fadb4847) make test monetdb5/tests/mapi/Tests/perl_dbi fail with several "Use of uninitialized value ..." errors; cf. e.g.,
http://monetdb.cwi.nl/testweb/tests/42325:373008bf840f/GNU-Fedora-x86_64-propcheck-assert/monetdb5/mTests/tests/mapi/perl_dbi.err.diff.html
http://monetdb.cwi.nl/testweb/tests/42325:373008bf840f/GNU-Fedora-x86_64-propcheck-assert/monetdb5/mTests/tests/mapi/perl_dbi.out.diff.html
## Comment 16557
Date: 2011-11-14 11:01:00 +0100
From: @drstmane
typo: it's the fix for bug #2891 (not bug #2819)
## Comment 16574
Date: 2011-11-26 16:23:33 +0100
From: @grobian
*** This bug has been marked as a duplicate of bug #2891 ***
| Fix for bug 2891 (MapiPP.pm) makes perl_dbi test fail | https://api.github.com/repos/MonetDB/MonetDB/issues/2931/comments | 0 | 2020-11-30T11:48:09Z | 2024-06-28T13:19:22Z | https://github.com/MonetDB/MonetDB/issues/2931 | 753,410,016 | 2,931 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-13 15:29:33 +0100
From: Alfred Nordman <<alfred.nordman>>
To: SQL devs <<bugs-sql>>
Version: 11.5.7 (Aug2011-SP2) [obsolete]
Last updated: 2011-11-20 14:15:06 +0100
## Comment 16555
Date: 2011-11-13 15:29:33 +0100
From: Alfred Nordman <<alfred.nordman>>
The query:
SELECT INRANGE(NOW(), NOW(), NOW());
results in the following error message:
"types timestamptz(7,0) and double(53,0) are not equal"
## Comment 16562
Date: 2011-11-20 14:15:06 +0100
From: Alfred Nordman <<alfred.nordman>>
I'm terribly sorry for the mistake, inrange() was a local custom function which was defined without a support for timestamptz.
| Error is returned when using inrange() with now() | https://api.github.com/repos/MonetDB/MonetDB/issues/2930/comments | 0 | 2020-11-30T11:48:06Z | 2024-06-28T13:19:21Z | https://github.com/MonetDB/MonetDB/issues/2930 | 753,409,986 | 2,930 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-10 18:41:28 +0100
From: sellam
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.11.7 (Jul2012-SP1)
CC: @njnes, @drstmane, @yzchang
Last updated: 2014-02-12 08:57:05 +0100
## Comment 16543
Date: 2011-11-10 18:41:28 +0100
From: sellam
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.23) Gecko/20110927 Fedora/3.6.23-1.fc14 Firefox/3.6.23
Build Identifier:
sql>select 10000000 * 100000 * 11.51 + 51.097 * 100000;
+-----------------------+
| sql_mul_single_value |
+=======================+
| -6936738964009.551616 |
+-----------------------+
Reproducible: Always
### Expected Results:
11510005109700
bash-4.1$ hg branch
Aug2011
bash-4.1$ hg id
[0299b070e196](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0299b070e196) (Aug2011) tip
## Comment 16544
Date: 2011-11-10 18:45:18 +0100
From: sellam
Workaround:
sql>select convert (10000000 * 100000 * 11.51,decimal(15)) +convert (51.097 * 100000, decimal(15));
+----------------------+
| sql_mul_single_value |
+======================+
| 11510005109700 |
+----------------------+
-> works fine
## Comment 16545
Date: 2011-11-10 18:49:58 +0100
From: @drstmane
Maybe a decimal overflow?
sql>select 10000000 * 100000 * 11.51 + 51.097 * 100000;
+-----------------------+
| sql_mul_single_value |
+=======================+
| -6936738964009.551616 |
+-----------------------+
1 tuple (1.268ms)
sql>select 10000000 * 100000 * 11.51 ;
+-----------------------+
| sql_mul_single_value |
+=======================+
| 11510000000000.0000 |
+-----------------------+
1 tuple (0.969ms)
sql>select 51.097 * 100000;
+-----------------------+
| sql_mul_single_value |
+=======================+
| 5109700.000000 |
+-----------------------+
1 tuple (0.324ms)
sql>select 11510000000000.0000 + 5109700.000000;
+-----------------------+
| scale_up_single_value |
+=======================+
| -6936738964009.551616 |
+-----------------------+
1 tuple (0.284ms)
sql>\f
Current formatter: sql
sql>\fraw
sql>select 10000000 * 100000 * 11.51 ;
% . table_name
% sql_mul_single_value name
% decimal type
% 21 length
[ 11510000000000.0000 ]
sql>select 51.097 * 100000;
% . table_name
% sql_mul_single_value name
% decimal type
% 21 length
[ 5109700.000000 ]
sql>select cast(11510000000000.0000 as double) + cast(5109700.000000 as double);
% . table_name
% sql_add_single_value name
% double type
% 24 length
[ 11510005109700 ]
sql>\fsql
sql>select cast(11510000000000.0000 as double) + cast(5109700.000000 as double);
+--------------------------+
| sql_add_single_value |
+==========================+
| 11510005109700 |
+--------------------------+
1 tuple (0.797ms)
## Comment 16547
Date: 2011-11-12 19:27:22 +0100
From: @njnes
time to detect overflow on lng multiply and addition.
## Comment 16548
Date: 2011-11-12 21:27:52 +0100
From: @sjoerdmullender
(In reply to comment 3)
> time to detect overflow on lng multiply and addition.
See gdk_calc.c in the headless branch for how that can be done.
## Comment 17084
Date: 2012-03-15 17:28:10 +0100
From: @sjoerdmullender
Changeset [c6affd6e5b32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=c6affd6e5b32) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c6affd6e5b32](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=c6affd6e5b32)
Changeset description:
Merged gdk-calc branch into default.
This marks the end of the gdk-calc branch. Overflow checking is now
default behavior.
This fixes bug #2929 by now giving an overwlow error.
## Comment 17662
Date: 2012-08-24 14:54:28 +0200
From: @sjoerdmullender
Jul2012-SP1 has been released.
## Comment 17907
Date: 2012-11-07 19:39:33 +0100
From: @grobian
(In reply to comment 5)
> This fixes bug #2929 by now giving an overwlow error.
## Comment 18088
Date: 2012-11-27 14:40:42 +0100
From: @yzchang
Test added in monetdb5/tests/BugTracker/Tests/large-number-operation-strange-results.Bug-2929.sql
## Comment 18199
Date: 2012-11-28 13:45:33 +0100
From: @yzchang
Changeset [e3d66953a533](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=e3d66953a533) made by Jennie Zhang <y.zhang@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e3d66953a533](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=e3d66953a533)
Changeset description:
Added test for Bug #2929
This bug is fixed in Jul2012-SP1, and produces expected results at the time of this commit
## Comment 19576
Date: 2014-02-12 08:57:05 +0100
From: MonetDB Mercurial Repository <<hg>>
Changeset [3e099b2ec0ff](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3e099b2ec0ff) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3e099b2ec0ff](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3e099b2ec0ff)
Changeset description:
large-number-operation-strange-results.Bug-2929: refined test for 38-digit decimals (128-bit integers)
| Very large number operations give strange results | https://api.github.com/repos/MonetDB/MonetDB/issues/2929/comments | 0 | 2020-11-30T11:48:03Z | 2024-06-27T11:53:54Z | https://github.com/MonetDB/MonetDB/issues/2929 | 753,409,935 | 2,929 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-09 17:48:40 +0100
From: sellam
To: SQL devs <<bugs-sql>>
Version: 11.5.1 (Aug2011) [obsolete]
CC: @njnes
Last updated: 2011-11-22 13:55:56 +0100
## Comment 16540
Date: 2011-11-09 17:48:40 +0100
From: sellam
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.23) Gecko/20110927 Fedora/3.6.23-1.fc14 Firefox/3.6.23
Build Identifier:
The bug can be created as follows:
create table ranges(
x_min decimal(8,5),
y_min decimal(7,5),
width integer,
nb integer
);
create table trips (
tripid bigint not null,
x decimal(8,5) not null,
y decimal(8,5) not null,
time bigint not null
);
UPDATE ranges SET nb = (SELECT count(*) FROM trips T
WHERE T.x between x_min and x_min + width
AND T.y between y_min and y_min + width);
Returns:
MAPI = (monetdb) /tmp/.s.monetdb.50000
ACTION= read_line
QUERY = UPDATE ranges SET nb = (SELECT count(*) FROM trips T
WHERE T.x between x_min and x_min + width
AND T.y between y_min and y_min + width);
ERROR = !Connection terminated
Reproducible: Always
### Actual Results:
Error message:
MAPI = (monetdb) /tmp/.s.monetdb.50000
ACTION= read_line
QUERY = UPDATE ranges SET nb = (SELECT count(*) FROM trips T
WHERE T.x between x_min and x_min + width
AND T.y between y_min and y_min + width);
ERROR = !Connection terminated
### Expected Results:
An update
bash-4.1$ hg branch
Aug2011
bash-4.1$ hg id
[0299b070e196](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0299b070e196) (Aug2011) tip
## Comment 16541
Date: 2011-11-09 19:41:38 +0100
From: @grobian
can you check your merovingian.log for some informative crash messages, perhaps?
## Comment 16542
Date: 2011-11-10 11:19:18 +0100
From: sellam
Hi, unfortunately the crash message is not very explicit:
2011-11-09 17:44:39 MSG merovingian[14936]: database 'testbug' (15460) was killed by signal SIGSEGV
2011-11-09 17:47:55 MSG merovingian[14936]: database 'testbug' has crashed after start on 2011-11-09 17:43:55, attempting restart, up min/avg/max: 0s/0s/0s, crash average: 1.00 0.30 0.10 (3-0=3)
Precision: in the provided example, the tables are empty before the UPDATE statement. The message is the same when the tables are populated
## Comment 16546
Date: 2011-11-10 22:31:30 +0100
From: @njnes
Changeset [4e67d716ffbf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4e67d716ffbf) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4e67d716ffbf](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4e67d716ffbf)
Changeset description:
fix bug #2928 (update using correlated count query)
## Comment 16549
Date: 2011-11-12 23:33:23 +0100
From: @njnes
fixed a bit cleaner now. In case of correlated updates (and subqueries) in the selection phase, we should use the inner query columns with in the aggregates.
## Comment 16550
Date: 2011-11-13 09:28:09 +0100
From: @njnes
Changeset [64efaa8ac570](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=64efaa8ac570) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=64efaa8ac570](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=64efaa8ac570)
Changeset description:
merge of fix for bug #2928
## Comment 16551
Date: 2011-11-13 09:29:30 +0100
From: @njnes
Changeset [5acc6a38e932](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=5acc6a38e932) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5acc6a38e932](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=5acc6a38e932)
Changeset description:
merged fix for bug #2928 (from Dec2011 branch)
## Comment 16552
Date: 2011-11-13 09:35:10 +0100
From: @njnes
added test file update_with_count.Bug-2928.sql
## Comment 16553
Date: 2011-11-13 09:36:28 +0100
From: @njnes
Changeset [b691e9f8af19](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b691e9f8af19) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b691e9f8af19](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b691e9f8af19)
Changeset description:
added test for bug #2928
## Comment 16568
Date: 2011-11-22 13:55:56 +0100
From: @grobian
Fixed in Aug2011-SP3
| Cannot update a field with a count query based on an other table | https://api.github.com/repos/MonetDB/MonetDB/issues/2928/comments | 0 | 2020-11-30T11:48:00Z | 2024-06-27T11:53:53Z | https://github.com/MonetDB/MonetDB/issues/2928 | 753,409,897 | 2,928 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-08 20:43:00 +0100
From: Tim Staley <<timstaley.donotreply.forspam>>
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2011-11-28 10:02:23 +0100
## Comment 16537
Date: 2011-11-08 20:43:00 +0100
From: Tim Staley <<timstaley.donotreply.forspam>>
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.1 (KHTML, like Gecko) Ubuntu/11.10 Chromium/14.0.835.202 Chrome/14.0.835.202 Safari/535.1
Build Identifier:
The tutorial at http://www.monetdb.org/Downloads/Tutorial is incorrect.
The line
shell> mclient -umonetdb voc
should read
shell> mclient -umonetdb -d voc
A tiny error, but it is meant to be the quickstart guide for newbies!
(I am using v11.5.7)
Reproducible: Always
### Steps to Reproduce:
1. Install monetdb
2. Attempt to use tutorial
## Comment 16538
Date: 2011-11-08 20:45:18 +0100
From: @grobian
ehm, the example should work, what release are you on?
## Comment 16580
Date: 2011-11-28 10:02:23 +0100
From: @grobian
You're right. Aug2011 doesn't support that syntax, so I changed the example throughout to have -d voc instead, since that works on all releases.
Thanks!
| Tutorial instructions are slightly wrong / out of date | https://api.github.com/repos/MonetDB/MonetDB/issues/2927/comments | 0 | 2020-11-30T11:47:58Z | 2024-06-27T11:53:52Z | https://github.com/MonetDB/MonetDB/issues/2927 | 753,409,862 | 2,927 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-07 18:10:47 +0100
From: @grobian
To: Documentation maintainers <<bugs-docs>>
Version: unspecified
Last updated: 2011-12-08 22:01:17 +0100
## Comment 16527
Date: 2011-11-07 18:10:47 +0100
From: @grobian
http://www.monetdb.org/Documentation/Manuals/SQLreference/Programming/ODBC
Not sure if we have the images still lingering around somewhere, probably should remove the refs otherwise.
## Comment 16639
Date: 2011-12-08 17:23:30 +0100
From: @sjoerdmullender
They're available at monetdb.project.cwi.nl:/home/monetdb/sites/monetdb.project.cwi.nl/WWW/monetdb/clients/src/odbc/doc/images/.
## Comment 16641
Date: 2011-12-08 19:49:40 +0100
From: @grobian
I don't think I have access to that machine.
## Comment 16644
Date: 2011-12-08 20:54:43 +0100
From: @sjoerdmullender
I copied them for now to ~sjoerd/src/MonetDB/
## Comment 16645
Date: 2011-12-08 22:01:17 +0100
From: @grobian
ok, thanks, fixed the guide.
| images missing from ODBC example | https://api.github.com/repos/MonetDB/MonetDB/issues/2926/comments | 0 | 2020-11-30T11:47:55Z | 2024-06-27T11:53:51Z | https://github.com/MonetDB/MonetDB/issues/2926 | 753,409,828 | 2,926 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-07 11:25:24 +0100
From: Yagiz <<y.kargin>>
To: clients devs <<bugs-clients>>
Version: 11.5.7 (Aug2011-SP2) [obsolete]
CC: y.kargin
Last updated: 2011-11-09 12:54:19 +0100
## Comment 16522
Date: 2011-11-07 11:25:24 +0100
From: Yagiz <<y.kargin>>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.23) Gecko/20110927 Fedora/3.6.23-1.fc14 Firefox/3.6.23
Build Identifier:
In this page (http://www.monetdb.org/Downloads/Tutorial), it is written that "mclient -umonetdb voc" works, but it doesn't. "monetdbd: please specify a database" is returned on the command line.
Secondary problem is that the same sentence is returned, whatever password you provide.
Reproducible: Always
### Steps to Reproduce:
1. The steps are the first 5 commands on this page: http://www.monetdb.org/Downloads/Tutorial
2.
3.
### Actual Results:
"monetdbd: please specify a database" is returned on the command line and quit.
### Expected Results:
should enter the "sql>" mode.
## Comment 16523
Date: 2011-11-07 11:29:47 +0100
From: @grobian
questions:
1) do you have a file called voc in your working directory?
2) do you get the password prompt?
## Comment 16524
Date: 2011-11-07 11:35:52 +0100
From: Yagiz <<y.kargin>>
(In reply to comment 1)
> questions:
> 1) do you have a file called voc in your working directory?
Yes.
> 2) do you get the password prompt?
Yes. Whatever password I write, I get the same answer.
The problem does not happen if I use a -d before database name.
## Comment 16525
Date: 2011-11-07 11:42:57 +0100
From: @grobian
correct:
--database=database (-d database)
Specify the name of the database to connect to. The -d can be
omitted if an equally named file does not exist in the current
directory. As such, the first non-option argument will be
interpreted as database to connect to if the argument does not
exist as file.
## Comment 16526
Date: 2011-11-07 12:50:06 +0100
From: Yagiz <<y.kargin>>
(In reply to comment 3)
> correct:
>
> --database=database (-d database)
> Specify the name of the database to connect to. The -d can be
> omitted if an equally named file does not exist in the current
> directory. As such, the first non-option argument will be
> interpreted as database to connect to if the argument does not
> exist as file.
Then it seems that the bug remains in the documentation, because I applied those 5 commands and it didn't work. In which case is the file actually there/not there?
## Comment 16539
Date: 2011-11-09 12:54:19 +0100
From: @grobian
you probably accidentially did
monetdbd create voc
iso
monetdb create voc
| mclient does not work with only a username and database name | https://api.github.com/repos/MonetDB/MonetDB/issues/2925/comments | 0 | 2020-11-30T11:47:53Z | 2024-06-28T13:19:20Z | https://github.com/MonetDB/MonetDB/issues/2925 | 753,409,798 | 2,925 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-07 08:43:23 +0100
From: taowen
To: SQL devs <<bugs-sql>>
Version: 11.5.7 (Aug2011-SP2) [obsolete]
CC: @njnes
Last updated: 2011-11-22 13:55:57 +0100
## Comment 16521
Date: 2011-11-07 08:43:23 +0100
From: taowen
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Ubuntu/10.04 Chromium/15.0.874.106 Chrome/15.0.874.106 Safari/535.2
Build Identifier:
When having two conditions, both of them is using escape. One condition is true another condition is false. But combining them with OR, the expression is evaluated as false.
Reproducible: Always
### Steps to Reproduce:
select * from schemas where name like '%ys' escape '=' or name like '%y' escape '=';
### Actual Results:
No row returned
### Expected Results:
The row with name "sys" should be returned
## Comment 16529
Date: 2011-11-08 07:55:18 +0100
From: @njnes
seems to work fine in the release candidate (Dec2011). Probably some back porting is needed.
## Comment 16530
Date: 2011-11-08 08:14:49 +0100
From: @njnes
Fixed. The problem was in range merging optimization step. The like expressions
were incorrectly handled as ranges. Added test like_or.Bug-2924
## Comment 16531
Date: 2011-11-08 08:21:14 +0100
From: @njnes
Changeset [0299b070e196](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=0299b070e196) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0299b070e196](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0299b070e196)
Changeset description:
fixed like or bug #2924, ie don't handle like's as range
## Comment 16570
Date: 2011-11-22 13:55:57 +0100
From: @grobian
Fixed in Aug2011-SP3
| use LIKE with ESCAPE ignores OR operator | https://api.github.com/repos/MonetDB/MonetDB/issues/2924/comments | 0 | 2020-11-30T11:47:50Z | 2024-06-27T11:53:49Z | https://github.com/MonetDB/MonetDB/issues/2924 | 753,409,765 | 2,924 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-03 10:01:31 +0100
From: sellam
To: SQL devs <<bugs-sql>>
Version: 11.5.1 (Aug2011) [obsolete]
CC: @njnes
Last updated: 2011-11-22 13:55:54 +0100
## Comment 16513
Date: 2011-11-03 10:01:31 +0100
From: sellam
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.23) Gecko/20110927 Fedora/3.6.23-1.fc14 Firefox/3.6.23
Build Identifier:
The following example illustrates the bug:
1. Working example:
DECLARE x decimal(3,2);
set x=-1.21;
SELECT cast((x *100) as integer);
-> this returns -121, OK
2. Bugging example:
create table test(t decimal(3,2));
insert into test (t) values (-1.21);
select cast((t * 100) as integer) from test;
-> this returns 0
Reproducible: Always
### Steps to Reproduce:
create table test(t decimal(3,2));
insert into test (t) values (-1.21);
select cast((t * 100) as integer) from test;
### Actual Results:
+-----------+
| sql_mul_t |
+===========+
| 0 |
+-----------+
### Expected Results:
-121
bash-4.1$ hg branch
Aug2011
bash-4.1$ hg id
[998089b7cdd7](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=998089b7cdd7) (Aug2011) tip
## Comment 16517
Date: 2011-11-04 11:00:54 +0100
From: @njnes
fixed by adding brackets (c - ? precedence rules :-(
Added test Multiplication_and_cast.Bug-2923.sql
## Comment 16518
Date: 2011-11-04 11:04:46 +0100
From: @njnes
Changeset [4e75b961d68a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=4e75b961d68a) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4e75b961d68a](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=4e75b961d68a)
Changeset description:
fixed bug #2923. Precedence rule problem...
## Comment 16564
Date: 2011-11-22 13:55:54 +0100
From: @grobian
Fixed in Aug2011-SP3
| Multiplication and cast from decimal to integer leads to 0 value | https://api.github.com/repos/MonetDB/MonetDB/issues/2923/comments | 0 | 2020-11-30T11:47:47Z | 2024-06-27T11:53:48Z | https://github.com/MonetDB/MonetDB/issues/2923 | 753,409,737 | 2,923 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-01 19:18:29 +0100
From: @drstmane
To: SQL devs <<bugs-sql>>
Version: 11.5.1 (Aug2011) [obsolete]
Last updated: 2011-11-01 21:21:52 +0100
## Comment 16505
Date: 2011-11-01 19:18:29 +0100
From: @drstmane
Since changeset [a9ec95c08325](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a9ec95c08325) (http://dev.monetdb.org/hg/MonetDB/rev/a9ec95c08325) 12 SQL prepare statement tests fail; cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42204:593ed97f337f,42195:7b1f094c4c47&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
These tests did not fail in the Aug2010-SP2 release (changeset 7b1f094c4c47); cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42195:7b1f094c4c47,42172:5ab2786b8f74&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
These tests still fail in the Aug2011 branch (changeset 593ed97f337f) and in the Dec2011 branch (changeset 444ccffd7751); cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42212:444ccffd7751,42204:593ed97f337f&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
and
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42225:448759006e01,42212:444ccffd7751&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
## Comment 16506
Date: 2011-11-01 19:21:11 +0100
From: @grobian
mine, needs approving
## Comment 16507
Date: 2011-11-01 19:25:21 +0100
From: @drstmane
Ok --- are the error messages also correct/expected?
## Comment 16508
Date: 2011-11-01 19:25:51 +0100
From: @drstmane
With fixed changeset IDs, the above bug report correctly reads as follows:
Since changeset 42203:a9ec95c08325
(http://dev.monetdb.org/hg/MonetDB/rev/a9ec95c08325), 12 SQL prepare statement
tests fail; cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42204:593ed97f337f,42195:7b1f094c4c47&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
These tests did not fail in the Aug2010-SP2 release (changeset
42172:5ab2786b8f74); cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42195:7b1f094c4c47,42172:5ab2786b8f74&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
These tests still fail in the Aug2011 branch (changeset 42212:444ccffd7751) and
in the Dec2011 branch (changeset 42225:448759006e01); cf.,
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42212:444ccffd7751,42204:593ed97f337f&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
and
http://monetdb.cwi.nl/testweb/web/testgrid.php?serial=42225:448759006e01,42212:444ccffd7751&module=sql&targets=GNU-Fedora-x86_64-propcheck-assert&order=platform,arch,compiler&diff
## Comment 16509
Date: 2011-11-01 19:27:37 +0100
From: @grobian
I meant "improved". Cannot just approve indeed :)
Problem is I don't know how to figure out the right ids...
## Comment 16510
Date: 2011-11-01 21:21:52 +0100
From: @grobian
think I fixed all of them now
## Comment 16511
Date: 2011-11-02 09:38:09 +0100
From: @drstmane
Changeset [b1eba5cae691](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b1eba5cae691) made by Stefan Manegold <Stefan.Manegold@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b1eba5cae691](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b1eba5cae691)
Changeset description:
approved new function & variable names in MAL plans
after changeset [a9ec95c08325;](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=a9ec95c08325;)
see also Bug #2922
| prepare statement tests fail after changeset a9ec95c08325 | https://api.github.com/repos/MonetDB/MonetDB/issues/2922/comments | 0 | 2020-11-30T11:47:45Z | 2024-06-27T11:53:47Z | https://github.com/MonetDB/MonetDB/issues/2922 | 753,409,713 | 2,922 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-11-01 15:19:33 +0100
From: @sjoerdmullender
To: SQL devs <<bugs-sql>>
Version: 11.5.7 (Aug2011-SP2) [obsolete]
CC: @njnes
Last updated: 2011-11-22 13:55:53 +0100
## Comment 16503
Date: 2011-11-01 15:19:33 +0100
From: @sjoerdmullender
The following query is generated by the ODBC driver for the call to SQLTables(stmt, NULL, 0, NULL, 0, NULL, 0, "LOCAL TEMPORARY,GLOBAL TEMPORARY", SQL_NTS);
select e."value" as table_cat, s."name" as table_schem, t."name" as table_name, case when t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" <> 'tmp' then cast('TABLE' as varchar(20)) when t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" = 'tmp' then cast('GLOBAL TEMPORARY' as varchar(20)) when t."type" = 0 and t."system" = true and t."temporary" = 0 then cast('SYSTEM TABLE' as varchar(20)) when t."type" = 1 then cast('VIEW' as varchar(20)) when t."type" = 0 and t."system" = false and t."temporary" = 1 then cast('LOCAL TEMPORARY' as varchar(20)) else cast('INTERNAL TABLE TYPE' as varchar(20)) end as table_type, cast(null as varchar(1)) as remarks from sys."schemas" s, sys."tables" t, sys."env"() e where s."id" = t."schema_id" and e.name = 'gdk_dbname' and ((t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" = 'tmp') or (t."type" = 0 and t."system" = false and t."temporary" = 1)) order by table_type, table_schem, table_name;
The problem is, this query only returns rows for LOCAL TEMPORARY tables. When the final (complex) AND clause is halved by using either one or the other part on either side of the OR, the query works as expected, returning either the local or global temporaries.
Initialize the database with:
create local temporary table loctemp (i int);
create global temporary table globtemp (i int);
Then try the above query. It should result in two rows but only produces one.
The following two queries both return a single row, the first for the global temporary, the second for the local temporary:
select e."value" as table_cat, s."name" as table_schem, t."name" as table_name, case when t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" <> 'tmp' then cast('TABLE' as varchar(20)) when t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" = 'tmp' then cast('GLOBAL TEMPORARY' as varchar(20)) when t."type" = 0 and t."system" = true and t."temporary" = 0 then cast('SYSTEM TABLE' as varchar(20)) when t."type" = 1 then cast('VIEW' as varchar(20)) when t."type" = 0 and t."system" = false and t."temporary" = 1 then cast('LOCAL TEMPORARY' as varchar(20)) else cast('INTERNAL TABLE TYPE' as varchar(20)) end as table_type, cast(null as varchar(1)) as remarks from sys."schemas" s, sys."tables" t, sys."env"() e where s."id" = t."schema_id" and e.name = 'gdk_dbname' and ((t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" = 'tmp')) order by table_type, table_schem, table_name;
select e."value" as table_cat, s."name" as table_schem, t."name" as table_name, case when t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" <> 'tmp' then cast('TABLE' as varchar(20)) when t."type" = 0 and t."system" = false and t."temporary" = 0 and s."name" = 'tmp' then cast('GLOBAL TEMPORARY' as varchar(20)) when t."type" = 0 and t."system" = true and t."temporary" = 0 then cast('SYSTEM TABLE' as varchar(20)) when t."type" = 1 then cast('VIEW' as varchar(20)) when t."type" = 0 and t."system" = false and t."temporary" = 1 then cast('LOCAL TEMPORARY' as varchar(20)) else cast('INTERNAL TABLE TYPE' as varchar(20)) end as table_type, cast(null as varchar(1)) as remarks from sys."schemas" s, sys."tables" t, sys."env"() e where s."id" = t."schema_id" and e.name = 'gdk_dbname' and ((t."type" = 0 and t."system" = false and t."temporary" = 1)) order by table_type, table_schem, table_name;
## Comment 16504
Date: 2011-11-01 15:26:53 +0100
From: @sjoerdmullender
Changeset [8e893c3336de](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=8e893c3336de) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8e893c3336de](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=8e893c3336de)
Changeset description:
Test for bug #2921.
## Comment 16514
Date: 2011-11-03 14:48:09 +0100
From: @njnes
fixed bug in handling 'in' and 'not in' properly for 'constant' left hands.
## Comment 16515
Date: 2011-11-03 14:49:19 +0100
From: @njnes
Changeset [f4eaa8850588](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f4eaa8850588) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f4eaa8850588](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f4eaa8850588)
Changeset description:
fixed problem with in lists with constants on the left hand side.
Fixes Bug #2921
## Comment 16516
Date: 2011-11-03 20:25:33 +0100
From: @sjoerdmullender
I'm not sure it's the same problem, but these two queries should, I believe, give the same result, but they don't. The first one uses a very similar pattern as the original query in the bug report. I rewrote the UNION of SELECTs to a SELECT with an OR.
with sk as (select t.id as table_id, k.type as "type", c.name from sys."schemas" s, sys."tables" t, sys."columns" c, sys."keys" k, sys."objects" kc, sys."env"() e where s."id" = t."schema_id" and t."id" = c."table_id" and t."id" = k."table_id" and c."name" = kc."name" and kc."id" = k."id" and k."type" in (0, 1) and e."name" = 'gdk_dbname'), tid as (select t.id as tid from sys._tables t, sys.keys k where t.id = k.table_id and k.type = 0) select sk.* from sk where (sk.type = 0 and sk.table_id in (select tid from tid)) or (sk.type = 1 and sk.table_id not in (select tid from tid));
with sk as (select t.id as table_id, k.type as "type", c.name from sys."schemas" s, sys."tables" t, sys."columns" c, sys."keys" k, sys."objects" kc, sys."env"() e where s."id" = t."schema_id" and t."id" = c."table_id" and t."id" = k."table_id" and c."name" = kc."name" and kc."id" = k."id" and k."type" in (0, 1) and e."name" = 'gdk_dbname'), tid as (select t.id as tid from sys._tables t, sys.keys k where t.id = k.table_id and k.type = 0) select sk.* from sk where sk.type = 0 and sk.table_id in (select tid from tid) union select sk.* from sk where sk.type = 1 and sk.table_id not in (select tid from tid);
## Comment 16519
Date: 2011-11-04 12:35:09 +0100
From: @njnes
this is a different problem. The problem is related to incorrect renaming of the internal column '%TID'.
## Comment 16520
Date: 2011-11-04 13:03:05 +0100
From: @njnes
Changeset [b2bfd42795eb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=b2bfd42795eb) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b2bfd42795eb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=b2bfd42795eb)
Changeset description:
fix bug #2921 (part 2). Internal expressions shouldn't be renamed.
## Comment 16563
Date: 2011-11-22 13:55:53 +0100
From: @grobian
Fixed in Aug2011-SP3
| complex where clause broken | https://api.github.com/repos/MonetDB/MonetDB/issues/2921/comments | 0 | 2020-11-30T11:47:42Z | 2024-06-27T11:53:46Z | https://github.com/MonetDB/MonetDB/issues/2921 | 753,409,676 | 2,921 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-31 11:27:29 +0100
From: @sjoerdmullender
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @njnes
Last updated: 2011-11-22 13:55:57 +0100
## Comment 16502
Date: 2011-10-31 11:27:29 +0100
From: @sjoerdmullender
After a database update from Aug2011 to Dec2011, the query
select * from sys._tables where name like '%';
fails with the message Subquery result missing.
The cause of this is that in the updated database the filter functions from createdb/09_like.sql are missing.
## Comment 16512
Date: 2011-11-02 22:20:26 +0100
From: @njnes
There is a new db-init file, ie 09_like.sql. This should be run. Should be run for upgrade from aug2011. We need a way to detect that at the sql level (ie after loading logs/catalog) but before first user starts.
## Comment 16533
Date: 2011-11-08 10:26:42 +0100
From: @njnes
fixed, ie we now add the filters when upgrading the db.
## Comment 16569
Date: 2011-11-22 13:55:57 +0100
From: @grobian
Fixed in Aug2011-SP3
| Updated database from Aug2011 to Dec2011 make queries fail with "Subquery result missing". | https://api.github.com/repos/MonetDB/MonetDB/issues/2920/comments | 0 | 2020-11-30T11:47:39Z | 2024-06-27T11:53:45Z | https://github.com/MonetDB/MonetDB/issues/2920 | 753,409,654 | 2,920 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-30 12:58:10 +0100
From: taowen
To: SQL devs <<bugs-sql>>
Version: 11.5.7 (Aug2011-SP2) [obsolete]
CC: @njnes
Last updated: 2011-11-22 13:55:55 +0100
## Comment 16501
Date: 2011-10-30 12:58:10 +0100
From: taowen
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Ubuntu/10.04 Chromium/15.0.874.102 Chrome/15.0.874.102 Safari/535.2
Build Identifier:
If in same transaction, new column being added to a existing table. Then you try to select the new column, exception will be raised. Is DDL and DML in same transaction supported, or I should always separate them in two transactions?
Reproducible: Always
### Steps to Reproduce:
1. Create table
START TRANSACTION;
CREATE TABLE a(b INTEGER);
COMMIT;
2. Add column and select
START TRANSACTION;
ALTER TABLE a ADD COLUMN c INTEGER;
INSERT INTO a(b) VALUES(1);
SELECT c FROM a;
### Actual Results:
auto commit mode: off
operation successful
1 affected row
SQLException:sql.bind:unable to find sys.a(c)
### Expected Results:
No exception raised
This bug does not use SAVEPOINT feature, just normal transaction.
## Comment 16532
Date: 2011-11-08 10:25:42 +0100
From: @njnes
fixed. The storage layer didn't create the update bat in this case.
Added test alter_table_add_column.Bug-2919.sql
## Comment 16536
Date: 2011-11-08 13:20:58 +0100
From: @njnes
Changeset [063cb8dc1fca](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=063cb8dc1fca) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=063cb8dc1fca](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=063cb8dc1fca)
Changeset description:
fixed bug #2919. We create all required bats for persistent tables again.
## Comment 16566
Date: 2011-11-22 13:55:55 +0100
From: @grobian
Fixed in Aug2011-SP3
| Add column and select in same transaction | https://api.github.com/repos/MonetDB/MonetDB/issues/2919/comments | 0 | 2020-11-30T11:47:37Z | 2024-06-27T11:53:44Z | https://github.com/MonetDB/MonetDB/issues/2919 | 753,409,621 | 2,919 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-30 09:14:35 +0100
From: taowen
To: SQL devs <<bugs-sql>>
Version: 11.5.7 (Aug2011-SP2) [obsolete]
CC: @mlkersten
Last updated: 2011-10-30 12:40:36 +0100
## Comment 16497
Date: 2011-10-30 09:14:35 +0100
From: taowen
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.2 (KHTML, like Gecko) Ubuntu/10.04 Chromium/15.0.874.102 Chrome/15.0.874.102 Safari/535.2
Build Identifier:
the content in bug.sql:
START TRANSACTION;
CREATE TABLE a(b INTEGER);
SAVEPOINT sc;
ALTER TABLE a ADD COLUMN c INTEGER;
INSERT INTO a(b, c) VALUES(1, 2);
SELECT c FROM a;
Reproducible: Always
### Steps to Reproduce:
1.mclient bug.sql
### Actual Results:
auto commit mode: off
operation successful
auto commit mode: off
operation successful
1 affected row
SQLException:sql.bind:unable to find sys.a(c)
### Expected Results:
Should not raise exception
I filed a similar bug previously, that was fixed by upgrading to SP2. This one is different.
## Comment 16499
Date: 2011-10-30 12:08:10 +0100
From: taowen
I tried to workaround this issue by not using SAVEPOINT. However, if I create the database in another transaction, and commit it. Start a new transaction, add some column, then select that column, same bug will appear again. Which means, I can not add new column and select from it in the same transaction. But if I create the table and add column in the same transaction, everything is fine.
## Comment 16500
Date: 2011-10-30 12:40:36 +0100
From: @mlkersten
SAVEPOINT is according to the latest knowledge not supported.
see http://www.monetdb.org/Documentation/Manuals/SQLreference/Features/unsupported
| Column added after savepoint can not be selected | https://api.github.com/repos/MonetDB/MonetDB/issues/2918/comments | 0 | 2020-11-30T11:47:34Z | 2024-06-28T13:19:20Z | https://github.com/MonetDB/MonetDB/issues/2918 | 753,409,593 | 2,918 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-30 09:12:01 +0100
From: @arjenpdevries
To: SQL devs <<bugs-sql>>
Version: 11.3.3 (Apr2011-SP1) [obsolete]
CC: @mlkersten
Last updated: 2011-10-30 09:16:08 +0100
## Comment 16495
Date: 2011-10-30 09:12:01 +0100
From: @arjenpdevries
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20100101 Firefox/7.0.1
Build Identifier:
Try
select * from sys.storage() where schema = 'sys';
The schema keyword occurrence triggers a syntax error, but the column name is schema (try select * from sys.storage() limit 1;)
Greetings,
Arjen
Reproducible: Always
### Steps to Reproduce:
1.select * from sys.storage() where schema = 'sys';
2.
3.
### Actual Results:
syntax error, unexpected SCHEMA in ...
### Expected Results:
show the storage() entries where column schema has value 'sys'.
## Comment 16496
Date: 2011-10-30 09:14:23 +0100
From: @mlkersten
SQL keywords used as attributes should be properly quoted.
select * from sys.storage() where "schema" = 'sys';
## Comment 16498
Date: 2011-10-30 09:16:08 +0100
From: @arjenpdevries
(In reply to comment 1)
> SQL keywords used as attributes should be properly quoted.
>
> select * from sys.storage() where "schema" = 'sys';
Tnx!
| sys.storage() result's schema column name triggers parsing error in select from sys.storage(); | https://api.github.com/repos/MonetDB/MonetDB/issues/2917/comments | 0 | 2020-11-30T11:47:32Z | 2024-06-28T13:19:19Z | https://github.com/MonetDB/MonetDB/issues/2917 | 753,409,568 | 2,917 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-29 14:58:42 +0200
From: ultan <<natlu2809>>
To: SQL devs <<bugs-sql>>
Version: 11.9.1 (Apr2012) [obsolete]
CC: @njnes
Last updated: 2012-05-25 12:58:45 +0200
## Comment 16494
Date: 2011-10-29 14:58:42 +0200
From: ultan <<natlu2809>>
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:7.0.1) Gecko/20100101 Firefox/7.0.1
Build Identifier:
Creating a user under monetdb and then connecting as that user, it would be expected that that user could create temporary tables in the tmp schema.
Reproducible: Always
### Steps to Reproduce:
connect with mclient to a database as admin user monetdb
create a user
disconnect
connect with mclient as new user
attempt to create a temporary table
### Actual Results:
receive error that user has insufficient privileges.
For example:
-----------------------------------------------------
ultan@boxer:~/$ mclient -d TEST2H
user:monetdb
password:
Welcome to mclient, the MonetDB/SQL interactive terminal (Aug2011-SP1)
Database: MonetDB v11.5.3, 'TEST2H'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create user TestUser with password 'password' name 'uoc' schema "sys";
operation successful (117.441ms)
sql>\q
ultan@boxer:~/$ mclient -u TestUser -d TEST2H
password:
Welcome to mclient, the MonetDB/SQL interactive terminal (Aug2011-SP1)
Database: MonetDB v11.5.3, 'TEST2H'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create global temporary table t (col1 int);
CREATE TABLE: insufficient privileges for user 'ultan' in schema 'tmp'
sql>create local temporary table t (col1 int);
CREATE TABLE: insufficient privileges for user 'ultan' in schema 'tmp'
sql>create temporary table t (col1 int);
CREATE TABLE: insufficient privileges for user 'ultan' in schema 'tmp'
sql>
### Expected Results:
expect temporary table creation to be allowed and successful
## Comment 16581
Date: 2011-11-29 10:45:31 +0100
From: @grobian
With the Dec2011 branch:
% $INSTALL_DIR/bin/mclient -u testuser test2h
password:
Welcome to mclient, the MonetDB/SQL interactive terminal (Dec2011-a3d851cb1775)
Database: MonetDB v11.7.0 (Dec2011-a3d851cb1775), 'mapi:monetdb://vomac.cwi.nl:54321/test2h'
Type \q to quit, \? for a list of available commands
auto commit mode: on
sql>create global temporary table t (col1 int);
CREATE TABLE: insufficient privileges for user 'testuser' in schema 'tmp'
sql>create local temporary table t (col1 int);
operation successful (17.816ms)
sql>create temporary table t (col1 int);
CREATE TABLE: name 't' already in use
sql>drop table t;
operation successful (4.918ms)
sql>create temporary table t (col1 int);
operation successful (3.024ms)
sql>
That is, we now allow creating temporary tables and local temporary tables, but not global temporary tables. Is this enough for you?
## Comment 17237
Date: 2012-05-07 20:42:02 +0200
From: @njnes
global temporaries change the global schema, as such by default new users do not have such rights.
| non admin user cannot create temporary tables | https://api.github.com/repos/MonetDB/MonetDB/issues/2916/comments | 0 | 2020-11-30T11:47:29Z | 2024-06-27T11:53:41Z | https://github.com/MonetDB/MonetDB/issues/2916 | 753,409,545 | 2,916 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-29 14:47:22 +0200
From: ultan <<natlu2809>>
To: SQL devs <<bugs-sql>>
Version: 11.7.5 (Dec2011) [obsolete]
CC: @njnes
Last updated: 2012-03-16 14:56:56 +0100
## Comment 16493
Date: 2011-10-29 14:47:22 +0200
From: ultan <<natlu2809>>
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:7.0.1) Gecko/20100101 Firefox/7.0.1
Build Identifier:
Trying to create adapter in Apache Jena for MonetDB. Jena unit test "Optional - 1" is failing while connected to newly created database with a particular query.
Merovingian log shows :
2011-10-27 20:53:45 MSG merovingian[7189]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2011-10-27 20:53:45 MSG merovingian[7189]: proxying client localhost.localdomain:55769 for database 'TEST2H' to mapi:monetdb:///home/ultan/my-farm/TEST2H/.mapi.sock?database=TEST2H
2011-10-27 20:53:45 MSG merovingian[7189]: target connection is on local UNIX domain socket, passing on filedescriptor instead of proxying
2011-10-27 20:53:48 MSG merovingian[7189]: database 'TEST2H' (8527) was killed by signal SIGSEGV
2011-10-27 21:00:59 MSG merovingian[7189]: caught SIGTERM, starting shutdown sequence
2011-10-27 21:00:59 MSG control[7189]: control channel closed
2011-10-27 21:00:59 ERR discovery[7189]: error while sending broadcast message: Network is unreachable
2011-10-27 21:00:59 MSG merovingian[7189]: Merovingian 1.5 stopped
JDBC debug log shows
--------------------
RD 1319891391488: inserting prompt
TD 1319891391488: write final block: 14 bytes
TX 1319891391488: Xauto_commit 1
RD 1319891391488: read final block: 0 bytes
RX 1319891391488:
RD 1319891391488: inserting prompt
TD 1319891391502: write final block: 574 bytes
TX 1319891391502: sSELECT R_1.lex AS V_1_lex, R_1.datatype AS V_1_datatype, R_1.lang AS V_1_lang, R_1.type AS V_1_type,
R_2.lex AS V_2_lex, R_2.datatype AS V_2_datatype, R_2.lang AS V_2_lang, R_2.type AS V_2_type
FROM
( SELECT T_1.s AS X_1
FROM Triples AS T_1
WHERE ( T_1.p = 2004134117598721274
AND T_1.o = 435905340492217258 )
) AS T_1
LEFT OUTER JOIN
Triples AS T_2
ON ( T_2.p = 3816485599920428794
AND T_1.X_1 = T_2.s )
LEFT OUTER JOIN
Nodes AS R_1
ON ( T_1.X_1 = R_1.hash )
LEFT OUTER JOIN
Nodes AS R_2
ON ( T_2.o = R_2.hash )
;
TD 1319891419345: write final block: 11 bytes
TX 1319891419345: Xrelease 11
-------------------------
Reproducible: Always
### Steps to Reproduce:
1. Create Jena Schema and contents using following Dump:
---------------------------------------------------------
START TRANSACTION;
SET SCHEMA "sys";
CREATE TABLE "sys"."prefixes" (
"prefix" VARCHAR(50) NOT NULL,
"uri" VARCHAR(500) NOT NULL,
CONSTRAINT "prefixes_prefix_pkey" PRIMARY KEY ("prefix")
);
COPY 1 RECORDS INTO "sys"."prefixes" FROM stdin USING DELIMITERS '\t','\n','"';
":" "http://example/"
CREATE TABLE "sys"."nodes" (
"hash" BIGINT NOT NULL,
"lex" CHARACTER LARGE OBJECT NOT NULL,
"lang" VARCHAR(10) NOT NULL,
"datatype" VARCHAR(200),
"type" INTEGER NOT NULL,
CONSTRAINT "nodes_hash_pkey" PRIMARY KEY ("hash")
);
COPY 13 RECORDS INTO "sys"."nodes" FROM stdin USING DELIMITERS '\t','\n','"';
-8395209716130787220 "http://example/x" "" "" 2
2004134117598721274 "http://example/p" "" "" 2
435905340492217258 "1" "" "http://www.w3.org/2001/XMLSchemainteger" 5
745852752491398227 "2" "" "http://www.w3.org/2001/XMLSchemainteger" 5
-5334307821581591471 "3" "" "http://www.w3.org/2001/XMLSchemainteger" 5
4788264553748351656 "http://example/a" "" "" 2
8936870869765386580 "http://example/b" "" "" 2
3816485599920428794 "http://example/q1" "" "" 2
-5216419694041718175 "http://example/z1" "" "" 2
-8287201118401564753 "http://example/q2" "" "" 2
7121703792433320712 "http://example/z2" "" "" 2
-4064636373028764940 "http://example/z" "" "" 2
-3401798235167296541 "abc" "" "" 3
CREATE TABLE "sys"."triples" (
"s" BIGINT NOT NULL,
"p" BIGINT NOT NULL,
"o" BIGINT NOT NULL,
CONSTRAINT "triples_s_p_o_pkey" PRIMARY KEY ("s", "p", "o")
);
CREATE INDEX "objsubj" ON "sys"."triples" ("o", "s");
CREATE INDEX "predobj" ON "sys"."triples" ("p", "o");
COPY 9 RECORDS INTO "sys"."triples" FROM stdin USING DELIMITERS '\t','\n','"';
-8395209716130787220 2004134117598721274 435905340492217258
-8395209716130787220 2004134117598721274 745852752491398227
-8395209716130787220 2004134117598721274 -5334307821581591471
-8395209716130787220 2004134117598721274 4788264553748351656
-8395209716130787220 2004134117598721274 8936870869765386580
4788264553748351656 3816485599920428794 -5216419694041718175
4788264553748351656 -8287201118401564753 7121703792433320712
8936870869765386580 3816485599920428794 -5216419694041718175
-4064636373028764940 2004134117598721274 -3401798235167296541
CREATE TABLE "sys"."quads" (
"g" BIGINT NOT NULL,
"s" BIGINT NOT NULL,
"p" BIGINT NOT NULL,
"o" BIGINT NOT NULL,
CONSTRAINT "quads_g_s_p_o_pkey" PRIMARY KEY ("g", "s", "p", "o")
);
CREATE INDEX "graobjsubj" ON "sys"."quads" ("g", "o", "s");
CREATE INDEX "grapredobj" ON "sys"."quads" ("g", "p", "o");
CREATE INDEX "objsubjpred" ON "sys"."quads" ("o", "s", "p");
CREATE INDEX "predobjsubj" ON "sys"."quads" ("p", "o", "s");
CREATE INDEX "subjpredobj" ON "sys"."quads" ("s", "p", "o");
COMMIT;
-----------------------------------------------------------
2. Connect with mclient or jdbc connection and issue following query
-----------------------------------------------------------
SELECT R_1.lex AS V_1_lex, R_1.datatype AS V_1_datatype, R_1.lang AS V_1_lang, R_1.type AS V_1_type,
R_2.lex AS V_2_lex, R_2.datatype AS V_2_datatype, R_2.lang AS V_2_lang, R_2.type AS V_2_type
FROM
( SELECT T_1.s AS X_1
FROM Triples AS T_1
WHERE ( T_1.p = 2004134117598721274
AND T_1.o = 435905340492217258 )
) AS T_1
LEFT OUTER JOIN
Triples AS T_2
ON ( T_2.p = 3816485599920428794
AND T_1.X_1 = T_2.s )
LEFT OUTER JOIN
Nodes AS R_1
ON ( T_1.X_1 = R_1.hash )
---------------------------------------------------------------
3.
### Actual Results:
Connection crash, possibly server ?
JDBC shows SQLException : java.sql.SQLException: Read from localhost:50000: End of stream reached (mserver still alive?)
### Expected Results:
Statment execution, result set returned.
## Comment 16646
Date: 2011-12-09 09:40:21 +0100
From: @sjoerdmullender
Changeset [6b880a8a3952](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6b880a8a3952) made by Sjoerd Mullender <sjoerd@acm.org> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6b880a8a3952](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=6b880a8a3952)
Changeset description:
Added test for bug #2915.
## Comment 16699
Date: 2011-12-27 15:28:34 +0100
From: @njnes
Changeset [57700c6efacb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=57700c6efacb) made by Niels Nes <niels@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=57700c6efacb](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=57700c6efacb)
Changeset description:
fix bug with semijoins. This partialy fixes bug #2933 and bug #2915.
## Comment 17012
Date: 2012-02-27 19:36:00 +0100
From: @njnes
test runs without problems
## Comment 17103
Date: 2012-03-16 14:56:56 +0100
From: @grobian
Released in Dec2011-SP2
| Query (mclient or jdbc) with 2 outer joins appears to crash connectionor server | https://api.github.com/repos/MonetDB/MonetDB/issues/2915/comments | 0 | 2020-11-30T11:47:26Z | 2024-06-27T11:53:40Z | https://github.com/MonetDB/MonetDB/issues/2915 | 753,409,516 | 2,915 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-27 21:12:19 +0200
From: @grobian
To: MonetDB5 devs <<bugs-monetdb5>>
Version: 11.5.9 (Aug2011-SP3) [obsolete]
CC: @yzchang
Last updated: 2012-11-27 14:42:12 +0100
## Comment 16491
Date: 2011-10-27 21:12:19 +0200
From: @grobian
Oct 27 20:31:53 vomac mserver5[91283]: [91283] detected buffer overflow
Oct 27 20:31:56 vomac ReportCrash[91285]: Formulating crash report for process mserver5[91283]
Process: mserver5 [91283]
Path: /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/bin/mserver5
Identifier: mserver5
Version: ??? (???)
Code Type: PPC (Native)
Parent Process: python2.7 [85933]
Date/Time: 2011-10-27 20:31:53.730 +0200
OS Version: Mac OS X 10.5.8 (9L31a)
Report Version: 6
Anonymous UUID: A7B69245-A140-45A6-AA70-780DF77B525C
Exception Type: EXC_CRASH (SIGABRT)
Exception Codes: 0x0000000000000000, 0x0000000000000000
Crashed Thread: 3
Thread 0:
0 libSystem.B.dylib 0x905dad30 read + 8
1 libSystem.B.dylib 0x90624090 _sread + 44
2 libSystem.B.dylib 0x90623ffc __srefill + 376
3 libSystem.B.dylib 0x90623d78 fgets + 124
4 libstream.2.dylib 0x0008ba14 bstream_next + 388
5 libmonetdb5.8.dylib 0x00a5acb4 MCreadClient + 388
6 libmonetdb5.8.dylib 0x00a70180 MALreader + 96
7 libmonetdb5.8.dylib 0x00a6ef1c runScenarioBody + 284
8 libmonetdb5.8.dylib 0x00a6f1d0 runScenario + 64
9 libmonetdb5.8.dylib 0x00a70404 MSserveClient + 244
10 mserver5 0x00003ad4 main + 3556
11 mserver5 0x00002990 start + 64
Thread 1:
0 libSystem.B.dylib 0x9063dc14 select$DARWIN_EXTSN + 12
1 libmonetdb5.8.dylib 0x01273544 SERVERlistenThread + 372
Thread 2:
0 libSystem.B.dylib 0x9063dc14 select$DARWIN_EXTSN + 12
1 libbat.4.dylib 0x014aaa98 MT_sleep_ms + 104
2 lib_sql.bundle 0x073f5e14 store_manager + 116
3 libSystem.B.dylib 0x9061bf70 _pthread_start + 316
Thread 3 Crashed:
0 libSystem.B.dylib 0x906e42ac __semwait_signal_nocancel + 8
1 libSystem.B.dylib 0x906e3d78 nanosleep$NOCANCEL$UNIX2003 + 176
2 libSystem.B.dylib 0x906dcfdc usleep$NOCANCEL$UNIX2003 + 68
3 libSystem.B.dylib 0x906f6b9c __abort + 88
4 libSystem.B.dylib 0x906b4238 __chk_fail + 116
5 libSystem.B.dylib 0x90695110 __strcat_chk + 64
6 lib_sql.bundle 0x073b58dc table_element + 1452
7 lib_sql.bundle 0x073b6994 rel_create_table + 724
8 lib_sql.bundle 0x073b746c rel_schemas + 828
9 lib_sql.bundle 0x07320b3c sql_symbol2stmt + 76
10 lib_sql.bundle 0x07305ed8 SQLparser + 2424
11 libmonetdb5.8.dylib 0x00a6ef70 runScenarioBody + 368
12 libmonetdb5.8.dylib 0x00a6f1d0 runScenario + 64
13 libmonetdb5.8.dylib 0x00a70404 MSserveClient + 244
14 libSystem.B.dylib 0x9061bf70 _pthread_start + 316
Thread 3 crashed with PPC Thread State 32:
srr0: 0x906e42ac srr1: 0x0200f030 dar: 0x00041000 dsisr: 0x42000000
r0: 0x000001a7 r1: 0xf03048a0 r2: 0xa00fd3c0 r3: 0x0000003c
r4: 0x00000000 r5: 0x00000001 r6: 0x00000001 r7: 0x00000000
r8: 0x00989680 r9: 0x00989680 r10: 0x0000800c r11: 0xa01017fc
r12: 0x906e3cc8 r13: 0x00000000 r14: 0x00000000 r15: 0x00000000
r16: 0x00000000 r17: 0x00000000 r18: 0x00000000 r19: 0xa00fc498
r20: 0x00000001 r21: 0x074263f4 r22: 0xa00fc498 r23: 0x0244b688
r24: 0x0742640c r25: 0x0743bebc r26: 0x0729c2f8 r27: 0x0296ae08
r28: 0x0244b808 r29: 0xf0304938 r30: 0x00000000 r31: 0x906e3cdc
cr: 0x48082448 xer: 0x00000000 lr: 0x906e3d7c ctr: 0x906e3cc8
vrsave: 0x00000000
Binary Images:
0x1000 - 0x4ff3 +mserver5 ??? (???) <64ab849dfb533c2ebc3e1b93b0ac5df3> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/bin/mserver5
0xe000 - 0xeffe +lib_sphinx.bundle ??? (???) <b4e2a14d2573329f88d52db090548919> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/monetdb5/lib_sphinx.bundle
0x15000 - 0x26ff3 +libmapi.3.dylib ??? (???) <8a89436a149f38cfba23a94b4892b6f8> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/libmapi.3.dylib
0x43000 - 0x73ff2 +libreadline.6.dylib ??? (???) <ed59f22c8c8cb43544e03afc38e1e913> /Volumes/Scratch/gentoo/lib/libreadline.6.dylib
0x86000 - 0x90ffe +libstream.2.dylib ??? (???) <d7da874a4eda35f0a6efb787b434339e> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/libstream.2.dylib
0x94000 - 0xa4ff9 +libbz2.1.dylib ??? (???) <b12c843f2eb2c4bd103c692a5f5129bb> /Volumes/Scratch/gentoo/lib/libbz2.1.dylib
0xa7000 - 0xd7fff +libpcre.0.dylib ??? (???) <e9cf5ed299689db043362d898e53d7a8> /Volumes/scratch/gentoo/lib/libpcre.0.dylib
0xeb000 - 0x13bff3 +libcurl.4.dylib ??? (???) <7e5a720dce933d5fd7e43420db89ff32> /Volumes/Scratch/gentoo/usr/lib/libcurl.4.dylib
0x147000 - 0x15bfff +libz.1.dylib ??? (???) <d785d0b8a8d3dcc3cdd83f433b10e1f8> /Volumes/Scratch/gentoo/lib/libz.1.dylib
0x15f000 - 0x16aff7 +libgcc_s.1.dylib ??? (???) <9b7348fc3e4a756e442af0f88b67f1bb> /Volumes/Scratch/gentoo/usr/lib/gcc/powerpc-apple-darwin9/4.2.1/libgcc_s.1.dylib
0x174000 - 0x17aff3 +lib_geom.bundle ??? (???) <82825c36a5fc30e98315995df0bd7f4d> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/monetdb5/lib_geom.bundle
0x1ab000 - 0x32cffb +libxml2.2.dylib ??? (???) <9201b6682c6a5a4fd27be19188b715f5> /Volumes/Scratch/gentoo/usr/lib/libxml2.2.dylib
0x356000 - 0x437ff7 +libiconv.2.dylib ??? (???) <08fb79206d003511b7cf48e604343557> /Volumes/Scratch/gentoo/lib/libiconv.2.dylib
0x441000 - 0x48affb +libssl.1.0.0.dylib ??? (???) <d89a5fb7ec8fb5529262a96eb572a6ab> /Volumes/Scratch/gentoo/usr/lib/libssl.1.0.0.dylib
0x4a0000 - 0x5e2ffb +libcrypto.1.0.0.dylib ??? (???) <bd5ddd0743c40de11305e6ed52fe404c> /Volumes/Scratch/gentoo/usr/lib/libcrypto.1.0.0.dylib
0x641000 - 0x683ff3 +libncurses.5.dylib ??? (???) <707a5e99b27e1bd558a77f2237115834> /Volumes/Scratch/gentoo/lib/libncurses.5.dylib
0x6a5000 - 0x6baff3 +libgeos_c.1.dylib ??? (???) <b391907839ef1ab526800696ba512736> /Volumes/Scratch/gentoo/usr/lib/libgeos_c.1.dylib
0x6c6000 - 0x6cefff +libsphinxclient-0.0.1.dylib ??? (???) <4b2bce3a56373279a669312b9ebf5d75> /Volumes/Scratch/gentoo/usr/lib/libsphinxclient-0.0.1.dylib
0xa4e000 - 0x13d9fff +libmonetdb5.8.dylib ??? (???) <3ad8ef039b743d87848f30ebac3ec17f> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/libmonetdb5.8.dylib
0x149c000 - 0x1e6bff3 +libbat.4.dylib ??? (???) <95ad4f3267f93d198617c88690159f35> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/libbat.4.dylib
0x6f00000 - 0x7024ffb +libgeos-3.2.2.dylib ??? (???) <153f3172dc7688cf3048be194f6f19ba> /Volumes/Scratch/gentoo/usr/lib/libgeos-3.2.2.dylib
0x70d6000 - 0x7173ff7 +libstdc++.6.dylib ??? (???) <6508b3616ed55d549fb775de7f97de48> /Volumes/scratch/gentoo/usr/lib/gcc/powerpc-apple-darwin9/4.2.1/libstdc++.6.dylib
0x7300000 - 0x7439fff +lib_sql.bundle ??? (???) <f43558df83ec32c787abb6179952f65e> /Volumes/Scratch/monet/GNU-Darwin-powerpc-444ccffd7751/install/lib/monetdb5/lib_sql.bundle
0x8fe00000 - 0x8fe30c23 dyld 97.1 (???) <89a0055b0e7ea2db881b73c6e63bc774> /usr/lib/dyld
0x905d9000 - 0x90779fe3 libSystem.B.dylib ??? (???) <7dc28e19e1aac16b29cbd7c5d9ce9638> /usr/lib/libSystem.B.dylib
0x90b58000 - 0x90b5dff6 libmathCommon.A.dylib ??? (???) /usr/lib/system/libmathCommon.A.dylib
0xffff8000 - 0xffff9703 libSystem.B.dylib ??? (???) /usr/lib/libSystem.B.dylib
## Comment 16571
Date: 2011-11-22 15:21:00 +0100
From: @grobian
This happens for test:
sql/test/BugTracker-2010/huge_nr_of_columns.Bug-2609
## Comment 16572
Date: 2011-11-22 17:26:23 +0100
From: @grobian
Changeset [f07428e6c155](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=f07428e6c155) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f07428e6c155](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=f07428e6c155)
Changeset description:
table_constraint_name: avoid potential buffer overflow
strncpy does not add \0 if it doesn't fit. strncat doesn't work
properly on Darwin9, so avoid using it to avoid a buffer overflow here
as well. Bug #2914.
## Comment 16828
Date: 2012-01-26 15:32:04 +0100
From: @sjoerdmullender
The Dec2011 version has been release, so declaring this bug as FIXED.
## Comment 18089
Date: 2012-11-27 14:42:12 +0100
From: @yzchang
Already covered by test sql/test/BugTracker-2010/huge_nr_of_columns.Bug-2609
| mserver5 crash for huge_nr_of_columns.Bug-2609 on ppc-macos | https://api.github.com/repos/MonetDB/MonetDB/issues/2914/comments | 0 | 2020-11-30T11:47:23Z | 2024-06-27T11:53:39Z | https://github.com/MonetDB/MonetDB/issues/2914 | 753,409,486 | 2,914 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-27 17:19:42 +0200
From: taowen
To: SQL devs <<bugs-sql>>
Version: 11.5.3 (Aug2011-SP1) [obsolete]
Duplicates: #2712
Last updated: 2011-10-27 20:55:23 +0200
## Comment 16485
Date: 2011-10-27 17:19:42 +0200
From: taowen
User-Agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/535.1 (KHTML, like Gecko) Ubuntu/10.04 Chromium/14.0.835.202 Chrome/14.0.835.202 Safari/535.1
Build Identifier:
Inside a savepoint, create and then drop a column. No matter commit or rollback, exception raised. However, if you do nothing (I think the connection just close), the transaction rolled back and no complain.
Reproducible: Always
### Steps to Reproduce:
START TRANSACTION;
SAVEPOINT sp;
CREATE TABLE a(b INTEGER);
ALTER TABLE a ADD COLUMN c INTEGER;
ALTER TABLE a DROP COLUMN c;
ROLLBACK;
### Actual Results:
When you commit it will complain:
auto commit mode: off
auto commit mode: off
operation successful
operation successful
operation successful
MALException:!ERROR: BBPdecref: tmp_1625 does not have logical references.
ERROR: BBPdecref: tmp_1626 does not have logical references.
ERROR: BBPdecref: tmp_27712 does not have logical references.
ERROR: BBPdecref: tmp_1631 does not have logical references.
When you rollback it will complain:
auto commit mode: off
auto commit mode: off
operation successful
operation successful
operation successful
MALException:!ERROR: BBPdecref: tmp_27733 does not have logical references.
ERROR: BBPdecref: tmp_27740 does not have logical references.
### Expected Results:
Exception should not be raised
Without savepoint, everything works fine.
Even the table is created before the savepoint, same behavior.
Event the column is created before the savepoint, same behavior.
START TRANSACTION;
CREATE TABLE a(b INTEGER);
ALTER TABLE a ADD COLUMN c INTEGER;
SAVEPOINT sp;
ALTER TABLE a DROP COLUMN c;
ROLLBACK;
## Comment 16486
Date: 2011-10-27 17:34:27 +0200
From: taowen
I tried to create the table in a separate transaction.
Then I simply do a drop column after savepoint, still save behavior.
So, the conclusion is, whenever you drop a column after savepoint, exception will be raised no matter you commit or rollback.
## Comment 16487
Date: 2011-10-27 17:40:59 +0200
From: @sjoerdmullender
I see you specified the Aug2011-SP1 release. Can you try this with the Aug2011-SP2 release? It contains a fix that was related to savepoints (bug #2712).
(When I try the first example, I don't get any errors.)
## Comment 16488
Date: 2011-10-27 17:52:20 +0200
From: taowen
(In reply to comment 2)
> I see you specified the Aug2011-SP1 release. Can you try this with the
> Aug2011-SP2 release? It contains a fix that was related to savepoints (bug
> 2712).
>
> (When I try the first example, I don't get any errors.)
Thanks! Confirmed, it is fixed in SP2.
## Comment 16489
Date: 2011-10-27 20:55:23 +0200
From: @sjoerdmullender
This problem was already fixed in Aug2011-SP2. Also, it's probably the same as bug #2712, so marking it as duplicate.
*** This bug has been marked as a duplicate of bug #2712 ***
| create and drop column inside savepoint causing error | https://api.github.com/repos/MonetDB/MonetDB/issues/2913/comments | 0 | 2020-11-30T11:47:20Z | 2024-06-28T13:19:18Z | https://github.com/MonetDB/MonetDB/issues/2913 | 753,409,460 | 2,913 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-26 14:34:32 +0200
From: sellam
To: SQL devs <<bugs-sql>>
Version: -- development
CC: @njnes
Last updated: 2011-11-22 13:55:55 +0100
## Comment 16471
Date: 2011-10-26 14:34:32 +0200
From: sellam
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.23) Gecko/20110927 Fedora/3.6.23-1.fc14 Firefox/3.6.23
Build Identifier:
Here is an example session. We create a table, create a query function over it. The query works. We then load data in the table: after this, the function crashes.
It seems that adding LIMIT 1 in the SELECT statement of inRange makes it work.
sql>create table trips (tripid bigint, lon decimal(8,5), lat decimal(7,5), time bigint);
sql>CREATE FUNCTION inRange(xx DECIMAL(8,5), yy DECIMAL(7,5), ww INTEGER, hh INTEGER)
more> RETURNS TABLE (tripid BIGINT, lon DECIMAL(8,5), lat DECIMAL(7,5), time BIGINT)
more>BEGIN
more>RETURN SELECT T.tripid, T.lon, T.lat, T.time
more> FROM trips T
more> WHERE T.lon > xx AND T.lon < xx + ww
more> AND T.lat > yy AND T.lat < yy + hh;
more>END;
operation successful (19.205ms)
sql>SELECT * FROM inRange(0,0,1,1);
+--------+-----+-----+------+
| tripid | lon | lat | time |
+========+=====+=====+======+
+--------+-----+-----+------+
0 tuples (35.449ms)
sql>DEBUG SELECT * FROM inRange(0,0,1,1);
mdb> mdb.start();
mdb>l user.inrange
0 function user.inrange(Axx,Ayy,Aww,Ahh);
1 X_21 := nil:bat[:oid,:lng];
2 X_36:bat[:oid,:int] := nil:bat[:oid,:int];
3 X_39:bat[:oid,:int] := nil:bat[:oid,:int];
4 X_58 := nil:bat[:oid,:int];
5 X_69:bat[:oid,:int] := nil:bat[:oid,:int];
6 X_77:bat[:oid,:int] := nil:bat[:oid,:int];
7 X_86 := nil:bat[:oid,:int];
8 X_106 := nil:bat[:oid,:oid];
9 barrier X_141 := language.dataflow();
10 X_9 := sql.mvc();
11 X_11:bat[:oid,:lng] := sql.bind(X_9,"sys","trips","tripid",1);
12 X_15:bat[:oid,:lng] := sql.bind(X_9,"sys","trips","tripid",2);
13 X_17:bat[:oid,:lng] := sql.bind(X_9,"sys","trips","tripid",0);
14 X_19 := algebra.kdifference(X_17,X_15);
15 X_20 := algebra.kunion(X_19,X_15);
16 X_21 := algebra.kunion(X_20,X_11);
17 X_22 := calc.lng(5,Axx,15,5);
18 X_25 := calc.lng(Aww,15,5);
etc...
**** Until here everything works fine. Now come the problems:
sql>copy 240540781 records into trips from '/net/rig.ins.cwi.nl/export/scratch0/manegold/TomTom/data.gz' using delimiters ',','\n';
Reproducible: Always
### Steps to Reproduce:
create table trips (tripid bigint, lon decimal(8,5), lat decimal(7,5), time bigint);
-- FUNCTION RETURNING POINTS IN RANGE
CREATE FUNCTION inRange(xx DECIMAL(8,5), yy DECIMAL(7,5), ww INTEGER, hh INTEGER)
RETURNS TABLE (tripid BIGINT, lon DECIMAL(8,5), lat DECIMAL(7,5), time BIGINT)
BEGIN
RETURN SELECT T.tripid, T.lon, T.lat, T.time
FROM trips T
WHERE T.lon > xx AND T.lon < xx + ww
AND T.lat > yy AND T.lat < yy + hh;
END;
SELECT * FROM inRange(0,0,1,1);
copy 240540781 records into trips from '/net/rig.ins.cwi.nl/export/scratch0/manegold/TomTom/data.gz' using delimiters ',','\n';
SELECT * FROM inRange(0,0,1,1);
### Actual Results:
SQLException:sql.resultSet:Cannot access descriptorSQLException:sql.resultSet:Cannot access descriptor
### Expected Results:
A result set
## Comment 16472
Date: 2011-10-26 14:37:42 +0200
From: sellam
After loading data :
mdb>l user.inrange
? 0 function user.inrange(Axx,Ayy,Aww,Ahh);
? 1 return (X_9,X_10,X_11,X_12);
? 2 end inrange;
The body disappeared
## Comment 16473
Date: 2011-10-26 17:04:56 +0200
From: @sjoerdmullender
Which version of MonetDB?
Since Fabian has been fixing stuff in this area, it is important to be exact here, so if you've built from source, what does "hg id" say?
## Comment 16474
Date: 2011-10-26 18:03:11 +0200
From: sellam
(In reply to comment 2)
> Which version of MonetDB?
> Since Fabian has been fixing stuff in this area, it is important to be exact
> here, so if you've built from source, what does "hg id" say?
hg id gives:
[6b371872cfe4](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=6b371872cfe4)
hg parent:
changeset: 42187:6b371872cfe4
parent: 42182:5b45d264631c
parent: 42186:9db8d45b28ee
user: Stefan Manegold <Stefan.Manegold@cwi.nl>
date: Sat Oct 22 10:38:33 2011 +0200
summary: Merge With Dec2011 branch.
## Comment 16475
Date: 2011-10-27 10:30:53 +0200
From: @sjoerdmullender
Can you try it on the Dec2011 branch? That branch contains Fabian's fixes. They have not yet been propagated to the default branch that you were using.
## Comment 16478
Date: 2011-10-27 10:57:06 +0200
From: @grobian
function user.inrange is empty apart from the return, I doubt my fixes to the query cache affect this at all.
## Comment 16528
Date: 2011-11-08 07:53:00 +0100
From: @njnes
This is probably fixed by the fix too return statements (ie make it a proper assignment). This fixed zones and datacell problems.
Could you repeat the test.
## Comment 16534
Date: 2011-11-08 11:39:22 +0100
From: sellam
I reproduced and it worked fine. Problem solved.
Thanks for your help
copy 240540781 records into trips from
more>'/export/scratch0/sellam/documents/data-sets/tomtom-example/data.gz' using delimiters
more>',','\n';
240540781 affected rows (11m 4s)
sql>SELECT * FROM inRange(0,0,1,1);
+--------+-----+-----+------+
| tripid | lon | lat | time |
+========+=====+=====+======+
+--------+-----+-----+------+
0 tuples (1m 30s)
## Comment 16535
Date: 2011-11-08 11:41:35 +0100
From: @njnes
fixed, tested by zones and datacell test, ie no extra test needed
## Comment 16567
Date: 2011-11-22 13:55:55 +0100
From: @grobian
Fixed in Aug2011-SP3
| Function body disappears after loading data - error message: SQLException:sql.resultSet:Cannot access descriptor | https://api.github.com/repos/MonetDB/MonetDB/issues/2912/comments | 0 | 2020-11-30T11:47:17Z | 2024-06-27T11:53:37Z | https://github.com/MonetDB/MonetDB/issues/2912 | 753,409,423 | 2,912 |
[
"MonetDB",
"MonetDB"
] | Date: 2011-10-25 16:04:34 +0200
From: Wouter Alink <<alink>>
To: clients devs <<bugs-clients>>
Version: -- development
Last updated: 2011-12-07 14:59:30 +0100
## Comment 16466
Date: 2011-10-25 16:04:34 +0200
From: Wouter Alink <<alink>>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.2.20) Gecko/20110816 Fedora/3.6.20-1.fc14 Firefox/3.6.20
Build Identifier:
To control a database on a remote machine, it would be useful to have a very lightweight package that at least contains 'monetdb' (create db/destroy db/etc) and 'mclient' (to interact with the individual databases). Preferably it should not contain the mserver, monetdbd, etc tools.
I think the lightweight package can already be achieved when creating a bundle of mero-control.jar and jdbc-client.jar.
Reproducible: Always
(Fabian asked me to make a feature request)
## Comment 16476
Date: 2011-10-27 10:39:33 +0200
From: @sjoerdmullender
The MonetDB-client package is already pretty minimal. It contains mclient, msqldump, and stethoscope. When you install it, you also get MonetDB-stream, but that's it.
Would it make sense and would it satisfy youif we were to move the monetdb program to MonetDB-client?
(These are the Fedora names, the Debian/Ubuntu situation is similar.)
## Comment 16477
Date: 2011-10-27 10:56:02 +0200
From: @grobian
monetdb must be in clients, then there must be a --enable-clients and --disable-gdk such that you can build only clients + java + monetdb, and also the inverse, to get a system that has only the servers, not the clients (from administrative point of view you might want that)
## Comment 16617
Date: 2011-12-07 14:58:15 +0100
From: @grobian
Changeset [3722e4299e26](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=3722e4299e26) made by Fabian Groffen <fabian@cwi.nl> in the MonetDB repo, refers to this bug.
For complete details, see [http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3722e4299e26](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=http//devmonetdborg/hg/MonetDB?cmd=changeset;node=3722e4299e26)
Changeset description:
monetdbd: make building/install optional based on monetdb5
Bug #2911: install monetdb tool as client (independent from monetdbd)
Because monetdb depends on sabaoth via utils, moved msabaoth to
common/utils, and changed Makefiles accordingly.
## Comment 16618
Date: 2011-12-07 14:59:30 +0100
From: @grobian
This (build-wise) feature is implemented in the Dec2011 branch
| client package: monetdb + mclient | https://api.github.com/repos/MonetDB/MonetDB/issues/2911/comments | 0 | 2020-11-30T11:47:14Z | 2024-06-27T11:53:36Z | https://github.com/MonetDB/MonetDB/issues/2911 | 753,409,393 | 2,911 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.