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: 2008-03-01 12:36:07 +0100
From: @lsidir
To: Lefteris Sidirourgos <<lsidir>>
Version: -- development
Last updated: 2008-03-01 12:51:32 +0100
## Comment 9111
Date: 2008-03-01 12:36:07 +0100
From: @lsidir
The algebra implementation of functions fn:substring(str,dbl) and fn:substring(str,dbl,dbl) is not correct. First, the double values must be round up to the closest integer and second, the XQuery function spec states that the first position of a string is 1 and not 0, however the MIL function string() starts from 0.
## Comment 9112
Date: 2008-03-01 12:51:32 +0100
From: @lsidir
Logged In: YES
user_id=1856546
Originator: YES
A fix has been committed. No tests are needed to be added since there are already enough tests for substring that were failing until now.
## Comment 9113
Date: 2008-03-01 15:00:27 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Let's document that this one was fixed in the development trunk, only ...
## Comment 9114
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1905252 at http://sourceforge.net/support/tracker.php?aid=1905252
| PF algebra does not compute correct output for fn:substring | https://api.github.com/repos/MonetDB/MonetDB/issues/1610/comments | 0 | 2020-11-30T09:48:55Z | 2024-06-27T11:29:00Z | https://github.com/MonetDB/MonetDB/issues/1610 | 753,324,820 | 1,610 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-28 04:59:55 +0100
From: David Wung <<fivecolor>>
To: clients devs <<bugs-clients>>
Version: 1.40.1 (Oct2010) [obsolete]
Last updated: 2011-01-25 12:39:12 +0100
## Comment 9108
Date: 2008-02-28 04:59:55 +0100
From: David Wung <<fivecolor>>
When I tried to modify the data source I had created, there was an interface bug.
Detail as follows:
1. To ensure that you have created a data source for MonetDB ODBC Driver.
2. Then, try to modify this data source, press button 'OK' of a dialog about MonetDB Data Source Configuration.
3. Unexpectedly, ODBC Data Source Manager, i.e. tab of User DSN, there is nothing displayed.
System I used is Windows XP sp2.
The install driver is 'MonetDB-ODBC-Installer-i386-20080214.msi'.
Otherwise, I suggest that we can add test connection function on dialog about MonetDB Data Source Configuration, for example, add a button named 'TEST' to test ODBC connection with information inputted.
Thanks.
## Comment 9109
Date: 2008-06-03 15:58:27 +0200
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
I can reproduce the problem, but I have no idea at all what's wrong. Presumably we do things slightly differently than what Microsoft expects, but they don't tell you what they expect. We do follow, as far as I can tell, the documentation, but Microsoft documentation is extremely bad in my experience.
A work around that I found is to click on Add... as if you want to create another datasource, and then in the pop-up window click on Cancel.
The Test button is a good idea, but that requires quite some reorganization of code, so isn't likely to happen in the short term (i.e. before the next release). In any case, that should be a Feature Request and not a Bug.
## Comment 9110
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1903643 at http://sourceforge.net/support/tracker.php?aid=1903643
## Comment 15417
Date: 2011-01-25 12:39:12 +0100
From: @grobian
closing due to comment 1
| ODBC: interface bug | https://api.github.com/repos/MonetDB/MonetDB/issues/1609/comments | 0 | 2020-11-30T09:48:52Z | 2024-06-28T13:30:02Z | https://github.com/MonetDB/MonetDB/issues/1609 | 753,324,794 | 1,609 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-27 10:55:28 +0100
From: lxwandy <<lxwandy>>
To: @sjoerdmullender
Version: -- development
Last updated: 2008-03-06 04:52:22 +0100
## Comment 9104
Date: 2008-02-27 10:55:28 +0100
From: lxwandy <<lxwandy>>
Hi..
When i connected to MonetDB(version 5.4) by ODBC Driver in operating system "Windows XP" , the erro message was :
"System error 126.(MonetDB ODBC Driver) Can't load the driver。"
The develop environment is VC++ 6.0.
And the files on odbc are all in the system directionary "C:/Windows/system32",including:
"libMapi.dll"
"libMonetODBC.dll"
"libMonetODBCs.dll"
"libstream.dll".
## Comment 9105
Date: 2008-03-06 10:41:51 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
This problem can be worked around by copying the files zlib1.dll and iconv.dll from C:\Program Files\CWI\MonetDB5\lib to C:\Windows\system32 .
We need to figure out a better way of installing the ODBC driver. Currently we need to copy a bunch of files to the system folder to get it to work, but I'd much rather not have to do that.
## Comment 9106
Date: 2008-03-06 16:52:22 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
The current CVS version (both stable and current branches) has the fix that the files are no longer copied to the system folder. Instead, the driver is used from the installation folder, and so no support dlls need to be copied to the system folder either anymore.
This fixes the bug properly.
The fix will be in the next release.
No test needed (actually, an automatic test of the installer would be nice but doesn't fit into our test setup).
## Comment 9107
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1902879 at http://sourceforge.net/support/tracker.php?aid=1902879
| Odbc: can't load the driver program. | https://api.github.com/repos/MonetDB/MonetDB/issues/1608/comments | 0 | 2020-11-30T09:48:49Z | 2024-06-27T11:28:58Z | https://github.com/MonetDB/MonetDB/issues/1608 | 753,324,751 | 1,608 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-26 02:13:14 +0100
From: @lsidir
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-27 01:18:52 +0100
## Comment 9101
Date: 2008-02-26 14:13:14 +0100
From: @lsidir
When I start mclient either with:
mclient -lxquery
and then run the \G switch
or
mclient -lxquery -G
in both cases, when I try to do \g to switch back to milprint_summer this switch doesn't take place and queries are still evaluated through the algebra version.
An easy way to check in which back-end you really are, is by using a function not implemented in the algebra yet, like pf:log(3).
Thank you.
## Comment 9102
Date: 2008-02-27 01:18:52 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
fixed in CVS (XQuery_0-22 branch; propagation to development trunk pending)
make sure we clear the previous state before setting the new one
(Interactive mclient feature; hence, no (simple) test possible.)
## Comment 9103
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1902131 at http://sourceforge.net/support/tracker.php?aid=1902131
| \g switch in mclient not switching | https://api.github.com/repos/MonetDB/MonetDB/issues/1607/comments | 0 | 2020-11-30T09:48:46Z | 2024-06-27T11:28:57Z | https://github.com/MonetDB/MonetDB/issues/1607 | 753,324,718 | 1,607 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-25 11:51:20 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-05-28 02:25:46 +0200
## Comment 9098
Date: 2008-02-25 11:51:20 +0100
From: @drstmane
This bug report subsumes previous bug reports
[ 1892560 ] PF: "insufficient memory" while compiling XMark module
https://sourceforge.net/tracker/index.php?func=detail&aid=1892560&group_id=56967&atid=482468
[ 1896890 ] PF/alg: module-tests segfault on Solaris 10/x86 (32-bit)
https://sourceforge.net/tracker/index.php?func=detail&aid=1896890&group_id=56967&atid=482468
[ 1895125 ] PF: 28 ALG tests still segfault
https://sourceforge.net/tracker/index.php?func=detail&aid=1895125&group_id=56967&atid=482468
that have been temporarily "fixed" in the XQuery_0-22 release branch.
However, not being satisfactory ultimate solutions, the "fixes" have not been propagated to the development trunk.
In the development trunk, the Pathfinder compiler --- at least the instance that is linked into the MonetDB/XQuery Mserver --- requires improved memory management, ranging from adeuqate allocation sizes over "garbage collection" to proper memory initialization.
## Comment 9099
Date: 2008-05-28 14:25:45 +0200
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
This improved memory management is in place now (by allocating space for PFarray a bit more conservatively).
## Comment 9100
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1901269 at http://sourceforge.net/support/tracker.php?aid=1901269
| PF: compiler requires improved memory management | https://api.github.com/repos/MonetDB/MonetDB/issues/1606/comments | 0 | 2020-11-30T09:48:44Z | 2024-06-27T11:28:56Z | https://github.com/MonetDB/MonetDB/issues/1606 | 753,324,693 | 1,606 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-23 08:15:21 +0100
From: lxwandy <<lxwandy>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
CC: @sjoerdmullender
Duplicates: #1603
Last updated: 2010-07-09 15:15:26 +0200
## Comment 9095
Date: 2008-02-23 08:15:21 +0100
From: lxwandy <<lxwandy>>
Hi..
When i connected to remote host by mclient , the error Message is as belown : "No route to host
Setup connection failed".
The version of mserver5 i tested was 5.4.
The connecting command i used was "mclient -lsql -dvss1 -h192.168.1.234" and the parameter "mapi_open=true" has been configed in "../MonetDB/etc/monetdb5.conf".
## Comment 9096
Date: 2008-02-23 08:21:11 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Please avoid reloading / re-posting bug tracker forms --- this will generate duplicate bug reports.
## Comment 9097
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1900189 at http://sourceforge.net/support/tracker.php?aid=1900189
Bug has invalid status, setting status to "NEW".
Previous status was "DELETED".
## Comment 14334
Date: 2010-07-09 15:15:26 +0200
From: @sjoerdmullender
*** This bug has been marked as a duplicate of bug #1603 ***
| failed while connecting to remote host | https://api.github.com/repos/MonetDB/MonetDB/issues/1605/comments | 0 | 2020-11-30T09:48:41Z | 2024-06-28T13:30:47Z | https://github.com/MonetDB/MonetDB/issues/1605 | 753,324,657 | 1,605 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-23 07:52:02 +0100
From: lxwandy <<lxwandy>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
CC: @sjoerdmullender
Duplicates: #1603
Last updated: 2010-07-09 15:15:36 +0200
## Comment 9093
Date: 2008-02-23 07:52:02 +0100
From: lxwandy <<lxwandy>>
Hi..
When i connected to remote host by mclient , the error Message is as belown : "No route to host
Setup connection failed".
The version of mserver5 i tested was 5.4.
The connecting command i used was "mclient -lsql -dvss1 -h192.168.1.234" and the parameter "mapi_open=true" has been configed in "../MonetDB/etc/monetdb5.conf".
## Comment 9094
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1900182 at http://sourceforge.net/support/tracker.php?aid=1900182
Bug has invalid status, setting status to "NEW".
Previous status was "DELETED".
## Comment 14336
Date: 2010-07-09 15:15:36 +0200
From: @sjoerdmullender
*** This bug has been marked as a duplicate of bug #1603 ***
| failed while connecting to remote host | https://api.github.com/repos/MonetDB/MonetDB/issues/1604/comments | 0 | 2020-11-30T09:48:38Z | 2024-06-28T13:30:47Z | https://github.com/MonetDB/MonetDB/issues/1604 | 753,324,625 | 1,604 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-23 07:32:32 +0100
From: lxwandy <<lxwandy>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2010-07-09 15:15:36 +0200
## Comment 9082
Date: 2008-02-23 07:32:32 +0100
From: lxwandy <<lxwandy>>
Hi..
When i connected to remote host by mclient , the error Message is as belown : "No route to host
Setup connection failed".
The version of mserver5 i tested was 5.4.
The connecting command i used was "mclient -lsql -dvss1 -h192.168.1.234" and the parameter "mapi_open=true" has been configed in "../MonetDB/etc/monetdb5.conf".
## Comment 9083
Date: 2008-02-23 07:47:19 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
[possibly a "stupid" question, but:]
What does `ping 192.168.1.234` say?
## Comment 9084
Date: 2008-02-23 09:38:13 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
> "No route to host"
That message usually indicates your routing tables are not setup correctly. Stefan's suggestion will most likely bring you closer to a solution.
## Comment 9085
Date: 2008-02-25 08:07:03 +0100
From: lxwandy <<lxwandy>>
Logged In: YES
user_id=2015459
Originator: YES
The "ping" command works well.
[root@server2 bin] ping 192.168.1.234
PING 192.168.1.234 (192.168.1.234) 56(84) bytes of data.
64 bytes from 192.168.1.234: icmp_seq=0 ttl=64 time=0.285 ms
64 bytes from 192.168.1.234: icmp_seq=1 ttl=64 time=0.232 ms
64 bytes from 192.168.1.234: icmp_seq=2 ttl=64 time=0.232 ms
## Comment 9086
Date: 2008-02-25 08:37:20 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Ok.
Then please try `host 192.168.1.234`,
and let us please know, what kind of operating system you're running on.
Stefan
## Comment 9087
Date: 2008-02-25 15:02:26 +0100
From: lxwandy <<lxwandy>>
Logged In: YES
user_id=2015459
Originator: YES
The operating system i running on is "Red hat Linux advanced server 4.0".
command of "host 192.168.1.234" seems to be not corrent.
[root@server2 ~] host 192.168.2.2
;; connection timed out; no servers could be reached
## Comment 9088
Date: 2008-02-25 15:10:25 +0100
From: lxwandy <<lxwandy>>
Logged In: YES
user_id=2015459
Originator: YES
Sorry.
The command "host 192.168.1.234" I described wrong in my last word.
It's that as below actually.
[root@server2 ~] host 192.168.1.234
;; connection timed out; no servers could be reached
## Comment 9089
Date: 2008-02-25 17:16:24 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Hm, as far as I can see, this rather seems to be a problem in you network (in particular name services) configuration than in MonetDB.
Could you please check & verify your network configuration?
From both machines involved (client & server) make sure that both pinging the other (by IP or name) and looking up the other's name and IP (e.g., via `host`) works correctly.
## Comment 9090
Date: 2008-02-26 12:16:47 +0100
From: lxwandy <<lxwandy>>
Logged In: YES
user_id=2015459
Originator: YES
I have resolved this problem. The network configuration is not corrent on the server i run .
Thanks for your suggestions!
## Comment 9091
Date: 2008-02-26 12:53:15 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Closed as it turned out to be a (local) network problem rather than a MonetDB bug.
## Comment 9092
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1900171 at http://sourceforge.net/support/tracker.php?aid=1900171
## Comment 14335
Date: 2010-07-09 15:15:26 +0200
From: @sjoerdmullender
*** Bug #1605 has been marked as a duplicate of this bug. ***
## Comment 14337
Date: 2010-07-09 15:15:36 +0200
From: @sjoerdmullender
*** Bug #1604 has been marked as a duplicate of this bug. ***
| failed while connecting to remote host | https://api.github.com/repos/MonetDB/MonetDB/issues/1603/comments | 0 | 2020-11-30T09:48:35Z | 2024-06-28T13:30:46Z | https://github.com/MonetDB/MonetDB/issues/1603 | 753,324,592 | 1,603 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-22 12:43:48 +0100
From: @yzchang
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-22 12:46:58 +0100
## Comment 9077
Date: 2008-02-22 12:43:48 +0100
From: @yzchang
Compiling module sql (current branch) on mac os 10.5.2 results in an error in merovingian.mx:
gcc -DHAVE_CONFIG_H -I. -I../../.. -I/Users/jennie/monet/dev/sql/src/backends/monet5 -I/Users/jennie/monet/dev/sql/src/backends/monet5 -I../../include -I/Users/jennie/monet/dev/sql/src/backends/monet5/../../include -I../../common -I/Users/jennie/monet/dev/sql/src/backends/monet5/../../common -I../../storage -I/Users/jennie/monet/dev/sql/src/backends/monet5/../../storage -I../../server -I/Users/jennie/monet/dev/sql/src/backends/monet5/../../server -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5/atoms -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5/compiler -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5/kernel -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5/mal -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5/optimizer -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5/scheduler -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB5 -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB/mapilib -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB/common -I/Users/jennie/scratch/monet-install/dev/debug/include/MonetDB/gdk -Wall -Wextra -std=c99 -g -Werror-implicit-function-declaration -Werror -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wp,-D_FORTIFY_SOURCE=2 -Wno-uninitialized -c merovingian.c
/Users/jennie/monet/dev/sql/src/backends/monet5/merovingian.mx:65:29: error: wait.h: No such file or directory
I searched for wait.h on my laptop, locate says:
[leopard:/Users/jennie/monet/dev]
$ locate wait.h
/Developer/SDKs/MacOSX10.4u.sdk/usr/include/sys/wait.h
/Developer/SDKs/MacOSX10.5.sdk/System/Library/Frameworks/Kernel.framework/Versions/A/Headers/sys/wait.h
/Developer/SDKs/MacOSX10.5.sdk/usr/include/sys/wait.h
/System/Library/Frameworks/Kernel.framework/Versions/A/Headers/sys/wait.h
/System/Library/Frameworks/Tcl.framework/Versions/8.4/Resources/Documentation/Reference/Tcl/TclCmd/vwait.htm
/System/Library/Frameworks/Tk.framework/Versions/8.4/Resources/Documentation/Reference/Tk/TkCmd/tkwait.htm
/usr/include/sys/wait.h
I simply edited merovingian.mx to include sys/wait.h:
[leopard:/Users/jennie/monet/dev/sql]
$ cvs diff src/backends/monet5/merovingian.mx
Index: src/backends/monet5/merovingian.mx
===================================================================
RCS file: /cvsroot/monetdb/sql/src/backends/monet5/merovingian.mx,v
retrieving revision 1.26
diff -r1.26 merovingian.mx
65c65
< include <wait.h> /* wait */
---
> include <sys/wait.h> /* wait */
To my surprise, module sql now compiles successfully.
It seems that we need to used a ifdef to include sys/wait.h instead of wait.h on mac os [10.5].
Jennie
## Comment 9078
Date: 2008-02-22 12:46:58 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
please cvs update, I already fixed this this morning.
## Comment 9079
Date: 2008-02-22 13:00:17 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: YES
I see, your fix went to stable branch, I was compiling current branch.
## Comment 9080
Date: 2008-02-22 13:03:10 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
sorry, I see.
Please apply http://article.gmane.org/gmane.comp.db.monetdb.sql.scm/2553 in current.
## Comment 9081
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1899535 at http://sourceforge.net/support/tracker.php?aid=1899535
| SQL: failed to compile on mac os 10.5.2 due to merovingian | https://api.github.com/repos/MonetDB/MonetDB/issues/1602/comments | 0 | 2020-11-30T09:48:32Z | 2024-06-27T11:28:52Z | https://github.com/MonetDB/MonetDB/issues/1602 | 753,324,563 | 1,602 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-22 11:54:10 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-22 01:23:20 +0100
## Comment 9074
Date: 2008-02-22 11:54:10 +0100
From: @drstmane
See e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-Fedora8/tests_BugTracker/NULL_BAT_in_serializer.SF-1694814b.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-Fedora8/tests_XQuery/union.out.00.html
## Comment 9075
Date: 2008-02-22 13:23:20 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Confirming the changes is an easy fix.
## Comment 9076
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1899508 at http://sourceforge.net/support/tracker.php?aid=1899508
| PF: 2 tests fail after checkins on Thu Feb 21 2008 | https://api.github.com/repos/MonetDB/MonetDB/issues/1601/comments | 0 | 2020-11-30T09:48:29Z | 2024-06-27T11:28:51Z | https://github.com/MonetDB/MonetDB/issues/1601 | 753,324,532 | 1,601 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-22 02:06:53 +0100
From: marspluto
To: @yzchang
Version: -- development
Last updated: 2009-08-06 02:48:42 +0200
## Comment 9056
Date: 2008-02-22 02:06:53 +0100
From: marspluto
When I go to web admin console at http://127.0.0.1:50001/admin/, I am
getting this error
(see the screenshot):
http://tinypic.com/view.php?pic=1ttoxf&s=3
It also has been attached. This occurs when I click on "List Collections" and "List All Documents".
I am also getting the error when I click on "Backup Database". To summarize, the web admin console is not working.
Could someone provide a fix, because I need to backup my index.
Thanks!
jz
## Comment 9057
Date: 2008-02-22 07:14:43 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jennie,
could you please have a look at this one (possibly together with Peter)?
In particular the backup functionality in MonetDB/XQuery 0.20 (and restore in 0.22) is required to keep existing databases when upgrading from 0.20 to 0.22.
Thanks!
Stefan
## Comment 9058
Date: 2008-02-22 12:57:02 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Sure, I'm looking at this problem. On my mac, I get neigher results nor error :(
Stefan, if you use the admin page, will you get similar error?
Jennie
## Comment 9059
Date: 2008-02-22 13:45:55 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jennie,
I currently only have the latest Stable (0.22) but not the former one (0.20) running.
With the latest Stable, I get the following (virgin DB with 1 document loaded):
- first click on "List Collections": No reaction
- second click on "List Collections": Attached Error1.png
- first click on "List All Documents": No reaction
- second click on "List All Documents": Attached Error2.png
- click on "Backup Database", fill in "Backup ID", click "Backup": Attached Error3.png
... I'm afraid, there is major need for testing this functionality --- preferably even automated (nightly!) testing, or mandatory "by hand" testing by a confirmed maintainer before each future release ...
File Added: Error1.png
## Comment 9060
Date: 2008-02-22 13:46:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
File Added: Error2.png
## Comment 9061
Date: 2008-02-22 13:47:15 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
File Added: Error3.png
## Comment 9062
Date: 2008-02-22 13:59:55 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Stefan,
Thanks for the pictures. I now also get similar problem. I'm searching for the reason why it can't find the module.
You are right, if it is used for such important task, it should be tested more often. Nightly test would be great, but I won't know how... I'm willing to test it manually before each release.
## Comment 9063
Date: 2008-02-22 19:12:52 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
After a lot of debugging, I found what is exactly happening, which is *very* strange.
The reason that clicking on buttons like "list collections" and "list all documents" (which triggers an xrpc request message to be send to the xrpc server to execute the corresponding function) does not respond, is that the HTTP server (sttpd) hangs in trying to receive the complete request message send by the function xmlParserInputBufferCreateFilename() (in PFurlcache()).
The error "could not load module" is given after the connection has expired.
The syntax error "unexpected $end on line 1, column 0" is caused by that xmlParserInputBufferCreateFilename() is not able to retrieve the module file, which results in an emtpy string to be parsed.
When xrpc server receives a request, e.g. "list collections", it first load the module in which the function "collections()" is defined. in my test case, it is: http://andorea.ins.cwi.nl:8080/admin/admin.xq
Loading the module is done by the function xmlParserInputBufferCreateFilename (in PFurlcache(), in compiler.c), which sends an HTTP GET request to andorea.ins.cwi.nl:8080 to retrieve the module file. Since this is a file request, it is handled by the HTTP server (shttpd) itself (not by the xrpc server).
What shttpd does is simple: buffer the full request message (i.e. the request message contains the chars '\n\r\n'), then handle it.
But here, strange thing happens: it seems that the request sent by xmlParserInputBufferCreateFilename() is not complete/valid !? After shttd has tried to read the request message several times, the buffer still only contains:
(gdb) p buf $21 = 0x2aaac0005258 "GET /admin/admin.xq HTTP/1.0\r\nHost: andorea.ins.cwi.nl:8080\r\nAccept-Encoding: gzip"
Note that the message is not ended by '\r\n\r\n', according to HTTP 1.1
To prove that shttpd works, using wget to retrieve the same file from shttpd succeeds:
$ wget http://andorea.ins.cwi.nl:8080/admin/admin.xq
...
18:41:36 (156.47 MB/s) - `admin.xq' saved [4950/4950]
$ cat admin.xq
module namespace admin = 'http://monetdb.cwi.nl/XQuery/admin/';
...
the request sent by wget is:
(gdb) p buf
$20 = 0x2aaac0005258 "GET /admin/admin.xq HTTP/1.0\r\nUser-Agent: Wget/1.10.2 (Red Hat modified)\r\nAccept: */*\r\nHost: andorea.ins.cwi.nl:8080\r\nConnection: Keep-Alive\r\n\r\n"
which is obviously complete/correct. However, another HTTP server (e.g. www.cwi.nl) does not have any problem with the request message sent by xmlParserInputBufferCreateFilename():
$ cat test.xq
import module namespace t = "http://www.monetdb.nl/basicfunctions" at
"http://homepages.cwi.nl/~zhang/xrpc/basicfunctions.xq";
t:convert(23)
$ mclient -lx test.xq
50.685330
For the query above, xmlParserInputBufferCreateFilename() is used to GET the file http://homepages.cwi.nl/~zhang/xrpc/basicfunctions.xq, which succeeded.
But, if I change the query to use a module file served by my Apache Tomcat/5.5.9 server, the same problem occurs. mclient hangs for a while (note the time in the prompt), and gives the same error:
$ cat test.xq
(:
import module namespace t = "http://www.monetdb.nl/basicfunctions" at
"http://homepages.cwi.nl/~zhang/xrpc/basicfunctions.xq";
:)
import module namespace t = "http://www.monetdb.nl/basicfunctions" at
"http://andorea.ins.cwi.nl:8080/basicfunctions.xq";
t:convert(23)
[Fri Feb 22 19:06:02 CET 2008]
[andorea:/ufs/zhang]
$ mclient -lx test.xq
MAPI = monetdb@localhost:50000
QUERY = (:
ERROR = !parse error: syntax error, unexpected $end on line 1, column 0
!parse error: XQuery parsing failed
! halted in /ufs/zhang/monet/stable/pathfinder/compiler/parser/parser.y (PFparse), line 2902
!parse error: syntax error, unexpected $end on line 1, column 0
!parse error: error parsing module from http://andorea.ins.cwi.nl:8080/basicfunctions.xq
! halted in /ufs/zhang/monet/stable/pathfinder/compiler/parser/parser.y (parse_module), line 2939
[Fri Feb 22 19:06:23 CET 2008]
[andorea:/ufs/zhang]
$
The code in shttpd.c to hand file request hasn't been changed since 2006, and the admin page has workt before.
So, am I able to conclude that: some thing is wrong with xmlParserInputBufferCreateFilename(), but maybe some HTTP server is not very strict with if the request message is correctly ended with a '\r\n\r\n'? Are we using a newer version of libxml2 and since when?
Thanks in advance for any suggestions/comments,
Jennie
## Comment 9064
Date: 2008-02-25 15:25:07 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
FYI:
With the following version of libxml2 it works fine:
% qlist -Iv libxml2
dev-libs/libxml2-2.6.30-r1
## Comment 9065
Date: 2008-02-25 15:43:45 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
It appears that we're hitting this bug in libxml2-2.6.31:
http://bugzilla.gnome.org/show_bug.cgi?id=514521
## Comment 9066
Date: 2008-02-25 17:40:08 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
The concequence of the bug in libxml2-2.6.31 is that MonetDB/XQuery is not able to load module definition files from an HTTP server that is running on a port other than 80 (this problem does not occur if libxml2-2.6.30 is used ).
What the admin console does is call pre-defined XQuery functions (defined in admin.xq) to do backup/restore work. Those functions can also be executed using other MonetDB clients, for example, 'mclient'.
A workaround for users of libxml2-2.6.31 is thus as the following:
(IMPORTANT: this workaround has been tested on the latest release, nl. version 0.22.
It is highly recommended to first test the following queries on a dummy database!)
The module definition file "admin.xq" can be found in '$prefix/share/MonetDB/xrpc/admin/', where '$prefix' is the prefix used to install MonetDB. I copied it to an easy to use place:
$ cp $prefix/share/MonetDB/xrpc/admin/admin.xq /tmp/
Call functions in "/tmp/admin.xq" a 'mclient' session to first backup your XML database:
$ mclient -lx
xquery>pf:collections()
more><>
<collection updatable="true" size="102 KiB" numDocs="1">test</collection>
xquery>pf:documents()
more><>
<document updatable="true" url="/tmp/test.xml" collection="test">test.xml</document>
xquery>pf:documents("test")
more><>
<document updatable="true" url="/tmp/test.xml" collection="test">test.xml</document>
xquery>import module namespace adm = "http://monetdb.cwi.nl/XQuery/admin/" at "/tmp/admin.xq"; adm:backup("backup1")
more><>
Here we see that with 'adm:backup', a backup of *all* collections in the database (in this case, one collection "test" with one document "test.xml" in the database 'demo') are made in directory '$prefix/var/MonetDB4/dbfarm/demo/backup/backup1/'.
(IMPORTANT: it is HIGHLY recommended to check now if all your documents are in the directory
'$prefix/var/MonetDB4/dbfarm/<dbname>/backup/<backup-name>/<coll-name>', before executing the next query!)
xquery>import module namespace adm = "http://monetdb.cwi.nl/XQuery/admin/" at "/tmp/admin.xq"; adm:del-col("test")
more><>
xquery>pf:documents()
more><>
Then, one can use 'adm:del-col()' to clean up the database.
Now the previous version of MonetDB/XQuery can be de-installed, and the new version can be installed, after which, the following query can be used to restore the database:
$ mclient -lx
xquery>import module namespace adm = "http://monetdb.cwi.nl/XQuery/admin/" at "/tmp/admin.xq"; adm:restore("backup1", 10)
more><>
xquery>pf:documents()
more><>
<document updatable="true" url="backup/backup1/test/10000/1.xml" collection="test">test.xml</document>
Hope this helps.
Regards,
Jennie
## Comment 9067
Date: 2008-02-25 17:47:19 +0100
From: marspluto
Logged In: YES
user_id=2016221
Originator: YES
Thanks Jennie (did I meet you at the SIGIR bag packing event?) and everyone else!
I will try out this work-around. I will let you if it works for me.
Best,
Junte
## Comment 9068
Date: 2008-06-20 06:59:25 +0200
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
This problem is solved in a newer version of libxml2 (> 2.6.31).
## Comment 9069
Date: 2008-07-14 14:15:47 +0200
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Reopen this bug, as it hasn't completely disappeared yet. Similar problem still happens when using the windows installer (MonetDB4-XQuery-Installer-i386-20080624.msi) of the latest release on 32 bits Windows XP, although this installer contains an *older* version of libxml2 (i.e., 2.6.30).
I overwrote the libxml2.dll that comes with the Windows installer with the one found in http://www.zlatkovic.com/libxml.en.html
(libxml2-2.6.32+.win32.zip), and then it works fine. So I'd like to suggest to upgrade the libxml2 included in MXQ Windows installer.
Regards,
Jennie
## Comment 9070
Date: 2008-09-12 16:35:52 +0200
From: @sjoerdmullender
The current version (not yet released) of the installer contains the 2.6.32 version of libxml2.
Reassigning to Jennie since the interface is broken in many respects. Just start clicking. View Database Statistics doesn't do anything, except perhaps causing other commands to then fail. Also try adding a document on Windows. It fails with a cryptic message saying that it can't open the file. Of course, you need to specify a URL (according to the label on the interface) but you can only specify a file since the text box is not editable. File names are not URLs. And on Windows they also don't look like one another.
## Comment 9071
Date: 2008-09-12 16:50:43 +0200
From: @drstmane
This is yet another reminder to no only thoroughly test the administrative interface interactively *before* a release, but also add (non-interactive) tests of both the functionality that the administrative interface provides and the functionality that the administrative interfaces requires from the MonetDB/XQuery server (incl. its built-in HTTP server and XRPC funtionality).
See also
https://sourceforge.net/mailarchive/forum.php?thread_name=20080618085309.GA2096%40cwi.nl&forum_name=monetdb-developers
## Comment 9072
Date: 2009-08-06 14:48:42 +0200
From: @yzchang
The problem with libxml2 is tested by the current admin Tests, and it has worked fine for sometime, so closing
## Comment 9073
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1899258 at http://sourceforge.net/support/tracker.php?aid=1899258
| PF: Error message with Administrative Console | https://api.github.com/repos/MonetDB/MonetDB/issues/1600/comments | 0 | 2020-11-30T09:48:26Z | 2024-06-27T11:28:50Z | https://github.com/MonetDB/MonetDB/issues/1600 | 753,324,494 | 1,600 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-21 04:19:06 +0100
From: Michael Schmidt <<michael_schmidt>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-22 07:24:01 +0100
## Comment 9046
Date: 2008-02-21 16:19:06 +0100
From: Michael Schmidt <<michael_schmidt>>
The command
> monetdb destroy $DATABASE
does not remove the sql log at
> $PREFIX/var/MonetDB5/sql_logs/$DATABASE
Although the subsequent recreation and restart of $DATABASE with
> monetdb create $DATABASE
> monetdb start $DATABASE
succeeds, the mserver5 process crashes as soon as the a client connects to the mserver5 $DATABASE process.
## Comment 9047
Date: 2008-02-21 17:51:29 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
For a database where monetdb(1) doesn't remove the SQL logdir, can you please post the output of the following:
% $PREFIX/bin/monetdb status $DATABASE
% ls -la $PREFIX/var/MonetDB5/dbfarm/$DATABASE
% ls -la $PREFIX/var/MonetDB5/sql_logs/$DATABASE
% ls -la $PREFIX/var/MonetDB5/sql_logs
% id
% $PREFIX/bin/monetdb destroy -f $DATABASE
% $PREFIX/bin/monetdb status $DATABASE
% ls -la $PREFIX/var/MonetDB5/dbfarm/$DATABASE
% ls -la $PREFIX/var/MonetDB5/sql_logs/$DATABASE
% ls -la $PREFIX/var/MonetDB5/sql_logs
## Comment 9048
Date: 2008-02-21 19:08:20 +0100
From: Michael Schmidt <<michael_schmidt>>
Logged In: YES
user_id=2015705
Originator: YES
OK,
here we go.
% INITIAL CONFIGURATION/STEPS
% - /usr/local/src/MonetDB/var/MonetDB5/dbfarm/ directory is empty
% - /usr/local/src/MonetDB/var/MonetDB5/sql_logs/ directory is empty
% ... then ...
% - start merovingian server
% - create and start database 'test' (monetdb create test + monetdb start test)
% - load some data into the database (otherwise no SQL logs will be written)
Here is the output produced by the commands:
% /usr/local/src/MonetDB/bin/monetdb status test
name state uptime crashes
test running 41s 0
% ls -la /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test
total 36
drwx------ 4 mschmidt mschmidt 4096 2008-02-21 19:43 .
drwxr-xr-x 3 mschmidt benchmark 4096 2008-02-21 19:43 ..
drwx------ 25 mschmidt mschmidt 4096 2008-02-21 19:44 bat
drwx------ 2 mschmidt mschmidt 4096 2008-02-21 19:43 box
-rw------- 1 mschmidt mschmidt 28 2008-02-21 19:43 .conn
-rw------- 1 mschmidt mschmidt 57 2008-02-21 19:43 .gdk_lock
-rw------- 1 mschmidt mschmidt 13 2008-02-21 19:43 .scen
-rw------- 1 mschmidt mschmidt 11 2008-02-21 19:43 .uplog
-rw------- 1 mschmidt mschmidt 48 2008-02-21 19:43 .vaultkey
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs/test
total 12
drwx------ 2 mschmidt mschmidt 4096 2008-02-21 19:44 .
drwx------ 3 mschmidt mschmidt 4096 2008-02-21 19:43 ..
-rw------- 1 mschmidt mschmidt 10 2008-02-21 19:44 log
-rw------- 1 mschmidt mschmidt 0 2008-02-21 19:44 log.3
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs
total 12
drwx------ 3 mschmidt mschmidt 4096 2008-02-21 19:43 .
drwxr-xr-x 4 mschmidt benchmark 4096 2008-02-21 19:43 ..
drwx------ 2 mschmidt mschmidt 4096 2008-02-21 19:43 test
% id
uid=1002(mschmidt) gid=1002(mschmidt) groups=117(admin),1002(mschmidt),1003(benchmark)
% /usr/local/src/MonetDB/bin/monetdb stop test
no output (this step was missing in your description)
% /usr/local/src/MonetDB/bin/monetdb destroy -f test
successfully destroyed database 'test'
% /usr/local/src/MonetDB/bin/monetdb status test
name state uptime crashes
status: no such database: test
% ls -la /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test
ls: /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test: No such file or directory
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs/test
total 16
drwx------ 2 mschmidt mschmidt 4096 2008-02-21 19:44 .
drwx------ 3 mschmidt mschmidt 4096 2008-02-21 19:43 ..
-rw------- 1 mschmidt mschmidt 10 2008-02-21 19:44 log
-rw------- 1 mschmidt mschmidt 2 2008-02-21 19:45 log.3
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs
total 12
drwx------ 3 mschmidt mschmidt 4096 2008-02-21 19:43 .
drwxr-xr-x 4 mschmidt benchmark 4096 2008-02-21 19:47 ..
drwx------ 2 mschmidt mschmidt 4096 2008-02-21 19:44 test
## Comment 9049
Date: 2008-02-22 11:27:30 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
> % /usr/local/src/MonetDB/bin/monetdb stop test
> no output (this step was missing in your description)
Hehe, it wasn't there on purpose, but anyway, have you tried tonight's nightly yet on the same order of things?
## Comment 9050
Date: 2008-02-22 13:22:41 +0100
From: Michael Schmidt <<michael_schmidt>>
Logged In: YES
user_id=2015705
Originator: YES
OK, I installed the current source super tarball build from /projects/monetdb/Stable/.DailyBuilds./20080222/ and rerun your commands, this time without stopping the server. (OT: monetdb start/stop -w does not yet work, is this correct?). Here is the output.
% INITIAL CONFIGURATION/STEPS
% - /usr/local/src/MonetDB/var/MonetDB5/dbfarm/ directory is empty
% - /usr/local/src/MonetDB/var/MonetDB5/sql_logs/ directory is empty
% ... then ...
% - start merovingian server
% - create and start database 'test' (monetdb create test + monetdb start
test)
% - load some data into the database (otherwise no SQL logs will be
written)
Here is the output produced by the commands:
% /usr/local/src/MonetDB/bin/monetdb status test
name state uptime crashes
test running 51s 0
% ls -la /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test
total 36
drwx------ 4 mschmidt mschmidt 4096 2008-02-22 14:13 .
drwxr-xr-x 3 mschmidt benchmark 4096 2008-02-22 14:13 ..
drwx------ 25 mschmidt mschmidt 4096 2008-02-22 14:14 bat
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 14:13 box
-rw------- 1 mschmidt mschmidt 28 2008-02-22 14:13 .conn
-rw------- 1 mschmidt mschmidt 57 2008-02-22 14:13 .gdk_lock
-rw------- 1 mschmidt mschmidt 13 2008-02-22 14:13 .scen
-rw------- 1 mschmidt mschmidt 11 2008-02-22 14:13 .uplog
-rw------- 1 mschmidt mschmidt 48 2008-02-22 14:13 .vaultkey
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs/test
total 12
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 14:14 .
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 14:13 ..
-rw------- 1 mschmidt mschmidt 10 2008-02-22 14:14 log
-rw------- 1 mschmidt mschmidt 0 2008-02-22 14:14 log.3
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs
total 12
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 14:13 .
drwxr-xr-x 4 mschmidt benchmark 4096 2008-02-22 14:13 ..
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 14:14 test
% id
uid=1002(mschmidt) gid=1002(mschmidt) groups=117(admin),1002(mschmidt),1003(benchmark
% /usr/local/src/MonetDB/bin/monetdb destroy -f test
destroy: database 'test' is still running, stop database first
% /usr/local/src/MonetDB/bin/monetdb status test
name state uptime crashes
test running 2m 22s 0
% ls -la /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test
total 36
drwx------ 4 mschmidt mschmidt 4096 2008-02-22 14:13 .
drwxr-xr-x 3 mschmidt benchmark 4096 2008-02-22 14:13 ..
drwx------ 25 mschmidt mschmidt 4096 2008-02-22 14:14 bat
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 14:13 box
-rw------- 1 mschmidt mschmidt 28 2008-02-22 14:13 .conn
-rw------- 1 mschmidt mschmidt 57 2008-02-22 14:13 .gdk_lock
-rw------- 1 mschmidt mschmidt 13 2008-02-22 14:13 .scen
-rw------- 1 mschmidt mschmidt 11 2008-02-22 14:13 .uplog
-rw------- 1 mschmidt mschmidt 48 2008-02-22 14:13 .vaultkey
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs/test
total 12
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 14:14 .
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 14:13 ..
-rw------- 1 mschmidt mschmidt 10 2008-02-22 14:14 log
-rw------- 1 mschmidt mschmidt 0 2008-02-22 14:14 log.3
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs
total 12
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 14:13 .
drwxr-xr-x 4 mschmidt benchmark 4096 2008-02-22 14:13 ..
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 14:14 test
## Comment 9051
Date: 2008-02-22 13:26:40 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
> OK, I installed the current source super tarball build from?
> /projects/monetdb/Stable/.DailyBuilds./20080222/ and rerun your commands,
> this time without stopping the server. (OT: monetdb start/stop -w does not
> yet work, is this correct?). Here is the output.
That's because you shouldn't use the daily superballs, but just use the monetdb-install.sh script in a clean directory with the argument --nightly=current.
could you execute:
% /usr/local/src/MonetDB/bin/monetdb stop test
% /usr/local/src/MonetDB/bin/monetdb destroy -f test
please?
## Comment 9052
Date: 2008-02-22 15:02:10 +0100
From: Michael Schmidt <<michael_schmidt>>
Logged In: YES
user_id=2015705
Originator: YES
OK, so I reinstalled with --nightly=current. Once more the log, including the stop command (notice that I used start/stop -w this time).
% INITIAL CONFIGURATION/STEPS
% - /usr/local/src/MonetDB/var/MonetDB5/dbfarm/ directory is empty
% - /usr/local/src/MonetDB/var/MonetDB5/sql_logs/ directory is empty
% ... then ...
% - start merovingian server
% - create and start database 'test' (monetdb create test + monetdb start -w
test)
% - load some data into the database (otherwise no SQL logs will be
written)
Here is the output produced by the commands:
% /usr/local/src/MonetDB/bin/monetdb status test
name state uptime crashes
test running 1m 51s 0
% ls -la /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test
total 36
drwx------ 4 mschmidt mschmidt 4096 2008-02-22 15:53 .
drwxr-xr-x 3 mschmidt benchmark 4096 2008-02-22 15:53 ..
drwx------ 5 mschmidt mschmidt 4096 2008-02-22 15:54 bat
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 15:53 box
-rw------- 1 mschmidt mschmidt 28 2008-02-22 15:53 .conn
-rw------- 1 mschmidt mschmidt 56 2008-02-22 15:53 .gdk_lock
-rw------- 1 mschmidt mschmidt 13 2008-02-22 15:53 .scen
-rw------- 1 mschmidt mschmidt 11 2008-02-22 15:53 .uplog
-rw------- 1 mschmidt mschmidt 48 2008-02-22 15:53 .vaultkey
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs/test
total 600
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 15:54 .
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 15:54 ..
-rw------- 1 mschmidt mschmidt 10 2008-02-22 15:54 log
-rw------- 1 mschmidt mschmidt 594376 2008-02-22 15:54 log.2
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs
total 12
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 15:54 .
drwxr-xr-x 4 mschmidt benchmark 4096 2008-02-22 15:54 ..
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 15:54 test
% id
uid=1002(mschmidt) gid=1002(mschmidt) groups=117(admin),1002(mschmidt),1003(benchmark)
% /usr/local/src/MonetDB/bin/monetdb stop -w test
(no feedback)
% /usr/local/src/MonetDB/bin/monetdb destroy -f test
successfully destroyed database 'test'
% /usr/local/src/MonetDB/bin/monetdb status test
name state uptime crashes
status: no such database: test
% ls -la /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test
ls: /usr/local/src/MonetDB/var/MonetDB5/dbfarm/test: No such file or directory
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs/test
total 600
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 15:54 .
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 15:54 ..
-rw------- 1 mschmidt mschmidt 10 2008-02-22 15:54 log
-rw------- 1 mschmidt mschmidt 594376 2008-02-22 15:54 log.2
% ls -la /usr/local/src/MonetDB/var/MonetDB5/sql_logs
total 12
drwx------ 3 mschmidt mschmidt 4096 2008-02-22 15:54 .
drwxr-xr-x 4 mschmidt benchmark 4096 2008-02-22 15:54 ..
drwx------ 2 mschmidt mschmidt 4096 2008-02-22 15:54 test
## Comment 9053
Date: 2008-02-22 15:57:49 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
The problem is reproducible here. Investigating...
## Comment 9054
Date: 2008-02-22 19:24:01 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
I figured out why I first couldn'r reproduce this. The problem was related to using the -f flag of destroy. Without -f it does what it should do. I fixed this now. Tomorrow's nightlies should have a fully working scenario for you now ;)
## Comment 9055
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898804 at http://sourceforge.net/support/tracker.php?aid=1898804
| "monetdb destroy" does not remove sql log | https://api.github.com/repos/MonetDB/MonetDB/issues/1599/comments | 0 | 2020-11-30T09:48:23Z | 2024-06-27T11:28:49Z | https://github.com/MonetDB/MonetDB/issues/1599 | 753,324,451 | 1,599 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-21 12:34:05 +0100
From: @arjenderijke
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-21 04:06:33 +0100
## Comment 9036
Date: 2008-02-21 12:34:05 +0100
From: @arjenderijke
./monetdb-install.sh --prefix=/export/scratch1//monetdb-cvs --build=/export/scratch1//tmp/monetdb-cvs --enable-sql --enable-xquery --cvs fails with the following error:
gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I. -I../include -I./../include -I/export/scratch1/rijke/monetdb-cvs/include/MonetDB -I/export/scratch1/rijke/monetdb-cvs/include/MonetDB/common -I/export/scratch1/rijke/monetdb-cvs/include/MonetDB/gdk -DLIBMIL -g -O2 -Wall -Wextra -std=c99 -Werror-implicit-function-declaration -Werror -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wp,-D_FORTIFY_SOURCE=2 -Wno-declaration-after-statement -c milgen.c -fPIC -DPIC -o .libs/libmil_la-milgen.o
cc1: warnings being treated as errors
milgen.brg: In function 'reduce':
milgen.brg:6339: warning: unused variable 'ty'
gmake[5]: *** [libmil_la-milgen.lo] Error 1
gmake[5]: Leaving directory `/export/scratch1/rijke/tmp/monetdb-cvs/MonetDB4-XQuery/pathfinder/compiler/mil'
gmake[4]: *** [all] Error 2
gmake[4]: Leaving directory `/export/scratch1/rijke/tmp/monetdb-cvs/MonetDB4-XQuery/pathfinder/compiler/mil'
gmake[3]: *** [all-recursive] Error 1
gmake[3]: Leaving directory `/export/scratch1/rijke/tmp/monetdb-cvs/MonetDB4-XQuery/pathfinder/compiler'
gmake[2]: *** [all] Error 2
gmake[2]: Leaving directory `/export/scratch1/rijke/tmp/monetdb-cvs/MonetDB4-XQuery/pathfinder/compiler'
gmake[1]: *** [all-recursive] Error 1
This was run today, 20080221, using the script from the Feb2008 tarball downloaded on 20080220
This also happens with --nightly={stable,current} instead of --cvs. This was run today on a fc6 machine, but earlier on f8 machines with the similar results
## Comment 9037
Date: 2008-02-21 12:37:59 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
do you happen to have any CFLAGS set in your environment?!? Could you tell me where I can find the logile?
## Comment 9038
Date: 2008-02-21 13:08:39 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Just as a "sanity check":
last nights TestWeb compilation (configured with --enable-strict --enable-optimize) seems to have worked fine:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.Makes3/index_short.html
## Comment 9039
Date: 2008-02-21 13:35:50 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
The problem is that even though configure is called like this:
./configure --prefix='/export/scratch1/rijke/monetdb-cvs' --disable-optimize -- disable-debug --disable-assert
it still thinks it is a good idea to compile with the following CFLAGS:
-g -O2 -Wall -Wextra -std=c99 -Werror-implicit-function-declaration -Werror -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wp,- D_FORTIFY_SOURCE=2 -Wno-declaration-after-statement
obviously, this is wrong, and should have been "-g -O2 -std=c99" alone.
## Comment 9040
Date: 2008-02-21 13:36:48 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
I guess adding --disable-strict to that list of disables will do the trick. Too bad this ended up in the release :(
## Comment 9041
Date: 2008-02-21 13:44:43 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
please test again with http://www.cwi.nl/~fabian/troep/monetdb-install.sh (version 1.32), and close this bug if it compiles.
## Comment 9042
Date: 2008-02-21 13:56:47 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Just for info:
- gcc issues "unused variable" warnings are only when optimization (at least -O1) is enabled.
- In the "Stable" branches (and hence in the release), --disable-strict --enable-optimize is default.
- In the development trunk ("Current"), --enable-strict is default (to enforce "clean" code).
- Nightly testing always compiles with --enable-strict --enable-optimize (to enforce "clean" code).
- (At least on our testing platforms (and our desktops)) the whole code base is supposed to compile with --enable-strict --enable-optimize (as checked/enforced by nightly testing).
## Comment 9043
Date: 2008-02-21 14:02:10 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
You just touched upon the reason why the monetdb-install.sh script needs to force configure not to use the strict flags, because the system used by Arjen is just a CST installed Fedora8 machine.
## Comment 9044
Date: 2008-02-21 16:06:33 +0100
From: @arjenderijke
Logged In: YES
user_id=1836138
Originator: YES
Compiling with the script from http://www.cwi.nl/~fabian/troep/monetdb-install.sh succeeds. Bug will be closed
## Comment 9045
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898643 at http://sourceforge.net/support/tracker.php?aid=1898643
| monetdb-install.sh -xquery failed | https://api.github.com/repos/MonetDB/MonetDB/issues/1598/comments | 0 | 2020-11-30T09:48:20Z | 2024-06-27T11:28:48Z | https://github.com/MonetDB/MonetDB/issues/1598 | 753,324,418 | 1,598 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-21 11:21:09 +0100
From: @grobian
To: @sjoerdmullender
Version: -- development
Last updated: 2008-02-22 11:33:28 +0100
## Comment 9033
Date: 2008-02-21 11:21:09 +0100
From: @grobian
Just because I notice them:
Making MonetDB BuildTools in /ufs/fabian/scratch/vtmp/build-stable-pegasus.ins.cwi.nl/buildtools
/ufs/fabian/scratch/monetdb/stable/buildtools/Mx/Form.c:412: warning: unused parameter ‘mod’
/ufs/fabian/scratch/monetdb/stable/buildtools/Mx/Form.c:424: warning: unused parameter ‘mod’
/ufs/fabian/scratch/monetdb/stable/buildtools/Mx/Form.c:424: warning: unused parameter ‘sec’
/ufs/fabian/scratch/monetdb/stable/buildtools/Mx/Display.c: In function ‘PrPrelude’:
/ufs/fabian/scratch/monetdb/stable/buildtools/Mx/Display.c:142: warning: unused variable ‘texDocStyle’
aparently X_CFLAGS is not used in buildtools or these would have been fatal in my setup.
## Comment 9034
Date: 2008-02-22 11:33:28 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Fixed.
No test needed.
## Comment 9035
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898594 at http://sourceforge.net/support/tracker.php?aid=1898594
| buildtools: compilation warnings | https://api.github.com/repos/MonetDB/MonetDB/issues/1597/comments | 0 | 2020-11-30T09:48:17Z | 2024-06-27T11:28:47Z | https://github.com/MonetDB/MonetDB/issues/1597 | 753,324,386 | 1,597 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-21 09:52:24 +0100
From: Johnson Liu <<tiger_liu>>
To: @njnes
Version: -- development
Last updated: 2008-03-06 10:37:04 +0100
## Comment 9030
Date: 2008-02-21 09:52:24 +0100
From: Johnson Liu <<tiger_liu>>
When I execute the SQL as below,some transaction error happen.
sql>create table t1(i int);
sql>create table t2(i int);
sql>start transaction;
sql>insert into t1 values(1);
Rows affected 1
sql>savepoint s1;
sql>insert into t2 values(1);
Rows affected 1
sql>commit;
sql>select * from t1;
+---+
| i |
+===+
| 1 |
+---+
sql>select * from t2;
+---+
| i |
+===+
| 1 |
+---+
sql>start transaction;
sql>select * from t1;
sql>select * from t2;
+---+
| i |
+===+
| 1 |
+---+
sql>commit;
sql>select * from t1;
sql>select * from t2;
+---+
| i |
+===+
| 1 |
+---+
We can find the t1 should have record but not now.
These phenoma is not so stable but very often.
## Comment 9031
Date: 2008-02-22 09:15:41 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
Fixed a but in commit code. Added the test to src/tests/BugTracker/Tests/savepoint-SF.1898533
## Comment 9032
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898533 at http://sourceforge.net/support/tracker.php?aid=1898533
| transaction error with savepoint | https://api.github.com/repos/MonetDB/MonetDB/issues/1596/comments | 0 | 2020-11-30T09:48:14Z | 2024-06-27T11:28:46Z | https://github.com/MonetDB/MonetDB/issues/1596 | 753,324,343 | 1,596 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-21 09:41:17 +0100
From: Maurice van Keulen <<mvankeulen>>
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2009-04-07 10:24:13 +0200
## Comment 9024
Date: 2008-02-21 09:41:17 +0100
From: Maurice van Keulen <<mvankeulen>>
Given the query
declare function getData($d) { $d//data[@num="1"] };
let $d := <b><c><data num="1">211098</data></c></b>
return <result>{getData($d)}</result>
MPS-version gives the error message
!type error: no variant of function pf:attribute accepts the given argument type(s): item*; string; string; string; string; string
!type error: maybe you meant: !type error: pf:attribute (node*, string, string, string, string, string) as node* !type error: illegal arguments for function pf:attribute ! halted in typecheck.brg (fun_not_found), line 1317
Algebra-version gives the error message "fatal error: Sorry, I cannot translate the test for type `node*'"
The problem lies in the fact that in this query, the parameter $d has no specified type, while in the body we refer to an attribute. Is we write
declare function getData($d as element()) { $d//data[@num="1"] };
let $d := <b><c><data num="1">211098</data></c></b>
return <result>{getData($d)}</result>
The query compiles and runs fine.
## Comment 9025
Date: 2008-02-21 10:04:39 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Maurice,
MPS gives you the error because Peters heuristic magic that transforms the path order always expects node* -- a bug in my opinion. The algebra variant on the other side misses the implementation for the runtime typeswitch and thus fails with a not-so-informative error message.
Perhaps Peter has an idea to solve this for MPS - maybe an additional typeswitch that ensures that pf:attribute is only called for nodes.
## Comment 9026
Date: 2009-02-17 13:45:37 +0100
From: @drstmane
added test in pathfinder/tests/BugTracker/Tests/function_parameter_without_type.SF-1898518.*
behaves as reported:
- works fine with ALG
- triggers error with MPS
## Comment 9027
Date: 2009-04-07 10:24:13 +0200
From: @peterboncz
will not be fixed
## Comment 9028
Date: 2009-11-14 23:40:35 +0100
From: @drstmane
2009/11/14 - stmane:
function_parameter_without_type.SF-1898518.stable.err,1.2.8.1
function_parameter_without_type.SF-1898518.stable.out,1.2.8.1
approved error message of first version of the query
after Peter's "will not be fixed" verdict;
cf.,
ID: 1898518 "PF pf:attribute error for function parameter without type"
https://sourceforge.net/tracker/?func=detail&aid=1898518&group_id=56967&atid=482468
Thus, we keep on testing the second version of the query.
## Comment 9029
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898518 at http://sourceforge.net/support/tracker.php?aid=1898518
| PF pf:attribute error for function parameter without type | https://api.github.com/repos/MonetDB/MonetDB/issues/1595/comments | 0 | 2020-11-30T09:48:11Z | 2024-06-27T11:28:45Z | https://github.com/MonetDB/MonetDB/issues/1595 | 753,324,308 | 1,595 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-21 07:48:04 +0100
From: David Wung <<fivecolor>>
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2008-02-21 08:30:38 +0100
## Comment 9018
Date: 2008-02-21 07:48:04 +0100
From: David Wung <<fivecolor>>
When calling the procedure including 'while', mserver seems to be in die circle, unless interrupt it.
Following query is my test script in mclient:
create table t1(a int);
create table t2(b int);
create procedure p1(a int) while (a>2) do select * into t1 from t2 ;end while;
call p1(1); -- OK
call p1(3); -- waiting all the times
Thanks.
## Comment 9019
Date: 2008-02-21 09:12:38 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
isn't this what you ask the server for, ie while (a>2) never ends. Once your code assigns a value <= 2 to a it stops.
## Comment 9020
Date: 2008-02-22 03:09:55 +0100
From: David Wung <<fivecolor>>
Logged In: YES
user_id=2013222
Originator: YES
Yes, you are right.
Furthermore, does procedure sustain ‘begin & end’? If it can, what syntax is it?
## Comment 9021
Date: 2008-02-22 07:35:20 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
create procedure p1(a int)
begin
statements
end;
Also a single statement works without begin/end as can be seen from your example.
## Comment 9022
Date: 2008-02-22 13:45:07 +0100
From: David Wung <<fivecolor>>
Logged In: YES
user_id=2013222
Originator: YES
Thank you.
## Comment 9023
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898473 at http://sourceforge.net/support/tracker.php?aid=1898473
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| call procedure including while | https://api.github.com/repos/MonetDB/MonetDB/issues/1594/comments | 0 | 2020-11-30T09:48:08Z | 2024-06-28T13:30:45Z | https://github.com/MonetDB/MonetDB/issues/1594 | 753,324,274 | 1,594 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-20 11:03:43 +0100
From: @yzchang
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-21 11:12:59 +0100
## Comment 9008
Date: 2008-02-20 23:03:43 +0100
From: @yzchang
Hi,
Compiling stable branch on andorea (fedora 8) results in:
monet/stable/sql/src/backends/monet5/merovingian.mx: In function ‘main’:
monet/stable/sql/src/backends/monet5/merovingian.mx:1355: warning: ISO C90 forbids mixed declarations and code
Configure options:
--enable-debug --enable-assert --disable-optimize --enable-strict
Regards,
Jennie
## Comment 9009
Date: 2008-02-21 10:42:20 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
1355: pthread_t ctid = 0;
I see. My compiler (gcc (GCC) 4.2.3 (Gentoo 4.2.3 p1.0)) just doesn't apparently, maybe because I don't have -O? in CFLAGS.
## Comment 9010
Date: 2008-02-21 10:45:55 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Fabian,
thanks for the fix.
gcc 4.2.3 should give the same warning --- however, it's considered an error only with --enable-strict, which is not the default in the Stable release branch ...
Stefan
## Comment 9011
Date: 2008-02-21 10:50:09 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
-Wall -Werror should do the trick, no?
I don't see any warnings or messages on stderr where warnings are sent to.
## Comment 9012
Date: 2008-02-21 10:57:45 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
my compilation statement is as follows:
x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I/ufs/fabian/scratch/monetdb/stable
/sql/src/backends/monet5 -I../../.. -I/ufs/fabian/scratch/monetdb/stable/sql/src
/backends/monet5 -I../../include -I/ufs/fabian/scratch/monetdb/stable/sql/src/ba
ckends/monet5/../../include -I../../common -I/ufs/fabian/scratch/monetdb/stable/
sql/src/backends/monet5/../../common -I../../storage -I/ufs/fabian/scratch/monet
db/stable/sql/src/backends/monet5/../../storage -I../../server -I/ufs/fabian/scr
atch/monetdb/stable/sql/src/backends/monet5/../../server -I/ufs/fabian/scratch/m
onetdb/stable/program-x86_64/include/MonetDB5/atoms -I/ufs/fabian/scratch/monetd
b/stable/program-x86_64/include/MonetDB5/compiler -I/ufs/fabian/scratch/monetdb/
stable/program-x86_64/include/MonetDB5/kernel -I/ufs/fabian/scratch/monetdb/stab
le/program-x86_64/include/MonetDB5/mal -I/ufs/fabian/scratch/monetdb/stable/prog
ram-x86_64/include/MonetDB5/optimizer -I/ufs/fabian/scratch/monetdb/stable/progr
am-x86_64/include/MonetDB5/scheduler -I/ufs/fabian/scratch/monetdb/stable/progra
m-x86_64/include/MonetDB5 -I/ufs/fabian/scratch/monetdb/stable/program-x86_64/in
clude/MonetDB -I/ufs/fabian/scratch/monetdb/stable/program-x86_64/include/MonetD
B/mapilib -I/ufs/fabian/scratch/monetdb/stable/program-x86_64/include/MonetDB -I
/ufs/fabian/scratch/monetdb/stable/program-x86_64/include/MonetDB/common -I/ufs/
fabian/scratch/monetdb/stable/program-x86_64/include/MonetDB/gdk -march=athlon64 -pipe -g -W -Wall -std=c99 -fgnu89-inline -Werror -D_FORTIFY_SOURCE=2 -c merovingian.c
I see -std=c99 in there, which makes me wonder if that doesn't allow me to even use // as comment separator? So in that case the compiler is right in not making any noise about this ISO C90 constraint...
Jennie, do you have -std=c99 in your compilation log?
## Comment 9013
Date: 2008-02-21 11:06:56 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: YES
This is my complete compilation statement:
gcc -DHAVE_CONFIG_H -I. -I../../.. -I/ufs/zhang/monet/stable/sql/src/backends/monet5 -I/ufs/zhang/monet/stable/sql/src/backends/monet5 -I../../include -I/ufs/zhang/monet/stable/sql/src/backends/monet5/../../include -I../../common -I/ufs/zhang/monet/stable/sql/src/backends/monet5/../../common -I../../storage -I/ufs/zhang/monet/stable/sql/src/backends/monet5/../../storage -I../../server -I/ufs/zhang/monet/stable/sql/src/backends/monet5/../../server -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5/atoms -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5/compiler -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5/kernel -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5/mal -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5/optimizer -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5/scheduler -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB5 -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB/mapilib -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB/common -I/export/scratch1/zhang/monet-install/stable/debug/include/MonetDB/gdk -Wall -Wextra -std=c99 -g -Werror-implicit-function-declaration -Werror -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wp,-D_FORTIFY_SOURCE=2 -c merovingian.c
so the answer is yes. I guess -Wdeclaration-after-statement doesn't like mixed declarations and code.
## Comment 9014
Date: 2008-02-21 11:08:29 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
see lines 971-1029 in buildtools/conf/MonetDB.m4
## Comment 9015
Date: 2008-02-21 11:10:41 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Ok, the MonetDB.m4 hint was for --std=c99
Jennie,
you're right, -Wdeclaration-after-statement triggers the warning and -Wdeclaration-after-statement is enabled only with --enable-strict.
Thanks.
## Comment 9016
Date: 2008-02-21 11:12:59 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
I'll add the extended flags to my CFLAGS.
## Comment 9017
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898219 at http://sourceforge.net/support/tracker.php?aid=1898219
| SQL: failed to compile due to merovingian | https://api.github.com/repos/MonetDB/MonetDB/issues/1593/comments | 0 | 2020-11-30T09:48:05Z | 2024-06-27T11:28:43Z | https://github.com/MonetDB/MonetDB/issues/1593 | 753,324,250 | 1,593 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-20 07:56:24 +0100
From: Michael Yang <<msyang>>
To: @njnes
Version: -- development
Last updated: 2008-03-14 02:08:14 +0100
## Comment 9004
Date: 2008-02-20 19:56:24 +0100
From: Michael Yang <<msyang>>
Using the newest x64 Feb2008 release of monetDB/SQL on Windows server 2003 x64.
This is a follow-up to issue [ 1892413 ] wrong or unexpected results in subquery - I'm instead attempting to use the row_number() over() to retrieve the next consecutive date.
sql>CREATE TABLE tradeDate(date int not null constraint pk_tradeDate primary key);
Timer 15.000 msec 0 rows
sql>COPY INTO tradeDate FROM 'T:/temp/dt.txt' USING DELIMITERS '\t';
Rows affected 32605
Timer 110.000 msec 0 rows
sql>SELECT COUNT(*) AS cnt from tradeDate;
+-------+
| cnt |
+=======+
| 32605 |
+-------+
Timer 0.000 msec 1 rows
sql>WITH e as (SELECT date, row_number() over (order by date) as rn from tradeDate) SELECT date, rn from e ORDER BY date limit 5;
+----------+---+
| date |rn |
+==========+===+
| 19251231 | 0 |
| 19260102 | 1 |
| 19260104 | 2 |
| 19260105 | 3 |
| 19260106 | 4 |
+----------+---+
Timer 0.000 msec 5 rows
sql>WITH e as (SELECT date, row_number() over (order by date) as rn from tradeDate) SELECT e1.date, e1.rn, e2.date, e2.rn from e as e1 inner join e as e2 ON e2.
rn = e1.rn limit 5;
+----------+---+---------+---+
| date |rn |date |rn |
+==========+===+=========+===+
| 19251231 | 0 |19251231 | 0 |
| 19260102 | 1 |19260102 | 1 |
| 19260104 | 2 |19260104 | 2 |
| 19260105 | 3 |19260105 | 3 |
| 19260106 | 4 |19260106 | 4 |
+----------+---+---------+---+
Timer 32.000 msec 5 rows
--------------------------------------------
At this point, everything is OK. I've loaded the dates, demonstrated that row_number() is working. However, when attempting to join the row-numbered table 'e' with itself based on getting the next day, the query returns no results:
sql>WITH e as (SELECT date, row_number() over (order by date) as rn from tradeDate) SELECT e1.date, e1.rn, e2.date, e2.rn from e as e1 inner join e as e2 ON e2.
rn + 1 = e1.rn limit 5;
Timer 0.000 msec 0 rows
Still doesn't work if I take the 'limit' clause off:
sql>WITH e as (SELECT date, row_number() over (order by date) as rn from tradeDa
te) SELECT e1.date, e1.rn, e2.date, e2.rn from e as e1 inner join e as e2 ON e2.
rn + 1 = e1.rn;
Timer 0.000 msec 0 rows
[data file dt.txt is attached]
## Comment 9005
Date: 2008-02-21 09:09:01 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
The bug seems to be related to the WITH statement. When we use a real view for e it gives the correct output.
## Comment 9006
Date: 2008-02-21 20:59:59 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
In the 'WITH' handling we didn't correctly handle the using a single local variable twice (once as e1 and once as e2). This is fixed now. The test was added to
sql/src/test/BugTracker/Tests/with_row_number.SF-1898089.sql
## Comment 9007
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1898089 at http://sourceforge.net/support/tracker.php?aid=1898089
| row_number() failing when used in join | https://api.github.com/repos/MonetDB/MonetDB/issues/1592/comments | 0 | 2020-11-30T09:48:02Z | 2024-06-27T11:28:42Z | https://github.com/MonetDB/MonetDB/issues/1592 | 753,324,213 | 1,592 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-19 01:43:38 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-25 11:54:32 +0100
## Comment 8996
Date: 2008-02-19 13:43:38 +0100
From: @drstmane
While working fine with MPS on all platforms and with ALG on most platforms,
all XQuery module test trigger a "Segmentation Fault" (or "Arithmetic Exception") on Solaris 10/x86 (32-bit); see e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_Music_mod_XQ/full_01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_Music_mod_XQ/full_02.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_Music_opt_XQ/full_01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_Music_opt_XQ/full_02.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_mod_XQ/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_mod_XQ/q02.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_opt_XQ/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_opt_XQ/q02.err.00.html
## Comment 8997
Date: 2008-02-22 19:27:05 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
For the first test (benchmarks/Music/mod/XQ/Tests/full_01.*),
it appears to be a devision by zero in
PFarray_at() (line 134 of compiler/mem/array.c);
"a->esize" appear to be "0" !??
See the attached GDB log for details.
Please let me know, in case you need more info and/or help.
File Added: SolarisSegFault
## Comment 8998
Date: 2008-02-22 19:34:48 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
small correction:
the division by zero occurs in line 131 of compiler/mem/array.c (rev. 1.24.6.3)
## Comment 8999
Date: 2008-02-22 21:20:04 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
I guess it could be the same reason as for bug [ 1895125 ] PF: 28 ALG tests still segfault.
It will interpret bogus data as array because of the missing initialization.
## Comment 9000
Date: 2008-02-22 21:29:06 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Thanks Jan!
In fact, I just did a simple test for the "ALGsegfault" bug as follows:
I simply ensure that arrays are initialized with zero as follows:
========
Index: compiler/mem/array.c
===================================================================
RCS file: /cvsroot/monetdb/pathfinder/compiler/mem/array.c,v
retrieving revision 1.24.6.3
diff -u -r1.24.6.3 array.c
--- compiler/mem/array.c 22 Feb 2008 15:13:15 -0000 1.24.6.3
+++ compiler/mem/array.c 22 Feb 2008 20:22:44 -0000
@@ -95,6 +95,9 @@
a->base = PFmalloc (nbytes);
+ if (a->base)
+ memset (a->base, 0, nbytes);
+
a->bound = nbytes / s;
a->esize = s;
========
The first result is that the "ALGsegfault" test still fails as before, but now not shows "weird" data, but rather "0xDBDBDBD..." meaning that memory is being accessed that has been GDKfree'ed (en hence wiped-out with our humble version of 0xDeadBeef) ... looks like a dangling pointer to obsolete memory is being maintained/used --- or non-initialized memory is read also outside PFarrays ...
Stefan
## Comment 9001
Date: 2008-02-23 08:27:45 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
The patch that I just suggested for
[ 1895125 ] PF: 28 ALG tests still segfault
https://sourceforge.net/tracker/index.php?func=detail&aid=1895125&group_id=56967&atid=482468
also seems to work for this one ...
## Comment 9002
Date: 2008-02-25 11:54:32 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Temporarily fixed in XQuery_0-22 branch.
Superseded by
[ 1901269 ] PF: compiler requires improved memory management
https://sourceforge.net/tracker/index.php?func=detail&aid=1901269&group_id=56967&atid=482468
## Comment 9003
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1896890 at http://sourceforge.net/support/tracker.php?aid=1896890
| PF/alg: module-tests segfault on Solaris 10/x86 (32-bit) | https://api.github.com/repos/MonetDB/MonetDB/issues/1591/comments | 0 | 2020-11-30T09:47:59Z | 2024-06-27T11:28:41Z | https://github.com/MonetDB/MonetDB/issues/1591 | 753,324,169 | 1,591 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-19 10:16:32 +0100
From: Thomas Vervest <<tvervest>>
To: @yzchang
Version: -- development
Last updated: 2008-02-19 12:24:23 +0100
## Comment 8988
Date: 2008-02-19 10:16:32 +0100
From: Thomas Vervest <<tvervest>>
When I call the function save-export (see below) via XRPC I get the following error: "!ERROR: err:FORG0005: function fn:exactly-one expects exactly one value."
This is the function i am calling:
------------------------------
declare updating function epv:save-export($title as xs:string
,$query as node()*)
{
let $seq := <sequence name="$title"></sequence>
return do insert $seq as last into doc("saved_exports.xml")/doc/saves
};
------------------------------
I've included the XRPC request I'm sending and the file which will be updated.
## Comment 8989
Date: 2008-02-19 10:30:14 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Thomas,
since we just finished the new MonetDB4 4.22 / MonetDB/XQuery 0.22 release,
there will most probably be no more bug fixes for the MonetDB4 4.20 / MonetDB/XQuery 0.20 release.
Hence, could you please check, whether how your query works with the new MonetDB4 4.22 / MonetDB/XQuery 0.22 release?
Please see http://monetdb.cwi.nl/Development/Releases/Version4.22/ for details.
Stefan
## Comment 8990
Date: 2008-02-19 10:49:51 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Thomas,
The target ('doc("saved_exports.xml")/doc/saves') of an insert-into statement must be exactly one node. I took a look at you input document, and see that the root node of your document is "saves". This is why you get the error: you documet doesn't have a root node "doc", so the expression 'doc("saved_exports.xml")/doc/saves' returns an empty result, while exactly one node is expected. So please change your query into: doc("saved_export.xml")/saves , and see if it will work.
If you then still get the same error, you can change you query into this: exactly-one(doc("saved_export.xml")/saves)
This is because the return type of an XPath expression (e.g. doc("saved_export.xml")/saves) is node()*, while insert-into expect a target node of the type node().
Due to the limitation in the typing system, MonetDB/XQuery might no able to detect at runtime that your expression doc("saved_export.xml")/saves actually only returns one node.
Please let us know if this helps.
Regards,
Jennie
## Comment 8991
Date: 2008-02-19 10:50:03 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Thomas,
The target ('doc("saved_exports.xml")/doc/saves') of an insert-into statement must be exactly one node. I took a look at you input document, and see that the root node of your document is "saves". This is why you get the error: you documet doesn't have a root node "doc", so the expression 'doc("saved_exports.xml")/doc/saves' returns an empty result, while exactly one node is expected. So please change your query into: doc("saved_export.xml")/saves , and see if it will work.
If you then still get the same error, you can change you query into this: exactly-one(doc("saved_export.xml")/saves)
This is because the return type of an XPath expression (e.g. doc("saved_export.xml")/saves) is node()*, while insert-into expect a target node of the type node().
Due to the limitation in the typing system, MonetDB/XQuery might no able to detect at runtime that your expression doc("saved_export.xml")/saves actually only returns one node.
Please let us know if this helps.
Regards,
Jennie
## Comment 8992
Date: 2008-02-19 10:52:32 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Sorry, changed header information by accident.
## Comment 8993
Date: 2008-02-19 12:20:10 +0100
From: Thomas Vervest <<tvervest>>
Logged In: YES
user_id=1949071
Originator: YES
Hey Jennie and Stefan,
Thanks for the quick resplies, I changed my query to insert data into ('doc("saved_exports.xml")/saves') and it worked!
I'm so ashamed nobody here saw that :$
Thanks for all the help and sorry for the false bug report!
Kind regards - Thomas
## Comment 8994
Date: 2008-02-19 12:24:23 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Thomas,
no problem.
I'll close it, then.
Stefan
## Comment 8995
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1896776 at http://sourceforge.net/support/tracker.php?aid=1896776
| XQ: error FORG0005 when updating xml file | https://api.github.com/repos/MonetDB/MonetDB/issues/1590/comments | 0 | 2020-11-30T09:47:56Z | 2024-06-28T13:30:45Z | https://github.com/MonetDB/MonetDB/issues/1590 | 753,324,127 | 1,590 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-19 09:45:56 +0100
From: Wouter Alink <<alink>>
To: @sjoerdmullender
Version: -- development
Last updated: 2008-03-20 05:37:05 +0100
## Comment 8984
Date: 2008-02-19 09:45:56 +0100
From: Wouter Alink <<alink>>
updates can still corrupt an xml-document.
in the attached 'bug.tgz' there is a tiny script that repeatedly does two things:
- check the validity of an xml-document under stress
- do a small update on an xml-document
if this script is run it will only output 0's, meaning that the document is still valid. if this script is run in 3 threads at the same time, it will (within short time) return non-'0' values, which indicate the document has been corrupted.
how to use the script:
- start bin/Mserver --dbinit="module(pathfinder);"
- insert start.xml as an updatable document named 'test.xml', for example with a free-percentage of 10
- start do.sh in 3 different shells
- wait
this was tested with the monetdb-install.sh script (on feb 12). monetdb says:
MonetDB Server v4.22.0
based on GDK v1.22.0
Copyright (c) 1993-2008, CWI. All rights reserved.
Compiled for x86_64-redhat-linux-gnu/64bit with 64bit OIDs; dynamically linked.
## Comment 8985
Date: 2008-02-19 13:06:13 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Perhaps this is a problem in maintaining the indices. I tried running the test, and at some point I got non-0 results, so I stopped the three scripts. Using a variation on the query that returns the 0, I get some strange results. Te node being returned doesn't actually match the query:
xquery>doc("test.xml")//aap[not(@xid)]
more><>
<mies prefix="dum"/>
xquery>doc("test.xml")//noot[not(@name)]
more><>
<aap xid="1"><noot name="hazel"><aap xid="1"><noot name="hazel"><aap xid="1"><noot name="hazel"/><mies prefix="dum"/></aap></noot><mies prefix="dum"><aap xid="1"><aap xid="1"><noot name="hazel"><aap xid="1"><noot name="hazel"/><mies prefix="dum"/></aap></noot><mies prefix="dum"/></aap><noot name="hazel"/><mies prefix="dum"/></aap></mies></aap></noot><mies prefix="dum"/></aap>
xquery>doc("test.xml")//mies[not(@prefix)]
more><>
<noot name="hazel"><aap xid="1"><noot name="hazel"><aap xid="1"><noot name="hazel"/><mies prefix="dum"/></aap></noot><mies prefix="dum"><aap xid="1"><aap xid="1"><noot name="hazel"><aap xid="1"><noot name="hazel"/><mies prefix="dum"/></aap></noot><mies prefix="dum"/></aap><noot name="hazel"/><mies prefix="dum"/></aap></mies></aap></noot>
xquery>
## Comment 8986
Date: 2008-03-20 17:37:05 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
By adding overlapping working sets to ws_overlaps_ws *before* cleaning up
page_ws (which uses ws_overlaps_ws) the bug seems to be solved.
Fix checked in on stable branch.
Test added as update-stress.SF-1896755.
## Comment 8987
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1896755 at http://sourceforge.net/support/tracker.php?aid=1896755
| XQ: updates under stress | https://api.github.com/repos/MonetDB/MonetDB/issues/1589/comments | 0 | 2020-11-30T09:47:53Z | 2024-06-27T11:28:39Z | https://github.com/MonetDB/MonetDB/issues/1589 | 753,324,091 | 1,589 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-17 05:13:48 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-18 02:05:03 +0100
## Comment 8981
Date: 2008-02-17 17:13:48 +0100
From: @drstmane
With assertions disabled Pathfinder fails to compile:
pathfinder/compiler/algebra/physical.c: In function 'PFpa_leftjoin':
compiler/algebra/physical.c:523: error: implicit declaration of function 'check_col'
## Comment 8982
Date: 2008-02-18 14:05:03 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
removed incorrect ifndef NDEBUG statement
## Comment 8983
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1895460 at http://sourceforge.net/support/tracker.php?aid=1895460
| PF fails to compile with assertions disabled | https://api.github.com/repos/MonetDB/MonetDB/issues/1588/comments | 0 | 2020-11-30T09:47:51Z | 2024-06-27T11:28:38Z | https://github.com/MonetDB/MonetDB/issues/1588 | 753,324,054 | 1,588 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-17 01:41:45 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-19 01:13:14 +0100
## Comment 8977
Date: 2008-02-17 13:41:45 +0100
From: @drstmane
On Windows (only) 719 Pathfinder tests fail using the Algebra version (only) with
ERROR = !fatal error: attribute `*' referenced in projection not found
See e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/Int.32.32.d-Windows5.1/benchmarks_MBench/qr01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/Int.64.32.d-Windows5.2/benchmarks_MBench/qr01.err.00.html
A complete list is attached.
## Comment 8978
Date: 2008-02-18 13:58:04 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
It seems that mixing types bool and unsigned int break the algebra optimizer.
A fix is checked in -- testing will show whether this is the only problem.
## Comment 8979
Date: 2008-02-19 13:13:14 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Testing confirmed the fix.
Thanks!
## Comment 8980
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1895379 at http://sourceforge.net/support/tracker.php?aid=1895379
| PF/alg on Windows 719x "attr. referenced in proj. not found" | https://api.github.com/repos/MonetDB/MonetDB/issues/1587/comments | 0 | 2020-11-30T09:47:48Z | 2024-06-27T11:28:37Z | https://github.com/MonetDB/MonetDB/issues/1587 | 753,324,019 | 1,587 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-17 10:36:30 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-06-02 10:42:18 +0200
## Comment 8970
Date: 2008-02-17 10:36:30 +0100
From: @drstmane
8 of the tests reported in
[ 1894891 ] PF: 334 tests fail with ALG that work fine with MPS
http://sourceforge.net/tracker/index.php?func=detail&aid=1894891&group_id=56967&atid=482468
fail with one of
ERROR = !fatal error: attribute 'item' with type '0x200' not found in environment
ERROR = !fatal error: attribute 'item' with type '0x4' not found in environment
ERROR = !fatal error: attribute `(NULL)' referenced in projection not found
ERROR = !fatal error: physical algebra equivalent for logical algebra node kind 44 not implemented, yet
ERROR = !fatal error: physical algebra equivalent for logical algebra node kind 45 not implemented, yet
ERROR = !fatal error: physical algebra equivalent for logical algebra node kind 54 not implemented, yet
ERROR = !fatal error: Schema of two arguments of set operation (union, difference, intersect) do not match
ERROR = !fatal error: Sorry, I cannot translate the test for type `attribute { atomic* }'
ERROR = !fatal error: Sorry, I cannot translate the test for type `document { item* }'
ERROR = !fatal error: Sorry, I cannot translate the test for type `node'
See attached log for details.
Jan,
I consider you the "dispatcher" for PF/alg bugs;
please re-assign if/as appropriate.
## Comment 8971
Date: 2008-02-18 15:14:21 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
The following errors are perhaps a little bit misleading but only report on missing features/functionality:
ERROR = !fatal error: physical algebra equivalent for logical algebra node kind *** not implemented, yet
ERROR = !fatal error: Sorry, I cannot translate the test for type ***
ERROR = !fatal error: Algebra implementation for function `***' unknown.
The other errors:
ERROR = !fatal error: attribute '***' with type '***' not found in environment
ERROR = !fatal error: attribute `***' referenced in projection not found
ERROR = !fatal error: Schema of two arguments of set operation (union, difference, intersect) do not match
are correct ones and should be fixed.
@Lefteris: please fix the MIL related errors -- The translation for alg_fun_upd_replace_element seems incorrect:
tests/Update/replacevaluetest, tests/Update/replacevaluetest2, tests/W3C_use_cases/XQUF/AddressBook/Q1x
@Manuel: please look at the `Schema of two arguments of set operation ...' error in opt_algebra_cse.c
I will look at the `ERROR = !fatal error: attribute `***' referenced in projection not found' errors. Whenever they appear some algebra optimization is probably broken.
## Comment 8972
Date: 2008-02-18 18:48:13 +0100
From: @lsidir
Logged In: YES
user_id=1856546
Originator: NO
Hi all,
the update functions are not working in the Algebra version because the a) correct MIL play tape function is missing and b) the QNames are not treated correctly. The update tests are correctly failing since they are not fully implemented yet.
## Comment 8973
Date: 2008-02-18 19:15:57 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Could you please all help to clean-up the TestWeb in the XQuery_0-22 release branch by listing all tests that are bound to fail with the released ALG version due to missing (or not yet fully supported) features.
Please attach your list(s) to this bug report (or one of the related ones, if more appropriate), or send it directly to me, or add the "NOT_ALGEBRA?" conditional to the respective "All" files yourself.
In either case, please clearly indicate the reason (missing feature(s)) and that the respective changes in the XQuery_0-22 branch should not be propagated to the development trunk.
Thanks you very much in advance for your cooperation!
## Comment 8974
Date: 2008-02-25 17:22:09 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
After fixing some bugs and disabling test the require features that are not yet supported by the Algebra back-end, the following "fatal error"s remain:
benchmarks/XMark/opt/XQ/q04.test.err-QUERY = import module namespace xmark = "http://www.cwi.nl/~boncz/xmark/opt/" at "/scratch/monet.GNU.64.64.d.32289/pathfinder/benchmarks/XMark/opt/XQ/Tests/xmark.xq";
benchmarks/XMark/opt/XQ/q04.test.err:ERROR = !fatal error: attribute `(NULL)' referenced in projection not found
benchmarks/XMark/opt/XQ/q04.test.err-
--
benchmarks/XBench/TC/MD/q19.test.err-QUERY = let $doc :=
benchmarks/XBench/TC/MD/q19.test.err:ERROR = !fatal error: attribute `(NULL)' referenced in projection not found
benchmarks/XBench/TC/MD/q19.test.err-
--
benchmarks/XBench/DC/MD/q04.test.err-QUERY = let $doc :=
benchmarks/XBench/DC/MD/q04.test.err:ERROR = !fatal error: attribute `(NULL)' referenced in projection not found
benchmarks/XBench/DC/MD/q04.test.err-
--
tests/XQuery/attr_err3.test.err-QUERY = doc("foo.xml")//@*/parent::*
tests/XQuery/attr_err3.test.err:ERROR = !fatal error: attribute 'item' with type '0x200' not found in environment
tests/XQuery/attr_err3.test.err-
--
tests/BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.test.err-QUERY = string(string(string(string(string(string(string(string(string(string( string(string(string(string(string(string(string(string(string(string(
tests/BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.test.err:ERROR = !fatal error: aborted too deep recursion
tests/BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.test.err-
--
tests/BugTracker/server-side_compilation_crash.SF-1607210.test.err-QUERY = <x/>
tests/BugTracker/server-side_compilation_crash.SF-1607210.test.err:ERROR = !fatal error: aborted too deep recursion
tests/BugTracker/server-side_compilation_crash.SF-1607210.test.err-
--
tests/BugTracker/crash_on_concatenated_query.SF-1730547.test.err-QUERY = "aap",
tests/BugTracker/crash_on_concatenated_query.SF-1730547.test.err:ERROR = !fatal error: aborted too deep recursion
tests/BugTracker/crash_on_concatenated_query.SF-1730547.test.err-
## Comment 8975
Date: 2008-06-02 22:42:18 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Most (if not all) of the bugs reported here have been fixed recently;
the remaining ones have been filed in individual/specific bug reports.
## Comment 8976
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1895322 at http://sourceforge.net/support/tracker.php?aid=1895322
| PF/alg: 20 tests fail with (various) "fatal error" | https://api.github.com/repos/MonetDB/MonetDB/issues/1586/comments | 0 | 2020-11-30T09:47:45Z | 2024-06-27T11:28:36Z | https://github.com/MonetDB/MonetDB/issues/1586 | 753,323,986 | 1,586 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-17 10:15:10 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-05-29 06:46:35 +0200
## Comment 8964
Date: 2008-02-17 10:15:10 +0100
From: @drstmane
8 of the tests reported in
[ 1894891 ] PF: 334 tests fail with ALG that work fine with MPS
http://sourceforge.net/tracker/index.php?func=detail&aid=1894891&group_id=56967&atid=482468
fail with
ERROR = !ERROR: fetch(tmp_271) illegal position 0.
See attached log for details.
Jan,
I consider you the "dispatcher" for PF/alg bugs;
please re-assign if/as appropriate.
## Comment 8965
Date: 2008-02-25 17:23:42 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
After fixing some bugs and disabling test the require features that are
not yet supported by the Algebra back-end, the following "illegal position"s
remain:
tests/WebSite/HelloWorld8.test.err-QUERY = let $f := doc("http://monetdb.cwi.nl/XQuery/files/HelloWorld.xml")/doc
tests/WebSite/HelloWorld8.test.err:ERROR = !ERROR: fetch(tmp_263) illegal position 0.
tests/WebSite/HelloWorld8.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/WebSite/HelloWorld6.test.err-QUERY = for $t in ( <doc>
tests/WebSite/HelloWorld6.test.err:ERROR = !ERROR: fetch(tmp_263) illegal position 0.
tests/WebSite/HelloWorld6.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/WebSite/voc2.test.err-QUERY = for $t in doc("voc.xml")//voyage[leftpage/boatname="ZEELANDIA"]
tests/WebSite/voc2.test.err:ERROR = !ERROR: fetch(tmp_460) illegal position 0.
tests/WebSite/voc2.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/XQuery/sum11.test.err-QUERY = for $a in (attribute foo {"11"}, element bar {text{"1"},text{"2"}}) return $a + 100
tests/XQuery/sum11.test.err:ERROR = !ERROR: fetch(tmp_452) illegal position 0.
tests/XQuery/sum11.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/XQuery/helper-functions.test.err-QUERY = <foo>{2, <bar/>, 3, text{"4"}}</foo>
tests/XQuery/helper-functions.test.err:ERROR = !ERROR: fetch(tmp_452) illegal position 0.
tests/XQuery/helper-functions.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/XQuery/order.test.err-QUERY = let $a := (1, 2.3, "test", element bar {text{"bl"},text{"ub"}}) return for $b in ("foo", "bar", "baz") return attribute {$b}{$a,$b}
tests/XQuery/order.test.err:ERROR = !ERROR: fetch(tmp_452) illegal position 0.
tests/XQuery/order.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/W3C_use_cases/XQ/SEQ/Q5a.test.err-QUERY = declare function local:precedes($a as node(), $b as node()) as xs:boolean
tests/W3C_use_cases/XQ/SEQ/Q5a.test.err:ERROR = !ERROR: fetch(tmp_363) illegal position 0.
tests/W3C_use_cases/XQ/SEQ/Q5a.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/W3C_use_cases/XQ/SEQ/Q5b.test.err-QUERY = <critical_sequence>
tests/W3C_use_cases/XQ/SEQ/Q5b.test.err:ERROR = !ERROR: fetch(tmp_363) illegal position 0.
tests/W3C_use_cases/XQ/SEQ/Q5b.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/W3C_use_cases/XQ/SEQ/Q5d.test.err-QUERY = <critical_sequence>
tests/W3C_use_cases/XQ/SEQ/Q5d.test.err:ERROR = !ERROR: fetch(tmp_363) illegal position 0.
tests/W3C_use_cases/XQ/SEQ/Q5d.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
--
tests/BugsViaSourgeforce/ID.1210173-2.test.err-QUERY = let $f := doc("HelloWorld.xml")/doc
tests/BugsViaSourgeforce/ID.1210173-2.test.err:ERROR = !ERROR: fetch(tmp_2467) illegal position 0.
tests/BugsViaSourgeforce/ID.1210173-2.test.err- !ERROR: CMDfetchConvert_oid: operation failed.
## Comment 8966
Date: 2008-02-25 18:57:52 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
In fact, the "illegal position" is (of course) not "0" but oid(nil) or int(nil) --- "0" occurred only due to a bug in the error message itself (just fixed that).
## Comment 8967
Date: 2008-05-29 17:01:07 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
All but one of the "illegal positions" seem to be fixed.
Still failing:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.64.64.d-Fedora8/tests_XQuery/order.err.00.html
MAPI = monetdb@beo-13:37381
QUERY = let $a := (1, 2.3, "test", element bar {text{"bl"},text{"ub"}}) return for $b in ("foo", "bar", "baz") return attribute {$b}{$a,$b}
ERROR = !ERROR: fetch(tmp_452) illegal position 9223372036854775808.
!ERROR: CMDfetchConvert_oid: operation failed.
!ERROR: interpret_params: mirror(param 1): evaluation error.
## Comment 8968
Date: 2008-05-29 18:46:35 +0200
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Fixed the last fetch problem: deleted text nodes (nil values) where not deleted by PROC merge_adjacent_text_nodes().
## Comment 8969
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1895311 at http://sourceforge.net/support/tracker.php?aid=1895311
| PF/alg: "fetch" fails in 8 tests | https://api.github.com/repos/MonetDB/MonetDB/issues/1585/comments | 0 | 2020-11-30T09:47:43Z | 2024-06-27T11:28:35Z | https://github.com/MonetDB/MonetDB/issues/1585 | 753,323,954 | 1,585 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-16 11:13:43 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-25 11:55:04 +0100
## Comment 8952
Date: 2008-02-16 23:13:43 +0100
From: @drstmane
After Manuel's fix of
[ 1894964 ] PF/ALG: 170 tests segfault in atom_eq() in opt_algebra_cse.c
http://sourceforge.net/tracker/index.php?func=detail&aid=1894964&group_id=56967&atid=482468
28 of the 170 tests still segfault in various places and/or for various reasons.
See `Mtest.py -G ...` and/or the attached list and log for details.
## Comment 8953
Date: 2008-02-16 23:16:54 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
File Added: ALG_segfault_2_GDB_log
## Comment 8954
Date: 2008-02-18 07:35:19 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Some of the segfaults with ALG occur because running more than 2 (two) queries
using the ALG version on the same `Mserver --dbinit="module(pathfinder);"`
make the Mserver segfault in "subexp_eq()" in "compiler/algebra/algebra_cse.c" .
E.g., see newly added test tests/BugTracker/Tests/ALGsegfault.SF-1894964{a,b}.*;
simply compare
`Mtest.py -q tests/BugTracker/ ALGsegfault.SF-1894964{a,b}`
and
`Mtest.py -G -q tests/BugTracker/ ALGsegfault.SF-1894964{a,b}`
See the attached log for details.
File Added: ALG_segfault_3_GDB_log
## Comment 8955
Date: 2008-02-18 14:51:24 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
1.) By producing an error message for the missing self step (instead of an assertion) we now have less segfault queries (and of course more with a FATAL ERROR.
2.) tests/BugTracker/Tests/ALGsegfault.SF-1894964{a,b}.*; works as expected on my setup
3.) Various queries listed in ALG_segfault_2 do not segfault on my machine anymore, but produce only an error message (mostly because of missing features). In fact all queries for which you attached a gdb profile now do not segfault anymore.
Stefan, what do you think we should do with this bug report? -- Would you call it fixed?
## Comment 8956
Date: 2008-02-18 19:56:48 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Jan,
thanks for the fixes and analysis!
1.)
We should conditionally disable (in the XQuery_0-22 release branch, only) all tests that are known/bound to fail due to not yet (fully) available features; see also
[ 1895322 ] PF/alg: 20 tests fail with (various) "fatal error"
http://sourceforge.net/tracker/index.php?func=detail&aid=1895322&group_id=56967&atid=482468
2.)
tests/BugTracker/Tests/ALGsegfault.SF-1894964{a,b}.*; still fails for me on my 64-bit Fedora8 desktop (both with "--enable-debug --enable-assert" and "--enable-optimize --disable-assert) --- let's see what tonights testing will reveal tomorrow morning; might be a 64-bit and/or platform-specific problem ...
3.)
See 1.) above.
## Comment 8957
Date: 2008-02-19 13:20:04 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
2.)
While working fine with MPS on all platforms
(cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora8/tests_BugTracker/ALGsegfault.SF-1894964a.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora8/tests_BugTracker/ALGsegfault.SF-1894964b.err.00.html
),
the segfault with the third query occurs with ALG only on 64-bit x86_64 machines running Fedora (6 & 8)
(cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.64.64.d-Fedora8/tests_BugTracker/ALGsegfault.SF-1894964a.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.64.64.d-Fedora8/tests_BugTracker/ALGsegfault.SF-1894964b.err.00.html
);
it works fine also with ALG on 32-bit systems as well as on 64-bit Windows (on x86_64) and on 64-bit RedHat (on IA-64 (Itanium)) ...
## Comment 8958
Date: 2008-02-19 17:26:08 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
It turned out that line 585 in algebra/algebra_cse.c <code>a = *(PFarray_t **) PFarray_at (subexps, n->kind);</code> relies on the fact that memory is allocated with calloc instead of malloc. This triggered queries ALGsegfault.SF-1894964{a,b} to fail on some platforms. (Similar problems arise in algebra/opt/opt_algebra_cse.c.)
Note that this is a general problem of all scenarios where PFarray is used as an array instead of a list.
## Comment 8959
Date: 2008-02-22 21:30:05 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
See also comment at
[ 1896890 ] PF/alg: module-tests segfault on Solaris 10/x86 (32-bit)
http://sourceforge.net/tracker/index.php?func=detail&aid=1896890&group_id=56967&atid=482468
## Comment 8960
Date: 2008-02-23 08:15:47 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
The attached patch, replacing the "block"-malloc's by calloc's in compiler/mem/mem.c, seems to "solve" (avoid) the segfaults --- by forcefully initializing memory before reading from it ...
Should I check it in?
If so, leave it free for propagation to the head, or rather not, forcing us to find a proper/better solution for the head?
File Added: malloc2calloc.patch
## Comment 8961
Date: 2008-02-24 17:49:04 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
I applied the patch in the XQuery_0-22 release branch, only (propagation prevented).
For the development trunk we should find a better/proper solution.
## Comment 8962
Date: 2008-02-25 11:55:04 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Temporarily fixed in XQuery_0-22 branch.
Superseded by
[ 1901269 ] PF: compiler requires improved memory management
https://sourceforge.net/tracker/index.php?func=detail&aid=1901269&group_id=56967&atid=482468
## Comment 8963
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1895125 at http://sourceforge.net/support/tracker.php?aid=1895125
| PF: 28 ALG tests still segfault | https://api.github.com/repos/MonetDB/MonetDB/issues/1584/comments | 0 | 2020-11-30T09:47:40Z | 2024-06-27T11:28:34Z | https://github.com/MonetDB/MonetDB/issues/1584 | 753,323,921 | 1,584 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-16 04:48:35 +0100
From: @drstmane
To: Manuel Mayr <<mayrm>>
Version: -- development
Last updated: 2008-02-16 09:18:51 +0100
## Comment 8946
Date: 2008-02-16 16:48:35 +0100
From: @drstmane
170 out of the 334 failing tests reported in
[ 1894891 ] PF: 334 tests fail with ALG that work fine with MPS
http://sourceforge.net/tracker/index.php?func=detail&aid=1894891&group_id=56967&atid=482468
fail with a segfault in atom_eq() in opt_algebra_cse.c, respectively, in case assertions are enabled,
assert (PFalg_atom_comparable (a,b));
fails in atom_eq() in opt_algebra_cse.c .
See attached GDB session log and, e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-Fedora8/tests_WebSite/HelloWorld1.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/GNU.32.32.d-Fedora8/benchmarks_XMark/q20.err.00.html
## Comment 8947
Date: 2008-02-16 17:32:28 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
File Added: ALG_segfault
## Comment 8948
Date: 2008-02-16 18:07:42 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Manuel can you please have a look please.
## Comment 8949
Date: 2008-02-16 21:18:51 +0100
From: Manuel Mayr <<mayrm>>
Logged In: YES
user_id=1629542
Originator: NO
Should working now,
the problem was that the equivalency check occurs even if the atoms (a and b) are not comparable due to
!comparable (a,b) ||Â !atom_eq(a,b)
I transformed it to
!(comparable(a,b) && atom_eq(a,b))
to benefit from short circuit mechanism.
Best regards,
Manuel
## Comment 8950
Date: 2008-02-16 23:14:34 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Hi Manuel,
that for the quick fix.
A simple quick check (`Mtest.py -G -r`) reveals that 109 of the 170 reported tests seem to work fine, now (on my 64-bit Fedora 8 machine); an other 33 do not produce any segfault anymore, but still produce "different" output (didn't check the details, yet, please check the TestWeb as of tomorrow); and the remaining 28 still produce some segfault.
Continued in
[ 1895125 ] PF: 28 ALG tests still segfault
http://sourceforge.net/tracker/index.php?func=detail&aid=1895125&group_id=56967&atid=482468
## Comment 8951
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1894964 at http://sourceforge.net/support/tracker.php?aid=1894964
| PF/ALG: 170 tests segfault in atom_eq() in opt_algebra_cse.c | https://api.github.com/repos/MonetDB/MonetDB/issues/1583/comments | 0 | 2020-11-30T09:47:38Z | 2024-06-27T11:28:34Z | https://github.com/MonetDB/MonetDB/issues/1583 | 753,323,892 | 1,583 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-16 01:43:17 +0100
From: @drstmane
To: XQuery devs <<bugs-xquery>>
Version: -- development
Last updated: 2008-06-02 10:40:01 +0200
## Comment 8936
Date: 2008-02-16 13:43:17 +0100
From: @drstmane
Attached you find a list of 334 Pathfinder / XQuery tests that work fine with the MPS version (on all "stable" platforms), but fail with the ALG version (on all "stable" platforms).
Note:
The list does neither include the tests that are bound to fail with ALG do to not yet available features (i.e., MIL modules, XRpc, PF/Tijah), nor tests that fail only on some platforms (in particular on Windows).
## Comment 8937
Date: 2008-02-16 13:44:29 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
ps:
for details see / compare
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/index_short.html
and
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/index_short.html
## Comment 8938
Date: 2008-02-16 16:49:26 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
see also
[ 1894964 ] PF/ALG: 170 tests segfault in atom_eq() in opt_algebra_cse.c
http://sourceforge.net/tracker/index.php?func=detail&aid=1894964&group_id=56967&atid=482468
## Comment 8939
Date: 2008-02-16 17:34:14 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
19 of the reported 334 tests fail with ALG due to
"Algebra implementation cannot cope with recursive functions yet."
See attached list for details.
File Added: ALG_no_recursion
## Comment 8940
Date: 2008-02-16 18:08:08 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
20 of the reported 334 tests fail with ALG due to
"Algebra implementation for function `*' unknown."
fn:root : 9 tests
pf:nid : 5 tests
fn:put : 2 tests
pf:log : 1 test
pf:pow : 1 test
pf:sqrt : 1 test
pf:collections : 1 tests
See attached list for details.
File Added: ALG_missing_functions
## Comment 8941
Date: 2008-02-17 10:20:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
After Manuel's recent fixes,
and after I disabled (only for the Algebra version, and only in the XQuery_0-22 release branch) the tests that fail with the Algebra version due to not yet available features (MIL modules; XRPC; PF/tijah; recursive functions; built-in functions: fn:root, fn:put, pf:collections, pf:collection, pf:nid, pf:sqrt, pf:pow, pf:log),
294 test remain failing with ALG that work fine with MPS.
## Comment 8942
Date: 2008-02-17 10:21:30 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
See also
[ 1895311 ] PF/alg: "fetch" fails in 8 tests
http://sourceforge.net/tracker/index.php?func=detail&aid=1895311&group_id=56967&atid=482468
## Comment 8943
Date: 2008-02-17 10:37:24 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
See also
[ 1895322 ] PF/alg: 20 tests fail with (various) "fatal error"
http://sourceforge.net/tracker/index.php?func=detail&aid=1895322&group_id=56967&atid=482468
## Comment 8944
Date: 2008-06-02 22:40:00 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Many of the Algebra-related bugs have been fixed recently;
the remaining bugs have been filed in individual/specific bug reports.
## Comment 8945
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1894891 at http://sourceforge.net/support/tracker.php?aid=1894891
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-xquery@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| PF: 334 tests fail with ALG that work fine with MPS | https://api.github.com/repos/MonetDB/MonetDB/issues/1582/comments | 0 | 2020-11-30T09:47:35Z | 2024-06-27T11:28:33Z | https://github.com/MonetDB/MonetDB/issues/1582 | 753,323,862 | 1,582 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-15 03:37:05 +0100
From: @drstmane
To: @sjoerdmullender
Version: -- development
Last updated: 2009-04-24 12:23:41 +0200
## Comment 8928
Date: 2008-02-15 15:37:05 +0100
From: @drstmane
Most (if not all) Patfinder / XQuery tests that currently fail on Windows, but not on other (Unix-like) systems, and have not been mentions in previously filed bug reports, --- there are tens of such cases; see
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/index_short.html
--- fail with
fatal error: aborted too deep recursion
See e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/runtime/escaped-string.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/runtime/escaped-string.out.00.html
## Comment 8929
Date: 2009-02-14 18:27:12 +0100
From: @drstmane
updated list of "fatal error: aborted too deep recursion" error that occur only on Windows:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/modules_pftijah/qmil0.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/runtime/printTypedMode.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/runtime/serialization.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/runtime/escaped-string.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugDay_2005-11-09_0.8/boolean_dm.SF-1204097.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugDay_2005-12-19_0.9.3/self_axis_not_supported.SF-1237444.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/corrupt-attributelist.SF-1483706.shred_doc.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/mposjoin_void_void.SF-1404023.q1a.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/mposjoin_void_void.SF-1404023.q1b.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/treat_as.SF-1586454.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/shredding-remap-error.SF-1487156.shred_doc.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d.1-Windows5.2/tests_BugTracker/insert_multiple.SF-1590580.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/accessing_renamed_inserted_deleted_node.SF-1718622-1718635-1718709.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugTracker/merged-union_error_with_fn-sum.SF-1751684.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_BugsViaSourgeforce/ID.1204965-c.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d.1-Windows5.2/tests_W3C_use_cases_XQUF_Parts/Q3a1.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_Parts/Q4a.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_Parts/Q6.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_Parts/Q6x.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q1.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q3.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q4.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q4x.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q5a.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q6a.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q7x.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d.1-Windows5.1/tests_W3C_use_cases_XQUF_R/Q9.out.00.html
## Comment 8930
Date: 2009-04-07 13:35:41 +0200
From: @peterboncz
maybe windows uses more stack space somehow..
I slightly increased the limit already (in src/pathfinder/main.c) to 7/8*1MB.
For going beyond 1MB, we should check the windows make options. Possibly the easiest way is to use the EDITBIN utility to e.g. specify a thread stack space of 2MB. Then the limit can be increased.
this thus depends on our windows build process.
## Comment 8931
Date: 2009-04-08 12:31:39 +0200
From: @sjoerdmullender
Peter wrote "maybe windows uses more stack space somehow.." This is absolutely true. The Intel compiler on Windows seems to allocate separate space for *all* variables declared in a function, even if the scopes of variables don't overlap. This is especially painful in the many deeply recursive functions in the burg code. I have already split (many of) those functions in a recursive function and one that does the work with all the local variables.
We can try to increase the stack space allocated per thread, but, especially on 32 bit systems, we have to be careful about our memory usage.
I'll give it a try.
## Comment 8932
Date: 2009-04-08 17:02:08 +0200
From: @sjoerdmullender
Using the editbin utility you can set the stack size of the main thread. All other threads are started using pthread_create, right after a call to pthread_attr_setstacksize where the new thread's stack size is set to a value we determine programmatically. So using editbin to change the stack size isn't going to do a whole lot.
However, there is a problem with the way we set the thread stack size.
There is a define for THREAD_STACK_SIZE which boils down to 1024*1024 on a 32 bit system and twice that on a 64 bit machine (why it does this using 2*512*1024 is beyond me ;-) ). The thread stack size is however set using 4*THREAD_STACK_SIZE. Why the value of THREAD_STACK_SIZE wasn't changed is, again, beyond me.
There is also code to check whether we're using too much stack space. This code differs for the MIL and MAL interpreters and SQL on the one hand and pathfinder on the other. Neither version was consistent with how much stack space was *actually* allocated.
I normalized the code, and I now don't get the "too deep recursion" error on 32 bit WIndows on the tests that I tried (that did give the error in the testweb).
We'll see how the testweb behaves tonight.
## Comment 8933
Date: 2009-04-17 15:47:13 +0200
From: @sjoerdmullender
I noticed that there were a bunch of pathfinder tests that failed on 64 bit Windows with too deep recursion that didn't fail on 32 bit Windows or other systems.
I localized the problem and will check in a (hacky) fix.
This particular problem will only occur with pathfinder tests in MIL scripts that are fed directly to Mserver (i.e. not through mclient).
## Comment 8934
Date: 2009-04-24 00:23:41 +0200
From: @drstmane
Sjoerd's (hacky) fix appears to work fine.
Thanks.
## Comment 8935
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1894477 at http://sourceforge.net/support/tracker.php?aid=1894477
| PF: "fatal error: aborted too deep recursion" on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1581/comments | 0 | 2020-11-30T09:47:32Z | 2024-06-27T11:28:32Z | https://github.com/MonetDB/MonetDB/issues/1581 | 753,323,829 | 1,581 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-15 03:31:21 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-16 10:35:23 +0100
## Comment 8924
Date: 2008-02-15 15:31:21 +0100
From: @drstmane
See e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/compiler_semantics_tests_ns/bad-006.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/compiler_semantics_tests_ns/bad-006.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/compiler_semantics_tests_recursion/recursion4.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/compiler_semantics_tests_recursion/recursion4.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/compiler_semantics_tests_recursion/recursion5.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/compiler_semantics_tests_recursion/recursion5.err.00.html
## Comment 8925
Date: 2008-02-15 17:49:07 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Hopefully fixed by propagating applying the changes that tsheyar did in the Unix shell scrip versions of the tests on 2007/07/26 in his check-in
"
-- Fix http://sourceforge.net/support/tracker.php?aid=1760467 by letting the compiler run a little bit longer.
(The changed behavior was triggered by introducing additonal
compiler phases.)
"
also to the Windows batch script versions of the tests.
## Comment 8926
Date: 2008-02-16 10:35:23 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Testing confirmed fix.
Closing.
## Comment 8927
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1894473 at http://sourceforge.net/support/tracker.php?aid=1894473
| PF: compiler tests do not produce expected errors on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1580/comments | 0 | 2020-11-30T09:47:30Z | 2024-06-27T11:28:31Z | https://github.com/MonetDB/MonetDB/issues/1580 | 753,323,789 | 1,580 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-14 11:05:00 +0100
From: @drstmane
To: @yzchang
Version: -- development
Last updated: 2009-11-27 09:00:06 +0100
## Comment 8914
Date: 2008-02-14 11:05:00 +0100
From: @drstmane
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/tests_XRpc/xrpcConvert.err.00.html
========
MAPI = monetdb@TURELUUR:32113
QUERY = import module namespace xrpc = "xrpc-fun" at
ERROR = !ERROR: Document is empty
!ERROR: during RPC call to "TURELUUR:49525"
!ERROR: shredder_parse: XML input not well-formed.
!ERROR: response2bat: invalid XRPC response received
!ERROR: CMDhttp_post: operation failed.
MAPI = monetdb@TURELUUR:32113
QUERY = import module namespace xrpc = "xrpc-fun" at
ERROR = !ERROR: doLoopLiftedRPC: execution failed at all destinations
========
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html
========
MAPI = monetdb@TURELUUR:32113
QUERY = import module namespace xrpc = "bar" at
ERROR = !parse error: syntax error, unexpected IntegerLiteral, expecting ) on line 7, column 36 (next token is `4')
!parse error: XQuery parsing failed
!parse error: syntax error, unexpected IntegerLiteral, expecting ) on line 7, column 36 (next token is `4')
!parse error: error parsing module from C:\\cygwin\\var\\tmp\\monet.Int.32.32.d.27057\\pathfinder\\NT\\mTests\\tests\\XRpc/xrpcConvert2.xrpcfun2.xq
========
## Comment 8915
Date: 2008-02-15 23:26:06 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
I took a look at the errors, but could not get a clue without debugging.
How urgent is this problem for the release? If urgent, I will ask Sjoerd for some debugging help next week, otherwise, I would like to lower the priority to solve it later.
Jennie
## Comment 8916
Date: 2008-02-16 01:44:52 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Jennie,
since the release is / will be based on the code base of 2008-02-13 (as tagged in CVS), there is no urgency to fix this before the release.
Stefan
## Comment 8917
Date: 2008-02-16 22:16:53 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Lower the priority, as I don't know how to debug on windows.
I will ask our windows expert for some help.
## Comment 8918
Date: 2008-05-29 15:43:25 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
With the code base for the upcoming Jun2008 release (XQuery_0-24 branch), the error for the first xrpcConvert test no seems to indicate some path matching / '\' quoting problem when checking for "allowed" ("trusted") module paths/locations:
========
MAPI = monetdb@TURELUUR:37285
QUERY = import module namespace xf = "xrpc-fun" at "C:\\cygwin\\var\\tmp\\monet.Int.32.32.d.20696\\pathfinder\\tests\\XRpc\\Tests/xrpcfun.xq";
ERROR = !ERROR: during RPC call to "TURELUUR:45558"
!ERROR: response2bat: invalid response from TURELUUR45558
!ERROR: !ERROR: during RPC call to "TURELUUR:45558
!ERROR: !ERROR: HTTP Error Code : 403 Forbidde
!ERROR: !ERROR: SOAP Fault Code : env:Sende
!ERROR: !ERROR: SOAP Fault Reason
!ERROR: !ERROR: Called module not allowed
!ERROR: !ERROR: 'C:cygwinvar mpmonet.Int.32.32.d.20696pathfinder estsXRpcTests/xrpcfun.xq' not i
!ERROR: !ERROR: 'http://monetdb.cwi.nl'
!ERROR: !ERROR: 'C:\cygwin\var\tmp\monet.Int.32.32.d.20696\MonetDB4\.Int.32.32.d-Windows5.1\share\MonetDB\xrpc\export'
!ERROR: !ERROR: 'http://127.0.0.1:45558'
!ERROR: !ERROR: 'C:\cygwin\var\tmp\monet.Int.32.32.d.20696\pathfinder\tests\XRpc\Tests'
!ERROR: !ERROR: 'C:\cygwin\var\tmp\monet.Int.32.32.d.20696\pathfinder\NT\mTests\tests\XRpc'
!ERROR: !ERROR: CMDhttp_post: operation failed
!ERROR: CMDhttp_post: operation failed.
MAPI = monetdb@TURELUUR:37285
QUERY = import module namespace xf = "xrpc-fun" at "C:\\cygwin\\var\\tmp\\monet.Int.32.32.d.20696\\pathfinder\\tests\\XRpc\\Tests/xrpcfun.xq";
ERROR = !ERROR: doLoopLiftedRPC: execution failed at all destinations
========
For xprcConvert2, the problem is still the same as initially reported.
## Comment 8919
Date: 2008-09-05 18:35:36 +0200
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Some status info about this bug:
The path matching problem has been fixed in milprint_summer.c, version 1.419.2.5, thanks to Sjoerd's debugger on windows.
Sjoerd and I had a look at the problem with xrpcConvert2, and unfortunetely, we could not reproduce the error with his version compiled with debugging on.
## Comment 8920
Date: 2009-02-14 18:31:23 +0100
From: @drstmane
(only) the xrpcConvert2 test still fails on Windows:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.32.32.d.1-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.64.32.d.1-Windows5.2/tests_XRpc/xrpcConvert2.err.00.html
## Comment 8921
Date: 2009-11-07 14:29:06 +0100
From: @drstmane
See also
ID: 2893815 "PF: xrpcConvert tests segfault since 2009/10/08"
https://sourceforge.net/tracker/?func=detail&aid=2893815&group_id=56967&atid=482468
## Comment 8922
Date: 2009-11-27 09:00:06 +0100
From: @drstmane
Test appaers to work fine also on Windows (at least) since 2009.11.16; cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.32.32.d.1-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.64.64.d.1-Windows6.0/tests_XRpc/xrpcConvert2.err.00.html
Closing.
## Comment 8923
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1893448 at http://sourceforge.net/support/tracker.php?aid=1893448
| PF: xrpcConvert2 fails on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1579/comments | 0 | 2020-11-30T09:47:27Z | 2024-06-27T11:28:30Z | https://github.com/MonetDB/MonetDB/issues/1579 | 753,323,754 | 1,579 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-13 02:29:22 +0100
From: @arjenderijke
To: @njnes
Version: -- development
Last updated: 2008-02-16 01:32:10 +0100
## Comment 8911
Date: 2008-02-13 14:29:22 +0100
From: @arjenderijke
Using the tarball from http://monetdb.cwi.nl/testing/projects/monetdb/Stable/.DailyBuilds./Feb2008/
The following query
CREATE FUNCTION CHECK_VIDEO()
RETURNS VARCHAR(24)
BEGIN
RETURN
SELECT BESTANDSNAAM
FROM VIDEO_PRODUKTIE
WHERE IMPALA_GEREED IS NULL
LIMIT 1;
END;
returns an error:
!syntax error, unexpected LIMIT, expecting SCOLON in: "create function check_video()
This is in my opinion correct sql and it worked in earlier versions.
It was tested on and F8 and FC6 machine.
## Comment 8912
Date: 2008-02-16 13:32:10 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
The parser had a small bug. Fixed now. Also added a test to src/tests/BugTracker/Tests/orderby_in_function.SF-1892708.sql
## Comment 8913
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1892708 at http://sourceforge.net/support/tracker.php?aid=1892708
| error in create function | https://api.github.com/repos/MonetDB/MonetDB/issues/1578/comments | 0 | 2020-11-30T09:47:25Z | 2024-06-27T11:28:29Z | https://github.com/MonetDB/MonetDB/issues/1578 | 753,323,722 | 1,578 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-13 11:21:37 +0100
From: @drstmane
To: @yzchang
Version: -- development
Last updated: 2008-02-14 10:01:59 +0100
## Comment 8903
Date: 2008-02-13 11:21:37 +0100
From: @drstmane
On Windows, all XRpc tests fail with
QUERY = import module namespace xrpc = "xrpc-fun" at "C:\\cygwin\\var\\tmp\\monet.Int.64.32.d.30065\\pathfinder\\tests\\XRpc\\Tests/xrpcfun.xq";
ERROR = !ERROR: during RPC call to "PARKIET:42437"
!ERROR: HTTP Error Code : 403 Forbidden
!ERROR: SOAP Fault Code : env:Sender
!ERROR: SOAP Fault Reason:
!ERROR: Called module not allowed:
!ERROR: 'C:\cygwin\var\tmp\monet.Int.64.32.d.30065\pathfinder\tests\XRpc\Tests/xrpcfun.xq' not in
!ERROR: 'http://monetdb.cwi.nl',
!ERROR: 'C:\cygwin\var\tmp\monet.Int.64.32.d.30065\MonetDB4\.Int.64.32.d-Windows5.2\share\MonetDB\xrpc\export',
!ERROR: 'http://127.0.0.1:42437',
!ERROR: 'C:\cygwin\var\tmp\monet.Int.64.32.d.30065\pathfinder/tests/XRpc/Tests',
!ERROR: 'C:\cygwin\var\tmp\monet.Int.64.32.d.30065\pathfinder\NT\mTests\tests\XRpc'.
!ERROR: CMDhttp_post: operation failed.
MAPI = monetdb@PARKIET:32550
QUERY = import module namespace xrpc = "xrpc-fun" at "C:\\cygwin\\var\\tmp\\monet.Int.64.32.d.30065\\pathfinder\\tests\\XRpc\\Tests/xrpcfun.xq";
ERROR = !ERROR: doLoopLiftedRPC: execution failed at all destinations
See
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/tests_XRpc/add_dbl-localhost.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/tests_XRpc/add_dbl.err.00.html
etc.
## Comment 8904
Date: 2008-02-13 13:24:59 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
The problem is caused by the prefix 'C:\cygwin\var\tmp\monet.Int.64.32.d.30065\pathfinder/tests/XRpc/Tests', where the '/' should be '\'.
This path has been added by the file 'add_trusted.mil' in the XRPC test directory, using MIL commands:
var TSTSRCDIR := getenv("TSTSRCBASE") + "/tests/XRpc/Tests";
add_xrpc_trusted(TSTSRCDIR);
which is wrong on Windows.
Does someone know how can I calculate the correct file separator (or check for OS) in my MIL script?
If this is not possible, I would suggest to use the simplest solution, nl. just add TSTSRCBASE (in this case 'C:\cygwin\var\tmp\monet.Int.64.32.d.30065\pathfinder') to xrpc_trusted.
Jennie
## Comment 8905
Date: 2008-02-13 13:51:36 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Maybe, we should consider letting Mtest set and export a system-specific environment variable DIRSEP that shoudl then be used in paths in such tests instead of a literal "/" ...
I'll consider that --- but it will in any case most probably be a post-release fix --- unless we want to postpone the release even longer ...
## Comment 8906
Date: 2008-02-13 14:07:05 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
I don't know how much work it is to add this DIRSEP variable...
I might temporarily just use TSTSRCBASE so that XRPC can be tested on Windows as well.
I think there is no need to postpone the release for this issue.
## Comment 8907
Date: 2008-02-13 15:09:28 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Oh, maybe replacing
var TSTSRCDIR := getenv("TSTSRCBASE") + "/tests/XRpc/Tests";
by
var TSTSRCDIR := getenv("TSTSRCDIR");
will already help ...
## Comment 8908
Date: 2008-02-13 15:52:10 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Fixed in CVS:
[ monetdb-Bugs-1892582 ] PF: XRpc does not work on Windows
Don't use hard coded directory name, '/' as directory separator will not
work on Windows.
Let's see tomorrow if this bug can be closed.
## Comment 8909
Date: 2008-02-14 10:01:59 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
The fix seems to work well.
Thanks.
## Comment 8910
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1892582 at http://sourceforge.net/support/tracker.php?aid=1892582
| PF: XRpc does not work on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1577/comments | 0 | 2020-11-30T09:47:20Z | 2024-06-27T11:28:28Z | https://github.com/MonetDB/MonetDB/issues/1577 | 753,323,664 | 1,577 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-13 10:57:23 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-25 11:54:05 +0100
## Comment 8894
Date: 2008-02-13 10:57:23 +0100
From: @drstmane
When compiling larger queries, such as e.g, the XMark module, pf crashes with
fatal error: insufficient memory (allocating 2048 bytes failed) in .\..\..\..\compiler\mem\array.c (PFarray), line 96
on Windows and Solaris/x86 (on machines that have 1 GB RAM ...)
See e.g.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_XMark_mod_XQ/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_mod_XQ/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_opt_XQ/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_XMark_opt_XQ/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_XMark_mod_MIL/xq2mil.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_mod_MIL/xq2mil.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_XMark_opt_MIL/xq2mil.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_opt_MIL/xq2mil.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_XMark_XRpc/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_XRpc/q01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMach-1/q1.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_XMach-1/q1.err.00.html
## Comment 8895
Date: 2008-02-13 11:53:26 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Just for info:
On Solaris/x86, compiling the XMark module worked fine until October 20 2007.
The "normal" (non-"optimized") version has been failing since October 21 2007:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_mod_MIL/xq2mil.err.00.html
The "optimized" version has been failing since November 16 2007:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d-SunOS5.10/benchmarks_XMark_opt_MIL/xq2mil.err.00.html
On Windows, both versions have been failing since nightly Windows testing has been started mid December 2007.
## Comment 8896
Date: 2008-02-13 13:04:10 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
I can only agree with Stefan -- In the testing I get a time out on my machine and a very high memory usage.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
9635 rittinge 15 0 1005m 662m 90m S 2.3 65.4 0:03.62 Mserver
If I use stand-alone Pathfinder everything works fine and consumes only 50MB of memory (thanks to garbage collection).
----------
After making the point about the garbage collection here comes the proposal to fix this. We use dynamic arrays for almost all generic structures. And because we do not want to resize (which maps to a new malloc+copy in the embedded variant) very often we set the initial size to 2KB. Of course a lot of these generic structures are used which leads to the reported memory problem. I think for the release the initial size can be decreased as in many cases we only store very few data in an array.
This change however only postpones the problem (of allocating memory only without freeing it) and is not necessary for the stand-alone variant. In the future the embedded variant thus needs a better and cleaner solution.
Reassigning it to the release people.
Index: mem/array.c
===================================================================
RCS file: /cvsroot/monetdb/pathfinder/compiler/mem/array.c,v
retrieving revision 1.24
diff -r1.24 array.c
76c76
< define ACHUNK 2048U
---
> define ACHUNK 256U
## Comment 8897
Date: 2008-02-13 13:54:35 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Indeed, lowering the default allocation size is just a HACK to cure symptoms, not to solve the problem.
We might consider doing it, anyway, as a post-release fix just to keep/make testing work reasonable (again).
It will not be in the initial release (which should be base on last night code base) --- unless we want to postpone the release even more ...
## Comment 8898
Date: 2008-02-13 22:00:04 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
In the XQuery_0-22 branch, only (propagation prevented):
as suggested by Jan:
reduce the default (inital) size of (dynamic) arrays
from 2048 byte to 256 byte
[post release-tag check-in (approved by Stefan ;-))]
## Comment 8899
Date: 2008-02-14 10:05:39 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
The HACK seems to work fine for testing.
I'm hesitant, though, to mark this bug-report as "Fixed", let alone close it, since we actually need a proper fix for the memory management in PF ...
(See also
[ 1889487 ] PF: Algebra compiler uses a lot of memory
http://sourceforge.net/tracker/index.php?func=detail&aid=1889487&group_id=56967&atid=482468
)
## Comment 8900
Date: 2008-02-22 12:08:10 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
With the "release HACK" in place, my work on this one is done.
A proper memory management solution for Pathfinder is still to be found (by someone else) ...
## Comment 8901
Date: 2008-02-25 11:54:05 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Temporarily fixed in XQuery_0-22 branch.
Superseded by
[ 1901269 ] PF: compiler requires improved memory management
https://sourceforge.net/tracker/index.php?func=detail&aid=1901269&group_id=56967&atid=482468
## Comment 8902
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1892560 at http://sourceforge.net/support/tracker.php?aid=1892560
| PF: "insufficient memory" while compiling XMark module | https://api.github.com/repos/MonetDB/MonetDB/issues/1576/comments | 0 | 2020-11-30T09:47:17Z | 2024-06-27T11:28:27Z | https://github.com/MonetDB/MonetDB/issues/1576 | 753,323,622 | 1,576 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-13 10:49:40 +0100
From: @drstmane
To: @sjoerdmullender
Version: -- development
Last updated: 2008-02-22 11:44:01 +0100
## Comment 8890
Date: 2008-02-13 10:49:40 +0100
From: @drstmane
ERROR = !ERROR: "proc_vid.insert("fnB3D04632_AssetByTrackNr_xs_string1_xs_integer1_xs_integer1", 1394161003LL);"
! ^
! can't help you here, sorry.
!module import: error loading module from C:\\cygwin\\var\\tmp\\monet.Int.64.32.d.30065\\pathfinder\\NT\\mTests\\benchmarks\\Music\\mod\\MIL/music.mil
See
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_Music_mod_MIL/full_01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_Music_opt_MIL/full_01.err.00.html
## Comment 8891
Date: 2008-02-16 11:56:35 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Some extra info:
The MIL module that fails to load is generated from XQuery module using `pf`, e.g., in this test
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/benchmarks_Music_mod_MIL/xq2mil.out.00.html
The generated MIL module that fails to load is available at
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.Int.64.32.d-Windows5.2/mTests10/benchmarks/Music/mod/MIL/music.mil
## Comment 8892
Date: 2008-02-22 11:44:01 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
There seems to a difference between Windows and Linux in the meaning of the second argument of snprintf and vsnprintf: in Linux this argument gives the size of the buffer to which the function writes, whereas in Windows it gives the number of non-NULL bytes that it can write (and then it adds a NULL). This means that on Windows you have to specify a size which is *smaller* than the actual size of the buffer.
To make the same code work on both flavors, in some cases we now overallocate the needed space by one byte so that we can use (sizeof(buf)-1) as size.
No additional test needed.
## Comment 8893
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1892554 at http://sourceforge.net/support/tracker.php?aid=1892554
| PF: loading pre-compiled MIL modules fails on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1575/comments | 0 | 2020-11-30T09:47:15Z | 2024-06-27T11:28:26Z | https://github.com/MonetDB/MonetDB/issues/1575 | 753,323,594 | 1,575 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-13 06:04:23 +0100
From: Michael Yang <<msyang>>
To: @njnes
Version: -- development
Last updated: 2008-03-14 02:29:08 +0100
## Comment 8886
Date: 2008-02-13 06:04:23 +0100
From: Michael Yang <<msyang>>
Using MonetDB/SQL v5 on Linux.
I'm getting simply wrong results or a !MALException:algebra.thetajoin:GDKerror when I try to run a subquery involving a self-join. Here's a small example:
sql>CREATE TABLE test(date int not null constraint pk_test primary key);
sql>INSERT into test values (19251231);
sql>INSERT into test values (19260102);
sql>INSERT into test values (19260104);
sql>select * from test;
+----------+
| date |
+==========+
| 19251231 |
| 19260102 |
| 19260104 |
+----------+
Timer 0.930 msec 3 rows
sql>SELECT date, (SELECT date from test where date > t1.date limit 1) as dtNext from test t1;
+----------+---------+
| date |dtnext |
+==========+=========+
| 19251231 |19260102 |
+----------+---------+
OK - not what I expected - since limit 1 is in the subquery, not the outer query. I was expecting that the query would at least return the next date as dtnext for each date in 'test'. I realize you can't specify an order by clause in the inner subquery - you should be able to.
now, without the limit clause:
sql>SELECT date, (SELECT date from test where date > t1.date) as dtNext from test
t1;
+----------+---------+
| date |dtnext |
+==========+=========+
| 19251231 |19260102 |
| 19260102 |19260104 |
+----------+---------+
Timer 1.673 msec 3 rows
OK, that's a little better. So I guess, unlike MS SQL Server where you must explicitly specify TOP(1) in a subquery to limit the subquery to returning only 1 value, monetDB/SQL implicitly does this, and orders by the default primary key ordering. Fine. However, when I try running this on a table with about 32605 dates, I get a GDKerror [the attached file of dates is an example].
sql>DELETE FROM test;
sql>COPY INTO test FROM '/home/michael/tmp/dt.txt' USING DELIMITERS '\t';
sql>select count(*) as cnt from test;
+-------+
| cnt |
+=======+
| 32605 |
+-------+
Timer 0.807 msec 1 rows
sql>select date, (select date from test where date > t1.date) as dtNext from test t1;
!MALException:algebra.thetajoin:GDKerror
Timer 1.142 msec 0 rows
## Comment 8887
Date: 2008-02-13 06:12:52 +0100
From: Michael Yang <<msyang>>
Logged In: YES
user_id=1952137
Originator: YES
furthermore, a simple self-join with GROUP BY fails as well
sql>select t1.date, min(t2.date) as dtNext from test t1, test t2 where t2.date > t1.date group by t1.date;
!MALException:algebra.thetajoin:GDKerror
Timer 1.141 msec 0 rows
## Comment 8888
Date: 2008-02-20 10:40:09 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
This bug report is a combination of problems.
1 we don't support limit in sub queries.
2 sub queries should return zero or one result. This wasn't checked correctly at runtime.
3 these queries need an outer join as some sub query results maybe null
4 there is a problem with the theta join, it claims to much memory (which on memory tight systems may well fail)
So far I fixed 2 and 3. 1 is at best a feature request, work around it will a aggregate (such as max and min) or use a row_number() over() solution. Fixing 4 is on the list (at least for this data it should behave much better), but the proper fix would be to use a windowing like solution which we still need to implement.
For 2 and 3 a test was added (src/tests/BugTracker/Tests/Attic/bug_in_selection.SF-1892413)
## Comment 8889
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1892413 at http://sourceforge.net/support/tracker.php?aid=1892413
| wrong or unexpected results in subquery | https://api.github.com/repos/MonetDB/MonetDB/issues/1574/comments | 0 | 2020-11-30T09:47:12Z | 2024-06-27T11:28:26Z | https://github.com/MonetDB/MonetDB/issues/1574 | 753,323,559 | 1,574 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-11 11:44:58 +0100
From: @drstmane
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2008-05-29 05:17:45 +0200
## Comment 8877
Date: 2008-02-11 23:44:58 +0100
From: @drstmane
After disabling vmalloc() in GDKmallocmax() tests MonetDB4/src/modules/plain/Tests/bat.py & MonetDB4/src/modules/plain/Tests/bat2.py segfault 64-bit Fedora (6 & 8) with 64-bit OIDs and dynamically-linked Mserver:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB4/.mTests103/GNU.64.64.d-Fedora6/src_modules_plain/bat.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB4/.mTests103/GNU.64.64.d-Fedora6/src_modules_plain/bat2.out.00.html
... investigating ...
## Comment 8878
Date: 2008-02-11 23:46:26 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
these two tests start Mserver with "--set gdk_mem_pagebi1970-01-01 01:00:16 +0100"
## Comment 8879
Date: 2008-02-11 23:54:43 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
... segfault does only occur when running with property checking enabled (default with Mtest.py), but not with property checking disabled (`Mtest.py -d0`) ...
## Comment 8880
Date: 2008-02-12 00:03:12 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
... segfault does only occur when the test script is run via input redirect (`Mserver --debug=10 --set gdk_mem_pagebi1970-01-01 01:00:16 +0100 < bat.milS`),
but not when the test script is given on the Mserver's commandline (`Mserver --debug=10 --set gdk_mem_pagebi1970-01-01 01:00:16 +0100 bat.milS`) ...
## Comment 8881
Date: 2008-02-12 00:12:07 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
the crash occurs with
var b3 := file_str.remap(map,false);
in strPut() in gdk_atoms.mx; it seems "dst" points to nirvana:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 46912536073280 (LWP 20781)]
0x00002aaaac543093 in strPut (h=0x69b468, dst=0x800000000107cf67, v=0x2aaaac854e00 "\200") at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB/src/gdk/gdk_atoms.mx:2070
2070 memcpy(h->base + (*dst = h->free), v, i);
(gdb) bt
0 0x00002aaaac543093 in strPut (h=0x69b468, dst=0x800000000107cf67, v=0x2aaaac854e00 "\200") at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB/src/gdk/gdk_atoms.mx:2070
1 0x00002aaaaaf867c9 in copy_segment (hdl=0x0, bn=0x72dff8, b=0x7be608, src=2, dst=2, cnt=1) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/modules/plain/bat.mx:1905
2 0x00002aaaaaf87d91 in CMDremap (res=0x7fff185ef040, b=0x7be608, map=0x794b18, force=0x7fff185ef070 "") at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/modules/plain/bat.mx:2048
3 0x00002aaaaaf7f9a8 in CMDremap_unpack1981300575 (argc=4, argv=0x7fff185ef040) at bat.glue.c:3048
4 0x00002aaaabbffd60 in interpret (stk=0, lt=0x7b0ff8, res=0x7fff185f0300) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1133
5 0x00002aaaabc0ae7e in interpret_assignment (stk=0, lt=0x7b0fa8, res=0x7fff185f0300) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1820
6 0x00002aaaabc02cdb in interpret_var (stk=0, lt=0x7b0f80, res=0x7fff185f0300) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1288
7 0x00002aaaabbfe3ed in interpret (stk=0, lt=0x7b0f80, res=0x7fff185f0300) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:791
8 0x00002aaaabc0b17d in interpret_seqblock (stk=0, lt=0x7b0f58, res=0x7fff185f0300, scope=0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:1870
9 0x00002aaaabbfd99b in interpret (stk=0, lt=0x7b0f58, res=0x7fff185f0300) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_interpreter.mx:729
10 0x00002aaaabc2e98c in handleRequest (t=0x2aaaacc9cee0, q=0x7952d8, res=0x7fff185f0300) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_queue.mx:537
11 0x00002aaaabc2ee89 in doRequest (t=0x2aaaacc9cee0, preference=0x0) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_queue.mx:563
12 0x00002aaaabc84475 in monetInterpreter (status=0x7fff185f0404) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/monet/monet_process.mx:111
13 0x0000000000402064 in main (argc=6, av=0x7fff185f0528) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/tools/Mserver.mx:328
(gdb) li
2065 }
2066
2067 if (!GDK_ELIMDOUBLES(h)) {
2068 if (GDK_STRHASHCREDIT(h) == 0) {
2069 /* if credits are gone, we do not hash insert at all */
2070 memcpy(h->base + (*dst = h->free), v, i);
2071 h->free += i; /* in this case, we do not round to var_t either */
2072 return *dst;
2073 }
2074 GDK_STRHASHCREDIT(h)--;
(gdb) print h
$1 = (Heap *) 0x69b468
(gdb) print *h
$2 = {maxsize = 1241896, free = 1034908, size = 1241889, base = 0x107cf68 "", filename = 0x649948 "66.theap", storage = 0 '\0', copied = 0 '\0', newstorage = 0 '\0', dirty = 0 '\0'}
(gdb) print *h->base
$3 = 0 '\0'
(gdb) print dst
$4 = (var_t *) 0x800000000107cf67
(gdb) print *dst
Cannot access memory at address 0x800000000107cf67
(gdb) print v
$5 = 0x2aaaac854e00 "\200"
(gdb) print i
$6 = 2
(gdb)
## Comment 8882
Date: 2008-02-12 00:31:36 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
[continuing the below debugging session]
(gdb) up
1 0x00002aaaaaf867c9 in copy_segment (hdl=0x0, bn=0x72dff8, b=0x7be608, src=2, dst=2, cnt=1) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB4/src/modules/plain/bat.mx:1905
1905 ATOMreplace(bn->ttype, bn->theap, BUNtail(bni,cur), nil);
(gdb) li
1900 ptr nil = ATOMnilptr(bn->ttype);
1901 BUN cur = BUNfirst(bn) + ((dst_off + partial)>>bn->T->shift);
1902 BUN end = BUNlast(bn);
1903
1904 for (; cur < end; cur++) {
1905 ATOMreplace(bn->ttype, bn->theap, BUNtail(bni,cur), nil);
1906 }
1907 }
1908 }
1909 ret = 1;
(gdb) print BUNtail(bni,cur)
No symbol "BUNtail" in current context.
(gdb) print bni.b->T->vheap->base+*(var_t*)(bni.b->T->heap.base+(cur<<bni.b->T->shift))
$29 = 0x800000000107cf67 <Address 0x800000000107cf67 out of bounds>
(gdb) print bni.b->T->vheap->base
$30 = 0x107cf68 ""
(gdb) print bni.b->T->heap.base
$31 = 0x11ac2a8 "�F\a"
(gdb) print *(var_t*)bni.b->T->heap.base
$33 = 476892
(gdb) print bni.b->T->shift
$32 = 3 '\003'
... hence, the tail's str-heap seems to be broken ??? ...
## Comment 8883
Date: 2008-05-29 17:09:40 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Tests seem to work fine, now:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB4/.mTests103/GNU.64.64.d-Fedora8/src_modules_plain/bat.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB4/.mTests103/GNU.64.64.d-Fedora8/src_modules_plain/bat.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB4/.mTests103/GNU.64.64.d-Fedora8/src_modules_plain/bat2.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB4/.mTests103/GNU.64.64.d-Fedora8/src_modules_plain/bat2.err.00.html
Closing.
## Comment 8884
Date: 2008-05-29 17:17:44 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
*Closing*.
## Comment 8885
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1891537 at http://sourceforge.net/support/tracker.php?aid=1891537
| disabling vmalloc() in GDKmallocmax() makes tests segfault | https://api.github.com/repos/MonetDB/MonetDB/issues/1573/comments | 0 | 2020-11-30T09:47:08Z | 2024-06-28T13:30:44Z | https://github.com/MonetDB/MonetDB/issues/1573 | 753,323,513 | 1,573 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-10 05:42:20 +0100
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2008-02-15 05:01:21 +0100
## Comment 8874
Date: 2008-02-10 17:42:20 +0100
From: @drstmane
In test sql/src/test/sql_xml/Tests/xml.sql, query
select
xmlelement(name "Customer",
xmlattributes(c.CustId as "id"),
xmlcomment('simple comment test'))
from Customers c;
produces
[ <Customer id="1"><!-- simple comment test -</Customer> ]
[ <Customer id="2"><!-- simple comment test -</Customer> ]
[ <Customer id="3"><!-- simple comment test -</Customer> ]
[ <Customer id="4"><!-- simple comment test -</Customer> ]
[ <Customer id="5"><!-- simple comment test -</Customer> ]
which is approved as correct in sql/src/test/sql_xml/Tests/xml.stable.out .
However, IMHO, this is not valid XML, and hence incorrect.
The XML comment end tags must not be "-", but rather "-->".
Hence, the correct result should IMHO be
[ <Customer id="1"><!-- simple comment test --></Customer> ]
[ <Customer id="2"><!-- simple comment test --></Customer> ]
[ <Customer id="3"><!-- simple comment test --></Customer> ]
[ <Customer id="4"><!-- simple comment test --></Customer> ]
[ <Customer id="5"><!-- simple comment test --></Customer> ]
## Comment 8875
Date: 2008-02-15 17:01:20 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Has been fixed by Niels.
Thanks.
## Comment 8876
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1890672 at http://sourceforge.net/support/tracker.php?aid=1890672
| SQL/XML: implementation of and test for xmlcomment() broken? | https://api.github.com/repos/MonetDB/MonetDB/issues/1572/comments | 0 | 2020-11-30T09:47:06Z | 2024-06-27T11:28:24Z | https://github.com/MonetDB/MonetDB/issues/1572 | 753,323,470 | 1,572 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-10 05:29:28 +0100
From: @drstmane
To: @mlkersten
Version: -- development
Last updated: 2008-02-11 10:30:35 +0100
## Comment 8870
Date: 2008-02-10 17:29:28 +0100
From: @drstmane
For details, see
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/MonetDB5/.mTests103/index_short.html
## Comment 8871
Date: 2008-02-10 19:08:44 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
the errors occurred with after enabling the property check in the interpreter.
A patch has been checked in and this error can be close after the next
test run.
Only remote00 remains an issue.
## Comment 8872
Date: 2008-02-11 10:30:35 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
fixed by Martin confirmed by testing.
## Comment 8873
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1890667 at http://sourceforge.net/support/tracker.php?aid=1890667
| M5: various tests fail after recent changes | https://api.github.com/repos/MonetDB/MonetDB/issues/1571/comments | 0 | 2020-11-30T09:47:03Z | 2024-06-27T11:28:23Z | https://github.com/MonetDB/MonetDB/issues/1571 | 753,323,442 | 1,571 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-09 01:54:38 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-10 11:47:23 +0100
## Comment 8866
Date: 2008-02-09 13:54:38 +0100
From: @drstmane
After yesterday's (Fri Feb 08 2008) changes,
two XQuery tests now fail with
ERROR = !WARNING: BATpropcheck: BAT ... was incorrectly marked sorted!
cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/swizzle-bug.SF-1760811.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_XQuery/orderby1.err.00.html
## Comment 8867
Date: 2008-02-10 21:54:51 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
investigating...
## Comment 8868
Date: 2008-02-10 23:47:23 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
fixed in CVS (MonetDB_4-22 branch of MonetDB4):
fixed sortedness propagation for "const [op] BAT" & "BAT [op] const",
in particular three cases:
- substracting a BAT from a constant
yields a result with the input'\''s sort order reversed
- dividing a BAT by a negative constant
yields a result with the input'\''s sort order reversed
- for modulo (%) (also on) float & double
we cannot predict any result order
## Comment 8869
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1890131 at http://sourceforge.net/support/tracker.php?aid=1890131
| PF: property error: incorrectly marked sorted! | https://api.github.com/repos/MonetDB/MonetDB/issues/1570/comments | 0 | 2020-11-30T09:47:01Z | 2024-06-27T11:28:22Z | https://github.com/MonetDB/MonetDB/issues/1570 | 753,323,421 | 1,570 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-08 05:05:45 +0100
From: Maurice van Keulen <<mvankeulen>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-06-03 05:54:06 +0200
## Comment 8860
Date: 2008-02-08 17:05:45 +0100
From: Maurice van Keulen <<mvankeulen>>
When trying to compile the attached query with the algebra backend, I get an error:
fatal error: attribute `iter1' referenced in join not found
It compiles and runs fine on the mps-backend. Although not needed for solving the compilation bug, the associated documents can be obtained with
(
pf:add-doc("http://library.cs.utwente.nl/xquery/nvdcve/nvdcve-2002.xml","nvdcve-2002.xml","nvdcve"),
pf:add-doc("http://library.cs.utwente.nl/xquery/nvdcve/nvdcve-2003.xml","nvdcve-2003.xml","nvdcve"),
pf:add-doc("http://library.cs.utwente.nl/xquery/nvdcve/nvdcve-2004.xml","nvdcve-2004.xml","nvdcve"),
pf:add-doc("http://library.cs.utwente.nl/xquery/nvdcve/nvdcve-2005.xml","nvdcve-2005.xml","nvdcve"),
pf:add-doc("http://library.cs.utwente.nl/xquery/nvdcve/nvdcve-2006.xml","nvdcve-2006.xml","nvdcve"),
pf:add-doc("http://library.cs.utwente.nl/xquery/nvdcve/nvdcve-2007.xml","nvdcve-2007.xml","nvdcve")
)
## Comment 8861
Date: 2008-05-28 14:29:13 +0200
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
This compiles fine with the current version.
## Comment 8862
Date: 2008-05-28 16:17:08 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
re-opened as we should consider adding a test (also) for this one ...
## Comment 8863
Date: 2008-06-03 09:23:44 +0200
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Maurice could you please add the test. Thanks!
## Comment 8864
Date: 2008-06-03 17:54:05 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
[I was faster ...]
Added test (compilation, only) in pathfinder/tests/BugTracker/Tests/join-query_compilation_error.SF-1889694.*
Seems to work fine with both MPS & ALG.
## Comment 8865
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1889694 at http://sourceforge.net/support/tracker.php?aid=1889694
| [PF] attribute `iter1' referenced in join not found | https://api.github.com/repos/MonetDB/MonetDB/issues/1569/comments | 0 | 2020-11-30T09:46:58Z | 2024-06-27T11:28:21Z | https://github.com/MonetDB/MonetDB/issues/1569 | 753,323,395 | 1,569 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-08 10:35:43 +0100
From: Maurice van Keulen <<mvankeulen>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-05-28 02:26:40 +0200
## Comment 8854
Date: 2008-02-08 10:35:43 +0100
From: Maurice van Keulen <<mvankeulen>>
When trying to compile the attached query (it doesn't matter what it does) with the algebra backend (pf -A b.xq), my machine (1GB) starts swapping considerably and "top" shows that all the while the pf-compiler is using >80% of my memory. The query only has about 100 lines. For my usage of MonetDB/XQuery, I would need to be able to compile and execute queries of at least 1000 lines.
## Comment 8855
Date: 2008-02-08 13:26:29 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Maurice, do you have the Boehm garbage collector library installed?
If yes -- please try compiling it stand-alone (pf -A) and then feeding the MIL output by hand into the Mserver.
Otherwise you will get punished for the fact that Pathfinder relies on garbage collection (which is not done by the Mserver in the MonetDB/XQuery context).
## Comment 8856
Date: 2008-02-10 14:17:59 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
tagged as "PF" in subject
## Comment 8857
Date: 2008-02-14 10:05:30 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
See also
[ 1892560 ] PF: "insufficient memory" while compiling XMark module
https://sourceforge.net/tracker/index.php?func=detail&aid=1892560&group_id=56967&atid=482468
## Comment 8858
Date: 2008-05-28 14:26:40 +0200
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
An improved memory management is in place now (by allocating space for
PFarray a bit more conservatively).
## Comment 8859
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1889487 at http://sourceforge.net/support/tracker.php?aid=1889487
| PF: Algebra compiler uses a lot of memory | https://api.github.com/repos/MonetDB/MonetDB/issues/1568/comments | 0 | 2020-11-30T09:46:56Z | 2024-06-27T11:28:20Z | https://github.com/MonetDB/MonetDB/issues/1568 | 753,323,371 | 1,568 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-06 02:56:07 +0100
From: Jan Flokstra <<jflokstra>>
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2008-02-06 04:35:44 +0100
## Comment 8846
Date: 2008-02-06 14:56:07 +0100
From: Jan Flokstra <<jflokstra>>
Since a couple of days the usage of any pftijah function crahses the compiler in both the STABLE and CURRENT(0.22) branches on SuSe9.3 and SuSe 10.3. It crahses in the PF/compiler function ns_lookup() looking up the "tijah" namespace. To reproduce just start the Mserver and query "tijah:ft-index-info()" result in a crash. The gdb stack trace reads:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1221921712 (LWP 14750)]
0x4614fca5 in ns_lookup (prefix=0x48d52048 "tijah") at ns.c:281
281 for (unsigned int i = PFarray_last (stat_known_ns); i; i--) {
(gdb) where
0 0x4614fca5 in ns_lookup (prefix=0x48d52048 "tijah") at ns.c:281
1 0x46150575 in ns_resolve (n=0x48d521b8) at ns.c:850
2 0x4614fdda in ns_resolve (n=0x48d52218) at ns.c:1019
3 0x4614fdda in ns_resolve (n=0x48d52248) at ns.c:1019
4 0x46131a57 in PFcompile_MonetDB (
xquery=0x8ab5981 "tijah:ft-index-info()\n",
url=0x8ab5981 "tijah:ft-index-info()\n", prologue=0x48d4f4ec,
query=0x48d4f454, epilogue=0x48d4f4f4, options=1177801568) at compile.c:889
5 0x4612eb6f in xquery_compile_exec (ctx=0x8c5e160,
url=0x8ab5981 "tijah:ft-index-info()\n", is_url=0, prologue=0x48d4f4ec,
query=0x48d4f4f0, epilogue=0x48d4f4f4, nsurl=0x0) at pathfinder.mx:4404
6 0x461306f0 in xquery_prepare (ctx=0x8c5e160, usec=The value of variable 'usec' is distributed across several
locations, and GDB cannot access its value.
) at pathfinder.mx:5746
7 0x461316d7 in xquery_client_engine (mc=0x467d1918) at pathfinder.mx:5895
8 0x467cf293 in mapi_client_engine (FC=0x467d1918) at mapi.mx:501
9 0x4096eaa7 in start_thread () from /lib/tls/libpthread.so.0
10 0x40b88c2e in clone () from /lib/tls/libc.so.6
(gdb)
## Comment 8847
Date: 2008-02-06 15:27:15 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
I was too quick in my previous bug report thinking it was only pftijah which suffered. Any namespace usage crashes the compiler when running in the server. The input
fn:count(((1))
crahses ns_lookup() the same way. When running the compiler standalone (eg. pf) everything works fine.
## Comment 8848
Date: 2008-02-06 16:04:49 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
After a better look at
[ 1886508 ] PF: many (most?) tests fail after recent checkins
I conclude this must be the same bug but after Jan's analysis 4/2 it still needs to be fixed because my checkouts were all today 6/2 after 14:00.
## Comment 8849
Date: 2008-02-06 16:09:28 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan (F), I assume you today's checkout were from the XQuery_0-22 Stable release branch, right?
Jan (R)'s fixes have not yet been propagated to the Current development trunk ...
## Comment 8850
Date: 2008-02-06 16:17:17 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
The checkout of XQuery_0-22 Stable by Djoerd this morning 10:00 caused the first problems. My version was indeed the Current from this afternoon. I did not see a fix from Jan yet, the last one I saw was the PFerrbuf fix [http://sourceforge.net/support/tracker.php?aid=1887791]
## Comment 8851
Date: 2008-02-06 16:26:13 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
After browsing SF I saw Jan's fix monday evening. So it should be fixed now. I will check Djoerds checkout this morning if it is there.
## Comment 8852
Date: 2008-02-06 16:35:44 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
OK problem solved. Djoerd mixed up a lot of scripts and ended up downloading the HEAD instead of 0.22. We will wait for the propagation and close the bug now.
## Comment 8853
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1887911 at http://sourceforge.net/support/tracker.php?aid=1887911
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| Any ns:fun() usage crashes compiler | https://api.github.com/repos/MonetDB/MonetDB/issues/1567/comments | 0 | 2020-11-30T09:46:53Z | 2024-06-27T11:28:19Z | https://github.com/MonetDB/MonetDB/issues/1567 | 753,323,330 | 1,567 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-06 11:47:22 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-06 01:31:47 +0100
## Comment 8843
Date: 2008-02-06 11:47:22 +0100
From: @drstmane
When provided an "invalid" (non-existing) option, `pf` segfaults:
in general:
$ pf -x
pf: invalid option -- x
Segmentation fault
in the Stable release branch, i.e, with assertion disabled:
$ pf -d
pf: invalid option -- d
Segmentation fault
For a test, see
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/compiler_semantics_tests_typechecking/internal.err.00.html
## Comment 8844
Date: 2008-02-06 13:31:46 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Since Niels change on 29.01. ('each xquery thread/client has its own PFerrbuf now. Fixes a bug, where the previous error of the previous client would repeat on the next client (even with a correct query).') the error buffer for the pf binary was initilized to late (in compile.c).
The initialization now happens in main().
## Comment 8845
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1887791 at http://sourceforge.net/support/tracker.php?aid=1887791
| PF: pf segfault on invalid option | https://api.github.com/repos/MonetDB/MonetDB/issues/1566/comments | 0 | 2020-11-30T09:46:51Z | 2024-06-27T11:28:18Z | https://github.com/MonetDB/MonetDB/issues/1566 | 753,323,302 | 1,566 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-05 03:52:58 +0100
From: Maurice van Keulen <<mvankeulen>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-08 05:34:16 +0100
## Comment 8831
Date: 2008-02-05 15:52:58 +0100
From: Maurice van Keulen <<mvankeulen>>
The attached query fails with
ERROR = !ERROR: err:FORG0005: function fn:exactly-one expects exactly one value.
Lines 104 and 105 is where the error is triggered:
for $w5 in $ctx4
let $pw5 := data(exactly-one($w5)/@prob)
Note that the argument to exactly-one is $w5 which is bound by the for just above it. So, it should always contain just one element ... but apparently it doesn't.
## Comment 8832
Date: 2008-02-06 19:21:29 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Maurice,
did/could you try whether it works with zero-or-one() instead of exactly-one()?
Stefan
## Comment 8833
Date: 2008-02-06 19:29:55 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Work for me with zero-or-one()!
Is $ctx4 and hence $w5 empty?
Or is "just" exactly-one() broken??
## Comment 8834
Date: 2008-02-07 09:23:06 +0100
From: Maurice van Keulen <<mvankeulen>>
Logged In: YES
user_id=654674
Originator: YES
If $ctx is empth, then the for-loop would have no iterations at all. In a for-loop, the loop variable, in this case $w5, is ALWAYS bound to exactly-one value or element. I think this bug is related to or may even have the same cause as my previous one (1886994). The behaviour I see if I leave out the exactly-one is that $w5 seems to be bound to two elements ... so I'm even more surprised that it works for you with zero-or-one().
## Comment 8835
Date: 2008-02-07 10:17:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan,
could you please have a quick look, whether this one rings a bell with you?
Feel free to re-/un-assign in case it's not "your business"...
Thanks!
Stefan
## Comment 8836
Date: 2008-02-07 18:34:16 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Here the for loop seems to be missing. It looks like $ctx4 is directly bound to $w5.
Again (as in bug report 1886994) the query is too large to say something useful.
Please provide a smaller example.
## Comment 8837
Date: 2008-02-08 10:24:24 +0100
From: Maurice van Keulen <<mvankeulen>>
Logged In: YES
user_id=654674
Originator: YES
I reduced the query to the one below. It returns a world-element with an attribute count, which should always be 1 (because it counts the loop variable). Inside the element is the loop-variable itself, i.e., should be 1 element. Expected result is 12 world elements with each a count of 1 and inside one world-element, but I get one world with a count of 12 and inside 12 world-elements. Funny thing is that the algebra-version also gets this one wrong, so perhaps the fault lies somewhere in the XQuery-code stages.
let $doc := exactly-one(doc("p.xml")//movies)
let $ctx0 := <world prob="1"><nid>{pf:nid($doc)}</nid></world>
let $ctx1 :=
for $w4 in $ctx0
for $poss4 in $doc//poss[./director]
return <world>{$poss4/@prob}</world>
return <bla>{
for $w5 in $ctx1
return <world count="{count($w5)}">{$w5}</world>
}</bla>
## Comment 8838
Date: 2008-02-08 14:51:39 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
> so perhaps the fault lies somewhere in the XQuery-code stages.
First I thought that this might be the query writers fault as $ctx0 is bound to a singleton and can be pruned. It was however necessary to trigger certain rewrites that forgot to update the type information.
A fix is checked into the stable branch (propagation is still pending).
It fixes both http://sourceforge.net/support/tracker.php?aid=1887055 and 1886994.
## Comment 8839
Date: 2008-02-08 15:00:58 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan,
THANKS for the fix(es)!
... we should recall to add some test(s), though ...
Stefan
## Comment 8840
Date: 2008-02-08 17:34:16 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Added test in pathfinder/tests/BugTracker/Tests/for_variable.SF-1887055.*
Maurice,
could your please validate the result (stable.out)?
## Comment 8841
Date: 2008-02-11 10:18:12 +0100
From: Maurice van Keulen <<mvankeulen>>
Logged In: YES
user_id=654674
Originator: YES
Checked and okay!
## Comment 8842
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1887055 at http://sourceforge.net/support/tracker.php?aid=1887055
| [PF] for variable contains more than one element | https://api.github.com/repos/MonetDB/MonetDB/issues/1565/comments | 0 | 2020-11-30T09:46:48Z | 2024-06-27T11:28:17Z | https://github.com/MonetDB/MonetDB/issues/1565 | 753,323,271 | 1,565 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-05 02:13:47 +0100
From: Maurice van Keulen <<mvankeulen>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-08 05:41:10 +0100
## Comment 8821
Date: 2008-02-05 14:13:47 +0100
From: Maurice van Keulen <<mvankeulen>>
(mps-version)
The attached query fails with
ERROR = !ERROR: batbat_dbl_mul: BATs l & r must be void-headed and head-aligned.
!ERROR: batbat_dbl_mul: operation failed.
if you change the "$p5*$pw5" into "$p5+$pw5", then the error message changes to
ERROR = !ERROR: batbat_dbl_add: BATs l & r must be void-headed and head-aligned.
!ERROR: batbat_dbl_add: operation failed.
so the problem lies in this neighbourhood.
## Comment 8822
Date: 2008-02-06 19:19:42 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Maurice,
did/could you by any chance try, whether it works better with property checking enabled, i.e., when starting Mserver with "--debug=10"?
Stefan
## Comment 8823
Date: 2008-02-06 19:55:36 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Ok, property checking does not help --- in fact, [*]() is called with two BAT of different length; hence, the bug is probably somewhere in the pathfinder-generated MIL code, or in the XQuery translation in general...
## Comment 8824
Date: 2008-02-07 10:17:12 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan,
could you please have a quick look, whether this one rings a bell with you?
Feel free to re-/un-assign in case it's not "your business"...
Thanks!
Stefan
## Comment 8825
Date: 2008-02-07 18:18:00 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
1.) I cannot follow the 26000 lines of MIL code.
2.) I cannot follow the 150 lines of XQuery (as it is not indented properly to recognize the nesting).
3.) I have no clue what the query does.
After looking at it for an hour I give up with absolutely no clue.
Maurice if you want to have this solved either use the algebra version or come up with a smaller example (< 20 lines of XQuery).
## Comment 8826
Date: 2008-02-08 14:52:30 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
A fix is checked into the stable branch that fixes both http://sourceforge.net/support/tracker.php?aid=1887055 and 1886994 (propagation is still pending).
## Comment 8827
Date: 2008-02-08 15:00:31 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan,
THANKS for the fix(es)!
... we should recall to add some test(s), though ...
Stefan
## Comment 8828
Date: 2008-02-08 17:41:10 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Added test in pathfinder/tests/BugTracker/Tests/Alignment_error.SF-1886994.*
Maurice,
could your please validate the result (stable.out)?
## Comment 8829
Date: 2008-02-11 10:17:41 +0100
From: Maurice van Keulen <<mvankeulen>>
Logged In: YES
user_id=654674
Originator: YES
Checked and okay!
## Comment 8830
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1886994 at http://sourceforge.net/support/tracker.php?aid=1886994
| [PF] Alignment error for batbat_dbl_mul | https://api.github.com/repos/MonetDB/MonetDB/issues/1564/comments | 0 | 2020-11-30T09:46:46Z | 2024-06-27T11:28:16Z | https://github.com/MonetDB/MonetDB/issues/1564 | 753,323,246 | 1,564 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-04 08:13:17 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-09 01:52:20 +0100
## Comment 8810
Date: 2008-02-04 20:13:17 +0100
From: @drstmane
As reported by the testing result emails on Friday, Feb 1 2008 (see attached files), lots of pathfinder tests fail after checkins on Thursday Jan 31 2008.
Since this affects both the "milprint_summer" and the "algebra version", these error prevent us from releasing MonetDB/XQuery 0.22.
For details, see the TestWeb at
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/index_short.html
and
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsG103/index_short.html
Assigning to Jan, as he did most of the (very much appreaciated!) checkins to pathfinder on Thursday Jan 31 2008.
## Comment 8811
Date: 2008-02-04 20:14:07 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
File Added: PF_ALG_res.2008-02-01.gz
## Comment 8812
Date: 2008-02-04 21:18:47 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Forgot to intialize namespaces in the server-side compilation (after splitting up namespace initialization and namespace resolve).
Sorry for the inconvenience.
## Comment 8813
Date: 2008-02-06 12:48:33 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Jan,
thanks for the quick fix!
Is get many of the test working, again, but various are still failing that did work fine until Jan 31 2008 (see below).
Most of these (now?) fail with
ERROR = !ERROR: CMDrevert: operation failed.
Possibly an attempt the revert a read-only BAT?
For examples, see
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/runtime/pf_docname.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/runtime/pf_docname.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/treat_as.SF-1586454.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/insert_multiple.SF-1590580.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/inserting_multiple_elements.SF-1590583.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/clear_attrs_on_delete.SF-1612739.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/clear_attrs_on_delete.SF-1612739-b.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/clear_attrs_on_delete.SF-1612739-c.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/clear_attrs_on_delete.SF-1612739-c.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/corrupt_after_update.SF-1706640.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/corrupt_after_update.SF-1706640.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/accessing_renamed_inserted_deleted_node.SF-1718622-1718635-1718709.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/NULL_BAT_in_serializer.SF-1694814a.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/NULL_BAT_in_serializer.SF-1694814a.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/NULL_BAT_in_serializer.SF-1694814b.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/NULL_BAT_in_serializer.SF-1694814b.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/insert_large_doc.SF-1726954.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/insert_large_doc.SF-1726954.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/replace-corrupts.SF-1758902.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/replace-corrupts.SF-1758902.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/swizzle-bug.SF-1760811.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/swizzle-bug.SF-1760811.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/swizzle-http://sourceforge.net/support/tracker.php?aid=2.SF-1763495.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/swizzle-http://sourceforge.net/support/tracker.php?aid=2.SF-1763495.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/insert_attribute_gives_ERROR_in_merged_union.SF-1763575.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/insert_attribute_gives_ERROR_in_merged_union.SF-1763575.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/repeated-insert.SF-1814911.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/repeated-insert.SF-1814911.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/insert-new-page.SF-1854215.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_StandOff_update/basic.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_StandOff_update/basic_insert.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_StandOff_update/basic_test.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/setattr-1.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/setattr-1.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/insert-1.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/insert-1.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/insert-2.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/insert-2.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/symm.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/symm.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/replacevaluetest.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/replacevaluetest.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/replacevaluetest2.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/replacevaluetest2.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/update.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_Update/update.err.00.html
...
Tests that worked on Jan 31 2008, but fail since Feb 01 2008:
F6 G32/32d runtime/pf_docname o o X X
F6 G32/32d tests/BugTracker/treat_as.SF-1586454 o o X o
F6 G32/32d tests/BugTracker/insert_multiple.SF-1590580 o o X o
F6 G32/32d tests/BugTracker/inserting_multiple_elements.SF-1590583 o o X o
F6 G32/32d tests/BugTracker/clear_attrs_on_delete.SF-1612739 o o o X
F6 G32/32d tests/BugTracker/clear_attrs_on_delete.SF-1612739-b o o o X
F6 G32/32d tests/BugTracker/clear_attrs_on_delete.SF-1612739-c o o X X
F6 G32/32d tests/BugTracker/corrupt_after_update.SF-1706640 o o X X
F6 G32/32d tests/BugTracker/accessing_renamed_inserted_deleted_node.SF-1718622-1718635-1718709 o o X o
F6 G32/32d tests/BugTracker/NULL_BAT_in_serializer.SF-1694814a o o X X
F6 G32/32d tests/BugTracker/NULL_BAT_in_serializer.SF-1694814b o o X X
F6 G32/32d tests/BugTracker/insert_large_doc.SF-1726954 o o X X
F6 G32/32d tests/BugTracker/replace-corrupts.SF-1758902 o o X X
F6 G32/32d tests/BugTracker/swizzle-bug.SF-1760811 o o X X
F6 G32/32d tests/BugTracker/swizzle-http://sourceforge.net/support/tracker.php?aid=2.SF-1763495 o o X X
F6 G32/32d tests/BugTracker/insert_attribute_gives_ERROR_in_merged_union.SF-1763575 o o X X
F6 G32/32d tests/BugTracker/repeated-insert.SF-1814911 o o X X
F6 G32/32d tests/BugTracker/preceding_following-sibling_bug.SF-1850772 o o X X
F6 G32/32d tests/BugTracker/insert-new-page.SF-1854215 o o o X
F6 G32/32d tests/StandOff/update/basic o o X o
F6 G32/32d tests/StandOff/update/basic_insert o o X o
F6 G32/32d tests/StandOff/update/basic_test o o X o
F6 G32/32d tests/Update/setattr-1 o o X X
F6 G32/32d tests/Update/insert-1 o o X X
F6 G32/32d tests/Update/insert-2 o o X X
F6 G32/32d tests/Update/symm o o X X
F6 G32/32d tests/Update/replacevaluetest o o X X
F6 G32/32d tests/Update/replacevaluetest2 o o X X
F6 G32/32d tests/Update/update o o X X
F6 G32/32d tests/Update/illegal-insert o o X X
F6 G32/32d tests/Update/illegal-delete o o X X
F6 G32/32d tests/W3C_use_cases/XQUF/AddressBook/Q1x o o o X
F6 G32/32d tests/W3C_use_cases/XQUF/AddressBook/check_docs o o X X
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q1 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q2 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q3a1 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q3a2 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q3b1 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q3b2 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q4a o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q4ax o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q4b o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q4bx o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q4c o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q4cx o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q6 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/Parts/Q6x o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q1 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q2 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q3 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q4 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q4x o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q5a o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q5b o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q6a o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q6b o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q7 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q7x o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q8 o o X o
F6 G32/32d tests/W3C_use_cases/XQUF/R/Q9 o o X o
F6 G32/32d tests/XRpc/echoAll o o X X
F6 G32/32d tests/XRpc/returnAncestor o o X X
F6 G32/32d tests/XRpc/returnAncestorOrSelf o o X X
F6 G32/32d tests/XRpc/returnChild o o X X
F6 G32/32d tests/XRpc/returnDescendant o o X X
F6 G32/32d tests/XRpc/returnDescendantOrSelf o o X X
F6 G32/32d tests/XRpc/returnFollowing o o X X
F6 G32/32d tests/XRpc/returnFollowingSibling o o X X
F6 G32/32d tests/XRpc/returnParent o o X X
F6 G32/32d tests/XRpc/returnPreceding o o X X
F6 G32/32d tests/XRpc/returnPrecedingSibling o o X X
F6 G32/32d tests/XRpc/returnSelf o o X X
F6 G32/32d tests/XRpc/insertNode o o X X
## Comment 8814
Date: 2008-02-06 14:03:10 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
This looks like a problem that occurs in the runtime -- namely in PROC ws_opendoc() -- and is limited to updateable documents (see below).
Can someone else please have a look!
MonetDB>shred_doc("/tmp/JTR.xml","JTR.xml","JTR",9LL);
shredding with holes: pagesize=16384, fillsize=14910
Elapsed time = 005ms 391us [002us/node]
Shredded 1 XML document (JTR.xml), total time after commit=0.100s
MonetDB>xquery("xml","doc(\"JTR.xml\")");
!ERROR: CMDrevert: operation failed.
!ERROR: interpret_params: leftfetchjoin(param 2): evaluation error.
MonetDB>delete_doc("JTR.xml");
MonetDB>shred_doc("/tmp/JTR.xml","JTR.xml");
Elapsed time = 019ms 550us [009us/node]
Shredded 1 XML document (JTR.xml), total time after commit=0.218s
MonetDB>xquery("xml","doc(\"JTR.xml\")");
MonetDB>
## Comment 8815
Date: 2008-02-06 19:03:15 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
The new error ("!ERROR: CMDrevert: operation failed.") could also be related to checkins on Friday Feb 01, since three tests (see below) did still worked on Feb 01, but fail since Feb 02.
On Feb 01 (as far as I can see), only Jan R checked in to pathfinder, and only Niels checked in to MonetDB (GDK; triggered by his "klocwork" analysis); there were no related changes in MonetDB4.
I don't know (yet?) whether or which of these changed might be related to the bug ...
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_StandOff_update/basic.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_StandOff_update/basic_insert.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_StandOff_update/basic_test.out.00.html
## Comment 8816
Date: 2008-02-07 18:30:38 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
All calls to "revert" that I could find in pathfinder are in MIL and preceded by a "access(BAT_WRITE)"; hence, it could be that either revert / CMDrevert / BATrevert fails itself, or the preceding "access(BAT_WRITE)" fails to properly set the BAT writable --- unfortunately there is no other (preceding) error message --- didn't have any time to GDB this, yet ...
## Comment 8817
Date: 2008-02-08 00:41:42 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
GDB found it:
Apparently, revert / CMDrevert / BATrevert is called with a BAT that has a void-nil (i.e., empty) tail.
This used to work fine.
However, after Niels' "klokwork"-fixes on Fri Feb 01 in MonetDB/src/gdk/gdk_batop.mx
========
@@ -1662,12 +1677,19 @@
if ((b->htype == TYPE_void && b->hseqbase != oid_nil) || (b->ttype == TYPE_void && b->tseqbase != oid_nil)) {
/* materialize void columns in-place */
b = BATmaterialize(b);
+ if (b == NULL)
+ return NULL;
}
ALIGNdel(b, "BATrevert", FALSE);
hs = (size_t) Hsize(b);
ts = (size_t) Tsize(b);
h = (char *) GDKmalloc(hs);
t = (char *) GDKmalloc(ts);
+ if (!h || !t) {
+ GDKfree(h);
+ GDKfree(t);
+ return NULL;
+ }
for (p = BUNlast(b)?BUNlast(b) - 1:0, q = BUNfirst(b); p > q; p--, q++){
char *ph = BUNhloc(bi,p);
char *qh = BUNhloc(bi,q);
========
BATrevert returns NULL on a BAT with a void-nil tail since
ts = (size_t) Tsize(b); ==> ts == 0
and
t = (char *) GDKmalloc(ts(==0)); ==> t == 0
(likewise for void-nil head).
I'll see, whether simply ignoring a void-nil head/tail in BATrevert works and solves the problem ...
## Comment 8818
Date: 2008-02-08 06:53:45 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Fixed in CVS (MonetDB_1-22 branch of MonetDB):
BATrevert does not have to do anything for an *empty* void-nil head or tail, thus avoiding the GDKmalloc(0).
To be closed, once testing has confirmed the fix.
## Comment 8819
Date: 2008-02-09 13:52:20 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Testing indicates that all related fixes seem to work properly.
Closing.
## Comment 8820
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1886508 at http://sourceforge.net/support/tracker.php?aid=1886508
| PF: many (most?) tests fail after recent checkins | https://api.github.com/repos/MonetDB/MonetDB/issues/1563/comments | 0 | 2020-11-30T09:46:43Z | 2024-06-27T11:28:15Z | https://github.com/MonetDB/MonetDB/issues/1563 | 753,323,218 | 1,563 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-04 03:05:33 +0100
From: Theodora Tsikrika <<theodora_t>>
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-20 09:02:26 +0100
## Comment 8805
Date: 2008-02-04 15:05:33 +0100
From: Theodora Tsikrika <<theodora_t>>
Hi all,
I am using the MonetDB JDBC driver in a java program similar to the sample program in http://monetdb.cwi.nl/projects/monetdb/XQuery/Documentation/The-JDBC-Library.htmlThe-JDBC-Library
to execute an XQuery query (instead of an SQL one).
When the query does not return any results, I get the following:
> Exception in thread "main" java.lang.ArithmeticException: / by zero
When the query returns results, it works fine.
I am using MonetDB/XQuery checked out from CVS HEAD on 5th November 2007:
MonetDB Server v4.21.0
based on GDK v1.21.0
Compiled for x86_64-redhat-linux-gnu/64bit with 32bit OIDs; dynamically linked.
MonetDB/XQuery module v0.21.0 loaded
> java -version
java version "1.7.0"
IcedTea Runtime Environment (build 1.7.0-b21)
IcedTea 64-Bit Server VM (build 1.7.0-b21, mixed mode)
Regards,
Theodora
## Comment 8806
Date: 2008-02-04 15:30:06 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
Do you happen to know a query which needs no documents which returns an empty result, so I can test?
## Comment 8807
Date: 2008-02-04 15:49:47 +0100
From: Theodora Tsikrika <<theodora_t>>
Logged In: YES
user_id=1785117
Originator: YES
for $d at $rank in (1,2,3) return $d[$rank=4]
## Comment 8808
Date: 2008-02-20 21:02:26 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
2 and a half hour later ...
Due to missing history (cvs moves) and missing documentation (on my
end) it took quite some while to figure out what was the "correct" way
to solve this bug. In the end it al boiled down on the limitation of
the XQuery backend to only support full-results-at-once, which means the
resource allocation of JDBC needs to allocate the entire result at once.
If the result is 0 rows big, however, then things went wrong when
calculating the current cache block. Anyway, for the future we now know
why we did this cache block calulation thing like this again...
The fix I checked in is slightly different from what I gave you, but yours shouldn't hurt in use.
TEST ADDED
xquery/tests/BugTracker/Tests/JDBC_0_results.SF-1886326
## Comment 8809
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1886326 at http://sourceforge.net/support/tracker.php?aid=1886326
| MonetDB JDBC driver exception for XQuery with no results | https://api.github.com/repos/MonetDB/MonetDB/issues/1562/comments | 0 | 2020-11-30T09:46:40Z | 2024-06-27T11:28:14Z | https://github.com/MonetDB/MonetDB/issues/1562 | 753,323,179 | 1,562 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-01 04:26:44 +0100
From: Milena Ivanova <<milena>>
To: @njnes
Version: -- development
Last updated: 2008-02-08 12:44:39 +0100
## Comment 8802
Date: 2008-02-01 16:26:44 +0100
From: Milena Ivanova <<milena>>
M5/SQL started with
mserver5 ... --dbinit="include sql;"
followed by
>sql.start();
leads to exit of the server;
mserver5 ... --dbinit="include sql;sql.start();"
works fine.
## Comment 8803
Date: 2008-02-08 12:44:39 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
the scenario re-initialization didn't call a clientInit function. This is fixed now. No test needed.
## Comment 8804
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1884497 at http://sourceforge.net/support/tracker.php?aid=1884497
| sql.start | https://api.github.com/repos/MonetDB/MonetDB/issues/1561/comments | 0 | 2020-11-30T09:46:37Z | 2024-06-27T11:28:13Z | https://github.com/MonetDB/MonetDB/issues/1561 | 753,323,153 | 1,561 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-02-01 01:42:44 +0100
From: @romulogoncalves
To: Romulo Goncalves <<goncalve>>
Version: -- development
Last updated: 2008-02-01 05:49:32 +0100
## Comment 8798
Date: 2008-02-01 13:42:44 +0100
From: @romulogoncalves
If I create a view over a table and the reference to this table is inside a union the dependency is not detected...
To understand better the problem run this SQL code:
create table t1 (id int);
create table t2 (id int);
create view v1 as (select * from t1
union
select * from t2);
drop table t1;
drop table t2;
drop view v1;
select * from t1;
drop table t1;
drop table t2;
The tables t1 and t2 should not be removed on the first drop statements because view v1 depend on them...
Romulo
## Comment 8799
Date: 2008-02-01 13:45:37 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Test added in:
sql/src/test/BugTracker/Tests/table_dependency_inside_union.SF-1884382.sql
## Comment 8800
Date: 2008-02-01 17:49:32 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The bug is fixed in the following check ins:
[Monetdb-sql-checkins] sql/src/server sql_statement.mx, , 1.159, 1.160
Log Message:
o statement st_sql only has
op1->stval
for statement st_column we also need to check the statements in
op1->stval (but only when we are looking for view dependencies)...
[Monetdb-sql-checkins] sql/src/server sql_select.mx,,1.218,1.219
Modified Files:
sql_select.mx
Log Message:
When I create a statement column for table_optname
I need to insert the table in this statement and not NULL
if not, I am not able to detect tables inside a union in the view definition...
For example:
create view as (select * from t1 UNION select * from t2);
## Comment 8801
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1884382 at http://sourceforge.net/support/tracker.php?aid=1884382
| table dependency inside union... | https://api.github.com/repos/MonetDB/MonetDB/issues/1560/comments | 0 | 2020-11-30T09:46:35Z | 2024-06-27T11:28:12Z | https://github.com/MonetDB/MonetDB/issues/1560 | 753,323,134 | 1,560 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-31 04:39:37 +0100
From: Wouter Alink <<alink>>
To: XQuery devs <<bugs-xquery>>
Version: -- development
Last updated: 2008-01-31 11:45:29 +0100
## Comment 8792
Date: 2008-01-31 16:39:37 +0100
From: Wouter Alink <<alink>>
I saw that the algebra was now runtime configurable... (yahoo!!)
I compiled it (last night's current), and ran some queries. Other than that query performance is much better, I noticed the following:
- The output format for algebra xqueries seems to be 'xml', no matter which option is supplied to '-f'. (milprint_summer had 'dm' as default)
- Typo: miprintsummer in pathfinder.mx (line: 4385)
"""(GDKgetenv("xquery_backend") == NULL) || (strcmp(GDKgetenv("xquery_backend"),"miprint_summer") != 0)))"""
- The function "pf:documents()" is not yet implemented?
- it cannot yet count to 3
QUERY = count(1 to 3)
ERROR = !fatal error: physical algebra equivalent for logical algebra node kind 31 not implemented, yet
## Comment 8793
Date: 2008-01-31 17:22:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
- Indeed, the algebra version for now only supports the "xml" serializer.
- Typo in pathfinder.mx is fixed. Thanks for notifying!
- pf:documents() is not yet available in the algebra version.
In fact, we also plan to implement fn:doc-available()
(cf., http://www.w3.org/TR/xpath-functions/func-doc-available)
which might be more suitable for your purposes than pf:documents() ...
- The algebra version does not yet suppor the "to" operator.
## Comment 8794
Date: 2008-01-31 22:52:03 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Just added support for op:to --- Now you can count until MonetDB crashes :D
But you have to come up with something more clever than
for $a in (1 to 1000000) return 1 to 1000000
This query only returns the useful error message '!ERROR: CMDcross: operation failed.' but does not crash MonetDB :-).
## Comment 8795
Date: 2008-01-31 23:45:29 +0100
From: Wouter Alink <<alink>>
Logged In: YES
user_id=621590
Originator: YES
thanks for the quick replies and fixes.
fn:doc-available() is indeed more useful in most of our cases. will there be a way in the upcoming version to obtain a list of documents/collections that are currently stored in the database?
i'll close the bug-report as there are enough tests with 'to' already.
Jan, good night.
## Comment 8796
Date: 2008-02-01 09:29:10 +0100
From: @lsidir
Logged In: YES
user_id=1856546
Originator: NO
the upcoming release is to soon to have the fn:doc-available() ready. However, it is high on our priority list so it will be there soon.
## Comment 8797
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1883569 at http://sourceforge.net/support/tracker.php?aid=1883569
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-xquery@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| XQ: Algebra | https://api.github.com/repos/MonetDB/MonetDB/issues/1559/comments | 0 | 2020-11-30T09:46:32Z | 2024-06-27T11:28:11Z | https://github.com/MonetDB/MonetDB/issues/1559 | 753,323,100 | 1,559 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-31 03:14:15 +0100
From: @grobian
To: @sjoerdmullender
Version: -- development
Last updated: 2008-02-14 11:22:54 +0100
## Comment 8780
Date: 2008-01-31 15:14:15 +0100
From: @grobian
sparcv9-sun-solaris2.10-gcc -DHAVE_CONFIG_H -I. -I/scratch/tmp/monetdb-fabian/current/common/src/gdk -I../.. -I/scratch/tmp/monetdb-fabian/current/common/src/gdk -I../common -I/scratch/tmp/monetdb-fabian/current/common/src/gdk/../common -DLIBGDK -mcpu=v9 -mtune=v9 -pipe -g -O2 -W -Wall -D__EXTENSIONS__ -std=c99 -fgnu89-inline -Werror -D_REENTRANT -c gdk_utils.c -fPIC -DPIC -o .libs/libbat_la-gdk_utils.o
cc1: warnings being treated as errors
/scratch/tmp/monetdb-fabian/current/common/src/gdk/gdk_utils.mx: In function 'GDKmem_heapcheck':
/scratch/tmp/monetdb-fabian/current/common/src/gdk/gdk_utils.mx:524: warning: comparison of unsigned expression < 0 is always false
/scratch/tmp/monetdb-fabian/current/common/src/gdk/gdk_utils.mx:524: warning: comparison of unsigned expression < 0 is always false
/scratch/tmp/monetdb-fabian/current/common/src/gdk/gdk_utils.mx:524: warning: comparison of unsigned expression < 0 is always false
make[5]: *** [libbat_la-gdk_utils.lo] Error 1
make[5]: *** Waiting for unfinished jobs....
(sparcv9 == 64-bits native target)
offending code block:
if ((SIZEOF_VOID_P==8) && defined(HAVE_SIGNED_MALLINFO))
if (m.usmblks < 0 || m.uordblks < 0 || m.hblkhd < 0)
mallinfo_ok = 0; /* incredible POSIX incompetence!! non-64-bit safe mallinfo */
else
endif
## Comment 8781
Date: 2008-01-31 15:16:56 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
The build system info:
(build-t1000-sol10:/scratch/tmp/monetdb-fabian) fabian% sparcv9-sun-solaris2.10-gcc --version
sparcv9-sun-solaris2.10-gcc (GCC) 4.2.2 (Gentoo 4.2.2 p1.0)
Copyright (C) 2007 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
## Comment 8782
Date: 2008-01-31 15:22:56 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
from MonetDB/src/gdk/gdk_posix.mx:
========
ifndef HAVE_STRUCT_MALLINFO
struct mallinfo {
int arena; /* total space in arena */
int ordblks; /* number of ordinary blocks */
int smblks; /* number of small blocks */
int hblks; /* number of holding blocks */
int hblkhd; /* space in holding block headers */
int usmblks; /* space in small blocks in use */
int fsmblks; /* space in free small blocks */
int uordblks; /* space in ordinary blocks in use */
int fordblks; /* space in free ordinary blocks */
int keepcost; /* cost of enabling keep option */
};
endif
========
so, apparently your system provides it's own mallinfo struct --- what does it look like, in particular, of which type are members "usmblks", "uordblks", and/or "hblkhd", and does it properly support 64-bit?
## Comment 8783
Date: 2008-01-31 15:27:27 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
s/your machine/CWI's Sun T1 Niagara machine running an up-to-date patched Solaris 10 installation/ :)
Yes, Sun seems to support 64-bits quite well.
(build-t1000-sol10:~) fabian% grep mallinfo /usr/include/*
/usr/include/malloc.h:struct mallinfo {
/usr/include/malloc.h:struct mallinfo mallinfo(void);
/usr/include/malloc.h:struct mallinfo mallinfo();
from malloc.h:
struct mallinfo {
unsigned long arena; /* total space in arena */
unsigned long ordblks; /* number of ordinary blocks */
unsigned long smblks; /* number of small blocks */
unsigned long hblks; /* number of holding blocks */
unsigned long hblkhd; /* space in holding block headers */
unsigned long usmblks; /* space in small blocks in use */
unsigned long fsmblks; /* space in free small blocks */
unsigned long uordblks; /* space in ordinary blocks in use */
unsigned long fordblks; /* space in free ordinary blocks */
unsigned long keepcost; /* cost of enabling keep option */
};
So the compiler is fine, and it is also LP64-safe.
## Comment 8784
Date: 2008-01-31 15:33:45 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
interestingly enough google seems to suggest with RedHat and IBM backing up that mallinfo is not in ANSI or POSIX.
IBM claims that "most platforms have mallinfo available", and that "The implementation of this function is specific to the implementation of dynamic memory on the z/TPF system."
## Comment 8785
Date: 2008-01-31 15:52:21 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
The configure check appears to be b0rkened, as HAVE_SIGNED_MALLINFO should not be 1 as it is in my case. It looks like configure never runs a check for this, so there is possibly some branch that isn't too obvious to me at the moment.
## Comment 8786
Date: 2008-01-31 17:11:19 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
$ grep -9 --color HAVE_SIGNED_MALLINFO MonetDB/configure.ag
AC_COMPILE_IFELSE(AC_LANG_SOURCE([int foo(int * __restrict__ p) { return *p; }]),
AC_DEFINE(HAVE___RESTRICT__, 1,
[Define if the compiler supports the __restrict__ keyword])))
AC_TRY_COMPILE([$ac_includes_default
include <malloc.h>], [struct mallinfo m;], AC_DEFINE(HAVE_STRUCT_MALLINFO, 1, [Define if you have struct mallinfo])
AC_TRY_COMPILE([$ac_includes_default
include <malloc.h>], [struct mallinfo m = {0};
/* compilers that don't like <0 comparisons on unsigned data *
* will fail here (but then not in src/gdk/gdk_utils.mx) */
int x = (m.usmblks < 0);], AC_DEFINE(HAVE_SIGNED_MALLINFO, 1, [Define if your mallinfo struct has signed elements])
))
Checks for library functions.
AC_FUNC_MEMCMP
AC_CHECK_FUNCS([memset])
AC_CHECK_FUNCS([fcntl ftime getopt getrlimit getuid kill lockf madvise nanosleep pipe posix_fadvise posix_madvise setenv shutdown strdup strtof sysconf uname strtoll fsync fdatasync])
AC_CHECK_DECLS([strdup, strtof])
## Comment 8787
Date: 2008-02-11 11:30:55 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Is there a config.log available from the configure run? It looks like the test expects that if the compiler complains about signed/unsigned mismatch when compiling gdk_utils.mx, it should do this also when compiling the test. But does it? And if so, does it only warn and not turn the warning into an error?
It would be possible to test this using a AC_TRY_RUN instead of AC_TRY_COMPILE. You can then assign -1 to a mallinfo field and test whether the value is > 0. But this requires running the test.
## Comment 8788
Date: 2008-02-11 12:22:20 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
See attach, or look around in /scratch/tmp/monetdb-fabian/build-current-build-t1000-sol10.cwi.nl/common on build-t1000-sol10
File Added: config.log
## Comment 8789
Date: 2008-02-12 10:36:34 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
I have changed the way we check for mallinfo(). Could you check whether this helped?
## Comment 8790
Date: 2008-02-14 11:22:54 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
Today's supersourcetarball passed this point with -Werror, so I consider this bug fixed. The target doesn't compile anyway, but that's another bug.
## Comment 8791
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1883505 at http://sourceforge.net/support/tracker.php?aid=1883505
| MonetDB Compilation fails on sparcv9-sun-solaris2.10 | https://api.github.com/repos/MonetDB/MonetDB/issues/1558/comments | 0 | 2020-11-30T09:46:29Z | 2024-06-27T11:28:10Z | https://github.com/MonetDB/MonetDB/issues/1558 | 753,323,057 | 1,558 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-29 09:46:00 +0100
From: @drstmane
To: @mlkersten
Version: -- development
Last updated: 2008-01-31 03:09:24 +0100
## Comment 8774
Date: 2008-01-29 09:46:00 +0100
From: @drstmane
Stating mserver5 on Windows fails with
MonetDB server v5.3.0, based on kernel v1.21.0
Serving database 'mTests'
Compiled for i686-pc-win32/32bit with 32bit OIDs dynamically linked
Copyright (c) 1993-2007 CWI, all rights reserved
Visit http://monetdb.cwi.nl/ for further information
!MALException:MAL.getAddress:address of 'replicator.REPprelude' not found
!MALException:MAL.getAddress:address of 'replicator.REPselect' not found
!MALException:MAL.getAddress:address of 'replicator.REPuselect' not found
Listening for connection requests on mapi:monetdb://tureluur:39506/
function user.main():void;
clients.quit();
end main;
For details, see (e.g.):
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/index_short.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.32.32.d-Windows5.1/test10.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Mic.32.32.d-Windows5.1/test10.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.64.64.d-Windows5.2/test10.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Mic.64.64.d-Windows5.2/test10.out.html
## Comment 8775
Date: 2008-01-29 10:17:15 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
added the missing export declarations, which should resolve the issue.
To be closed after next nightly run.
No additional test needed.
## Comment 8776
Date: 2008-01-30 11:41:02 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Now, MonetDB5 compilation fails on Windows with
opt_replicator.obj : error LNK2019: unresolved external symbol __imp__ALGselectInclusive referenced in function _REPselect
For details, see
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.32.32.d-Windows5.1/make.out.extrerrmsg.htmll2070
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Mic.32.32.d-Windows5.1/make.out.extrerrmsg.htmll1878
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.32.32.d-Windows5.1/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Mic.32.32.d-Windows5.1/make.out.html
## Comment 8777
Date: 2008-01-30 14:07:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
(Hopefully) fixed in CVS:
opt_replicators calls symbols from module algebra,
hence, we need to link with ../modules/kernel/lib_algebra
## Comment 8778
Date: 2008-01-31 15:09:24 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Both loading and the compilation/linking problems appear to be fixed.
Closing.
(Obviously no extra test needed.)
## Comment 8779
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1881656 at http://sourceforge.net/support/tracker.php?aid=1881656
| M5: starting mserver5 fails on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1557/comments | 0 | 2020-11-30T09:46:26Z | 2024-06-27T11:28:09Z | https://github.com/MonetDB/MonetDB/issues/1557 | 753,323,025 | 1,557 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-28 03:36:16 +0100
From: Jan Flokstra <<jflokstra>>
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2009-02-15 10:28:19 +0100
## Comment 8766
Date: 2008-01-28 15:36:16 +0100
From: Jan Flokstra <<jflokstra>>
Next too the difficult to reproduce preceding/following http://sourceforge.net/support/tracker.php?aid=1850772 ([preceding and following]-sibling:: bug) there is another problem with the following() axis step .
On the small 1M xmark document the queries listed below all result in a memory allocation failure of BAT's sized 200M to several G's. I think this example is so simple it should run on any machine.
count(doc("xmark1.xml")//street[./following::keyword[./following::current]])
count(doc("xmark1.xml")//*[./following::incategory[./following::interest]])
count(doc("xmark1.xml")//*[./following::category]/following::interest)
count(doc("xmark1.xml")//*[./following::incategory[./following::interest]])
count(doc("xmark1.xml")//itemref[./following::phone[./following::africa]])
The first query for instance running the CVS HEAD (28/1-14:00) on SuSe9.3 (1G RAM)
flokstra@ewi581:~/scripts/DATA> mclient -lx
xquery>count(doc("xmark1.xml")//street[./following::keyword[./following::current]])
more>MAPI = monetdb@localhost:50000
QUERY = count(doc("xmark1.xml")//street[./following::keyword[./following::current]])
ERROR = !ERROR: GDKload failed: name=06/646, ext=head.priv
!ERROR: PFll_following: could not allocate a result BAT of size 555611000.
!ERROR: PFll_following: operation failed.
xquery>
I added the used xmark document bzipped2 to this report,
Jan.
## Comment 8767
Date: 2008-01-28 18:25:11 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
I just tried the fourth query ('count(doc("xmark1.xml")//*[./following::incategory[./following::interest]])') on the 580k auctionG.xml from the testset using the algebra variant (pf -A) and added a count before and after the following steps:
before following::incategory:
[ 8517 ]
before following::incategory:
[ 614915 ]
before following::interest:
[ 614915 ]
during following::interest:
!ERROR: PFll_following: could not allocate a result BAT of size 3069254744.
I think this is perfectly fine as you asked (almost) for a Cartesian between all descendants of the documents (8517), the incategory nodes (436), and the interest nodes (211) which gives you around 783 million nodes. The guess of the system (3.069 million) thus was not far off.
I assume the query writer was looking for the axis following-sibling instead of the 'globabl' following path check.
## Comment 8768
Date: 2008-01-28 21:03:24 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan (R.), thanks for the analysis!
The problem is "particularly bad", since the following- & preceding-step (unlike descendant & child) do not (yet?) perform a "name-test push-down", i.e., they effectively first need to materialize following::* before applying the name-test.
I'll try to check how difficult it will be to add the "name-test push-down" also for following & preceding ...
(Obviously, even with the "name-test push-down", Jan R.'s analysis/comment still holds.)
## Comment 8769
Date: 2008-01-29 10:33:41 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
Jan and Stefan thanks for the effort. These following() queries were (again:-) created by a student of ours, Gerben Broenink, who does a statistical analysis project of XML docs for Riham and Maurice. He told me he will just skip these queries so it is not important to do the push-down optimizations in the MPS version.
JanF.
## Comment 8770
Date: 2008-01-29 10:50:02 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Hoi Jan,
thanks for the info!
The push-down are actually for both MPS & ALG (as long as we still have staircasejoins), hence, they should eventually be done in any case --- most probably not "this weeK', though ...
Stefan
## Comment 8771
Date: 2008-02-10 14:14:25 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
tagged as "XQ" in subject
## Comment 8772
Date: 2009-02-15 22:28:19 +0100
From: @drstmane
turned the request for name-test push-down in preceding & following steps into a feature request:
[ 2603533 ] XQ: name-test push down for preceding and following step
https://sourceforge.net/tracker/index.php?func=detail&aid=2603533&group_id=56967&atid=482471
## Comment 8773
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1881181 at http://sourceforge.net/support/tracker.php?aid=1881181
| XQ: following uses too much memory | https://api.github.com/repos/MonetDB/MonetDB/issues/1556/comments | 0 | 2020-11-30T09:46:23Z | 2024-06-28T13:30:43Z | https://github.com/MonetDB/MonetDB/issues/1556 | 753,322,994 | 1,556 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-28 10:16:42 +0100
From: @sjoerdmullender
To: Lefteris Sidirourgos <<lsidir>>
Version: -- development
Last updated: 2008-01-29 01:17:33 +0100
## Comment 8761
Date: 2008-01-28 10:16:42 +0100
From: @sjoerdmullender
The values for aat_path, aat_docnm and aat_colnm are the same as those for aat_pfrag1, aag_pre1 and aat_afrag1 respectively causing a (correct) warning on Gentoo.
## Comment 8762
Date: 2008-01-28 10:30:26 +0100
From: @lsidir
Logged In: YES
user_id=1856546
Originator: NO
This problem was caused from a commit I did last week. I was trying to save bits in the type system but the semantics of this solution is wrong.
## Comment 8763
Date: 2008-01-29 13:17:33 +0100
From: @lsidir
Logged In: YES
user_id=1856546
Originator: NO
Bug fixed and now Gentoo is compiling. I had to use 3 more bits from the type system of pathfinder to make this work correct.
## Comment 8764
Date: 2008-01-29 13:35:35 +0100
From: @lsidir
Logged In: YES
user_id=1856546
Originator: NO
Resolution is fixed.
## Comment 8765
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1881018 at http://sourceforge.net/support/tracker.php?aid=1881018
| PF: value clash in compiler/include/algebra.h | https://api.github.com/repos/MonetDB/MonetDB/issues/1555/comments | 0 | 2020-11-30T09:46:21Z | 2024-06-27T11:28:07Z | https://github.com/MonetDB/MonetDB/issues/1555 | 753,322,968 | 1,555 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-26 07:11:41 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-12 10:03:30 +0100
## Comment 8755
Date: 2008-01-26 19:11:41 +0100
From: @drstmane
On 32-bit Windows, compilation of MonetDB-Java fails with
'choice' is not recognized as an internal or external command,
cf.
http://monetdb.cwi.nl/testing/projects/monetdb/Current/java/.Int.32.32.d-Windows5.1/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/java/.Mic.32.32.d-Windows5.1/make.out.html
## Comment 8756
Date: 2008-01-26 19:13:51 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
On 64-bit Windows, we get
ERROR: The file is either empty or does not contain the valid choices.
but the compilation seem to succeed, anyway; cf.
http://monetdb.cwi.nl/testing/projects/monetdb/Current/java/.Int.64.64.d-Windows5.2/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/java/.Mic.64.64.d-Windows5.2/make.out.html
and
http://monetdb.cwi.nl/testing/projects/monetdb/Current/java/.Makes3/index_short.html
## Comment 8757
Date: 2008-02-11 11:19:40 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
We call the program ant.bat from the Apache Ant installation to do the actual work. This script contains the line:
if not "%OS%"=="Windows_NT" echo 1 > nul | choice /n /c:1
It is this line that causes the error.
The reason is, in TestTools we set the variable OS to (in this case) Windows. The default value is, however, Windows_NT.
## Comment 8758
Date: 2008-02-11 12:29:15 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Hopefully fixed in CVS (latest Stable release branches):
To avoid the clash (thanks, Sjoerd, for finding out!),
we rename our own variables OS & OSVER (in TestTools and related code) into SYST & SYSTVER, respectively, and hope they do not clash with existing ones...
To be closed once testing confirms the fix.
## Comment 8759
Date: 2008-02-12 10:03:30 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Testing confirmed fix.
Closing.
## Comment 8760
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1880371 at http://sourceforge.net/support/tracker.php?aid=1880371
| Java: compilation fails on 32-bit Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1554/comments | 0 | 2020-11-30T09:46:18Z | 2024-06-27T11:28:06Z | https://github.com/MonetDB/MonetDB/issues/1554 | 753,322,938 | 1,554 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-26 06:59:01 +0100
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2008-01-28 08:39:12 +0100
## Comment 8752
Date: 2008-01-26 18:59:01 +0100
From: @drstmane
Compilation of SQL with asserts disabled (e.g., with `make rpm`) fails with
gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -I. -I.. -I./.. -I../../include -I./../../include -I../../common -I./../../common -I../bat -I./../bat -I/scratch/monet.GNU.64.64.d.2617/MonetDB/.GNU.64.64.d-Fedora6/include/MonetDB -I/scratch/monet.GNU.64.64.d.2617/MonetDB/.GNU.64.64.d-Fedora6/include/MonetDB/common -I/scratch/monet.GNU.64.64.d.2617/MonetDB/.GNU.64.64.d-Fedora6/include/MonetDB/gdk -DLIBRESTRICTSTORE -O2 -Wall -Wextra -std=c99 -O6 -fomit-frame-pointer -finline-functions -falign-loops=4 -falign-jumps=4 -falign-functions=4 -fexpensive-optimizations -funroll-loops -frerun-cse-after-loop -frerun-loop-opt -Werror-implicit-function-declaration -Werror -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wp,-D_FORTIFY_SOURCE=2 -c restrict_storage.c -fPIC -DPIC -o .libs/librestrictstore_la-restrict_storage.o
cc1: warnings being treated as errors
restrict_storage.c: In function 'bind_col':
restrict_storage.c:8: warning: unused parameter 'access'
restrict_storage.c: In function 'bind_idx':
restrict_storage.c:22: warning: unused parameter 'access'
make[8]: *** [librestrictstore_la-restrict_storage.lo] Error 1
icc -no-gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -I. -I.. -I./.. -I../../include -I./../../include -I../../common -I./../../common -I../bat -I./../bat -I/scratch/monet.Int.64.64.d.2617/MonetDB/.Int.64.64.d-Fedora6/include/MonetDB -I/scratch/monet.Int.64.64.d.2617/MonetDB/.Int.64.64.d-Fedora6/include/MonetDB/common -I/scratch/monet.Int.64.64.d.2617/MonetDB/.Int.64.64.d-Fedora6/include/MonetDB/gdk -DLIBRESTRICTSTORE -O2 -we140 -wd1418 -Wall -w2 -c99 -mp1 -O3 -restrict -unroll -tpp7 -axWP -we266 -Werror -wd1418,1419,279,981,810,193,111,1357,1572,1599 -c restrict_storage.c -KPIC -DPIC -o .libs/librestrictstore_la-restrict_storage.o
icc: Command line warning: overriding '-O2' with '-O3'
restrict_storage.c(8): error 869: parameter "access" was never referenced
bind_col(sql_trans *tr, sql_column *c, int access)
^
See (e.g.)
http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.GNU.64.64.d-Fedora6/rpm.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.Int.64.64.d-Fedora6/rpm.out.html
## Comment 8753
Date: 2008-01-28 08:39:12 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Fixed by Niels.
Thanks!
## Comment 8754
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1880361 at http://sourceforge.net/support/tracker.php?aid=1880361
| SQL: compilation without asserts (e.g. `make rpm`) fails | https://api.github.com/repos/MonetDB/MonetDB/issues/1553/comments | 0 | 2020-11-30T09:46:16Z | 2024-06-27T11:28:05Z | https://github.com/MonetDB/MonetDB/issues/1553 | 753,322,902 | 1,553 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-26 06:18:14 +0100
From: @drstmane
To: Stefan Manegold <<Stefan.Manegold>>
Version: -- development
Last updated: 2008-02-22 01:28:10 +0100
## Comment 8748
Date: 2008-01-26 18:18:14 +0100
From: @drstmane
`make install` and `make rpm` for various modules on various platforms trigger
libtool: install: warning: `*.la' has not been installed in `...'
Should we have a look at whether they appear due to (e.g.) not properly created Makefiles (by autogen?), or are they "harmless", "unavoidable", and should hence be ignored?
See (e.g.):
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB/.GNU.64.64.d-Fedora6/rpm.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/clients/.GNU.64.64.d-Fedora6/rpm.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB4/.GNU.64.64.d-Fedora6/install.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB4/.GNU.64.64.d-Fedora6/rpm.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.GNU.64.64.d-Fedora6/install.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.GNU.64.64.d-Fedora6/rpm.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.GNU.64.64.d-Fedora6/install.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.GNU.64.64.d-Fedora6/rpm.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.GNU.64.64.d-Fedora6/install.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.GNU.64.64.d-Fedora6/rpm.out.html
## Comment 8749
Date: 2008-02-22 13:17:54 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
I say: ignore this warning. The .la file is so that you can use libtool to link against the library, but when the library is installed properly (as it is), we don't need this.
Now the test tools need to also ignore the warning.
## Comment 8750
Date: 2008-02-22 13:28:10 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
TestTools now ignore this warning.
Closing.
## Comment 8751
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1880346 at http://sourceforge.net/support/tracker.php?aid=1880346
| libtool: install: warning: `*.la' has not been installed in | https://api.github.com/repos/MonetDB/MonetDB/issues/1552/comments | 0 | 2020-11-30T09:46:13Z | 2024-06-28T13:30:42Z | https://github.com/MonetDB/MonetDB/issues/1552 | 753,322,858 | 1,552 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-22 03:23:40 +0100
From: @drstmane
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-01-22 03:41:25 +0100
## Comment 8745
Date: 2008-01-22 15:23:40 +0100
From: @drstmane
Since 2007.12.08, i.e., after chckins on 2007.12.07, tests
pathfinder/tests/BugTracker/Tests/bogus_text_step.SF-1537500.mps
pathfinder/tests/BugTracker/Tests/bogus_text_step.SF-1537500.alg
pathfinder/tests/XQuery/Tests/attr_err2
fail with
.../pathfinder/compiler/core/coreopt.brg:441: reduce: Assertion `rule' failed.
cf.
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/bogus_text_step.SF-1537500.mps.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/bogus_text_step.SF-1537500.mps.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/bogus_text_step.SF-1537500.alg.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/bogus_text_step.SF-1537500.alg.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder_algebra/.mTests103/GNU.64.64.d-Fedora6/tests_XQuery/attr_err2.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_XQuery/attr_err2.err.00.html
## Comment 8746
Date: 2008-01-22 15:41:25 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Fixed in CVS.
A new state label for the empty sequence constructor (introduced in a bogus-step-rewrite) was missing.
## Comment 8747
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1877300 at http://sourceforge.net/support/tracker.php?aid=1877300
| XQ: reduce: Assertion `rule' failed | https://api.github.com/repos/MonetDB/MonetDB/issues/1551/comments | 0 | 2020-11-30T09:46:11Z | 2024-06-27T11:28:03Z | https://github.com/MonetDB/MonetDB/issues/1551 | 753,322,824 | 1,551 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-21 10:48:13 +0100
From: Bob Ippolito <<etrepum>>
To: @njnes
Version: -- development
Last updated: 2008-03-14 02:25:30 +0100
## Comment 8741
Date: 2008-01-21 22:48:13 +0100
From: Bob Ippolito <<etrepum>>
MonetDB SR5 behaves unexpectedly when we've tried to use COALESCE with
aggregates, here's an example:
sql>CREATE TABLE tmp(a int);
sql>SELECT SUM(a) FROM tmp;
+-------+
| sum_a |
+=======+
| 0 |
+-------+
sql>SELECT MIN(a) FROM tmp;
+-------------+
| min_a |
+=============+
| null |
+-------------+
sql>SELECT COALESCE(MIN(a), 0) FROM tmp;
sql>
For the first SELECT I would expect NULL, the second SELECT is okay,
but I expect a result for the third!
Here's what PosgreSQL does:
bob= CREATE TABLE tmp(a int);
CREATE TABLE
bob= SELECT SUM(A) FROM tmp;
sum
-----
(1 row)
bob= SELECT MIN(a) FROM tmp;
min
-----
(1 row)
bob= SELECT COALESCE(MIN(a), 0) FROM tmp;
coalesce
----------
0
## Comment 8742
Date: 2008-02-20 11:29:44 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
The coalesce part of the problem is fixed now .
Test added to src/test/BugTracker/Tests/case_with_aggr.SF-1876779.sql
The sum result of 0 for empty columns is still open.
## Comment 8743
Date: 2008-03-14 14:25:29 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
fixed the sum implementation.
## Comment 8744
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1876779 at http://sourceforge.net/support/tracker.php?aid=1876779
| Unexpected results with COALESCE and aggregates | https://api.github.com/repos/MonetDB/MonetDB/issues/1550/comments | 0 | 2020-11-30T09:46:08Z | 2024-06-27T11:28:02Z | https://github.com/MonetDB/MonetDB/issues/1550 | 753,322,787 | 1,550 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-21 05:11:13 +0100
From: @yzchang
To: @sjoerdmullender
Version: -- development
Last updated: 2008-01-23 05:31:32 +0100
## Comment 8737
Date: 2008-01-21 17:11:13 +0100
From: @yzchang
When the option "--enable-oid32" is used, build all modules of monet from scratch on Darwin 9 results in errors.
In dev branch when compiling module MonetDB:
cc1: warnings being treated as errors
/Users/jennie/monet/dev/MonetDB/src/gdk/gdk_align.mx: In function
'VIEWreset':
/Users/jennie/monet/dev/MonetDB/src/gdk/gdk_align.mx:623: warning:
format '%zu' expects type 'size_t', but argument 5 has type 'BUN'
/Users/jennie/monet/dev/MonetDB/src/gdk/gdk_align.mx:623: warning:
format '%zu' expects type 'size_t', but argument 6 has type 'BUN'
/Users/jennie/monet/dev/MonetDB/src/gdk/gdk_align.mx:623: warning:
format '%zu' expects type 'size_t', but argument 7 has type 'BUN'
In stable branch when compiling module pathfinder:
cc1: warnings being treated as errors
/Users/jennie/monet/stable/pathfinder/runtime/xrpc_server.mx: In
function 'getCallValues':
/Users/jennie/monet/stable/pathfinder/runtime/xrpc_server.mx:944:
warning: format '%zu' expects type 'size_t', but argument 6 has type
'oid'
/Users/jennie/monet/stable/pathfinder/runtime/xrpc_server.mx:999:
warning: format '%zu' expects type 'size_t', but argument 6 has type
'oid'
System info:
$ uname -a
Darwin leopard.mshome.net 9.1.0 Darwin Kernel Version 9.1.0: Wed Oct
31 17:46:22 PDT 2007; root:xnu-1228.0.2~1/RELEASE_I386 i386
Configure options:
--enable-debug
--enable-assert
--disable-optimize
--enable-bi1970-01-01 01:00:32 +0100
--enable-oid32
--enable-strict
Both errors disappear if the (redundant) option "--enable-oid32" is removed.
Although it doesn't make sense to pass both options "--enable-bi1970-01-01 01:00:32 +0100" and "--enable-oid32" to configure, it should not cause compilation problems.
I would suggest to just ignore "--enable-oid32" in such cases.
Regards,
Jennie
## Comment 8738
Date: 2008-01-22 11:03:26 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jennie,
I made sure (in the buildtools; Stable & Current) that we handle --enable-oid32 and set MONET_OID32 only with 64-bit builds.
Hence, could you please check, whether this solves your problem?
Stefan
## Comment 8739
Date: 2008-01-23 17:31:32 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: YES
Compile both stable branch and current branch with the same set of options now succeeds.
Hence, close the bug.
No test needed/Compilation issue
## Comment 8740
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1876580 at http://sourceforge.net/support/tracker.php?aid=1876580
| Compile MonetDB on Darwin 9: failed by --enable-oid32 | https://api.github.com/repos/MonetDB/MonetDB/issues/1549/comments | 0 | 2020-11-30T09:46:05Z | 2024-06-27T11:28:01Z | https://github.com/MonetDB/MonetDB/issues/1549 | 753,322,744 | 1,549 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-21 01:25:59 +0100
From: @yzchang
To: Jan Flokstra <<jflokstra>>
Version: -- development
Last updated: 2008-01-23 05:34:39 +0100
## Comment 8728
Date: 2008-01-21 13:25:59 +0100
From: @yzchang
Hello, starting Mserver with module pathfinder results in following errors. The same problem occurs with both stable brach and dev branch.
$ Mserver --dbinit="module(pathfinder);"
MonetDB Server v4.20.1
based on GDK v1.20.1
Copyright (c) 1993-2007, CWI. All rights reserved.
Compiled for i686-apple-darwin9.1.0/32bit with 32bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
!ERROR: pathfinder: when trying to load module 'pftijah':
!ERROR: DL_open: dlopen(/Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib,10) -> Error: dlopen(/Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib, 10): Symbol not found: _tj_extend_plane
!ERROR: Referenced from: /Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib
!ERROR: Expected in: flat namespace
!ERROR: .
!ERROR: TBL_openmod: DL_open: dlopen(/Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib, 10): Symbol not found: _tj_extend_plane
!ERROR: Referenced from: /Users/jennie/scratch/monet-install/stable/debug/lib/MonetDB4/lib/lib_pftijah.dylib
!ERROR: Expected in: flat namespace
!ERROR: .
!ERROR: TBL_loadmod: module load _pftijah cancelled.
!ERROR: moduleClient: module(_pftijah) load error.
MonetDB>exit();
!ERROR: interpret: unknown variable 'pf_wal'.
!ERROR: interpret_params: lock_set(param 1): evaluation error.
MonetDB>^D
I'm using Mac OS 10.5 on Intel Duo core.
$ uname -a
Darwin leopard.local 9.1.0 Darwin Kernel Version 9.1.0: Wed Oct 31 17:46:22 PDT 2007; root:xnu-1228.0.2~1/RELEASE_I386 i386
The following configure options are used:
--enable-debug
--enable-assert
--disable-optimize
--enable-strict
Regards,
Jennie
## Comment 8729
Date: 2008-01-21 14:55:04 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
side-note: we should try not to use flat namespace
## Comment 8730
Date: 2008-01-21 15:28:12 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: NO
This is a very strange error. The linker tries to find a symbol which should not be there because the function is INLINE.
$PFROOT/modules/pftijah/serialize_pftijah.mx:
INLINE oid tj_extend_plane(struct tijahContextStruct *tjctx) {
oid base = tjctx->tijahPre - tjctx->dbat_collPre.seqbase;
if ( base >= tjctx->dbat_collPre.raw_max ) {
if ( dbat_extend(&tjctx->dbat_collPre,base, 0) < 0 )
return oid_nil;
/* IMPORTANT: the size of the two bats is synchronized by the use
* of the forced size (last) parameter of dbat_extend
*/
if ( dbat_extend(&tjctx->dbat_collSize,base,tjctx->dbat_collPre.raw_max) < 0 )
return oid_nil;
}
return tjctx->tijahPre++;
}
My first reaction there must be an error in the Darwin INLINE function handling. Maybe you could replace INLINE by static to see if this works. If so the INLINE function handling in Darwin has a bug.
## Comment 8731
Date: 2008-01-21 15:29:23 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: YES
I see that pftijah can be at least be loaded on darwin 8, so this might be a darwin 9 specific problem...
## Comment 8732
Date: 2008-01-21 15:35:18 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
The linker from Xcode 3.1 (only available for Leopard) is completely different than the linker from the last Xcode for Tiger. On Tiger a prelease of this linker was available as ld64, which is distinct from ld, whereas on Leopard ld == ld64. This ld64 linker is more strict, and more precise on where what is wrong.
I'd have to look into it into detail to figure out what it is, and how to fix it.
## Comment 8733
Date: 2008-01-21 16:36:32 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
There are many more INLINEd functions in the MonetDB code base (that seem to work fine?); hence, comparing this one to all the other ones --- consider the way it is declared, how it's called, how the code is compiled, how the object/library is built, how the object/library is linked-in where it is called, how the shared library (module) is loaded, etc. --- might shed some light on this problem...
## Comment 8734
Date: 2008-01-22 23:18:12 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
I declared INLINEd function tj_extend_plane() static (just like all other INLINEd functions in this file).
Please check, whether it helps ...
## Comment 8735
Date: 2008-01-23 17:34:39 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: YES
Yes, it helps. Reported error is gone in both stable branch and current branch. Thanks!
Close the bug.
No test needed/compilation issue
## Comment 8736
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1876422 at http://sourceforge.net/support/tracker.php?aid=1876422
| PF: could not load module pftijah on Darwin 9 | https://api.github.com/repos/MonetDB/MonetDB/issues/1548/comments | 0 | 2020-11-30T09:46:01Z | 2024-06-27T11:28:00Z | https://github.com/MonetDB/MonetDB/issues/1548 | 753,322,712 | 1,548 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-16 03:27:55 +0100
From: @yzchang
To: Wouter Alink <<alink>>
Version: -- development
Last updated: 2008-01-16 11:10:29 +0100
## Comment 8722
Date: 2008-01-16 15:27:55 +0100
From: @yzchang
Querying fn:id() on my own xml document returns weird results.
The query doc("http://homepages.cwi.nl/~zhang/sheep.xml")/id("s1") returns empty result (there do exist a node with id "s1"), while the query doc("http://homepages.cwi.nl/~zhang/sheep.xml")/id("001") returns the whole document.
However, the query doc("http://www.functx.com/input/book.xml")/id("preface") returns correct result.
Maybe my document is incorrect?
Below is a copy of my query session. I'm using the current brach, recompiled this morning, but I got the same behaviour with the stable version.
Regards,
Jennie
==============
$ mclient -lx
xquery>doc("http://homepages.cwi.nl/~zhang/sheep.xml")/id("s1")
more><>
xquery>doc("http://homepages.cwi.nl/~zhang/sheep.xml")/id("001")
more><>
<?xml version="1.0" encoding="utf-8"?>
<sheepdb>
<sheep id="s1">
<name>Baby Sheep</name>
<price>100</price>
<stock>98</stock>
<picture>pictures/BabySheep.jpg</picture>
<description>
My favourite sheep with extreme high cuddle factor
</description>
</sheep>
<sheep id="s2">
<name>Lost Sheep</name>
<price>1234</price>
<stock>50</stock>
<picture>pictures/DSC01689.jpg</picture>
<description>
Oops, wrong site ;-)
</description>
</sheep>
<sheep id="001">
<name>Baby Sheep</name>
<price>100</price>
<stock>98</stock>
<picture>pictures/BabySheep.jpg</picture>
<description>
My favourite sheep with extreme high cuddle factor
</description>
</sheep>
<sheep id="002">
<name>Lost Sheep</name>
<price>1234</price>
<stock>50</stock>
<picture>pictures/DSC01689.jpg</picture>
<description>
Oops, wrong site ;-)
</description>
</sheep>
</sheepdb>
xquery>doc("http://www.functx.com/input/book.xml")/id("preface")
more><>
<section id="preface">This book introduces XQuery...
The examples are downloadable<fnref ref="fn1"/>...
</section>
xquery>
## Comment 8723
Date: 2008-01-16 22:33:41 +0100
From: Wouter Alink <<alink>>
Logged In: YES
user_id=621590
Originator: NO
hello jenny,
the results seem to me to be completely as one would expected.
The 'id(xs:string)' function is not per definition tied to a 'id' attribute. normally you have to specify in a dtd or xmlschema which attribute name denotes a 'id' or 'idref' relation. MonetDB had (at least a while back) a compile-time option in the shredder denoting all id-attribute nodes to be 'id's.
Therefore your first query doesn't give you a result.
The second query has another problem. Your query is malformed. 'id's may never start with a digit. However, this fact has been misused in monetdb. monetdb _does_ allow the parameter given to the "id(xs:string)" function to start with a digit. When it detects the first character of the string to be a digit, it will try to convert the value to a number, and will return the node with the corresponding 'nid' value (an internal numbering scheme of monetdb's xml representation).
Therefore your second query will return the document root of the first document in your query. (When you would have queried id("0"), you probably would have gotten the whole collection) This feature is specific to monetdb. "pf:nid() as xs:string" is the inverse function. for example: 'pf:nid(doc("x.xml")/id("1"))' should return 1
i hope this helps (and that i explained it correctly :)
p.s. as i am not a developer i cant change the status of this bugreport.
## Comment 8724
Date: 2008-01-16 22:37:57 +0100
From: Wouter Alink <<alink>>
Logged In: YES
user_id=621590
Originator: NO
p.s. i cannot explain why the third query did return a result. i would have expected an empty result.
## Comment 8725
Date: 2008-01-16 23:10:29 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
The third query (correctly) returns the expected result as the respective document does have a(n inlined) DTD.
Hence, given the documents, all three queries work as expected and correctly.
## Comment 8726
Date: 2008-01-17 11:07:57 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: YES
Hi Stefan en Wouter, thanks a lot for your explaination!
I also had the impression that the number "001" was treated as a NID, but could not understand why the third query does return the expected node, while the document seems similar with my document. The problem is that I opened book.xml in firefox and it didn't tell me that the document has a DTD!
## Comment 8727
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1872806 at http://sourceforge.net/support/tracker.php?aid=1872806
| PF: weird results of fn:id() | https://api.github.com/repos/MonetDB/MonetDB/issues/1547/comments | 0 | 2020-11-30T09:45:59Z | 2024-06-28T13:30:42Z | https://github.com/MonetDB/MonetDB/issues/1547 | 753,322,682 | 1,547 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-16 12:38:01 +0100
From: @swingbit
To: GDK devs <<bugs-common>>
Version: -- development
Last updated: 2008-02-13 03:12:37 +0100
## Comment 8711
Date: 2008-01-16 12:38:01 +0100
From: @swingbit
.../gdk/gdk_bat.mx:2501: HEAPsetmmap: Assertion `p' failed.
This happens while indexing a (not small) xml file in pftijah.
It is relatively new, it did not happen before Xmas (same data, same operations).
Not completely sure whether this is a M4 or a pftijah bug, but I saw some recent changes to gdk_heap..
I enclose the script used to index (a part of) wikipedia collection in pftijah.
To reproduce the bug:
- run the server with pathfinder module
- MODIFY the 'DST' variable in the script
- run the scrip
I am running the head version of everything, compiled with default options and 32bit OID's.
## Comment 8712
Date: 2008-01-21 16:47:11 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Indicate in the Summary, that this bug is triggered by Pathfinder; PFtijah to be precise.
## Comment 8713
Date: 2008-01-22 11:03:29 +0100
From: @swingbit
Logged In: YES
user_id=883451
Originator: YES
After some investigation,
I found out that the bug is triggered by the
commit in gdk_utils.mx, 1.200 -> 1.201, dated 14 Dec 2007:
- lower the threshold for creation of mmap-backed heaps to 64MB
[...]
Unfortunately this doesn't say much (to me, at least)
on the real cause of the problem.
Lowering the GDK_vm_minsize has probably had the effect of
triggering this bug earlier (i.e. with smaller data size)
than before the change.
About PF / PfTijah, it is indeed triggered by PfTijah and not by PF alone:
Server:
$ Mserver --dbinit="module(pathfinder);"
Client:
$ echo 'pf:add-doc("/export/scratch0/roberto/wikipedia_XML/part-0.xml", "part-0.xml")' | mclient -lx
[Succeeds]
$ echo "tijah:create-ft-index()" | mclient -lx
[Fails]
So, the bug shows up during PfTijah indexing.
## Comment 8714
Date: 2008-01-26 16:30:11 +0100
From: marspluto <<juntezhang>>
Logged In: YES
user_id=1992393
Originator: NO
I have encountered this problem when building an index. I am using a default
installation of MonetDB + Pathfinder module. Compiled for
x86_64-suse-linux-gnu/64bit with 64bit OIDs; dynamically linked. If anyone
knows a work-around for this, please let me know.
## Comment 8715
Date: 2008-02-10 12:58:26 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Just a quick update:
also after disabling GDKvmalloc for sizes that exceed GDK_mem_bigsize, this probelm persists on the latest Stable release branch (see below; segfault iso. assertion as assertions are disabled); i.e., the problem is indeed related to GDK_vm_minsize, not to GDK_mem_bigsize ...
========
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1115703632 (LWP 30275)]
0x00002aaaac6dafc4 in HEAPsetmmap (hp=0x1288c40, committed=32, remap=0x427f098c) at /ufs/manegold/_/scratch0/Monet/Testing/Stable/source/MonetDB/src/gdk/gdk_bat.mx:2536
2536 *p = 0; /* cut off .priv */
2517 /* transition heap from writable (or appendable) to readonly */
2518 static int
2519 HEAPsetmmap(Heap *hp, int committed, int *remap)
2520 {
2521 assert(hp);
2522
2523 if (hp->base == NULL || hp->storage == STORE_MEM || hp->filename == NULL || !committed) {
2524 return hp->storage;
2525 }
2526 *remap = TRUE;
2527 if (hp->storage == STORE_MMAP) {
2528 /* state(2), goto state (3) */
2529 long_str src, dst;
2530 str p = strstr(hp->filename, ".priv");
2531 int ret;
2532
2533 assert(p); /* absence of .priv would indicate an illegal heap state */
2534
2535 GDKfilepath(src, BATDIR, hp->filename, NULL);
2536 *p = 0; /* cut off .priv */
2537 GDKfilepath(dst, BATDIR, hp->filename, NULL);
(gdb) print hp->filename
$3 = 0x1b895d8 "07/743.tail"
(gdb) print p
$4 = 0x0
(gdb) print hp->storage
$5 = 1 '\001'
========
## Comment 8716
Date: 2008-02-10 13:00:52 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
some more debug info:
========
(gdb) print committed
$6 = 32
(gdb) print *hp
$7 = {maxsize = 268435456, free = 153429276, size = 268435456, base = 0x2aaafb320000 "FII\002EII\002", filename = 0x1b895d8 "07/743.tail", storage = 1 '\001', copied = 0 '\0', newstorage = 1 '\001', dirty = 0 '\0'}
========
## Comment 8717
Date: 2008-02-10 15:04:39 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Increasing vm_minsize to <=512 MB (`Mserver --set gdk_vm_minsize=$[512*1024*1024]`) does NOT avoid the assertion / segfault.
Increasing vm_minsize to >=768 MB (`Mserver --set gdk_vm_minsize=$[768*1024*1024]`) DOES avoid the assertion / segfault.
## Comment 8718
Date: 2008-02-11 01:36:35 +0100
From: @peterboncz
Logged In: YES
user_id=591107
Originator: NO
Recent lowering te gdk_vm_minsize threshold, that determines when BAT heap allocs
(due to a new or extending BAT) start directly mapping on disk,
exposed this bug that has been there for 4 years already.
The gdk_vm_minsize threshold is now 64MB on Linux (it is higher on Windows,
because of high cost).
Direct-mmaps usually are only allowed for non-writable bats, but in this
case we get it in a temporary, but non-persistent bat.
Such cases are (DIRTY) recognized by filename extension: the mode
is STORE_MMAP, it the filename ends with .priv.
The bug occurred when making the bat non-writable again. The assert
goes off because it expects the .priv extension, but it is gone.
Turns out it was gone for two reasons:
(1) the initial HEAPalloc/HEAPextend failed to attach the privext
extension in the first place
(2) on a subsequent HEAPextend, the .priv extension would be
lost due to the behavior of decompose_filename()
## Comment 8719
Date: 2008-02-11 08:36:31 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Peter,
Thanks for the fix. Works fine for me.
Roberto.
could you please check and report, whether the latest Stable release branches also work for you, and if so, close this bug report?
Unfortunately, Roberto's example is to big / resource demanding to be added as test to CVS.
Hence, unless we find a "reasonable" test case, we might have to close this one without having a test monitoring it...
## Comment 8720
Date: 2008-02-13 15:12:37 +0100
From: @swingbit
Logged In: YES
user_id=883451
Originator: YES
The assertion doesn't pop up anymore in my test, so I'll close this bug report.
## Comment 8721
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1872685 at http://sourceforge.net/support/tracker.php?aid=1872685
The original assignee of this bug does not have
an account here. Reassigning to the default assignee
for the component, bugs-common@monetdb.org.
Previous assignee was nobody@users.sourceforge.net.
| PF(tijah): HEAPsetmmap: Assertion `p' failed. | https://api.github.com/repos/MonetDB/MonetDB/issues/1546/comments | 0 | 2020-11-30T09:45:56Z | 2024-06-27T11:27:58Z | https://github.com/MonetDB/MonetDB/issues/1546 | 753,322,644 | 1,546 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-15 09:38:29 +0100
From: Wouter Alink <<alink>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-01-22 02:19:33 +0100
## Comment 8706
Date: 2008-01-15 09:38:29 +0100
From: Wouter Alink <<alink>>
the following query returns 'fatal error: incorrect comparison' using "pf -A" (on a "--nightly=current" of last week)
"""
for $x in doc("x.xml")//a
return
if (exactly-one($x/a/@c) != exactly-one($x/b/@c)) then
$x
else
()
"""
## Comment 8707
Date: 2008-01-15 10:29:33 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
The reason is that the MIL thetajoin can not cope with not equal predicates I fixed the planner to generate a different physical thetajoin that evaluates all inequality as normal MIL selections.
## Comment 8708
Date: 2008-01-15 11:41:01 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Re-opened to remind us that we should add a test (also) for this one.
(Should be trivial ...)
## Comment 8709
Date: 2008-01-22 14:19:33 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Added test (compiler only) in
pathfinder/tests/BugTracker/Tests/uncompilable_comparison.SF-1871790.*
## Comment 8710
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1871790 at http://sourceforge.net/support/tracker.php?aid=1871790
| XQ: uncompilable comparison | https://api.github.com/repos/MonetDB/MonetDB/issues/1545/comments | 0 | 2020-11-30T09:45:53Z | 2024-06-27T11:27:57Z | https://github.com/MonetDB/MonetDB/issues/1545 | 753,322,624 | 1,545 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-15 09:35:48 +0100
From: Wouter Alink <<alink>>
To: Wouter Alink <<alink>>
Version: -- development
Last updated: 2008-01-22 02:20:08 +0100
## Comment 8698
Date: 2008-01-15 09:35:48 +0100
From: Wouter Alink <<alink>>
The following query did not compile (on a "--nightly=current" build of last week) using "pf -A"
element { "a" }{
for $a in doc("x.xml")//a
let $m := distinct-values($a//d)
for $q in distinct-values($a//c[d])
for $r in $m
where some $x in $a//c satisfies $x[e/d[contains(.,$r)] and a=$q]
return element { "a" } { $a//c }
}
the query returns: "fatal error: attribute `(NULL)' referenced in selection not found"
however it does compile if the query is rewritten to: (in which $m is substituted):
element { "a" }{
for $a in doc("x.xml")//a
for $q in distinct-values($a//c[d])
for $r in distinct-values($a//d)
where some $x in $a//c satisfies $x[e/d[contains(.,$r)] and a=$q]
return element { "a" } { $a//c }
}
## Comment 8699
Date: 2008-01-15 11:39:17 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Fixed in CVS. This bug was due to a bug in the thetajoin optimization phase.
@Wouter: For our compiler the two queries are different as 'distinct-values($a//d)' is evaluated in different scopes ($a vs $q).
## Comment 8700
Date: 2008-01-15 11:41:24 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Re-opened to remind us that we should add a test (also) for this one.
(Should be trivial ...)
## Comment 8701
Date: 2008-01-15 12:48:23 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Fixed in CVS. This bug was due to a bug in the thetajoin optimization phase.
@Wouter: For our compiler the two queries are different as 'distinct-values($a//d)' is evaluated in different scopes ($a vs $q).
## Comment 8702
Date: 2008-01-15 12:52:57 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
hmm sf still has the back button problems...
Wouter could you please provide the XML document and the output (perhaps even add the test :-)).
## Comment 8703
Date: 2008-01-16 22:21:11 +0100
From: Wouter Alink <<alink>>
Logged In: YES
user_id=621590
Originator: YES
i don't have a document for the query (i thought it was just a query-compiler issue, which i narrowed down to the mentioned query. could we at a test in pathfinder/src/compiler (only checking compilation). (same goes for the other bug)
lately i haven't checked out and compiled monetdb from cvs (i try to use nightly builds as much as possible). jan, if you could just add the query to the repository (as a compilation test), i would very much appreciate it.
## Comment 8704
Date: 2008-01-22 14:20:08 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Added test (compiler only) in
pathfinder/tests/BugTracker/Tests/substitution_fails.SF-1871789.*
## Comment 8705
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1871789 at http://sourceforge.net/support/tracker.php?aid=1871789
| XQ: substitution fails | https://api.github.com/repos/MonetDB/MonetDB/issues/1544/comments | 0 | 2020-11-30T09:45:51Z | 2024-06-27T11:27:56Z | https://github.com/MonetDB/MonetDB/issues/1544 | 753,322,605 | 1,544 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-15 07:10:17 +0100
From: @drstmane
To: Stratos Idreos <<idreos>>
Version: -- development
Last updated: 2008-01-21 04:43:22 +0100
## Comment 8694
Date: 2008-01-15 07:10:17 +0100
From: @drstmane
For details, see
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Makes3/index_short.html
and in particular
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.GNU.32.32.d-CentOS4.4/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.GNU.32.32.d-Darwin8.11.0/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.GNU.64.64.s-Fedora6/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.64.32.d-Fedora6/make-k.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.64.64.d-Fedora6/make-k.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.64.64.d-RedHat4WS/make-k.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Mic.32.32.d-Windows5.1/make.out.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Mic.64.64.d-Windows5.2/make.out.html
## Comment 8695
Date: 2008-01-17 10:16:12 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
or here:
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.64.32.d-Fedora6/make.out.extrerrmsg.htmll2806
(just noticed that my html pages got overwritten by the "newer" ones from testweb)
## Comment 8696
Date: 2008-01-21 16:43:22 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Seems to be fixed.
Closing.
## Comment 8697
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1871748 at http://sourceforge.net/support/tracker.php?aid=1871748
| M5: new crackers code fails to compile on various platforms | https://api.github.com/repos/MonetDB/MonetDB/issues/1543/comments | 0 | 2020-11-30T09:45:49Z | 2024-06-27T11:27:55Z | https://github.com/MonetDB/MonetDB/issues/1543 | 753,322,579 | 1,543 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-13 04:44:02 +0100
From: @drstmane
To: @sjoerdmullender
Version: -- development
Last updated: 2008-01-21 04:39:44 +0100
## Comment 8690
Date: 2008-01-13 16:44:02 +0100
From: @drstmane
Since checkins on Fri Jan 11 2008, mclient/MAPI only echo the first line of failing queries, while the whole query (or at least as much as fit in some internal buffer) was echoed before; cf.
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugDay_2005-12-19_0.9.3/zero-or-one_merged-union.SF-1221336.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugDay_2005-12-19_0.9.3/cast_as_anyNode.SF-1280519.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugDay_2005-12-19_0.9.3/type_error_in_cast.SF-1281176.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugTracker/call2builtin_function_crashes_XRPC_client.SF-1655857.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugTracker/type_check.SF-1705898.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-Darwin8.11.0/tests_BugTracker/crash_on_concatenated_query.SF-1730547.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugsViaSourgeforce/invisible_error_messages.SF-1409122.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_BugsViaSourgeforce/ID.1652527.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_XQuery/typing3.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_XQuery/matches01.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-CentOS4.4/tests_XQuery/matches02.err.00.html
In case the new behaviour is "desired" / "not easily avoided", we should simply approve the new output.
I suspect this to be related to one of the following checkins:
===================================================================
2008/01/11 - sjoerd:
clients/src/mapiclient/MapiClient.mx,1.103
Do not use a calculated offset with fseek. It is not portable.
Instead, we read one line at a time in XQuery mode to find lines that
contain only "<>".
===================================================================
===================================================================
2008/01/11 - tsheyar: pathfinder/compiler/parser/scanner.l,1.55
-- Do not reset the error buffer inside the parser.
(Messages generated by main.c get swallowed this way.)
===================================================================
## Comment 8691
Date: 2008-01-13 16:53:17 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
It's definitely due to my checkin to MapiClient.mx. Echoing the failed query is done completely within mclient (MapiLib actually). The difference is due to the fact that XQuery queries are now read line by line instead of wholesale because we can't use fseek to seek back to the first occurrence of <> in the data.
We can indeed approve the changed output, but we can also change MapiLib to remember more of the query. Let's do the approval.
## Comment 8692
Date: 2008-01-21 16:39:44 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Fixed by Sjoerd in CVS by approving the new output as desired and correct.
## Comment 8693
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1870379 at http://sourceforge.net/support/tracker.php?aid=1870379
| XQ: mclient/MAPI only echo first line of failing query | https://api.github.com/repos/MonetDB/MonetDB/issues/1542/comments | 0 | 2020-11-30T09:45:46Z | 2024-06-27T11:27:54Z | https://github.com/MonetDB/MonetDB/issues/1542 | 753,322,550 | 1,542 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-08 05:47:11 +0100
From: @lsidir
To: Torsten Grust <<teggy>>
Version: -- development
Last updated: 2008-01-09 10:22:41 +0100
## Comment 8687
Date: 2008-01-08 17:47:11 +0100
From: @lsidir
When compiling with "warnings being treated as errors" I get the following message:
/home/lsidir/develop//pathfinder/src/sqlhelpers/xmlshred/encoding.c: In function ‘start_element’:
/home/lsidir/develop//pathfinder/src/sqlhelpers/xmlshred/encoding.c:516: warning: field precision should have type ‘int’, but argument 7 has type ‘long int’
/home/lsidir/develop//pathfinder/src/sqlhelpers/xmlshred/encoding.c:525: warning: field precision should have type ‘int’, but argument 7 has type ‘long int’
I guess this has to do with today's commit from Torsten Grust, but I didn't want (and didn't know how) to mess with this part of the code so I opened a bug.
## Comment 8688
Date: 2008-01-09 10:22:41 +0100
From: @Teggy
Logged In: YES
user_id=642632
Originator: NO
Thanks, Lefteris. I did not see this warning on
my platform (Mac OS X 10.5). Bad, bad.
This should be fixed now.
Cheers,
--Torsten
## Comment 8689
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1866944 at http://sourceforge.net/support/tracker.php?aid=1866944
| Failed to compile current pathfinder | https://api.github.com/repos/MonetDB/MonetDB/issues/1541/comments | 0 | 2020-11-30T09:45:43Z | 2024-06-27T11:27:53Z | https://github.com/MonetDB/MonetDB/issues/1541 | 753,322,523 | 1,541 |
[
"MonetDB",
"MonetDB"
] | Date: 2008-01-07 04:32:21 +0100
From: @romulogoncalves
To: Romulo Goncalves <<goncalve>>
Version: -- development
Last updated: 2008-01-07 06:12:44 +0100
## Comment 8684
Date: 2008-01-07 16:32:21 +0100
From: @romulogoncalves
If I start a mclient and I execute the query:
create local temp table t1(id int);
and then I restart my mclient the mserver chrashes.
>[New Thread 1107310928 (LWP 18319)]
mserver5: /ufs/goncalve/MonetDB/sql/src/storage/store.mx:2922: sys_drop_table: Assertion `rid != -1' failed.
Program received signal SIGABRT, Aborted.
[Switching to Thread 1107310928 (LWP 18319)]
0x0000003c4bc30ec5 in raise () from /lib64/libc.so.6
(gdb)
Regards,
Romulo
## Comment 8685
Date: 2008-01-07 18:12:44 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The bug can be closed. The local temp tables are destroyed once the client closes its session so we do not need to clean them again once we start a new session (restart session). The fix is in the check in:
[Monetdb-sql-checkins] sql/src/storage store.mx,,1.4,1.5
Log Message:
The local temp tables where already removed when the user closed its session. So during the session_reset, called in the new session, the local temp tables do not need to removed again.
The bug is covered by the test:
sql/src/test/BugTracker/Tests/local_temp_table.SF-1865953
## Comment 8686
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1865953 at http://sourceforge.net/support/tracker.php?aid=1865953
| local temp table | https://api.github.com/repos/MonetDB/MonetDB/issues/1540/comments | 0 | 2020-11-30T09:45:41Z | 2024-06-27T11:27:52Z | https://github.com/MonetDB/MonetDB/issues/1540 | 753,322,503 | 1,540 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-21 09:16:02 +0100
From: mharrison <<matt_harrison>>
To: @njnes
Version: -- development
Last updated: 2008-01-27 10:55:29 +0100
## Comment 8676
Date: 2007-12-21 21:16:02 +0100
From: mharrison <<matt_harrison>>
I've got a query that returns some 260K rows. Every time I run it, my memory use goes up 10-20Megs (and doesn't come back down).
Am attaching valgrind logs for the same query (but returns a smaller set). It is showing leakage.
## Comment 8677
Date: 2007-12-21 21:46:35 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
Although your valgrind output indeed shows we leak it doesn't tell where. Probably you didn't run with memcheck and all of the required flags. Could you re-run valgrind?
## Comment 8678
Date: 2007-12-30 20:40:24 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
we fixed a number of leaks in the cvs head version. Could you re-run your experiments with that version? I'm both interested in known if the leaking stopped and if the performance is better on the head version.
## Comment 8679
Date: 2008-01-09 00:47:26 +0100
From: mharrison <<matt_harrison>>
Logged In: YES
user_id=1164236
Originator: YES
Hey Niels, I can't get the latest current to build. Using this command:
monetdb-install.sh --nightly=current --enable-sql
I get the following error:
...
unpacking MonetDB-client.tar.gz
mv: cannot stat `clients-[0-9]*': No such file or directory
failed to rename sources to the right name
Please refer to the full log at /var/tmp/MonetDB-1199834607/log
If you believe this is an error in the script or the software,
file a bug and attach the logfile.
Terminated
## Comment 8680
Date: 2008-01-09 09:53:29 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
Please try again with the latest monetdb-install.sh script from http://www.cwi.nl/~fabian/troep/monetdb-install.sh
## Comment 8681
Date: 2008-01-13 16:30:57 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
we fixed an leak in the cvs head version. Could you re-run your
experiments with that version? I'm both interested in known if the leaking
stopped and if the performance is better on the head version.
## Comment 8682
Date: 2008-01-27 10:55:29 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
On behalf of the Niels/Harrison, this bug can be closed.
Many leakage problems have been resolved and no
new complaint has been received.
The testweb does not need additional tests.
## Comment 8683
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1856038 at http://sourceforge.net/support/tracker.php?aid=1856038
| Memory leaks when querying | https://api.github.com/repos/MonetDB/MonetDB/issues/1539/comments | 0 | 2020-11-30T09:45:38Z | 2024-06-27T11:27:51Z | https://github.com/MonetDB/MonetDB/issues/1539 | 753,322,476 | 1,539 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-21 09:10:43 +0100
From: mharrison <<matt_harrison>>
To: @njnes
Version: -- development
Last updated: 2008-01-27 10:55:56 +0100
## Comment 8668
Date: 2007-12-21 21:10:43 +0100
From: mharrison <<matt_harrison>>
While attempting to load 1Mil rows into a table memory use goes 700+Megs.
Loading the same tables in My/postgre/sql doesn't ever take more than 20Megs!
After bulk load is done memory comes down a bit (100M or so) but still stays high.
Am attaching valgrind log.
## Comment 8669
Date: 2007-12-22 09:21:04 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
Are you referring to the on-disk size?
or the memory consumption as seen by 'top'
Note that the memory consumption of the ^sql
are based on the memory buffer pool set aside
for processing. Not the table sizes.
In MonetDB you see the table size as a large
memory mapped file.
## Comment 8670
Date: 2007-12-22 14:42:58 +0100
From: mharrison <<matt_harrison>>
Logged In: YES
user_id=1164236
Originator: YES
Memory consumed by ram/top.
Memory get's mmapped during bulk load? Then there's no way I'm every going to get 100+Mil rows loaded.
## Comment 8671
Date: 2007-12-22 14:56:11 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
If you are running on a 32-bit
you might indeed run into problems :(
And may have to perform an incremental
load, such that the individual columns
fit your machine limits.
We successfully loaded a 100 Gbyte
snapshot of the SkyServer database
and are awaiting the hardware for
the 3TB load.
## Comment 8672
Date: 2007-12-22 17:47:42 +0100
From: mharrison <<matt_harrison>>
Logged In: YES
user_id=1164236
Originator: YES
Am running 32bit, but can run 64bit.
How much mem was required to load 100G snapshot? Did you have to load it in steps?
Am still confused as I thought bulk load would only put the data in the filesystem. Again my|postgres/sql use no more than 20M when doing the same bulk load....
Yes I care about performance and if you are going to give it to me during queries, excellent. But using up all the memory for bulk loading seems like an architectural flaw to me (a simple minded guy, who perhaps doesn't understand the benefits of it).
## Comment 8673
Date: 2007-12-30 20:42:19 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
we now only need 400MB. Not sure yet why we need way more then the on disk size (which is about 180MB).
## Comment 8674
Date: 2008-01-27 10:55:56 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
On behalf of the Niels/Harrison, this bug can be closed.
Many leakage problems have been resolved and no
new complaint has been received.
The testweb does not need additional tests.
## Comment 8675
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1856034 at http://sourceforge.net/support/tracker.php?aid=1856034
| Memory leaks when bulk loading | https://api.github.com/repos/MonetDB/MonetDB/issues/1538/comments | 0 | 2020-11-30T09:45:36Z | 2024-06-27T11:27:50Z | https://github.com/MonetDB/MonetDB/issues/1538 | 753,322,447 | 1,538 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-21 12:48:29 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2008-01-05 12:38:42 +0100
## Comment 8664
Date: 2007-12-21 12:48:29 +0100
From: @romulogoncalves
The following query was crashing the M5-SQL server:
SELECT xmlelement(name element,
xmlattributes ('unnamed and wrong'));
The crash is gone after the Niels check in:
[Monetdb-sql-checkins] sql/src/server sql_parser.mx, SQL_2-20, 1.256.2.6, 1.256.2.7
Modified Files:
Tag: SQL_2-20
sql_parser.mx
Log Message:
fixed bug in xml attribute parsing.
However, the select output is not correct. So the bug will continue open.
Note: This is a mix of SQL and XML so I hope that I am opening the bug in the correct place. If not please correct me.
## Comment 8665
Date: 2007-12-21 13:17:09 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Test added in:
sql/src/BugTracker/Tests/xml_attribute_parsing.SF-1855673.sql
## Comment 8666
Date: 2008-01-05 12:38:42 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
fixed bug in handling non-named attributes. Also fixed a bug in empty elements.
## Comment 8667
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1855673 at http://sourceforge.net/support/tracker.php?aid=1855673
| bug in xml attribute parsing | https://api.github.com/repos/MonetDB/MonetDB/issues/1537/comments | 0 | 2020-11-30T09:45:33Z | 2024-06-27T11:27:49Z | https://github.com/MonetDB/MonetDB/issues/1537 | 753,322,416 | 1,537 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-20 11:26:58 +0100
From: @drstmane
To: @sjoerdmullender
Version: -- development
Last updated: 2007-12-21 10:34:34 +0100
## Comment 8661
Date: 2007-12-20 11:26:58 +0100
From: @drstmane
MonetDB server v5.3.0, based on kernel v1.21.0
Serving database 'mTests'
Compiled for i686-pc-win32/32bit with 32bit OIDs dynamically linked
Copyright (c) 1993-2007 CWI, all rights reserved
Visit http://monetdb.cwi.nl/ for further information
!MALException:MAL.getAddress:address of 'mmath.MATHunary_COTflt' not found
!MALException:MAL.getAddress:address of 'mmath.MATHunary_COTdbl' not found
Listening for connection requests on mapi:monetdb://tureluur:34269/
function user.main():void;
clients.quit();
end main;
see also
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.Int.32.32.d-Windows5.1/test10.out
## Comment 8662
Date: 2007-12-21 10:34:34 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Fixed.
## Comment 8663
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1854753 at http://sourceforge.net/support/tracker.php?aid=1854753
| mserver 5 fails to run on Windows: missing export? | https://api.github.com/repos/MonetDB/MonetDB/issues/1536/comments | 0 | 2020-11-30T09:45:31Z | 2024-06-27T11:27:48Z | https://github.com/MonetDB/MonetDB/issues/1536 | 753,322,401 | 1,536 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-19 05:14:59 +0100
From: Wouter Alink <<alink>>
To: Wouter Alink <<alink>>
Version: -- development
Last updated: 2007-12-21 04:23:39 +0100
## Comment 8657
Date: 2007-12-19 17:14:59 +0100
From: Wouter Alink <<alink>>
when doing many small updates in a single query on a single document, a (very) large amount of memory is used.
this can be reproduced by rewriting the many subsequenced-queries in the script from http://sourceforge.net/support/tracker.php?aid=1854215 to a single query. MonetDB will then use at least 11GB (with oid64) of memory (that's when i killed mserver, i don't know if it will use more). the source document in which the inserts are being made is about 1M in size (plain text). peter and lefteris have seen this happen and might have more info on this.
i guess it is a separate bug (or feature).
note: the work-around "chopping the query into statements with less inserts per query" works for me.
## Comment 8658
Date: 2007-12-21 15:08:08 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
The insert queries are all of the form:
let $todo := subsequence(doc("prepared5123.data")/XIRAF/item[*], 0,128)
let $dest := for $i in doc("test9.xml")//*[@xid]
where some $k in $todo/@xid satisfies $k = $i/@xid
return $i
for $item in $todo
return
do insert $item/*
into exactly-one($dest[@xid[. = $item/@xid]])
If you want to do them all at once, it becomes:
let $todo := doc("prepared5123.data")/XIRAF/item[*]
let $dest := for $i in doc("test9.xml")//*[@xid]
where some $k in $todo/@xid satisfies $k = $i/@xid
return $i
for $item in $todo
return
do insert $item/*
into exactly-one($dest[@xid[. = $item/@xid]])
However, why do you use this complex query. Isn't this query equivalent to:
let $todo := doc("prepared5123.data")/XIRAF/item[*]
for $item in $todo
return
do insert $item/*
into exactly-one(doc("test9.xml")//*[@xid = $item/@xid])
The advantage of this last query is that it runs in finite time using a finite amount of memory (it completes on my 2GB system within 4 minutes.
The problem with the original complete query is in the actual querying part, i.e. before it gets to updating the document.
Wouter, please comment and then reassign.
## Comment 8659
Date: 2007-12-21 16:23:39 +0100
From: Wouter Alink <<alink>>
Logged In: YES
user_id=621590
Originator: YES
hello sjoerd, thanks for looking at this. you are right. i guess i added it during debugging. i will close this bug.
## Comment 8660
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1854224 at http://sourceforge.net/support/tracker.php?aid=1854224
| XQ: large memory usage during multiple inserts | https://api.github.com/repos/MonetDB/MonetDB/issues/1535/comments | 0 | 2020-11-30T09:45:28Z | 2024-06-28T13:30:41Z | https://github.com/MonetDB/MonetDB/issues/1535 | 753,322,365 | 1,535 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-19 05:03:24 +0100
From: Wouter Alink <<alink>>
To: @sjoerdmullender
Version: -- development
Last updated: 2008-01-03 04:46:36 +0100
## Comment 8652
Date: 2007-12-19 17:03:24 +0100
From: Wouter Alink <<alink>>
a recent nightly snapshot (last week, dec 13th) of the stable pathfinder 0.20 branch still has problems when doing many inserts. the bug message is similar to the "immune for updates" bug, namely: CMDfind: value not found.
the attached xquery script reproduces this bug (at least on a fedora core 5, 64bit machine).
## Comment 8653
Date: 2007-12-21 13:35:13 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
This bug is fixed.
The test is too large to add, but I don't have a better one. I think the trick is to create a document that ends exactly at a page boundary and then trying to insert a node at the end.
## Comment 8654
Date: 2008-01-03 14:23:20 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
While coming up with a test for this bug (see attachment--will add to repository), I found another off-by-one bug with the same symptoms: CMDfind: value not found.
The test creates a document which complete fills exactly one page and then adds a node to the end.
I'm looking into this.
File Added: insert-new-page.SF-1854215.xq.in
## Comment 8655
Date: 2008-01-03 16:46:36 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Fixed, and added a test which should test both versions of this bug: pathfinder/tests/BugTracker insert-new-page.SF-1854215
## Comment 8656
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1854215 at http://sourceforge.net/support/tracker.php?aid=1854215
| XQ: another insert bug | https://api.github.com/repos/MonetDB/MonetDB/issues/1534/comments | 0 | 2020-11-30T09:45:25Z | 2024-06-27T11:27:46Z | https://github.com/MonetDB/MonetDB/issues/1534 | 753,322,335 | 1,534 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-18 05:20:22 +0100
From: @peterboncz
To: @sjoerdmullender
Version: -- development
Last updated: 2007-12-19 10:43:46 +0100
## Comment 8649
Date: 2007-12-18 17:20:22 +0100
From: @peterboncz
the file pathfinder/runtime/xrpc/export/xrpc.html
should also be installed on windows.
(this is the XRPC demo page)
it is mentioned in the documentation, and is/was part of the .20 release (on Linux it is)
## Comment 8650
Date: 2007-12-19 10:43:46 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Fixed by fixing the bug in Makefile.ag where the new file xrpc.html was added in the wrong place and by adding the file (and xrpc_mod.xq) to the installer.
No test needed.
## Comment 8651
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1853242 at http://sourceforge.net/support/tracker.php?aid=1853242
| XQ: XRPC test page missing | https://api.github.com/repos/MonetDB/MonetDB/issues/1533/comments | 0 | 2020-11-30T09:45:22Z | 2024-06-27T11:27:45Z | https://github.com/MonetDB/MonetDB/issues/1533 | 753,322,304 | 1,533 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-17 05:39:45 +0100
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2007-12-21 01:04:11 +0100
## Comment 8643
Date: 2007-12-17 17:39:45 +0100
From: @drstmane
Since Nov 15 2007 (i.e., after checkins on Nov 14 2007) PF/XQ tests
pathfinder/modules/pftijah/Tests/cleanup.milS
pathfinder/tests/BugTracker/Tests/pf_add-doc_property_error.SF-1706438.xq
fail with
Mserver: .../MonetDB/src/gdk/gdk_bbp.mx:1284: incref: Assertion `BBP[(((int)(i))>0?((int)(i)):-((int)(i)))].refs+BBP[(((int)(i))>0?((int)(i)):-((int)(i)))].lrefs' failed.
Cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/modules_pftijah/cleanup.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/pf_add-doc_property_error.SF-1706438.err.00.html
Since there was only one Pathfinder checkin on Nov 14 2007 by Niels related to his GDK_init and (un)aligned memory changes in GDK on the same day, I (for now) assume this a GDK bug, and assign it to Niels.
I haven't done any further analysis, yet.
## Comment 8644
Date: 2007-12-17 17:43:43 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Better look at (e.g.)
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests10
3/GNU.32.32.d-Fedora6/modules_pftijah/cleanup.err.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests10
3/GNU.32.32.d-Fedora6/tests_BugTracker/pf_add-doc_property_error.SF-1706438
.err.00.html
as the GNU-compiled 64-bit, 64-bit OID version of Pathfinder (also when compiled with optimization enabled) seems to suffer from the gcc bug reported in
[ 1830366 ] gcc 4.1/4.2 produces wrong code on FC6 & RHEL4WS
https://sourceforge.net/tracker/index.php?func=detail&aid=1830366&group_id=56967&atid=482468
## Comment 8645
Date: 2007-12-21 08:41:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
For the second test the problem can easily be reproduced by having a MonetDB/XQuery server (`Mserver --dbinit='module(pathfinder);'`) running and then executing the following command in .../pathfinder/tests/BugTracker/Tests/:
sed "s|\$TSTSRCDIR|$PWD|" pf_add-doc_property_error.SF-1706438.xq.in | mclient -lx
## Comment 8646
Date: 2007-12-21 08:46:09 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
For the first test the problem can easily be reproduced by executing the following commands in .../pathfinder/modules/pftijah/Tests/:
export TSTSRCDIR=$PWD
Mserver loadthesis.milS cleanup.milS
## Comment 8647
Date: 2007-12-21 12:48:27 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
It seems the assert is incorrect. It is allowed too access (ie incref) a transient bat which used to be persistent (even if all references too it are removed). I'll check in a better assert after testing finishes.
## Comment 8648
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1852518 at http://sourceforge.net/support/tracker.php?aid=1852518
| GDK/M4/PF: GDK assertion in PF/XQ tests since 2007/11/15 | https://api.github.com/repos/MonetDB/MonetDB/issues/1532/comments | 0 | 2020-11-30T09:45:20Z | 2024-06-27T11:27:44Z | https://github.com/MonetDB/MonetDB/issues/1532 | 753,322,271 | 1,532 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-17 03:42:43 +0100
From: @drstmane
To: @njnes
Version: -- development
Last updated: 2007-12-21 04:40:54 +0100
## Comment 8639
Date: 2007-12-17 15:42:43 +0100
From: @drstmane
Tests
MonetDB5/src/mal/Tests/tst285.mal
MonetDB5/src/mal/Tests/tst960.mal
fail on Windows with
!MALException:str.match:No implementation available
see
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Int.32.32.d-Windows5.1/src_mal/tst285.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Mic.32.32.d-Windows5.1/src_mal/tst285.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Int.64.64.d-Windows5.2/src_mal/tst285.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Mic.64.64.d-Windows5.2/src_mal/tst285.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Int.32.32.d-Windows5.1/src_mal/tst960.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Mic.32.32.d-Windows5.1/src_mal/tst960.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Int.64.64.d-Windows5.2/src_mal/tst960.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB5/.mTests103/Mic.64.64.d-Windows5.2/src_mal/tst960.out.00.html
## Comment 8640
Date: 2007-12-21 13:43:24 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
For some reason, str.match (in src/modules/kernel/batstr.mx) is implemented using the regex library which is not available on Windows (hence the message). At the very least, there should be an alternative implementation using PCRE (which is available on Windows). The PCRE implementation should probably be the primary implementation with regex as a backup.
## Comment 8641
Date: 2007-12-21 16:40:54 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
removed this function. The functionality is available via pcre and batstr.like.
## Comment 8642
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1852339 at http://sourceforge.net/support/tracker.php?aid=1852339
| M5: tst285.mal & tst960.mal in src/mal/Tests fail on Windows | https://api.github.com/repos/MonetDB/MonetDB/issues/1531/comments | 0 | 2020-11-30T09:45:17Z | 2024-06-27T11:27:43Z | https://github.com/MonetDB/MonetDB/issues/1531 | 753,322,239 | 1,531 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-17 03:27:34 +0100
From: @drstmane
To: Peter Boncz <<boncz>>
Version: -- development
Last updated: 2008-01-05 05:23:05 +0100
## Comment 8631
Date: 2007-12-17 15:27:34 +0100
From: @drstmane
Test MonetDB4/src/modules/plain/Tests/bat.py fails on 64-bit Windows:
var b4 := mmap_int.remap(map,false);
!ERROR: MT_mmap: CreateFile('bat\46.tail', -2147483648, 1, &sa, 4, 268435456, NULL) failed
!OS: Permission denied
!ERROR: CMDremap: operation failed.
Cf:
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB4/.mTests103/Int.64.64.d-Windows5.2/src_modules_plain/bat.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB4/.mTests103/Mic.64.64.d-Windows5.2/src_modules_plain/bat.out.00.html
(See also
[ 1640197 ] tests bat & bat2 in src/modules/plain/Tests/ fail on 32-bit
https://sourceforge.net/tracker/index.php?func=detail&aid=1640197&group_id=56967&atid=482468
)
## Comment 8632
Date: 2007-12-18 17:34:04 +0100
From: @peterboncz
Logged In: YES
user_id=591107
Originator: NO
On Vista 64-bits I have seen that the CreateFile this.
this is the bug that went away when Mserver.exe is run 'as administrator' on Vista
but looking again, the value passed on mode0 (second param of CreateFile) is strange, so I have now tried with a less ambitious falg setting. It may be that windows64 adds some strange flags to GENERIC_READ/WRITE that do not fit in an integer and/or are likely to fail on windows64.
## Comment 8633
Date: 2007-12-19 15:09:18 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
The "strange" mode0 value is 0x80000000L which fits fine in a DWORD (unsigned long). The problem is the printing of the value which uses %d (since fixed).
But the less ambitious flags don't seem to help on XP64. The test still fails.
## Comment 8634
Date: 2007-12-20 17:24:16 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
I added some debug prints and this is the result of a failing run. It doesn't make a difference whether we run a 32-bit binary (compiled on a 32-bit system where the binary works fine) or a 64-bit binary on the 64-bit system. The failure is the same. Also, up to the failing call, the calls are also the same on a 32-bit system, but then the call does not fail and the programs just continues on.
MonetDB>var b4 := mmap_int.remap(map,false);
CreateFile("bat\46.buns", 391, 3, SA, 4, 2415919104, 0) -> 264
CreateFileMapping(264, A, 4, 0, 1048576, "(null)") -> 260
CloseHandle(264) -> 1
MapViewOfFileEx(260, 2, 0, 0, 1048576, 0) -> 8bc0000
CloseHandle(260) -> 1
CreateFile("bat\46.buns", 129, 1, SA, 4, 268435456, 0) -> [ffffffffffffffff](https://dev.monetdb.org/hg/MonetDB?cmd=changeset;node=ffffffffffffffff)
ERROR 32: The process cannot access the file because it is being used by another process.
!ERROR: MT_mmap: CreateFile('bat\46.buns', 129, 1, &sa, 4, 268435456, NULL) failed
!OS: The process cannot access the file because it is being used by another process.
!ERROR: CMDremap: operation failed.
## Comment 8635
Date: 2007-12-20 18:13:02 +0100
From: @peterboncz
Logged In: YES
user_id=591107
Originator: NO
Hi Sjoerd,
I have asked a question about this on the microsoft.public.win32.programmer.kernel newsgroup. Hopefully a Microsoft kernel guru can shed light on this.
watch this thread:
http://msdn.microsoft.com/newsgroups/managed/default.aspx?query=boncz&dg=microsoft.public.win32.programmer.kernel
One thing that could be tried is what would happen if we, rather than doing CreateFile() twice, would re-use the file handle. Would CreateFileMapping() or MapViewofFile() then fail?
It is really hard to imagine that mapping a file twice would be forbidden, because copy-on-write maps are almost designed for that use pattern.
Massaging our code to retain file handles for re-use could be done, although it is ugly. Actually, in Linux with the VMtrim thread we already do a thing like that (to be able to call posix_fadvise on the handle), but we only do it for huge memory maps, so it does not increase the number of open file handles by that much, there.
Peter
PS I removed the M4 tag, because it is a MonetDB (aka GDK) issue, not M4 specific.
## Comment 8636
Date: 2007-12-27 00:52:31 +0100
From: @peterboncz
Logged In: YES
user_id=591107
Originator: NO
Hi,
I made an attempt at a fix (checked into the MonetDB_1-20 *branch* now) after reading the second comment of Ben Voigt in the microsoft.public.win32.programmer.kernel newsgroup.
After some discussion on microsoft.public.win32.programmer.kernel it appears that *not* passing FILE_SHARE_WRITE in the third paraneter of CreateFile() will make a read-only open on a file fail if that file is also opened in write-mode.
It perfectly explains why the second CreateFile fails (except for the question why the problem does not exist in Windows32)
The other suggestion from the community was to keep the file handles open as long as a mmap on the file is open, but doing so changes the API and would be a hassle in MonetDB.
So let's hope this does it. I cannot test on windows64, so this is not tested.
## Comment 8637
Date: 2008-01-05 17:23:05 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
With Peter's fix, the test seems to be working fine now.
Closing.
## Comment 8638
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1852327 at http://sourceforge.net/support/tracker.php?aid=1852327
| remap in src/modules/plain/Tests/bat.py fails on Win64 | https://api.github.com/repos/MonetDB/MonetDB/issues/1530/comments | 0 | 2020-11-30T09:45:14Z | 2024-06-27T11:27:42Z | https://github.com/MonetDB/MonetDB/issues/1530 | 753,322,206 | 1,530 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-14 03:23:01 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-18 01:25:51 +0100
## Comment 8626
Date: 2007-12-14 15:23:01 +0100
From: @romulogoncalves
The following SQL code crash the M5-SQL server:
Note: You need to run in this order and you need to run all the queries.
CREATE TEMP TABLE temptest(col int) ON COMMIT DELETE ROWS;
INSERT INTO temptest VALUES (1);
DROP TABLE temptest;
CREATE TEMP TABLE temptest(col int) ON COMMIT DROP;
INSERT INTO temptest VALUES (3);
The but it is really weird... If I remove the first insertion the mserver does not crash and I get:
MAPI = monetdb@localhost:50000
QUERY = INSERT INTO temptest VALUES (3);
ERROR = !INSERT INTO: no such table 'temptest'
If I do not remove any line I get:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1107310912 (LWP 1184)]
0x00002aaabc747410 in cs_find_name (cs=0x48, name=0x1aa2e88 "col") at /ufs/goncalve/MonetDB/Stable/sql/src/storage/sql_catalog.mx:25
25 return list_find_name(cs->set, name);
(gdb) up
1 0x00002aaabc747727 in find_sql_column_node (t=0x0, cname=0x1aa2e88 "col", id=-1) at /ufs/goncalve/MonetDB/Stable/sql/src/storage/sql_catalog.mx:139
139 return cs_find_name(&t->columns, cname);
(gdb)
It seems the cleaning is not done properly ;)
Regards,
Romulo
## Comment 8627
Date: 2007-12-16 23:38:09 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
O test to cover this bug is:
sql/src/test/BugTracker/Tests/create_same_temp_tab_after_its_drop.SF-1850809
## Comment 8628
Date: 2007-12-18 00:15:14 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
It seems we have here a cache problem...
It seems the first insert stays in cache and then it is not removed properly when the table is dropped.
So in the next insertion the table does not exist (because after the commit this temp table is removed). This error is seen at the MAL code (produce by the SQL level) execution. When I do the bind of a column the table is missing and at this stage we expect that the checks were already done, so that's why the mserver crashes...
I decide to change the syntax of second insertion to avoid the cache:
Insert into temptest(col) values(3);
And it works...
Niels?
## Comment 8629
Date: 2007-12-18 13:25:51 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
When a temp table has the commit action different than DROP the cache should be refreshed...
opt_on_commit: /* only for temporary tables */
/* empty */ { $$ = CA_DELETE; }
| ON COMMIT sqlDELETE ROWS { $$ = CA_DELETE; }
| ON COMMIT PRESERVE ROWS { $$ = CA_PRESERVE; }
| ON COMMIT DROP { $$ = CA_DROP; }
;
The fix is in the check in :
[Monetdb-sql-checkins] sql/src/storage store.mx, SQL_2-20, 1.1.4.1, 1.1.4.2
Ro....
## Comment 8630
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1850809 at http://sourceforge.net/support/tracker.php?aid=1850809
| create the same temp table after its drop... | https://api.github.com/repos/MonetDB/MonetDB/issues/1529/comments | 0 | 2020-11-30T09:45:12Z | 2024-06-27T11:27:42Z | https://github.com/MonetDB/MonetDB/issues/1529 | 753,322,170 | 1,529 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-14 02:57:00 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-18 11:20:39 +0100
## Comment 8620
Date: 2007-12-14 14:57:00 +0100
From: @romulogoncalves
The following sql queries crash the M5-SQL server:
create table atacc1 (test int not null);
alter table atacc1 add constraint "atacc1_pkey" primary key (test);
alter table atacc1 drop constraint "atacc1_pkey";
drop table atacc1;
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1107310912 (LWP 32194)]
0x00002aaabc6e740c in insert_check_ukey (sql=0x16a64d8, inser1970-01-01 01:00:00 +0100x1bcbd38, k=0x1be27a8, idx_inser1970-01-01 01:00:00 +0100x0, alter=0)
at /ufs/goncalve/MonetDB/Stable/sql/src/server/sql_updates.mx:99
99 stmt *ts = stmt_basetable(k->t, k->t->base.name);
(gdb)
Regards,
Romulo
## Comment 8621
Date: 2007-12-16 23:39:26 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The test to cover this bug is:
sql/src/test/BugTracker/Tests/add_and_drop_the_same_constraint.SF-1850789
## Comment 8622
Date: 2007-12-18 00:21:45 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Niels did you fix this bug?
Because it is not crashing anymore...
I want to close it, but I would like to know what was the fix.
Regards,
Romulo
## Comment 8623
Date: 2007-12-18 11:20:39 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The bug can be closed. Unfortunately I can say which check in fixed the bug...
The test for this bug was already added and approved...
Ro...
## Comment 8624
Date: 2007-12-18 11:27:43 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Arggg I am always forgetting the "not"
Unfortunately I can not say which check in fixed the
bug...
Regards,
Romulo
## Comment 8625
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1850789 at http://sourceforge.net/support/tracker.php?aid=1850789
| add and then drop the same constraint... | https://api.github.com/repos/MonetDB/MonetDB/issues/1528/comments | 0 | 2020-11-30T09:45:09Z | 2024-06-27T11:27:40Z | https://github.com/MonetDB/MonetDB/issues/1528 | 753,322,135 | 1,528 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-14 02:46:00 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-18 12:45:23 +0100
## Comment 8616
Date: 2007-12-14 14:46:00 +0100
From: @romulogoncalves
The following sql queries crash the M5-SQL
create table atacc1 (test int not null);
alter table atacc1 alter bar set not null;
drop table atacc1;
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1107310912 (LWP 31808)]
0x00002aaabc74523f in sql_trans_alter_null (tr=0x14cdec8, col=0x0, isnull=0) at /ufs/goncalve/MonetDB/Stable/sql/src/storage/store.mx:3429
3429 if (col->null != isnull) {
(gdb)
Regards,
Romulo
## Comment 8617
Date: 2007-12-16 23:40:58 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The test to cover this bug is:
sql/src/test/BugTracker/Tests/alter_col_to_set_same_constraint.SF-1850781
## Comment 8618
Date: 2007-12-18 00:45:23 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Fixed in the check in:
[Monetdb-sql-checkins] sql/src/server sql_schema.mx, SQL_2-20, 1.132.2.4, 1.132.2.5
Modified Files:
Tag: SQL_2-20
sql_schema.mx
Log Message:
Check if the column exist, if not we cannot alter the column constraint ;)
## Comment 8619
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1850781 at http://sourceforge.net/support/tracker.php?aid=1850781
| alter column to set the same constraint | https://api.github.com/repos/MonetDB/MonetDB/issues/1527/comments | 0 | 2020-11-30T09:45:05Z | 2024-06-27T11:27:39Z | https://github.com/MonetDB/MonetDB/issues/1527 | 753,322,087 | 1,527 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-14 02:28:05 +0100
From: Jan Flokstra <<jflokstra>>
To: Jan Rittinger <<jan.rittinger>>
Version: -- development
Last updated: 2008-02-22 09:08:42 +0100
## Comment 8596
Date: 2007-12-14 14:28:05 +0100
From: Jan Flokstra <<jflokstra>>
Im posting this bug on behalf of a student of ours,
e.g.broenink@student.utwente.nl . I had a quick look at the problem and it looks like a real bug to me.
Attached is a file "xmark01.xml" which is a small (33K)XMARK document. When we run the query:
count(doc("xmark01.xml")//*/following-sibling::regions)
the system reponds with:
MAPI = monetdb@localhost:50000
QUERY = count(doc("xmark01.xml")//*/following-sibling::regions)
ERROR = !ERROR: BATfetchjoin(tmp_746,1000000000_rid_prop) does not hit always (|bn|=245 != 246=|l|) => can't use fetchjoin.
!ERROR: CMDleftfetchjoin: operation failed.
!ERROR: interpret_params: leftjoin(param 1): evaluation error.
!ERROR: interpret_params: hmark(param 1): evaluation error.
It does so in the Stable- (on Windows) and the Current (on Suse10.0) branch. Furthermore we get an error with the preceding-sibling. But this is the same problem I suppose.
I will not be there the coming 2 weeks so if somebody fixes this it would be nice to inform
e.g.broenink@student.utwente.nl the problem is fixed,
Thanx, Jan Flokstra.
## Comment 8597
Date: 2007-12-14 15:33:58 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Did/could you try with `Mserver -d10 ...`?
## Comment 8598
Date: 2007-12-14 15:52:24 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
Yep, I tried -d10 and the result was exactly the same. There were no warnings in the server window:
flokstra@ewi581:~/scripts> Mserver -d10 --config=/usr/local/flokstra/etc/MonetDB.conf --dbname test
MonetDB Server v4.21.0
based on GDK v1.21.0
Copyright (c) 1993-2007, CWI. All rights reserved.
Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
MonetDB>module(pathfinder);
PF/Tijah module v0.3.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.21.0 loaded
XRPC administrative console at http://localhost:50001/admin
MonetDB>
And the log of the client window was:
flokstra@ewi581:~/scripts> mclient -lx
xquery>count(doc("xmark01.xml")//*/following-sibling::regions)
more>MAPI = monetdb@localhost:50000
QUERY = count(doc("xmark01.xml")//*/following-sibling::regions)
ERROR = !ERROR: BATfetchjoin(tmp_565,1000000000_rid_prop) does not hit always (|bn|=245 != 246=|l|) => can't use fetchjoin.
!ERROR: CMDleftfetchjoin: operation failed.
!ERROR: interpret_params: leftjoin(param 1): evaluation error.
!ERROR: interpret_params: hmark(param 1): evaluation error.
xquery>
## Comment 8599
Date: 2007-12-14 16:14:19 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
With a Current checkout of Wednesday Dec 12 2007, configure with --enable-debug, compiled with gcc 4.1.2 on 64-bit Fedora Core 6 on an Athlon 64 X2, it seems to work for me --- provided the expected result is "0":
========
$ Mserver --dbinit='module(pathfinder);'
MonetDB Server v4.21.0
based on GDK v1.21.0
Copyright (c) 1993-2007, CWI. All rights reserved.
Compiled for x86_64-redhat-linux-gnu/64bit with 64bit OIDs; dynamically linked.
Visit http://monetdb.cwi.nl/ for further information.
PF/Tijah module v0.3.0 loaded. http://dbappl.cs.utwente.nl/pftijah
MonetDB/XQuery module v0.21.0 loaded
XRPC administrative console at http://localhost:50001/admin
MonetDB>
========
$ ls -l /tmp/xmark01.xml
-rw-r--r-- 1 manegold ins 33928 Dec 14 16:03 /tmp/xmark01.xml
========
$ mclient -lx
xquery>count(doc("/tmp/xmark01.xml")//*/following-sibling::regions)
more><>
0
xquery>
========
Hence, which "Current" version did/do you use?
(I currently don't have a running Stable available, and no time to build/install one ...)
(In case JanF is not available for discussions/feedback, it would be handy if "e.g.broenink@student.utwente.nl" (sorry, Jan did not reveal your name) could follow/join the discussion, here.)
## Comment 8600
Date: 2007-12-14 16:54:00 +0100
From: Jan Flokstra <<jflokstra>>
Logged In: YES
user_id=1054297
Originator: YES
Hi Stefan,
I used a pretty up-to-date head version but to be sure I recompiled on my own SuSe9.3 notebook and used the Current of today (14 dec 16:15). The bug was still there. I'm using a 32 bit Pentium (with 32 bit oid:-) running SuSe9.3 and gcc 3.3.5. I do not know which Windows and Pathfinder version the student, who's first name is Gerben (sorry), is running. He has no SF account and I only showed him how to follow this discussion. This will probably be my last e-mail on this issue the coming 2 weeks.
Jan.
## Comment 8601
Date: 2007-12-14 21:57:18 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Added test in pathfinder/tests/BugTracker/Tests/preceding_following-sibling_bug.SF-1850772.*
Works fine for me with both the latest Stable and the latest Current sources from CVS.
Testing will tell us more about other platforms tomorrow.
## Comment 8602
Date: 2007-12-15 11:22:59 +0100
From: Pseudo user for Sourceforge import <<sfimport>>
---- Original comment by: broenink@users.sourceforge.net ----
Logged In: YES
user_id=1960510
Originator: NO
I'm the student Jan spoke about,
I'm using windows XP, and Thursday I downloaded the latest monetDB/xquery installer for windows. (32bit): http://prdownloads.sourceforge.net/monetdb/MonetDB4-XQuery-Installer-i386-20071017.msi
## Comment 8603
Date: 2007-12-16 16:49:56 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Nightly Testing seems to confirm the bug for following-sibling, but not for preceding sibling:
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/preceding_following-sibling_bug.SF-1850772.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/preceding_following-sibling_bug.SF-1850772.err.00.html
"By hand" I was not (yet?) able to reproduce the bug consistently to investigate it. In fact, I saw it only once after a fresh virgin re-build on a virgin database (dbfarm). After restarting Mserver, everything seems to work fine, both on the previous database, on a new virgin database and even after removing `monetdb4-config --localstatedir` (i.e., <prefix>/var) completely.
Gerben,
does the problem occur for you consistently *every* time you run the given query (with following- & preceding-sibling), both within the same Mserver session and after restarting Mserver?
How do/did you load the document? Explicitly via pf:add-doc() or implicitly via fn:doc() ?
## Comment 8604
Date: 2007-12-17 10:48:02 +0100
From: Pseudo user for Sourceforge import <<sfimport>>
---- Original comment by: broenink@users.sourceforge.net ----
Logged In: YES
user_id=1960510
Originator: NO
The problem occurs every time I run the query. Even when I reinstall monetDB, and have a new database. However, the preceding-sibling error gives another error message the the following-sibling error:
xquery>pf:add-doc("D:\XML-DATA\Xmark\xmark0.1.xml", "xmark01.xml")
more>^Z
xquery>count(doc("xmark01.xml")//*/following-sibling::regions)
more>^Z
MAPI = monetdb@localhost:50000
QUERY = count(doc("xmark01.xml")//*/following-sibling::regions)
ERROR = !ERROR: BATfetchjoin(tmp_426,1000000000_rid_prop) does not hit always (|
bn|=245 != 246=|l|) => can't use fetchjoin.
!ERROR: CMDleftfetchjoin: operation failed.
!ERROR: interpret_params: leftjoin(param 1): evaluation error.
!ERROR: interpret_params: hmark(param 1): evaluation error.
xquery>count(doc("xmark01.xml")//*/preceding-sibling::regions)
more>^Z
MAPI = monetdb@localhost:50000
QUERY = count(doc("xmark01.xml")//*/preceding-sibling::regions)
ERROR = !ERROR: BATmalleftfetchjoin: Cannot perform leftfetchjoin: right head is
no dense OID column.
!ERROR: CMDleftfetchjoin: operation failed.
!ERROR: interpret_params: chk_order(param 1): evaluation error.
xquery>
As you can see, I load the document explicitly, however when I try to load it implicitly, I got the same error.
Gerben
## Comment 8605
Date: 2007-12-17 11:06:37 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
The preceding-sibling bug has been fixed in CVS, cf.,
[ 1841747 ] preceding-sibling failure
http://sourceforge.net/tracker/index.php?func=detail&aid=1841747&group_id=56967&atid=482468
Note, though, that there has not yet been a bug-fix release since this fix; hence, it is only available in CVS and in nightly built source tarballs.
(We are busy with extending our nigtly build & testing system with the ability to build Windows installers, but this exercise is not finished, yet.)
## Comment 8606
Date: 2008-02-10 14:14:52 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
tagged as "XQ" in subject
## Comment 8607
Date: 2008-02-11 10:00:23 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Fixed in CVS (XQuery_0-22 branch):
Fixed off-by-one error in ll_following.
(Test already available)
## Comment 8608
Date: 2008-02-21 21:17:40 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
The preceding-sibling axis is still broken.
The attached query working on the Romeo and Juliet play at http://www.cafeconleche.org/examples/shakespeare/r_and_j.xml fails with latest stable version 0.22 in loop_lifted_preceding_sibling_step_with_nsloc_test():
!ERROR: PFll_child: end_ctx must be head-aligned with iter_bat & ctx_bat, i.e., have equal head seqbases and length.
!ERROR: PFll_child: operation failed.
File Added: romeo.delta.mil
## Comment 8609
Date: 2008-02-21 21:21:59 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Jan,
do you also happen to have the respective XQuery query?
Stefan
## Comment 8610
Date: 2008-02-21 21:28:57 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
Attached you find the query -- I compiled with the HEAD version.
File Added: romeo.mxq
## Comment 8611
Date: 2008-02-21 21:31:51 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
How can I (successfully) load the document?
========
$ wget http://www.cafeconleche.org/examples/shakespeare/r_and_j.xml http://www.cafeconleche.org/examples/shakespeare/play.dtd
--21:28:27-- http://www.cafeconleche.org/examples/shakespeare/r_and_j.xml
=> `r_and_j.xml.1'
Resolving www.cafeconleche.org... 152.46.7.81
Connecting to www.cafeconleche.org|152.46.7.81|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 218,508 (213K) [text/xml]
100%[===========================================================================================================================================================================================================>] 218,508 306.08K/s
21:28:28 (305.10 KB/s) - `r_and_j.xml.1' saved [218508/218508]
--21:28:28-- http://www.cafeconleche.org/examples/shakespeare/play.dtd
=> `play.dtd.1'
Reusing existing connection to www.cafeconleche.org:80.
HTTP request sent, awaiting response... 200 OK
Length: 1,188 (1.2K) [text/html]
100%[===========================================================================================================================================================================================================>] 1,188 --.--K/s
21:28:28 (130.83 MB/s) - `play.dtd.1' saved [1188/1188]
FINISHED --21:28:28--
Downloaded: 219,696 bytes in 2 files
$ perl -i -p -e 's|"play.dtd"|"/tmp/play.dtd"|' /tmp/r_and_j.xml
$ grep play.dtd /tmp/r_and_j.xml
<!DOCTYPE PLAY SYSTEM "/tmp/play.dtd">
$ l /tmp/r_and_j.xml /tmp/play.dtd
-rw-r----- 1 manegold ins 1188 1999-08-16 21:00 /tmp/play.dtd
-rw-r----- 1 manegold ins 218513 2008-02-21 21:29 /tmp/r_and_j.xml
$ echo 'pf:add-doc("/tmp/r_and_j.xml","r_and_j.xml")' | mclient -lx
MAPI = monetdb@localhost:50000
QUERY = pf:add-doc("/tmp/r_and_j.xml","r_and_j.xml")
ERROR = !ERROR: xmlParseEntityRef: no name
!ERROR: Entity 'amp' failed to parse
!ERROR: [shred_url]: 1 times inserted nil due to errors at tuples 0@0.
!ERROR: [shred_url]: first error was:
!ERROR: fail to install ENTITY("amp")
!ERROR: shredder_parse: XML input not well-formed.
!ERROR: CMDshred_url: operation failed.
$ grep amp /tmp/r_and_j.xml /tmp/play.dtd
/tmp/r_and_j.xml:<LINE>We waste our lights in vain, like lamps by day.</LINE>
/tmp/r_and_j.xml:<LINE>As daylight doth a lamp; her eyes in heaven</LINE>
/tmp/r_and_j.xml:<LINE>Two such opposed kings encamp them still</LINE>
/tmp/r_and_j.xml:<STAGEDIR>Enter Citizens, &c</STAGEDIR>
/tmp/r_and_j.xml:mattock, &c</STAGEDIR>
/tmp/play.dtd:<!ENTITY amp "&">
========
## Comment 8612
Date: 2008-02-21 23:26:48 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
fixed in CVS:
The "unique iters" / "unique items" optimization
must NOT be done at all in case we have end_ctx nodes,
i.e., in case we "mis-use" a limited child-step
to evaluate preceding-sibling(ctx) as child(parent(ctx))!
... test(s) will be added, once it/they have been adapted to the Stable version of Pathfinder ...
## Comment 8613
Date: 2008-02-22 18:07:21 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Added Jan's tests in pathfinder/tests/BugTracker/Tests/preceding_following-sibling_bug.SF-1850772-2.*
Jan,
could you please verify the stable output, and close this bug report in case it is correct?
Thanks!
Stefan
## Comment 8614
Date: 2008-02-22 21:08:42 +0100
From: Jan Rittinger <<jan.rittinger>>
Logged In: YES
user_id=993208
Originator: NO
It looks correct. Thanks Stefan!
## Comment 8615
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1850772 at http://sourceforge.net/support/tracker.php?aid=1850772
| XQ: [preceding and following]-sibling:: bug | https://api.github.com/repos/MonetDB/MonetDB/issues/1526/comments | 0 | 2020-11-30T09:45:02Z | 2024-06-27T11:27:39Z | https://github.com/MonetDB/MonetDB/issues/1526 | 753,322,050 | 1,526 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-13 06:03:40 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-13 07:29:21 +0100
## Comment 8588
Date: 2007-12-13 18:03:40 +0100
From: @romulogoncalves
The following SQL execution will show the problem:
sql>create table test ( str VARCHAR(20), str2 VARCHAR(20));
sql>insert into test values (' test ','');
Rows affected 1
sql>select length(str), str from test;
+------------+-------+
| length_str |str |
+============+=======+
| 5 | test |
+------------+-------+
sql>
The length of the string is 6 and not 5 ;)
Regards,
Romulo
## Comment 8589
Date: 2007-12-13 18:06:38 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The problem is the space used in string: Look to the following example:
sql>insert into test values (' ','');
Rows affected 1
sql>select length(str), str from test;
+------------+-------+
| length_str |str |
+============+=======+
| 5 | test |
| 0 | |
+------------+-------+
sql>
## Comment 8590
Date: 2007-12-13 18:42:38 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
looks like there is an ltrim going on.
does ' test ' result in 5 too?
## Comment 8591
Date: 2007-12-13 19:29:21 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
does ' test ' result in 5 too?
No its result in 6...
we currently do a special length(rtrim(x)).
The white spaces, after the last work, are not counted...
Niels just told me that, so I will close the bug.
The resolution will be rejected.
If it is not the proper choice please let me know.
Regards,
Romulo
## Comment 8592
Date: 2007-12-13 19:33:22 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
Hmmm, sure it is an rtrim, because the rendering of mclient shows it is an ltrim instead... (note the space on the right of test)
## Comment 8593
Date: 2007-12-13 19:47:06 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
rtrim as it needs to remove trailing whitespace.
## Comment 8594
Date: 2007-12-13 20:01:35 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
then there are couple more bugs in mclient's rendering, and/or this doesn't happen currently. The code sample shows a space at the end. Here is JdbcClient's output, which shows that mclient's rendering is broken.
So length returns the strlen(rtrim())? If the SQL standard is useless like that, ok. The space is retained in the output though.
monetdb-> start transaction;
Operation successful
auto commit mode: off
monetdb-> create table test ( str VARCHAR(20), str2 VARCHAR(20));
Operation successful
monetdb-> insert into test values (' test ','');
1 affected row
monetdb-> select length(str), str from test;
+------------+--------+
| length_str | str |
+============+========+
| 5 | test |
+------------+--------+
1 row
monetdb->
## Comment 8595
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1850187 at http://sourceforge.net/support/tracker.php?aid=1850187
| length function is not working properly | https://api.github.com/repos/MonetDB/MonetDB/issues/1525/comments | 0 | 2020-11-30T09:44:59Z | 2024-06-28T13:30:40Z | https://github.com/MonetDB/MonetDB/issues/1525 | 753,322,014 | 1,525 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-13 10:27:37 +0100
From: @drstmane
To: @sjoerdmullender
Version: -- development
Last updated: 2007-12-13 08:27:01 +0100
## Comment 8577
Date: 2007-12-13 10:27:37 +0100
From: @drstmane
On Windows (unlike Unix), a plain `(n)make` does not seem to extract modules' .mil/.mal files from their .mx sources, making it impossible to run `RunMtest.bat <options>` by hand.
Only `(n)make check` (as well as `(n)make install`) triggers the extraction of modules' .mil/.mal files from their .mx sources. However, `(n)make check` does not allow the specification of any options for `RunMtest.bat`, which is desirable for nightly testing.
## Comment 8578
Date: 2007-12-13 12:58:21 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
I believe I have fixed this with a patch to autogen. Testing will tell.
## Comment 8579
Date: 2007-12-13 12:58:29 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
In particular, this does not allow to choose (easily/conveniently) between SQL/4 & SQL/5 testing; SQL/4 testing is default.
## Comment 8580
Date: 2007-12-13 13:00:22 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Sorry, I didn't see Sjoerd message & checking before my last comment.
Sjoerd message & checking render my last comment "useless".
I'll test the fix later today...
## Comment 8581
Date: 2007-12-13 14:00:09 +0100
From: @mlkersten
Logged In: YES
user_id=490798
Originator: NO
To comment further. In the testing priority is given to M5/SQL.
M4/SQL is only (primarily) used for sanity checks and solely
under control of Niels.
so, please, make M5/SQL the default on Windows
## Comment 8582
Date: 2007-12-13 16:46:29 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
The patch seems to work, i.e., .mil/.mal files are now also on Windows during plain `(n)make` extracted in the build tree.
However, since the patch, header files get not only installed in <prefix>/include/<PKG>/<srcsubdir>/, but also in <prefix>/include/<srcsubdir>/;
likewise, .mil/.mal files get installed not only in <prefix>/lib/<PKG>/, but also in <prefix>/include/<srcsubdir>/.
## Comment 8583
Date: 2007-12-13 17:31:11 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
I have reason to believe that the latest fix fixes the latest problem.
## Comment 8584
Date: 2007-12-13 17:34:36 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Assuming that noone outsite room C0.12 @ CWI has ever called `(n)make check` for any MonetDB packages on Windows, I don't consider the default setting in MonetDB/SQL's RunMtest.bat.in a crucial/urgent issue. Once I find a way how to detect in sql's RunMtest.bat.in (on Windows) AND RunMtest.in (on Unix) whether M4 or M5 or both are available, I'll change the default setting in BOTH RunMtest.bat.in AND RunMtest.in of sql to use M5, if available, and fall-back to M4, otherwise.
## Comment 8585
Date: 2007-12-13 17:54:36 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Re-set Sjoerd's "Resolution" choice --- was unset due to concurrent changes with non transaction save SF tracker --- second concurrency conflict on this report today ...
## Comment 8586
Date: 2007-12-13 20:27:01 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Indeed, Sjoerd's latest fix seems to fix the latest problems.
Thanks!
## Comment 8587
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1849909 at http://sourceforge.net/support/tracker.php?aid=1849909
| Windows: plain `(n)make` does not extract .m[ia]l from .mx | https://api.github.com/repos/MonetDB/MonetDB/issues/1524/comments | 0 | 2020-11-30T09:44:56Z | 2024-06-27T11:27:37Z | https://github.com/MonetDB/MonetDB/issues/1524 | 753,321,975 | 1,524 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-11 07:19:32 +0100
From: @grobian
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-19 10:30:32 +0100
## Comment 8574
Date: 2007-12-11 19:19:32 +0100
From: @grobian
the -a switch for the start, stop and kill commands is implemented in such a way that merovingian will always ignore anything after the first database.
## Comment 8575
Date: 2008-02-19 22:30:31 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
Fixed in stable.
## Comment 8576
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848808 at http://sourceforge.net/support/tracker.php?aid=1848808
| monetdb {start,stop,kill} -a doesn't work | https://api.github.com/repos/MonetDB/MonetDB/issues/1523/comments | 0 | 2020-11-30T09:44:53Z | 2024-06-27T11:27:36Z | https://github.com/MonetDB/MonetDB/issues/1523 | 753,321,946 | 1,523 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-11 12:51:01 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2008-06-04 11:31:51 +0200
## Comment 8567
Date: 2007-12-11 12:51:01 +0100
From: @romulogoncalves
The following queries crash the Mserver5-SQL:
create table s ( snr int, sname varchar(30) );
create table sp ( snr int, pnr varchar(30) );
select 24;
SELECT DISTINCT S.SNAME
FROM S, SP
GROUP BY S.SNR, S.SNAME, SP.SNR, SP.PNR
HAVING SP.SNR = S.SNR
AND SP.PNR = 'P2';
drop table s;
drop table sp;
Regards,
Romulo
## Comment 8568
Date: 2007-12-14 12:49:54 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Another query which crashes the mserver in the same place...
create table crash_me (a integer not null,b char(10) not null);
insert into crash_me (a,b) values (1,'a');
This is the query:
select a from crash_me where 5 between 4 and 6;
drop table crash_me;
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1107310912 (LWP 31989)]
0x00002aaabc702fe0 in optimize (c=0x16a6508, s=0x1a68928)
at /ufs/goncalve/MonetDB/Stable/sql/src/server/sql_optimize.mx:1542
1542 s->optimized = np->optimized = 1;
(gdb) up
1 0x00002aaabc702dc3 in optimize (c=0x16a6508, s=0x1a418b8)
at /ufs/goncalve/MonetDB/Stable/sql/src/server/sql_optimize.mx:1487
1487 stmt *ns = optimize(c, os);
(gdb) up
2 0x00002aaabc702f11 in optimize (c=0x16a6508, s=0x1a419a8)
at /ufs/goncalve/MonetDB/Stable/sql/src/server/sql_optimize.mx:1523
1523 stmt *ns = optimize(c, n->data);
(gdb)
## Comment 8569
Date: 2007-12-18 16:59:35 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
After some debug lets pump some information...
These two queries crash for different reasons. The firs one presented creates an infinite loop due the pivot stmt...
The second query crashes after 4 iterations (in the optimization). A pivot is missing so the find_pivot returns a empty stmt.
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1107310912 (LWP 23037)]
0x00002aaabc70308c in optimize (c=0x1ace498, s=0x1aa1af8) at /ufs/goncalve/MonetDB/Stable/sql/src/server/sql_optimize.mx:1542
1542 s->optimized = np->optimized = 1;
(gdb) l
0 0x00002aaabc70308c in optimize (c=0x1ace498, s=0x1aa1af8) at /ufs/goncalve/MonetDB/Stable/sql/src/server/sql_optimize.mx:1542
1542 s->optimized = np->optimized = 1;
(gdb) l
1537 stmt *ns = optimize(c, s->op2.stval); /* optimize ptable */
1538 stmt *np = find_pivot(ns, s->t);
1539
(gdb) p np
$2 = (stmt *) 0x0
(gdb)
Looking to the query it is easy to see that no table or column reference is used in the where clause. If I do:
...a between 4 and 6;
instead of:
...5 between 4 and 6;
The Mserver does not crash.
Regards,
Romulo
## Comment 8570
Date: 2008-06-02 17:24:14 +0200
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
This bug is from a old stable release, I decided to assign it current stable release and had a test:
sql/src/tests/BugTracker/Tests/select_crashes_in_the_optimization.SF-1848553
If I am doing something wrong please let me know...
Romulo
## Comment 8571
Date: 2008-06-02 17:33:35 +0200
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Just an aditional information:
This bug is solved in the stable branch Algebra version, but not in the normal version.
The bug will remain open...
Romulo
## Comment 8572
Date: 2008-06-04 11:31:51 +0200
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Test now also works with "normal" (non-Algebra) SQL/4 & SQL/5.
Closing.
## Comment 8573
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848553 at http://sourceforge.net/support/tracker.php?aid=1848553
| select crashes in the optimization | https://api.github.com/repos/MonetDB/MonetDB/issues/1522/comments | 0 | 2020-11-30T09:44:51Z | 2024-06-27T11:27:34Z | https://github.com/MonetDB/MonetDB/issues/1522 | 753,321,915 | 1,522 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-11 12:33:28 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-13 04:34:17 +0100
## Comment 8564
Date: 2007-12-11 12:33:28 +0100
From: @romulogoncalves
This query crashes the mserver5-sql:
select sign(-5) ;
Error:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 46912516520208 (LWP 26140)]
0x00002aaabc6d397b in export_length (s=0x621100, mtype=6881387, eclass=6,
digi1970-01-01 01:00:08 +0100, scale=0, bid=0, p=0xfb)
at /ufs/goncalve/MonetDB/Stable/sql/src/backends/monet5/sql_result.mx:863
863 val = *((lng *)p);
(gdb)
Regards,
Romulo
## Comment 8565
Date: 2007-12-13 16:34:17 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The bug can be closed it was fixed by the check in:
[Monetdb-checkins] MonetDB4/src/modules/plain arith.mx, MonetDB_4-20, 1.3, 1.3.6.1
[Monetdb-checkins] MonetDB5/src/modules/kernel calc.mx, MonetDB_5-2, 1.63.2.1, 1.63.2.2
Log Message:
added function sign (sql allready expected this to exist)
returns 1 for positive numbers
0 for 0
-1 for negative
The example used to open the bug was extracted from the test:
sql/src/test/crashme.sql
Regards,
Romulo
## Comment 8566
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848541 at http://sourceforge.net/support/tracker.php?aid=1848541
| select sign... | https://api.github.com/repos/MonetDB/MonetDB/issues/1521/comments | 0 | 2020-11-30T09:44:48Z | 2024-06-27T11:27:34Z | https://github.com/MonetDB/MonetDB/issues/1521 | 753,321,885 | 1,521 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-11 11:46:07 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-14 09:35:59 +0100
## Comment 8559
Date: 2007-12-11 11:46:07 +0100
From: @romulogoncalves
The following query crashes the mserver5-sql....
select insert('abcd',2,2,'ef') ;
Regards,
Romulo
## Comment 8560
Date: 2007-12-11 11:50:10 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
The following queries create the same crash (at least it is in the same place)...
select replace('abbaab','ab','ba') ;
select soundex('hello') ;
Regards,
Romulo
## Comment 8561
Date: 2007-12-11 12:40:49 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
More two queries that create the same crash..
select cot(1) ;
select patindex('%a%','crash') ;
This query does not create a crash anymore because soundex() is not a valid operator in the Stable branch...
select soundex('hello') ;
Regards,
Romulo
## Comment 8562
Date: 2007-12-14 09:35:59 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
All the functions reported in this bug do not crash the Mserver5-SQL.
The fix was due the check ins:
For the insert
[Monetdb-sql-checkins] sql/src/common sql_types.mx, SQL_2-20, 1.95.2.6, 1.95.2.7
For the other 2 functions:
[Monetdb-checkins] MonetDB5/src/modules/kernel mmath.mx, MonetDB_5-2, 1.35, 1.35.2.1
[Monetdb-checkins] MonetDB5/src/modules/mal pcre.mx, MonetDB_5-2, 1.49, 1.49.2.1
[Monetdb-sql-checkins] sql/src/common sql_types.mx, SQL_2-20, 1.95.2.7, 1.95.2.8
All the examples were extracted from the test:
sql/src/test/crashme.sql
## Comment 8563
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848511 at http://sourceforge.net/support/tracker.php?aid=1848511
| select insert.... | https://api.github.com/repos/MonetDB/MonetDB/issues/1520/comments | 0 | 2020-11-30T09:44:45Z | 2024-06-27T11:27:33Z | https://github.com/MonetDB/MonetDB/issues/1520 | 753,321,856 | 1,520 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-11 11:21:23 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-12 04:06:53 +0100
## Comment 8556
Date: 2007-12-11 11:21:23 +0100
From: @romulogoncalves
This query crashes the MServer5-SQL
select octet_length('abc') ;
Regards,
Romulo
## Comment 8557
Date: 2007-12-12 16:06:53 +0100
From: @romulogoncalves
Logged In: YES
user_id=1498628
Originator: YES
Well, I need to close this bug because M5-SQL does not crash anymore.
The crash was happening in the current branch and since the last update it does not crash anymore.
I do not know which check in fixed it or maybe I had old sources (which I do not real believe).
Sjoerd did a propagation:
propagated changes of Friday Nov 30 2007 - Monday Dec 03 2007
from the SQL_2-20 branch to the development trunk
And in this check in, the module used for octet_length was corrected. That might be the reason, but I am not sure.
Anyway, the example was extracted from the crashme.sql test in sql/src/test/
So it is covered....
Regards,
Romulo
## Comment 8558
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848500 at http://sourceforge.net/support/tracker.php?aid=1848500
| octet_length... | https://api.github.com/repos/MonetDB/MonetDB/issues/1519/comments | 0 | 2020-11-30T09:44:43Z | 2024-06-27T11:27:31Z | https://github.com/MonetDB/MonetDB/issues/1519 | 753,321,824 | 1,519 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-11 11:20:27 +0100
From: @romulogoncalves
To: @njnes
Version: -- development
Last updated: 2007-12-18 11:37:14 +0100
## Comment 8553
Date: 2007-12-11 11:20:27 +0100
From: @romulogoncalves
If I run the following queries the mserver5-sql crashes:
create table test ( str VARCHAR(20), str2 VARCHAR(20));
insert into test values ('', 'test');
insert into test values ('test', '');
insert into test values ('','');
insert into test values (' test ','');
select str,str2,soundex(str),soundex(str2), editdistance2(soundex(str),soundex(str2)), difference(str,str2) from test;
drop table test;
Note: This bug was detected by the execution of the test string.sql at
sql/src/test/Tests
Regards,
Romulo
## Comment 8554
Date: 2007-12-18 23:37:14 +0100
From: @njnes
Logged In: YES
user_id=43556
Originator: NO
fixed the levenshtein implemenation, it returned the distance in the return value instead via the first argument 'result'. Also fixed a couple of m5 interface bugs.
## Comment 8555
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848499 at http://sourceforge.net/support/tracker.php?aid=1848499
| difference function crashes | https://api.github.com/repos/MonetDB/MonetDB/issues/1518/comments | 0 | 2020-11-30T09:44:40Z | 2024-06-27T11:27:30Z | https://github.com/MonetDB/MonetDB/issues/1518 | 753,321,795 | 1,518 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-10 06:30:56 +0100
From: @drstmane
To: @yzchang
Version: -- development
Last updated: 2007-12-10 06:46:41 +0100
## Comment 8548
Date: 2007-12-10 18:30:56 +0100
From: @drstmane
On my FC6 desktop with
MonetDB-Java fails to compile with
compile_xrpc_wrapper:
[echo] Compiling XRPC Wrapper
[javac] Compiling 2 source files to /net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/build/java
[javac] /net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/source/java/src/nl/cwi/monetdb/xquery/xrpc/wrapper/XRPCWrapperWorker.java:33: cannot find symbol
[javac] symbol : class CmdLineOpts
[javac] location: class nl.cwi.monetdb.xquery.xrpc.wrapper.XRPCWrapperWorker
[javac] private CmdLineOpts opts;
[javac] ^
[javac] /net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/source/java/src/nl/cwi/monetdb/xquery/xrpc/wrapper/XRPCWrapperWorker.java:40: cannot find symbol
[javac] symbol : class CmdLineOpts
[javac] location: class nl.cwi.monetdb.xquery.xrpc.wrapper.XRPCWrapperWorker
[javac] XRPCWrapperWorker(Socket s, CmdLineOpts o)
[javac] ^
[javac] /net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/source/java/src/nl/cwi/monetdb/xquery/xrpc/wrapper/XRPCWrapperWorker.java:41: cannot find symbol
[javac] symbol : class OptionsException
[javac] location: class nl.cwi.monetdb.xquery.xrpc.wrapper.XRPCWrapperWorker
[javac] throws OptionsException
[javac] ^
[javac] /net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/source/java/src/nl/cwi/monetdb/xquery/xrpc/wrapper/XRPCWrapperWorker.java:177: cannot find symbol
[javac] symbol : class OptionsException
[javac] location: class nl.cwi.monetdb.xquery.xrpc.wrapper.XRPCWrapperWorker
[javac] } catch (OptionsException oe) {
[javac] ^
[javac] /net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Current/source/java/src/nl/cwi/monetdb/xquery/xrpc/wrapper/XRPCWrapperWorker.java:328: cannot find symbol
[javac] symbol : class OptionsException
[javac] location: class nl.cwi.monetdb.xquery.xrpc.wrapper.XRPCWrapperWorker
[javac] } catch (OptionsException oe){
[javac] ^
[javac] 5 errors
## Comment 8549
Date: 2007-12-10 18:33:54 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
that is a missing import nl.cwi.monetdb.util.*; statement.
## Comment 8550
Date: 2007-12-10 18:36:41 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
which was introduced in revision 1.3 of XRPCWrapperWorker.java
Log Message:
Since NamespaceContextImpl.java has been removed from the util package,
we don't need to import this package any more.
## Comment 8551
Date: 2007-12-10 18:46:41 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: YES
Fixed by Jennie in CVS.
Thanks.
## Comment 8552
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1848027 at http://sourceforge.net/support/tracker.php?aid=1848027
| Java fails to compile | https://api.github.com/repos/MonetDB/MonetDB/issues/1517/comments | 0 | 2020-11-30T09:44:38Z | 2024-06-27T11:27:29Z | https://github.com/MonetDB/MonetDB/issues/1517 | 753,321,767 | 1,517 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-10 10:22:28 +0100
From: Manuel Mayr <<mayrm>>
To: @sjoerdmullender
Version: -- development
Last updated: 2007-12-10 02:25:48 +0100
## Comment 8541
Date: 2007-12-10 10:22:28 +0100
From: Manuel Mayr <<mayrm>>
Hi all,
with my new Kubuntu (gutsy gibbon 2.6.22-14-generic) bootstrap fails with the following messages:
./bootstrap
configure.ac:6: installing `./missing'
configure.ac:6: installing `./install-sh'
Makefile.am: installing `./INSTALL'
Makefile.am: installing `./depcomp'
configure.ac:6: installing `./missing'
configure.ac:6: installing `./install-sh'
rules.mk:23: `%'-style pattern rules are a GNU make extension
Makefile.am:129: `rules.mk' included from here
rules.mk:26: `%'-style pattern rules are a GNU make extension
Makefile.am:129: `rules.mk' included from here
rules.mk:29: `%'-style pattern rules are a GNU make extension
Makefile.am:129: `rules.mk' included from here
rules.mk:32: `%'-style pattern rules are a GNU make extension
Makefile.am:129: `rules.mk' included from here
...
Ok, that's simply a problem with gnu make extensions
but since KUbuntu is a gnu-based system, gnutools are
active by default.
$ aclocal --version
aclocal (GNU automake) 1.10
Written by Tom Tromey <tromey@redhat.com>
and Alexandre Duret-Lutz <adl@gnu.org>.
$ autoconf --version
autoconf (GNU Autoconf) 2.61
Copyright (C) 2006 Free Software Foundation, Inc.
$ automake --version
automake (GNU automake) 1.10
Written by Tom Tromey <tromey@redhat.com>
and Alexandre Duret-Lutz <adl@gnu.org>.
$ gcc --version
gcc (GCC) 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ g++ --version
g++ (GCC) 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
A known workaround is to set the -Wno-portability flag with each automake in bootstrap, mel/bootstrap, Mx/bootstrap, burg/bootstrap.
After this bootstrap runs through without complaining,
but now it fails with make in "mel":
make[3]: Entering directory `/home/mayrm/cvsroot/buildtools/mel'
g++ -g -o mel any_arg.o atom_arg.o atom.o atomops.o bat_arg.o builtin.o command.o debug.o depend.o dependency.o epilogue.o FileInfo.o glue.o html.o iterator.o link.o ListIterator.o mel.o mel.tab.o mel.yy.o mil.o module.o object.o operator.o ops.o prelude.o proto.o symbol.o symtable.o type_arg.o use.o var_arg.o
mel.yy.o: In function `main':
/home/mayrm/cvsroot/buildtools/mel/mel.yy.cc:2: multiple definition of `main'
mel.o:/home/mayrm/cvsroot/buildtools/mel/mel.mx:443: first defined here
mel.o: In function `main':
mel.cc:(.text+0x7cf): undefined reference to `yyin'
mel.tab.o: In function `semerror()':
/home/mayrm/cvsroot/buildtools/mel/mel.yy:473: undefined reference to `melget_lineno()'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:474: undefined reference to `melget_lineno()'
mel.tab.o: In function `yyparse()':
/home/mayrm/cvsroot/buildtools/mel/y.tab.c:1508: undefined reference to `yylex()'
mel.tab.o: In function `yyparse()':
/home/mayrm/cvsroot/buildtools/mel/mel.yy:61: undefined reference to `melset_lineno(int)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:62: undefined reference to `yyin'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:69: undefined reference to `melset_lineno(int)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:72: undefined reference to `close_and_change(void*, _IO_FILE*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:83: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:86: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:94: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:132: undefined reference to `get_cur_buffer()'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:133: undefined reference to `melget_lineno()'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:137: undefined reference to `new_buffer(_IO_FILE*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:141: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:189: undefined reference to `get_signature(char*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:190: undefined reference to `script'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:248: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:307: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:315: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:323: undefined reference to `mel_yyerror(char const*)'
/home/mayrm/cvsroot/buildtools/mel/mel.yy:331: undefined reference to `mel_yyerror(char const*)'
mel.tab.o:/home/mayrm/cvsroot/buildtools/mel/mel.yy:397: more undefined references to `mel_yyerror(char const*)' follow
collect2: ld returned 1 exit status
make[3]: *** [mel] Error 1
make[3]: Leaving directory `/home/mayrm/cvsroot/buildtools/mel'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/home/mayrm/cvsroot/buildtools/mel'
make[1]: *** [all-recursive] Error 1
Would be nice if you have a look at it.
Best regards,
Manuel
## Comment 8542
Date: 2007-12-10 10:27:14 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: NO
> rules.mk:23: `%'-style pattern rules are a GNU make extension
> Makefile.am:129: `rules.mk' included from here
those are warnings, not errors.
> mel.yy.o: In function `main':
> /home/mayrm/cvsroot/buildtools/mel/mel.yy.cc:2: multiple definition of
> `main'
> mel.o:/home/mayrm/cvsroot/buildtools/mel/mel.mx:443: first defined here
> mel.o: In function `main':
> mel.cc:(.text+0x7cf): undefined reference to `yyin'
> mel.tab.o: In function `semerror()':
what bison/yacc do you have installed?
## Comment 8543
Date: 2007-12-10 10:41:20 +0100
From: Manuel Mayr <<mayrm>>
Logged In: YES
user_id=1629542
Originator: YES
I currently run bison with the following version
$ bison --version
bison (GNU Bison) 2.3
Written by Robert Corbett and Richard Stallman.
## Comment 8544
Date: 2007-12-10 13:30:44 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
compilation work fine with similarly "up-to-date" tools on Fedora 8 & Debian 4
$ libtool --version
ltmain.sh (GNU libtool) 1.5.24 (1.1220.2.456 2007/06/24 02:25:32)
Copyright (C) 2007 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ automake --version
automake (GNU automake) 1.10
Written by Tom Tromey <tromey@redhat.com>
and Alexandre Duret-Lutz <adl@gnu.org>.
Copyright 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ autoconf --version
autoconf (GNU Autoconf) 2.61
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software. You may redistribute copies of it under the terms of
the GNU General Public License <http://www.gnu.org/licenses/gpl.html>.
There is NO WARRANTY, to the extent permitted by law.
Written by David J. MacKenzie and Akim Demaille.
$ gcc --version
gcc (GCC) 4.1.2 20070925 (Red Hat 4.1.2-33)
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ bison --version
bison (GNU Bison) 2.3
Written by Robert Corbett and Richard Stallman.
Copyright (C) 2006 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ flex --version
flex 2.5.33
## Comment 8545
Date: 2007-12-10 13:37:53 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
We do indeed require GNU make (cf.,
http://sourceforge.net/mailarchive/forum.php?thread_name=20071209110800.GA32684%40cwi.nl&forum_name=monetdb-developers
MonetDB/HowToStart.rst
MonetDB4/HowToStart
MonetDB5/HowToStart
buildtools/README
http://monetdb.cvs.sourceforge.net/*checkout*/monetdb/MonetDB4/HowToStart
http://monetdb.cvs.sourceforge.net/*checkout*/monetdb/MonetDB5/HowToStart
http://monetdb.cvs.sourceforge.net/*checkout*/monetdb/MonetDB/HowToStart.rst
http://monetdb.cvs.sourceforge.net/*checkout*/monetdb/buildtools/README
)
--- I don't know why automake 1.10 (in fact also on Debian 4.0, Fedora 8 & Gentoo 1.12.9) feels the need to point this out. However, ignoring these *remarks*, compilation works fine on these systems.
Hence, please try again *without* -Wno-portability .
## Comment 8546
Date: 2007-12-10 14:25:47 +0100
From: Manuel Mayr <<mayrm>>
Logged In: YES
user_id=1629542
Originator: YES
Ok, right, it works even without the no-portability flag.
I close this bug-report.
Cheers,
Manuel
## Comment 8547
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1847739 at http://sourceforge.net/support/tracker.php?aid=1847739
| bootstrap fails on KUbuntu (automake 1.10) | https://api.github.com/repos/MonetDB/MonetDB/issues/1516/comments | 0 | 2020-11-30T09:44:35Z | 2024-06-28T13:30:40Z | https://github.com/MonetDB/MonetDB/issues/1516 | 753,321,737 | 1,516 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-09 01:55:34 +0100
From: @drstmane
To: @sjoerdmullender
Version: -- development
Last updated: 2007-12-10 10:11:38 +0100
## Comment 8538
Date: 2007-12-09 13:55:34 +0100
From: @drstmane
Since Sjoerd's check-in
"
Remove line directives from all packaged files to avoid confusion.
"
adding
"
find MonetDB-* -type f -exec sed -n -e "/^line/!w $${TMPDIR:-/var/tmp}/mn-dist-$$$$" {} \; -exec mv -f $${TMPDIR:-/var/tmp}/mn-dist-$$$$ {} \;
"
to */rpm.mk.in,
`make rpm` fails with
"
+ ./configure --build=x86_64-unknown-linux-gnu --host=x86_64-unknown-linux-gnu --target=x86_64-redhat-linux-gnu --program-prefix= --prefix=/usr --exec-prefix=/usr --bindir=/usr/bin --sbindir=/usr/sbin --sysconfdir=/etc --datadir=/usr/share --includedir=/usr/include --libdir=/usr/lib64 --libexecdir=/usr/libexec --localstatedir=/var --sharedstatedir=/usr/com --mandir=/usr/share/man --infodir=/usr/share/info --enable-assert=no --disable-optimize --enable-debug --enable-noexpand=all --enable-bi1970-01-01 01:01:04 +0100 CC=gcc
/tmp/rpm-tmp.68964: line 35: ./configure: Permission denied
"
since the above line does not maintain executable bits on files when creating a copy to remove line directives.
## Comment 8539
Date: 2007-12-10 10:11:38 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Fixed by using chmod + cp.
## Comment 8540
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1847243 at http://sourceforge.net/support/tracker.php?aid=1847243
| `make rpm` fails for all packages ("Current") | https://api.github.com/repos/MonetDB/MonetDB/issues/1515/comments | 0 | 2020-11-30T09:44:33Z | 2024-06-27T11:27:27Z | https://github.com/MonetDB/MonetDB/issues/1515 | 753,321,709 | 1,515 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-09 11:33:50 +0100
From: @grobian
To: @njnes
Version: -- development
Last updated: 2007-12-09 01:32:19 +0100
## Comment 8535
Date: 2007-12-09 11:33:50 +0100
From: @grobian
Compiling clients (Current) failed on GNU.32.32.d-Debian4.0
1767 encoding = nl_langinfo(CODESET);
1768 endif
1769 if (encoding != NULL && strcasecmp(encoding, "utf-8") == 0)
1770 encoding = NULL;
1771 if (encoding != NULL) {
/home/monet/monet.GNU.32.32.d.23543/clients/src/mapiclient/MapiClient.mx: In
function ‘main’:
/home/monet/monet.GNU.32.32.d.23543/clients/src/mapiclient/MapiClient.mx:1769:
error: implicit declaration of function ‘strcasecmp’
make[5]: *** [MapiClient.o] Error 1
make[5]: Leaving directory
`/home/monet/monet.GNU.32.32.d.23543/clients/Debian4.0/src/mapiclient'
## Comment 8536
Date: 2007-12-09 13:32:19 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Fixed by Niels in CVS:
strings.h contains the non conditional definition of strcasecmp
(in string.h its only available if we define _BSD_SOURCE which we correctly
don't)
This fixes the compilation issue on debian
## Comment 8537
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1847191 at http://sourceforge.net/support/tracker.php?aid=1847191
| clients: Debian doesn't have strcasecmp | https://api.github.com/repos/MonetDB/MonetDB/issues/1514/comments | 0 | 2020-11-30T09:44:30Z | 2024-06-27T11:27:26Z | https://github.com/MonetDB/MonetDB/issues/1514 | 753,321,685 | 1,514 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-08 04:16:59 +0100
From: @grobian
To: Fabian Groffen <<fabian>>
Version: -- development
Last updated: 2008-02-20 01:12:04 +0100
## Comment 8532
Date: 2007-12-08 16:16:59 +0100
From: @grobian
merovingian doesn't report an error on the console when being started and its port is already in use. The error message only appears in the log. While merovingian immediately terminates upon this, it returns success to the caller.
## Comment 8533
Date: 2008-02-20 13:12:03 +0100
From: @grobian
Logged In: YES
user_id=963970
Originator: YES
fixed in cvs stable
## Comment 8534
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1846883 at http://sourceforge.net/support/tracker.php?aid=1846883
| Merovingian: successful exit when port is already in use | https://api.github.com/repos/MonetDB/MonetDB/issues/1513/comments | 0 | 2020-11-30T09:44:28Z | 2024-06-27T11:27:25Z | https://github.com/MonetDB/MonetDB/issues/1513 | 753,321,662 | 1,513 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-08 01:20:29 +0100
From: @grobian
To: @yzchang
Version: -- development
Last updated: 2007-12-11 01:10:51 +0100
## Comment 8528
Date: 2007-12-08 13:20:29 +0100
From: @grobian
nl/cwi/monetdb/util/NamespaceContextImpl.java
doesn't sound like it is an "util", why is it in this package? If it is here for a valid reason build.xml has to be changed such that jdbc can compile again without Java 1.5. If it is not, the file should be moved somewhere under the xml/xquery package.
## Comment 8529
Date: 2007-12-10 11:58:29 +0100
From: @yzchang
Logged In: YES
user_id=341633
Originator: NO
Fixed in CVS by moving NamesapceContextImpl.java to xquery/xrpc/api:
======
[monetdb-Bugs-1846797] Java: nl/cwi/monetdb/util/NamespaceContextImpl.java
Moved NamespaceContextImpl.java from the package nl.cwi.monetdb.util, to
nl.cwi.monetdb.xquery.xrpc.api, so that JDBC can compile again without
requiring Java 1.5. NamespaceContextImpl.java is needed for evaluating
XPath expressions on XRPC messages.
======
Will close the bug tomorrow, if the testweb says OK.
## Comment 8530
Date: 2007-12-11 13:10:51 +0100
From: @drstmane
Logged In: YES
user_id=572415
Originator: NO
Appears to be fixed:
http://monetdb.cwi.nl/testing/projects/monetdb/Current/java/.Makes3/index_short.html
## Comment 8531
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1846797 at http://sourceforge.net/support/tracker.php?aid=1846797
| Java: nl/cwi/monetdb/util/NamespaceContextImpl.java | https://api.github.com/repos/MonetDB/MonetDB/issues/1512/comments | 0 | 2020-11-30T09:44:26Z | 2024-06-27T11:27:24Z | https://github.com/MonetDB/MonetDB/issues/1512 | 753,321,645 | 1,512 |
[
"MonetDB",
"MonetDB"
] | Date: 2007-12-06 03:37:40 +0100
From: mharrison <<matt_harrison>>
To: @sjoerdmullender
Version: -- development
Last updated: 2007-12-11 10:09:40 +0100
## Comment 8525
Date: 2007-12-06 03:37:40 +0100
From: mharrison <<matt_harrison>>
converters.py should change Unicode2Str from:
def Unicode2Str(s, d):
"""Convert a unicode object to a string using latin1 encoding."""
return s.encode('latin')
to:
def Unicode2Str(s, d):
"""Convert a unicode object to a string using latin1 encoding."""
return string_literal(s.encode('latin'), d)
If you pass in a u"unicodestring" it doesn't get quoted the way ascii does.
## Comment 8526
Date: 2007-12-11 10:09:40 +0100
From: @sjoerdmullender
Logged In: YES
user_id=43607
Originator: NO
Fixed in CVS in the stable branch. This will be propagated to the development branch within a few days.
## Comment 8527
Date: 2010-05-04 09:32:09 +0200
From: Pseudo user for Sourceforge import <<sfimport>>
This bug was previously known as tracker item 1845266 at http://sourceforge.net/support/tracker.php?aid=1845266
| Python converters don't quote unicode strings... | https://api.github.com/repos/MonetDB/MonetDB/issues/1511/comments | 0 | 2020-11-30T09:44:23Z | 2024-06-27T11:27:23Z | https://github.com/MonetDB/MonetDB/issues/1511 | 753,321,611 | 1,511 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.