threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "This is a proposal of fast data loading using batch update of indexes for 8.4.\nIt is a part of pg_bulkload (http://pgbulkload.projects.postgresql.org/) and\nI'd like to integrate it in order to cooperate with other parts of postgres.\n\nThe basic concept is spooling new coming data, and merg... |
[
{
"msg_contents": "I got around to testing PG 8.3 on HPUX on Itanium (feel free to play\nalong at www.testdrive.hp.com) ... and was dismayed to find that it\ndoesn't work. If compiled with HP's C compiler, the regression tests\ndump core. Investigation shows that the problem occurs where\ntuptoaster.c tries t... |
[
{
"msg_contents": "I would like to have your opinion about the following behavior of TO_DATE.\nIs this correct or a hidden feature?\n\ncase:\n'mon 21-feb-2008' is obviously a bad date. It should be 'thu 21-feb-2008'\n\ntest:\ntestdb=# select to_date('mon 21-feb-2008','dy dd-mon-yyyy');\n to_date\n------------\... |
[
{
"msg_contents": "regarding http://archives.postgresql.org/pgsql-hackers/2008-02/msg00826.php\n\nI was wondering why we have a custom written parser to parse the date/time\nfrom/to string and vice versa?\nIs there a historical reason? Would a yacc/lex parser be better?\n\nRegards,\nGevik \n\n",
"msg_date":... |
[
{
"msg_contents": "This is a postgresql 8.2.5 backend. connected to a jdbc client.\n\n2008-02-21 06:12:30 EST [18880] 10.21.0.21 ERROR: insufficient data \nleft in message\n2008-02-21 06:12:30 EST [18880] 10.21.0.21 STATEMENT: select * from \nuser_profile,user_profile_access where user_profile.uid=user_prof... |
[
{
"msg_contents": "Hi all.\n\nThe door was opened by Mr. Ralf S. Engelschall of a great developer.\nhttp://www.ossp.org/pkg/lib/uuid/\nIt can be used from Version 1.6.1. \n\nPlease see,\nhttp://winpg.jp/~saito/pg_work/OSSP_win32/\n\nwe can include in the next release.:-)\nThanks!\n\nRegards,\nHiroshi Saito\n",
... |
[
{
"msg_contents": "\n\nMark Woodward wrote:\n> I have been looking at this thread for a bit and want to interject an idea.\n>\n> A couple years ago, I offered a patch to the GUC system that added a\n> number of abilities, two left out were:\n>\n> (1) Specify a configuration file on the command line.\n> (2) Allo... |
[
{
"msg_contents": "Consider this test case:\n\n create table baz (a int, b int);\n insert into baz (select 1, generate_series(1, 3000000));\n\n select baz.*, (select 1 from generate_series(1, 1) g(a)\n where g.a < baz.b)\n from baz;\n\nThe final query consumes ~800 MB of mem... |
[
{
"msg_contents": "Here is a patch so that the backend is linked in one piece instead of using \nthe SUBSYS.o files.\n\nThe question is how we want to activate that. I currently used make \nBIGLINK=1, which is obviously just for testing. Should we just turn it on by \ndefault and see if anyone complains?\n\n-... |
[
{
"msg_contents": "I've been looking at a problem someone encountered with ts_headline:\nhttp://archives.postgresql.org/pgsql-general/2008-02/msg01035.php\n\nIt turns out the problem was mixing ts_headline(<no specified config>) \nwith to_tsquery(<specified config>) where <specified config> wasn't the \ndefault... |
[
{
"msg_contents": "Motivation:\nTo enable probes to work with Mac OS X Leopard and other OSes that will \nsupport DTrace in the future. Bug filed on this issue \nhttp://archives.postgresql.org/pgsql-bugs/2007-10/msg00201.php\n\nThe problem:\nThere are two different mechanisms to implementing application level \... |
[
{
"msg_contents": "I have added support in readfuncs.c to write out Query and PlannedStmt objects \nusing nodeToString() and then read them back in. We needed this so that we \ncould use PREPARE and write the arguments to StorePreparedStatement() out to \na file. We are going then read them back into Postgres o... |
[
{
"msg_contents": "The purpose of CopyReadLineText is to scan the input buffer, and find \nthe next newline, taking into account any escape characters. It \ncurrently operates in a loop, one byte at a time, searching for LF, CR, \nor a backslash. That's a bit slow: I've been running oprofile on COPY, \nand I've... |
[
{
"msg_contents": "\n[This message is mostly for the benefit of the list -- he and I already talked\na bit about this here at FOSDEM. Ishii-san, if you have a chance we should sit\ndown and talk about this in more detail before we leave!]\n\n\nTatsuo Ishii wrote:\n>> On Tue, Feb 19, 2008 at 3:36 AM, Tatsuo Ishi... |
[
{
"msg_contents": "Hi\n\nI just stumbled over the following behaviour, introduced with 8.3, and\nwondered if this is by design or an oversight.\n\nIf you define a domain over some existing type, constrain it to\nnon-null values, and use that domain as a field type in a table \ndefinition, it seems to be impossi... |
[
{
"msg_contents": "Hello,\n\nI found easy implementation of variadic functions. It's based on\nadapation FuncnameGetCandidates. When I found variadic function, then\nI should create accurate number of last arguments (diff between\npronargs and nargs). Variadic function can be signed via flag or via\nsome pseudo... |
[
{
"msg_contents": "Hi all:\n\nI'm at PostgreSQL 8.3 for my production database and everything is \nworking great. I had no problems converting free text search from 8.2 to \n8.3, and I really like the improvements.\n\nI tried using insert ... delete ... return ... and get a syntax error:\n\n> pccyber=# insert i... |
[
{
"msg_contents": "Hi\n\ndblink in 8.3 blocks without any possibility of interrupting it while\nwaiting for an answer from the remote server. Here is a strace\n[pid 27607] rt_sigaction(SIGPIPE, {SIG_IGN}, {SIG_IGN}, 8) = 0\n[pid 27607] sendto(56, \"Q\\0\\0\\0008lock table travelhit.booking_code in \nexclusive m... |
[
{
"msg_contents": "Hi \n\nThe database is initialized with utf8, so in order for LIKE to use the index \non a text field, I used text_pattern_ops when I created it. So far so good. \n\nIt's in the documentation, but there's no explanation of why this index will \nonly work for LIKE searches. How come that I hav... |
[
{
"msg_contents": "Hi all,\nOn the 6th of february, there's been a thread about adding new options to \npg_dump, but it is now too late for me to add comments to this thread, since \nall that was said wouldn't be readable at this time, so I add an new thread \nhere.\nI haven't found any option to dump any user-... |
[
{
"msg_contents": "I saw this strange behavior due to a customer problem who managed to get \ndumps which aren't restorable anymore:\n\nCREATE TABLE foo(datum date);\n\nINSERT INTO foo VALUES('0000-02-29');\nINSERT 0 1\n\nSELECT * FROM foo;\n datum\n---------------\n 0001-02-29 BC\n(1 row)\n\nCOPY foo TO ST... |
[
{
"msg_contents": "I need to run about 1000 PostgreSQL connections on a server that I can\nuse about 4 GB of the total 16 GB of total RAM. It seems that each\nsession creates a process that uses about 15 MB of RAM just for\nconnecting so I'm running out of RAM rather quickly.\n\n \n\nI have these non-default s... |
[
{
"msg_contents": "Log Message:\n-----------\nLink postgres from all object files at once, to avoid the error-prone\nSUBSYS.o step and allow for better optimization by the linker.\n\nInstead of partial linking into SUBSYS.o, the list of object files is\nassembled in objfiles.txt files that are expanded when the... |
[
{
"msg_contents": "The libpq.rc make rule says:\n\n# depend on Makefile.global to force rebuild on re-run of configure\n$(srcdir)/libpq.rc: libpq.rc.in $(top_builddir)/src/Makefile.global\n sed -e 's/\\(VERSION.*\\),0 *$$/\\1,'`date '+%y%j' | sed 's/^0*//'`'/' < $< > $@\n\nHowever, libpq.rc is also inclu... |
[
{
"msg_contents": "Hi, \n \n \nRefering to this request http://momjian.us/mhonarc/patches_hold/msg00022.html\nI have created a patch. The output doesn't exaclty match with what is stated here \nhttp://momjian.us/mhonarc/patches_hold/msg00023.html. \n \nHowever, it does tell the required details in a similar for... |
[
{
"msg_contents": "I am working on some beginner level patches to help clean up the\ntimestamp code in PostgreSQL.\n\nBasically, my first few patches are aimed at removing the dependence\non the HAVE_INT_TIMESTAMP to choose types for variables. I will\neventually try to remove the use of HAVE_INT64_TIMESTAMP to... |
[
{
"msg_contents": "I'm looking at ways to reduce the number of network calls and/or the\nwaiting time while we perform network COPY.\n\nThe COPY calls in libpq allow asynchronous actions, yet are coded in a\nsynchronous manner in pg_dump, Slony and psql \\copy.\n\nDoes anybody have any experience with running C... |
[
{
"msg_contents": "In pgAgent.cpp, I would like to add LIMIT as shown below:\n\nLogMessage(_(\"Checking for jobs to run\"), LOG_DEBUG);\nDBresult *res=serviceConn->Execute(\n wxT(\"SELECT J.jobid \")\n wxT(\" FROM pgagent.pga_job J \")\n wxT(\" WHERE jobenabled \")\n wxT(\" AND jobagentid IS NULL \")\n wxT(\... |
[
{
"msg_contents": "Log Message:\n-----------\nDon't build the win32 support files in the all target, only in distprep and\nwhen they are actually needed as prerequisites.\n\nModified Files:\n--------------\n pgsql/src:\n Makefile.shlib (r1.112 -> r1.113)\n (http://developer.postgresql.org/cvswe... |
[
{
"msg_contents": "Looking at the profile for COPY and then a profile for CREATE TABLE AS\nSELECT (CTAS) there is clearly too much overhead caused by inserting the\nrows one at a time. \n\nFlat profile of CTAS: (2 cols of this output removed for clarity)\n\nEach sample counts as 0.01 seconds.\n % cumulative ... |
[
{
"msg_contents": "Hi,\n\nI'm working on a GiST opclass to support prefix searching as presented here:\n http://pgsql.tapoueh.org/site/html/prefix/index.html\n http://prefix.projects.postgresql.org/README.html\n http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/prefix/prefix/\n\nIn order to have a much more effici... |
[
{
"msg_contents": "Sooner or later we ought to completely reimplement LISTEN/NOTIFY,\nbut while thinking about Joel Stevenson's performance issue I had a\nsudden idea for a very simple change that would buy some improvement.\nCurrently, all operations in async.c take ExclusiveLock on pg_listener,\nbut it strike... |
[
{
"msg_contents": "Hackers,\n\nAs you may have read, Coverity is running their static analysis tool (\"Scan\") \nagainst the PostgreSQL codebase daily: http://scan.coverity.com/\n\nWe need two (or more) PostgreSQL hackers to volunteer to regularly check the \nCoverity reports and either fix/forward the bugs fou... |
[
{
"msg_contents": "There is a build farm failure now because (apparently) an old make version:\n\nhttp://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=dawn_bat&dt=2008-02-26%2019:00:01\n\nThe new amended backend linking code needs GNU make 3.80, released 2002-10-03. \nShould we just require that? I think th... |
[
{
"msg_contents": "As far as I can see the main difficulty in making COPY run faster (on\nthe server) is that pretty involved conversion from plain-text lines\ninto tuples. Trying to get rid of this conversion by having the client\nsend something that resembles the data stored in on-disk tuples is not a\ngood a... |
[
{
"msg_contents": "I'm having trouble compiling the current cvs version on windows xp (msvc 2005 express). Compile errors below.\n\nI have bison 1.875 (I can't find 2.2+ for windows) and flex 2.5.4. These tools seem to generate correct outputs.\n\nIt looks like it might be including parse.h from include/parser/... |
[
{
"msg_contents": "The default parser doesn't allow commas in numbers (I can see why, I think).\n\nSELECT ts_parse('default', '123,000');\n ts_parse\n----------\n (22,123)\n (12,\",\")\n (22,000)\n\nOne option of course is to pre-process the text, but since we can \nsupport custom parsers I thought I'd take... |
[
{
"msg_contents": "I'd like to extend the libpq service file by allowing\r\nwildcards, e.g. like this:\r\n\r\n[%]\r\nhost=dbhost.mycompany.com\r\ndbname=%\r\n\r\nSuch an entry would match all service parameters,\r\nand all ocurrences of the wildcard right of a = would\r\nbe replaced with the service parameter.\... |
[
{
"msg_contents": "I have the following table:\n\nObjeto\t\tValor\n\nob1\t\t10\nob1\t\t20\nob2\t\t50\nob2\t\t10\nob3\t\t50\n\nWith the following command:\n\n\"select distinct Objeto, sum(valor) from tb\ngroup by Objeto;\"\n\n\nI have to return:\n\nObjeto\t\tValor\n\nob1\t\t30\nob2\t\t60\nob3\t\t50\n\nWhat you n... |
[
{
"msg_contents": "Hello\n\nI thing RETURN QUERY is successful idea. It should be completed with\nsupport of dynamic SQL.\n\nSyntax:\n\nRETURN EXECUTE sqlstring [USING];\n\nThis is shortcut for\nFOR r IN EXECUTE sqlstring USING LOOP\n RETURN NEXT r;\nEND LOOP;\n\nRegards\nPavel Stehule\n",
"msg_date": "We... |
[
{
"msg_contents": "Hi,\n\nI'm toying around with the idea of tracking snaphots more accurately to\nbe able to advance Xmin for read committed transactions.\n\nI think it's relatively easy to do it in the straightforward way, which\nis to just add \"destroy snapshots\" in the spots where a snapshot\nvariable goe... |
[
{
"msg_contents": "I notice that several of the call sites of tuplestore_puttuple() start\nwith arrays of datums and nulls, call heap_form_tuple(), and then switch\ninto the tstore's context and call tuplestore_puttuple(), which\ndeep-copies the HeapTuple into the tstore. ISTM it would be faster and\nsimpler to... |
[
{
"msg_contents": "Referring to tuplesort.c and tuplestore.c\n\nBACKGROUND: Starting from dumptuples() [ tuplesort.c ] write functions move \nthe tuple from a buffer to another in order to finally write it in a logical \ntape. Is there a way (even the most inefficient way) to use current \nread/write func... |
[
{
"msg_contents": "On IRC today someone brought up a problem in which users were still able\nto connect to a database after a \"REVOKE CONNECT ... FROM theuser\". The\nreason theuser is still able to connect is because PUBLIC still has\nprivileges to connect by default (AndrewSN was the one who answered\nthis).... |
[
{
"msg_contents": "In Read Committed transactions we take snapshots much more frequently\nthan transactions begin and commit. It would be help scalability if we\ndidn't need to re-take a snapshot. That's only helpful if the chances of\nseeing the snapshot is relatively high.\n\nNow that we have virtual transact... |
[
{
"msg_contents": "Hello.\n\nI am currently playing with UUID data type and try to use it to store provided\nby third party (Hewlett-Packard) application. The problem is they\nformat UUIDs as\n0000-0000-0000-0000-0000-0000-0000-0000, so I have to\nreplace(text,'-','')::uuid for\nthis kind of data.\n\nNooow, th... |
[
{
"msg_contents": "It looks like gypsy_moth has been failing like this:\n\ncreating directory /export/home/tmp/pg-test/build-suncc/HEAD/pgsql.21325/src/test/regress/./tmp_check/data ... ok\ncreating subdirectories ... ok\nselecting default max_connections ... 100\nselecting default shared_buffers/max_fsm_pages ... |
[
{
"msg_contents": "Hello All,\n\n \n\nWe are facing some problems while downloading the Postgresql 8.2.4\nversion of 64bit processor for both Windows and SUSE Linux ES-7000\npartitions. The entire download URL's of PostgreSQL 8.2.4 version are\ndisplaying the error message \"The webpage can't found\". \n\nCan... |
[
{
"msg_contents": "Hello,\n\nWe see error logs something like following on deadlocks:\n\n| ERROR: deadlock detected\n| DETAIL: Process 1612 waits for ShareLock on transaction 407; blocked by process 3388.\n| Process 3388 waits for ShareLock on transaction 406; blocked by process 1612.\n| STATEMENT: U... |
[
{
"msg_contents": "I had this idea sometime back. Not sure if this has been discussed before\n\nIn a typical scenario, user might create a table and load data in the table as\npart of a single transaction (e.g pg_restore). In this case, it would help if we\ncreate the tuples in the *frozen* state to avoid any w... |
[
{
"msg_contents": "To all,\n\nI am brand new to writing functions for PostgreSQL and understand that this question has been asked one hundred and one times. Please be patient with me, I do learn quickly.\n\nMy question is:\n\nI wish to create a an aggregate function and have read and re-read the 34.10 section ... |
[
{
"msg_contents": "Where can I see a list of projects?\n\nThanks!\nJoey\n\nWhere can I see a list of projects?Thanks!Joey",
"msg_date": "Thu, 28 Feb 2008 10:28:11 -0800",
"msg_from": "\"Joey K.\" <pguser@gmail.com>",
"msg_from_op": true,
"msg_subject": "Google SoC 2008?"
},
{
"msg_conten... |
[
{
"msg_contents": "Hackers,\n\nOver the last 2 years, Google Summer of Code has been responsible for \nseveral new features of PostgreSQL: XML, ghost XID, phpPgAdmin upgrade, \nand more, as well as some new long-term contributors to the project. We \nwant to have an equally good summer this year.\n\nSo: we nee... |
[
{
"msg_contents": "Architecture: Intel Core 2 Duo\nOS: linux-2.6.20-gentoo-r8\nFilesystem: ext3\nPostgres v8.2.3 compiled with gcc 4.1.1-r3\nRAM - 2GB\nShared buffers - 24MB\n[All other Postgres configuration parameters are default values]\n\nProblem description:\nCOPY into temp table fails using a specific com... |
[
{
"msg_contents": "Don't forget that our first \"commit fest\" for 8.4 development starts\nSaturday March 1. If you've got a patch that you'd like reviewed in\nthis fest, it's time to send it in. (But there's plenty more fests\nto come, so don't panic about it.)\n\nAlso, 8.3.0 has been out for almost a month,... |
[
{
"msg_contents": "I remember Heikki mentioned improving redo recovery in one of the\nemails in the past, so I know people are already thinking about this.\nI have some ideas and just wanted to get comments here.\n\nISTM that its important to keep the redo recovery time as small as possible\nin order to reduce ... |
[
{
"msg_contents": "I'm now looking into toast code and I found following code in \npglz_decompress:\n\n00704 if (destsize != source->rawsize)\n00705 elog(destsize > source->rawsize ? FATAL : ERROR,\n00706 \"compressed data is corrupt\");\n\n\nI'm surprise why we there panic? By my opini... |
[
{
"msg_contents": "Hi, all\n\nseb=> select version();\n version\n--------------------------------------------------------------------------------------------------------------------\n PostgreSQL 8.3.0 on x86_64-pc-linux-gnu, compiled by GCC \nx86_64-pc-linux-... |
[
{
"msg_contents": "Sometimes, I have to add column to table, then adding the column to some view.\n\nCurrently it impossible to alter a view to add a new column, which is really \nannoying. The only solution is to drop the view, and all dependented view, \nthen recreating everything.\n\nSomeone on #postgresqlfr... |
[
{
"msg_contents": "\nFlorian G. Pflug wrote:\n> The same holds true for index scans, though. Maybe we can find a\n> solution that benefits both cases - something along the line of a\n> bgreader process\n\nI posted a patch to do readahead for bitmap index scans using posix_fadvise.\nExperiments showed it works g... |
[
{
"msg_contents": "Hi -\n\nI was perusing the todo list to see some easy items that I might help out on (and get up to speed on postgres hacking)... one of them (with %) seems to lead to another:\n\n o %Have ALTER TABLE RENAME rename SERIAL sequence names\n o Have ALTER SEQUENCE RENAME rename the sequ... |
[
{
"msg_contents": "Oops: meant pg_type.typname \nand \npg_class.relname.\n\n----- Original Message ----\nFrom: craigp <craigp98072@yahoo.com>\nTo: pgsql-hackers@postgresql.org\nSent: Sunday, March 2, 2008 2:36:33 AM\nSubject: newbie: renaming sequences task\n\n\nHi \n-\n\nI \nwas \nperusing \nthe \ntodo \nlist ... |
[
{
"msg_contents": "Oops: meant pg_type.typname \nand \npg_class.relname.\n\n----- Original Message ----\nFrom: craigp <craigp98072@yahoo.com>\nTo: pgsql-hackers@postgresql.org\nSent: Sunday, March 2, 2008 2:36:33 AM\nSubject: newbie: renaming sequences task\n\n\nHi \n-\n\nI \nwas \nperusing \nthe \ntodo \nlist ... |
[
{
"msg_contents": "Hi!\n\nDoes anybody have a good pointer for an example of how to construct an \narray in a SRF? Or if not such a pointer, hints on which way is best to \ndo it? (it'll be an array of text strings, one-dimensional)\n\n(This is from C, if that's not obvious)\n\n//Magnus\n",
"msg_date": "Sun... |
[
{
"msg_contents": "Reminder to patch appliers,\n\nIf you apply a patch that was discussed on patches or hackers, please\nreply to the email indicating you have applied the patch. I can connect\nthe commit message to the email discussion but often the patch submitter\ndoes not read committers so it would be goo... |
[
{
"msg_contents": "Hi,\n\nI read about Google Summer of Code recently and being involved with pgAdmin\nsince a few weeks, would like to add the Query Builder feature to pgAdmin,\nas a part of this programme.\n\nI have been reading GSoC related emails in the past year on various PG\nlists, and believe that keepi... |
[
{
"msg_contents": "Hi PostgreSQL!\n\nAlthough this year's GSoC is just starting, I thought getting in touch a bit\nearlier would only be of benefit.\n\nI study Computer Science in Faculty of Mathematics, Informatics\nand Mechanics of Warsaw University. I'm currently in my fourth year of\nstudies. Having chosen ... |
[
{
"msg_contents": "Hi,\n\nI'm seeing a 8.3 database whose autovacuum-initiated vacuums are being\ncancelled and I am not sure of the cause. I am wondering if a HOT\ncleanup round on a page could try to acquire locks on it that would\nconflict with autovacuum (basically anything that conflicts with a\nvacuum lo... |
[
{
"msg_contents": "The latest version of libpq type system is available here:\nhttp://www.esilo.com/projects/postgresql/libpq/typesys-0.9a.tar.gz\n\nThe following modifications where made:\n*) documentation fixes\n*) parameter resets are no longer automatic\n*) updated to patch vs. REL8_3_STABLE\n\nMerlin Moncu... |
[
{
"msg_contents": "Following the discussion on making UUID data type to be much more liberal\n( http://archives.postgresql.org/pgsql-hackers/2008-02/msg01214.php )\nI have decided to try to approach it from more general perspective.\n\nThe current state of code is available at:\nhttp://qnex.net/hexstring-0.1-20... |
[
{
"msg_contents": "Dear Magnus,\n\nI've sent another xml file to your e-mail. Let's see if this one is Ok...\n\nBest regards,\n\nJomar\n\n",
"msg_date": "Mon, 3 Mar 2008 23:01:50 -0300",
"msg_from": "\"jomarba\" <jomarba@uol.com.br>",
"msg_from_op": true,
"msg_subject": "pt_br translation"
}
... |
[
{
"msg_contents": "Sorry for sending this directly to the hackers mailing list. But, i think it did not belong in any other as it involves internals about the files in the data directory.\n\n\nPostgreSQL version: 8.1.2\nOperating system: Windows XP/2003\n\nThe PostgreSQL service is not starting anymore. When ... |
[
{
"msg_contents": "I'm working on implementing pg_check functions. Because I want to test \nwhole table I need catch a error and handle it myself. I use following \nconstruct:\n\n\n\nPG_TRY();\n{\n\t...\n\tereport(ERROR, (errmsg(\"Error test\")));\n\t...\n}\nPG_CATCH();\n{\n\terrcontext(\"Context error\");\n ... |
[
{
"msg_contents": "This morning I had a postgres 8.3 install core this morning while multiple\nvacuum fulls were taking place. I saved the core file, would anyone be\ninterested in dissecting it? I've otherwise had no issues with this machine\nor pgsql install.\nGavin\n\nThis morning I had a postgres 8.3 insta... |
[
{
"msg_contents": "bruce wrote:\n> Joshua D. Drake wrote:\n> > -----BEGIN PGP SIGNED MESSAGE-----\n> > Hash: SHA1\n> > \n> > On Mon, 3 Mar 2008 20:33:08 -0500 (EST)\n> > Bruce Momjian <bruce@momjian.us> wrote:\n> > \n> > > Added to TODO based on this discussion:\n> > > \n> > > o Allow pg_dump to utilize... |
[
{
"msg_contents": "I've got a patch to psql that offers a \"no wider than terminal\" option, \npatched against cvs head. Would anyone be willing to test this before I \nsubmit the patch?\n\n# \\pset format aligned-wrapped\n# \\pset border 2\n# select * from distributors order by did;\nWord wrap debug: rows=11 ... |
[
{
"msg_contents": "Yesterday, we notified -hackers of the latest version of the libpq\ntype system. Just to be sure the right people are getting notified,\nwe are posting the latest patch here as well. Would love to get some\nfeedback on this.\n\nThe latest version of libpq type system is available here:\nhtt... |
[
{
"msg_contents": "Hi,\n\nRecent version of NetBSD/MIPS support dlopen. This patch makes the \nnetbsd dynloader tests availability of dlopen on HAVE_DLOPEN rather \nthan on __mips__\n\nTested on NetBSD 4.99.20 on mips\n\nI plan on registering a buildfarm member once this patch is in (and \nmaybe after upgrad... |
[
{
"msg_contents": "It says\n\n (1 row)\n \n SELECT import_succeed();\n+ NOTICE: ('import failed -- No module named whrandom',)\n import_succeed \n! ----------------------------------------\n! failed, that wasn't supposed to happen\n (1 row)\n \n -- test import and simple argume... |
[
{
"msg_contents": "While working on the previously discussed refactoring of\nheap_page_prune, I came to the realization that its use of\nCacheInvalidateHeapTuple is not just a PANIC risk but simply wrong :-(\nThe semantics aren't right: inval.c assumes that it's dealing with\ntransactional invalidations, but wh... |
[
{
"msg_contents": "I'm not sure what value a tsquery has if it's composed from stopwords \nonly, but it doesn't seem to be null or equal to itself.\n\nThat strikes me as ... unintuitive, although I'm happy to be re-educated \non this.\n\nI think it's because CompareTSQ (tsquery_op.c, line 142) doesn't have a \n... |
[
{
"msg_contents": "=> SHOW search_path;\n search_path\n-------------\n beta\n(1 row)\n\n=> CREATE OR REPLACE FUNCTION func_b() RETURNS SETOF int AS $$\n SELECT id FROM table_a;\n$$ LANGUAGE sql SET search_path = alpha;\nERROR: relation \"table_a\" does not exist\nCONTEXT: SQL function \"func_b\"\n\n... |
[
{
"msg_contents": "Peter complained awhile back about unrealistically small selectivity\nestimates for LIKE with a fixed-prefix pattern:\nhttp://archives.postgresql.org/pgsql-hackers/2007-12/msg00939.php\nI had suspected that this was related to the locale-specific problems\nwe fixed a few months ago, but havin... |
[
{
"msg_contents": "I have an idea for this TODO item:\n\n\t* Allow administrators to safely terminate individual sessions either\n\t via an SQL function or SIGTERM\n\t\n\t Lock table corruption following SIGTERM of an individual backend\n\t has been reported in 8.0. A possible cause was fixed in 8.1, but\n\... |
[
{
"msg_contents": "Hi all,\nWe have a web app that is using a 32 bit 8.1.4 (I know but upgrading\nis not an option for another couple of months...) running on Suse 10.\nWe just increased from 3GO to 6GO of RAM, and I increased the various\nmemory related values...\n\nFirst info - the server ONLY does one pg db ... |
[
{
"msg_contents": "The first commitfest has been on for about a week now, but I haven't \nseem much festivities going on.\n\nDo we plan to drain Bruce's patch queue completely during this \ncommitfest? Or the items on the developer wiki TODO:PatchStatus list? \nWhen do we declare the commitfest to be over?\n\nB... |
[
{
"msg_contents": "It's not clear how this commitfest thing is supposed to work in \npractice. May I suggest that:\n\n1. When a patch author wants to have a patch reviewed in the next \ncommitfest, he posts it to pgsql-patches as usual, and then adds it to \nthe list on the Todo:PatchStatus page (or perhaps eve... |
[
{
"msg_contents": "Log Message:\n-----------\nAdd:\n\n> \to Add SQLSTATE severity to PGconn return status\n> \n> \t http://archives.postgresql.org/pgsql-interfaces/2007-11/msg00015.php\n\nModified Files:\n--------------\n pgsql/doc:\n TODO (r1.2274 -> r1.2275)\n (http://anoncvs.postgresql.org/... |
[
{
"msg_contents": "Related to the concurrent discussion about selectivity estimations ...\n\nWhat is the reason the statistics target is limited to 1000? I've seen more \nthan one case where increasing the statistics target to 1000 improved results \nand one would have wanted to increase it further.\n\nWhat's ... |
[
{
"msg_contents": "Hello,\n I would like to be able to add CONSTRAINT, or DEFAULT with out\naffecting old rows. Yes, it sounds strange, but... Let's say I have\nbig table, I want to add new column, with DEFAULT and NOT NULL.\nNormally it means long exclusive lock. So - right now I'm adding plain\nnew column, t... |
[
{
"msg_contents": "PosgreSQL hackers,\n\nHere's an initial bit of my attempt at cleaning up the the timestamp datatype.\nI have gone through the backend and made a couple small changes to stop using\nthe HAVE_INT64_TIMESTAMP define to select a type in code by creating typedefs\nin a header and using the typedef... |
[
{
"msg_contents": "PosgreSQL hackers,\n\nHere's an initial bit of my attempt at cleaning up the the timestamp datatype.\nI have gone through the backend and made a couple small changes to stop using\nthe HAVE_INT64_TIMESTAMP define to select a type in code by creating typedefs\nin a header and using the typedef... |
[
{
"msg_contents": "Hello,\n I would like to be able to add CONSTRAINT and/or DEFAULT with out\naffecting old rows. Yes, it sounds strange, but... Let's say I have\nbig table, I want to add new column, with DEFAULT and NOT NULL.\nNormally it means long exclusive lock. So - right now I'm adding plain\nnew column,... |
[
{
"msg_contents": "I've gone through the patch queue and marked all or at least most of\nthe threads that don't contain any commit-fest material. (The remaining\nthreads contain actual patches, or at least design discussions that seem\nworth getting more eyeballs on at the present time.) Could you pull out\na... |
[
{
"msg_contents": "\nA while back Tom Lane presented the potential problem with gcc signed \noverflow handling, but now it's not just a theoretical problem.\n\nhttp://archives.postgresql.org/pgsql-hackers/2005-12/msg00635.php\n\nGcc 4.3 has started to perform optimizations based on the denial of the \nexistence... |
[
{
"msg_contents": "== Postgres Weekly News - March 09 2008 ==\n\n== Postgres Product News ==\n\nBenetl 1.6 for Windows released.\nhttp://www.benetl.net\n\nContinuent uni/cluster for PostgreSQL 2007.1 Update 2 released.\nhttp://www.continuent.com\n\nImage2db 2.2 released.\nhttp://www.vive.net/products/image2db.h... |
[
{
"msg_contents": "Hi,\n\nwhat's your opinion on this?\nI saw response only from Alvaro on the -patches list.\n\nThanks in advance,\nZoltán Böszörményi\n\n-------- Eredeti üzenet --------\nTárgy: \tRe: [PATCHES] 64-bit CommandIds\nDátum: \tTue, 04 Mar 2008 21:52:25 +0100\nFeladó: \tZoltan Boszormenyi <zb@cybert... |
[
{
"msg_contents": "I'm reviewing this patch\nhttp://archives.postgresql.org/pgsql-patches/2007-04/msg00531.php\nwhich contains this configure.in code to decide if it's safe to allow\nuse of non-segmented files:\n\n if test $ac_cv_func_fseeko = yes; then\n AC_SYS_LARGEFILE\n fi\n \n+ if test \"$ac_cv_sys_lar... |
[
{
"msg_contents": "Does it make any sense to allow LISTEN or UNLISTEN in a prepared\ntransaction?\n\nIt's certainly not sensical for these actions to affect the backend that\nactually executes the COMMIT PREPARED, in the sense of creating or\ndestroying pg_listener entries for it. But how can we say that they\... |
[
{
"msg_contents": "Hi,\n\nI have encountered a strange pg_ctl's behavior in 8.3.\n\npg_ctl -w -o \"-p 5432\" start\t<-- works\npg_ctl -w -o \"-i\" start\t<-- works\npg_ctl -w -o \"-p 5432 -i\" start\t<-- doesn't work\n\nIn the last case, even postmaster starts successfully, pg_ctl keeps\ntrying to make sure tha... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.