threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "A very minor point, but I found when reading dumps it makes more sense\nthat in the CREATE SEQUENCE command MINVALUE comes before MAXVALUE.\nObjections to this patch?",
"msg_date": "Sat, 19 Dec 2009 23:15:22 +0200",
"msg_from": "Peter Eisentraut <peter_e@gmx.net>",
"msg_from_op": ... |
[
{
"msg_contents": "Here's a patch:\n\ndiff --git a/doc/src/sgml/backup.sgml b/doc/src/sgml/backup.sgml\nindex 6750db8..1276c39 100644\n*** a/doc/src/sgml/backup.sgml\n--- b/doc/src/sgml/backup.sgml\n*************** if (!triggered)\n*** 2018,2024 ****\n <itemizedlist>\n <listitem>\n <para>\n! ... |
[
{
"msg_contents": "In the event that my plpython3 patch does not make it, it seems prudent to try and get a *much* smaller patch in to allow the PL to easily exist out of core.\n\nI added a couple SPI functions in order to support the database access functionality in plpython3u. Also, a getelevel() function for... |
[
{
"msg_contents": "Alpha3 has been bundled and is available at\n\nhttp://developer.postgresql.org/~petere/alpha/\n\nPlease check that it is sane.\n\nNo one has written an announcement yet. I will do it unless someone\nfeels inspired. Release should be Monday or Tuesday.\n\n",
"msg_date": "Sun, 20 Dec 2009... |
[
{
"msg_contents": "It appears that the function for validating generic options to a FDW,\nSERVER and USER MAPPING is always passed a catalog oid of 0. Whereas\nit should actually be passed the oid of the catalog that the options\napply to.\n\nAttached patch fixes the issue by passing the proper catalog id from\... |
[
{
"msg_contents": "I just got done fixing a different problem in that area, and then I\nnoticed this:\n\nregression=# create table src (f1 text);\nCREATE TABLE\nregression=# create index srclower on src(lower(f1));\nCREATE INDEX\nregression=# comment on column srclower.pg_expression_1 is 'a comment';\nCOMMENT\n... |
[
{
"msg_contents": "Hi\n\nI've completed a (first) working version of a extension that allows\neasier introspection of composite types from SQL and pl/PGSQL.\n\nThe original proposal and ensuing discussion can be found here:\nhttp://archives.postgresql.org/pgsql-hackers/2009-11/msg00695.php\n\nThe extension can ... |
[
{
"msg_contents": "Hello\n\nI am thinking about implementation of median function. This function\nshould be implemented in two ways:\n\na) direct entering an ORDER BY clause for median funcall in gram.y\nb) general support for \"preordered aggregates\".\n\nI prefer plan b, because there are more similar aggrega... |
[
{
"msg_contents": "Greetings,\n\nFollowing up on my experimental patch last month to revamp the BKI\ninfrastructure, I am proposing a less invasive set of changes with the\nhope of offering something committable. Some of these were discussed\nby Robert Haas and others last summer.\n\n1. Remove all DATA() declar... |
[
{
"msg_contents": "Hi Simon, Hi all,\n\n\nHS currently does the following in GetConflictingVirtualXIDs\n\n\nTransactionId pxmin = proc->xmin;\n\n/*\n * We ignore an invalid pxmin because this means that backend\n * has no snapshot and cannot get another one while we hold exclusive lock.\n */\nif (TransactionIdI... |
[
{
"msg_contents": "Attached is a WIP patch for addressing the problems mentioned in this\nthread:\nhttp://archives.postgresql.org/pgsql-hackers/2009-12/msg01764.php\n\nThe main things that it does are (1) consider all index columns, not\njust the first one as formerly; and (2) try to generate a usable name\nfor... |
[
{
"msg_contents": "-----BEGIN PGP SIGNED MESSAGE-----\nHash: SHA1\n\nHello\n\nI wonder why the function pg_relation_size(text) does not take into\naccount the space used by toast data in a table when returning the space\nused by the table.\n\nAs an administrator I would expect pg_total_relation_size() to return... |
[
{
"msg_contents": "Hiroyuki Yamada <yamada@kokolink.net> wrote:\n \n> 4. Startup process tries to redo XLOG_HEAP2_CLEAN record, calls\n> LockBufferForCleanup() and freezes until the Xact 1 ends.\n \nI think they word you're searching for is \"blocks\". Blocking to\nprotect integrity doesn't sound like a b... |
[
{
"msg_contents": "At least I think it's a segfault. This function returns a vstring:\n\nCREATE OR REPLACE FUNCTION wtf(\n) RETURNS text LANGUAGE plperl IMMUTABLE STRICT AS $X$\n return $^V;\n$X$;\n\nHere's what happens when I call it:\n\ntry=# select wtf();\nserver closed the connection unexpectedly\n\tThis... |
[
{
"msg_contents": "I'm about ready to post the next draft of my plperl feature patch.\n\nWhen I looked at the diff I saw ~7800 lines of it were just due to\nupdating ppport.h. So I've broken that out into this large but totally\ntrivial patch.\n\nTim.",
"msg_date": "Mon, 21 Dec 2009 21:19:12 +0000",
"ms... |
[
{
"msg_contents": "Building on my earlier plperl refactoring patch, and the follow-on\nppport.h patch, here's a second draft of my first plperl feature patch.\n\nSignificant changes in this patch:\n\n- New GUC plperl.on_perl_init='...perl...' for admin use.\n- New GUC plperl.on_trusted_init='...perl...' for plp... |
[
{
"msg_contents": "Hi,\n\nI found there is no \"primary\" tag for the HS parameters\nin config.sgml. Attached patch adds that tag.\n\nRegards,\n\n-- \nFujii Masao\nNIPPON TELEGRAPH AND TELEPHONE CORPORATION\nNTT Open Source Software Center",
"msg_date": "Tue, 22 Dec 2009 11:25:27 +0900",
"msg_from": "Fu... |
[
{
"msg_contents": "With regards to this bug report:\nhttp://archives.postgresql.org/pgsql-bugs/2009-12/msg00241.php\n\nI think we should change tuplestore code so that callers of\ntuplestore_put* don't need to switch to the correct memory context (and\nresource owner, after this patch) before call. Instead,\ntu... |
[
{
"msg_contents": "Some ideas to improve current behaviour of SR\n\nhttp://wiki.postgresql.org/wiki/Streaming_Replication\n\nThe current startup process is copied below. (7) gives some issues if it\nis a very long step, notably that the master may fill with data and then\nbreak off the connection before replica... |
[
{
"msg_contents": "Hi,\n\nI've been playing with using a Simulated Annealing-type algorithm for\ndeterminig join ordering for relations. To get into context see\nhttp://archives.postgresql.org/pgsql-hackers/2009-05/msg00098.php\n(there's also a TODO in the wiki). There's a nice paper on that in\nhttp://referen... |
[
{
"msg_contents": "There is code in src/port/win32env.c to update the environment within\nthe msvcrt.dll runtime even when running in msvcr70.dll (visual C++\n2005). It updates that one, the local one *and* the system\nenvironment.\n\nHowever, in case there is no msvcrt.dll linked into the process, it\nwill abo... |
[
{
"msg_contents": "Log Message:\n-----------\nRemove code that attempted to rename index columns to keep them in sync with\ntheir underlying table columns. That code was not bright enough to cope with\ncollision situations (ie, new name conflicts with some other column of the\nindex). Since there is no functi... |
[
{
"msg_contents": "In connection with the operator-exclusion patch,\nBrendan Jurd <direvus@gmail.com> wrote:\n> * What to do about INCLUDING INDEXES EXCLUDING CONSTRAINTS --\n> Postgres gets this wrong for unique indexes currently. Should we\n> persist with the existing behaviour or fix it as part of this pat... |
[
{
"msg_contents": "Hello dear hackers,\n\nI wonder what is the future of \"_PG_fini\", documentation say at [1]:\n\"Note that _PG_fini will only be called during an unload of the file,\nnot during process termination. (Presently, unloads are disabled and\nwill never occur, but this may change in the future.)\"\... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupt WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit stru... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupt WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit stru... |
[
{
"msg_contents": "Hi,\n\nIn UpdateMinRecoveryPoint() and XLogNeedsFlush(), updateMinRecoveryPoint\nis used for us to short-circuit future checks only during a crash recovery.\nBut it doesn't seem to play its role in a crash recovery that follows an\narchive recovery. Because such crash recovery always starts f... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupt WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit stru... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupt WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit stru... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupted WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit st... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupted WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit st... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupted WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit st... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupted WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit st... |
[
{
"msg_contents": "Hello,\n\nI was reading GiST core codes when I found an XLogInsert()\ncall that can produce a corrupted WAL record.\n\n== Summary ==\nThere is an execution path that produces a WAL record whose\nxl_info indicates XLOG_GIST_PAGE_UPDATE while the record\nactually contains a gistxlogPageSplit st... |
[
{
"msg_contents": "Hello\n\nI try to explain my motivation for creating this patch\nhttps://commitfest.postgresql.org/action/patch_view?id=224 .\n\nParametrised queries are supported in PostgreSQL long time. Using the\nparametrised queries is point of all advices about good programming\nstyle. On application le... |
[
{
"msg_contents": "Dear PG hackers,\n\nI have two question regarding Unicode support in PG:\n\n1) If I set my database and connection encoding to UTF-8, does pg (and\nfuture versions of it) guarantee that unicode code points are stored\nunmodified? or could it be that pg does some unicode\nnormalization/manipul... |
[
{
"msg_contents": "Hi,\n\npg_hba.conf accepts some keywords like 'all', 'sameuser' and so on.\nLikewise, I'd like to support the same keywords for .pgpass file.\nThought?\n\nIn Streaming Replication, new keyword 'replication' has been introduced\ninto pg_hba.conf to authenticate the standby server. If my propos... |
[
{
"msg_contents": "Dear Sir,\n\nI am a 3rd year B.Tech. (Computer Science Engineering) student at IIIT,\nHyderabad (India). For the past 5 months I've been trying to implement a\nmodule in postgresSQL and I am stuck at a point.\nI'll be highly grateful to you if you could help me out.\n\nThe module is about con... |
[
{
"msg_contents": "Hi everyone,\n\nI've been playing with the 8.5alpha3 in the last few days. Among other \nthings, I'm making sure that the pgsql PHP extensions still work \ncorrectly with the new version. It would seems so, as all the errors in \nthe standard pgsql extension test suite come from the fact that... |
[
{
"msg_contents": "Hello\n\nI am returning to discussion\nhttp://archives.postgresql.org/pgsql-hackers/2009-12/msg01378.php\n\nI propose a new aggregate function - listagg. This function\nconcatenate values to string. If this function is used with two\nparameters, then second parameter is used as delimiter. NUL... |
[
{
"msg_contents": "Hello\n\nIt is maybe a bug. I cannot use ORDER BY clause in two parameter aggregate.\n\ncreate aggregate la(text, text) (SFUNC=listagg2_transfn,\nSTYPE=internal, FINALFUNC=listagg_finalfn);\n\npostgres=# select la(town,',') from country;\n la\n-------------------------------\n Pr... |
[
{
"msg_contents": "I'm not sure whether we ever posted this schedule anywhere official -\nif so, I can't find it - but my understanding is that we have\nconsensus on the release schedule described here:\n\nhttp://archives.postgresql.org/pgsql-hackers/2009-09/msg01251.php\n\nUnder this schedule, the last CommitF... |
[
{
"msg_contents": "Hi everyone,\n\nI'm glad to announce that the pgsql and PDO PHP extensions test suites \nare now passing when used with 8.5-cvs. Mostly it was just a matter of \nupdating the tests themselves, but a bug in PDO_PgSQL dealing with the \nnew hex format affecting only the 5.2.x branch was fixed d... |
[
{
"msg_contents": "On Sat, Nov 14, 2009 at 6:36 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:\n> There's an example in parse_oper.c of a specialized cache that's about\n> as complicated as this would be.\n\nI was just taking a look at find_oper_cache_entry() and noticed\nsomething odd. When we discover that OprCache... |
[
{
"msg_contents": "The JDBC driver's regression test suite has revealed a change in behavior \nintroduced by the hot standy patch. Previously when a client sent a \ncancel request on an idle connection, nothing happened. Now it sends an \nerror message \"ERROR: canceling statement due to user request\". This... |
[
{
"msg_contents": "Hi,\n\nWhile unlikely to cause issues two new LWLockAcquire calls use the wrong \nlocking mode.\nPatch attached.\n\nAndres",
"msg_date": "Sun, 27 Dec 2009 20:12:32 +0100",
"msg_from": "Andres Freund <andres@anarazel.de>",
"msg_from_op": true,
"msg_subject": "Small locking bugs... |
[
{
"msg_contents": "We can define variables with any names in pgbench,\nbut only can refer them with names that consist of [A-Za-z0-9_]+ .\nIt could cause confusion discussed here:\nhttp://archives.postgresql.org/message-id/4B272833.8080500@2ndquadrant.com\n\nThe attached patch verifies variable names at definit... |
[
{
"msg_contents": "Hi,\n\nattached is a small patch that makes it possible for clients\nto receive row count for SELECT ... INTO ... and CREATE TABLE ... AS ...\n\nComments?\n\nBest regards,\nZolt�n B�sz�rm�nyi\n\n-- \nBible has answers for everything. Proof:\n\"But let your communication be, Yea, yea; ... |
[
{
"msg_contents": "In 8.3, running \\c from a file prints something like\n\nYou are now connected to database \"postgres\".\n\nIn 8.4 it prints\n\npsql (8.4.1)\nYou are now connected to database \"postgres\".\n\nIs it intentional/sensible to repeat the startup banner every time the\nconnection changes, or was t... |
[
{
"msg_contents": "For perspective: Since our current application framework is about\nten years old now, the Wisconsin State Courts System has put\ntogether a \"Long-Term Application Development Group\" to review all\naspects of our development and production runtime environments. We\nstarted by reviewing var... |
[
{
"msg_contents": "In ExecTidReScan, we have the following:\n\n /* If we are being passed an outer tuple, save it for runtime\nkey calc */\n if (exprCtxt != NULL)\n node->ss.ps.ps_ExprContext->ecxt_outertuple =\n exprCtxt->ecxt_outertuple;\n\nIs this dead code... |
[
{
"msg_contents": "This paper has a brief but interesting discussion of Admission\nControl in section 2.4:\n \nArchitecture of a Database System. (Joseph M. Hellerstein, Michael\nStonebraker and James Hamilton). Foundations and Trends in Databases\n1(2).\n \nhttp://db.cs.berkeley.edu/papers/fntdb07-architecture... |
[
{
"msg_contents": "Has anyone built postgresql (or just libpq.a) on Windows SFU/SUA?\nWould prefer to not reinvent any wheels....\n\nWe have a number of Unix/Linux applications that are also compiled\nunder Microsoft Windows SFU 3.5. We need to have a SFU-compatible\nlibpq.a, not a complete install. The regul... |
[
{
"msg_contents": "Is there some way to export the postgresql query parse tree in XML format? I can not locate the API/Tool etc to do that...\n\nthanks\n-Matt\n\n\n \n",
"msg_date": "Mon, 28 Dec 2009 13:30:44 -0800 (PST)",
"msg_from": "matt <nuc233@yahoo.com>",
"msg_from_op": true,
"msg_sub... |
[
{
"msg_contents": "Following up on the discussion here\nhttp://archives.postgresql.org/message-id/4B3875C6020000250002D97D@gw.wicourts.gov\nI'd like to propose making the following changes that would allow saner\nplanning for queries involving inheritance:\n\n1. Currently the primary key of pg_statistic is (sta... |
[
{
"msg_contents": "I was thinking about true serializability, and started thinking that\nthere are a lot of special cases where true serializability can be\nachieved without modification.\n\nFor instance, the following is problematic:\n\n BEGIN;\n SELECT count(*) FROM mytable;\n ...\n\nbecause future insert/... |
[
{
"msg_contents": "While inspecting a complain from a pgpool user, I found that\nPostgreSQL crushes with following statck trace:\n\n#0 0x0826436a in list_length (l=0xaabe4e28)\n at ../../../src/include/nodes/pg_list.h:94\n#1 0x08262168 in IsTransactionStmtList (parseTrees=0xaabe4e28)\n at postgres.c:242... |
[
{
"msg_contents": "I found the struct IntArray defined in c.h is actually used only in\nexecQual.c. ISTM the definition should be at least moved to the right\nplace.\n\nAttached is a trivial fix. Addition to the explanation above, I\nreplaced IntArray by simple int array bounded with MAXDIM and remove\nlocal v... |
[
{
"msg_contents": "hi\n\nwhen a superuser is logged in to the postgres server ,he has priveledge to access all the databases.\n\ncan we hide the structure of some of the databases from the superuser?\nlike if he gives command to open such a database the postgres may ask for access password again. \n\nas in Orac... |
[
{
"msg_contents": "Hi all,\n\nSince moving PostGIS over to PGXS for the 1.4 release, we've been \nlooking at how we can support multiple versions of PostGIS being \ninstalled in different databases within the same cluster.\n\nWe are able to version the .so file produced by PGXS without too much \ndifficulty, ho... |
[
{
"msg_contents": "Hello\n\nI am working on new patch. I see a problem with copy quote_ident on\nclient side. This function call ScanKeywordLookup function.\n\n const ScanKeyword *keyword = ScanKeywordLookup(ident,\n\n ScanKeywords,\n\n ... |
[
{
"msg_contents": "Hi,\n\nI try to build msvc version of postgres using the current cvs and\n get the following error.\n\n Unable to determine vcbuild version from first line of output! at\n src/tools/msvc/Solution.pm line 70.\n\nThe error comes from the following output of vcbuild /? in my\n environment.\n... |
[
{
"msg_contents": "All,\n\nOne of our clients is having query plan issues with a table with a\ncontinuously advancing timestamp column (i.e. one with default now()).\nThe newest rows, which are the most in demand, are always estimated to\nbe fewer than they are or even non-existant. As a result, the user has\n... |
[
{
"msg_contents": "With the update_process_title parameter set to off some PostgreSQL\nprocesses still change their ps title to a different name than the\ndefault. I appreciate this setting came about for performance reasons\nwhich the logger, wal writer, autovacuum, and stats collector would\nnot have but I ac... |
[
{
"msg_contents": "Hi.\n\ntest/example does not support win32.\nAlthough I posted also in the past, I am slightly persistent.\nI think whether it also needs correction of a document. \n\n== CVS-HEAD on as for MinGW + gcc ==\n\ntestlibpq2.c: In function `main':\ntestlibpq2.c:98: error: `fd_set' undeclared (first... |
[
{
"msg_contents": "hi,\n \ni assigned super user privelledge to a user by specifing entries in pg_hba.conf file as \n\nhost all newuser 127.1.1.1 md5\n\nand the default postgres user is made access to only the default postgres databse and is no more the super user.\n\nthe problem is that this conf file... |
[
{
"msg_contents": "hi,\n\ni assigned super user privelledge to a user by specifing entries in pg_hba.conf file as \n\nhost all newuser 127.1.1.1 md5\n\nand the default postgres user is made access to only the default postgres databse and is no more the super user.\n\nthe problem is that this conf file ... |
[
{
"msg_contents": "Hi,\n\nAs far as I tested, the krb_server_keyfile setting\nin postgres.conf doesn't work on Windows.\n\nBecause gssapi32.dll(krb5_32.dll) seems to call\ngetenv(\"KRB5_KTNAME\") in msvcr71, postgres.exe\nshould call putenv(\"KRB5_KTNAME=...\") in msvcr71.\nThe attached patch fixes the problem ... |
[
{
"msg_contents": "We got an interesting report on the testers list today:\n\nhttp://archives.postgresql.org/pgsql-testers/2009-12/msg00000.php\n\nBasically, configure failed on their OpenBSD system because thread \nsafety is on but the libxml2 wasn't compiled with threaded support: \nhttp://xmlsoft.org/thread... |
[
{
"msg_contents": "While waiting for feedback on my earlier plperl refactor and feature\npatches I'm working on a further patch that adds, among other things,\nfast inter-plperl-sp calling.\n\nI want to outline what I've got and get some feedback on open issues.\n\nTo make a call to a stored procedure from plpe... |
[
{
"msg_contents": "I've been reviewing code to get a better handle on the scope of\nchanges to support serializable transactions, in preparation for\nnext month's meetings with our CIO. My posts should start getting\nprogressively less hand-wavy. :-)\n \nI've come to a few conclusions:\n \n(1) The notions of... |
[
{
"msg_contents": "Hi,\n\nPostgreSQL RPM Building Project released RPM sets for 3rd Alpha of the\nupcoming 8.5 release.\n\nPlease note that these packages are **not** production ready. They are\nfor Fedora 7,8,9,11,12 and RHEL/CentOS 4,5.\n\nThese packages *do* require a dump/reload, even from the second alpha\... |
[
{
"msg_contents": "Back when we put in the ability to use \"x IS NULL\" as a btree search\ncondition, we intentionally left out \"x IS NOT NULL\", on the grounds\nthat it is comparable to \"x <> something\" which is not btree-searchable\neither. However, it occurs to me that we missed a bet here. The NOT\nNUL... |
[
{
"msg_contents": "Bruce Momjian wrote:\n \n> I must be missing something but I thought the only problem with our\n> existing snapshot system was that you could see a row updated after\n> your snapshot was created, and that the solution to that was to\n> abort the transaction that would see the new row. Can y... |
[
{
"msg_contents": "pg_migrator has become more popular recently, so it seems time to look\nat some enhancements that would improve pg_migrator. None of these are\nrequired, but rather changes that would be nice to have:\n\n1) Right now pg_migrator preserves relfilenodes for TOAST files because\nthis is requir... |
[
{
"msg_contents": "Hi - Does anyone know how to add a xml support function into postgresql?\n\nI want to add a function named xmlquery into postgresql.\n\nI modify gram.y by adding xmlquery relatedcode wherever other xml support\nfunctions appear.\nbut the parser can not find xmlquery, the makeXmlExpr is not ca... |
[
{
"msg_contents": "Attached is the fix pointed out in the previous CommitFest plus RANGE\noffset support.\n\n*fix\n- move window regression test to another parallel group, but regarding\nthe limitation of 20 per group the union test goes to the group the\nwindow test belonged to.\n- allow NULL iswindowagg as an... |
[
{
"msg_contents": "Hi All,\n\n We are seeking to enable SET ROLE in security-definer functions, since\nD.E Shaw has a usage for this, and I think you'd also agree that SET ROLE is\nsecurity definer functions has it's uses.\n\n As the code stands right now, I see that the only concern against\nenabling SET... |
[
{
"msg_contents": "Hello all,\nI'm creating gist index for testing purpuses :)\n\nI created index element structure:\n\ntypedef struct moving_object\n{\n double x_high;\n double y_high;\n double x_low;\n double y_low;\n time_t mov_time;\n double x_plus;\n double y_plus;\n double x_minus;... |
[
{
"msg_contents": "Hi All,\n\n We are seeking to enable SET ROLE in security-definer functions, since @\nD.E Shaw there are scripts from the past that used this feature, and I think\nyou'd also agree that SET ROLE is security definer functions has it's uses.\n\n As the code stands right now, I see that th... |
[
{
"msg_contents": "\"Albe Laurenz\" wrote:\n\n> See the example I concocted in\n> http://archives.postgresql.org/pgsql-hackers/2009-05/msg00316.php\n \nSure, let's look at that example. Of course, *any* transaction run\nby itself won't show differences from true serializable behavior\n*regardless* of the mode... |
[
{
"msg_contents": "\"Albe Laurenz\" wrote:\n \n> But with \"snapshot isolation\" (what our \"serializable\" corresponds\n> to) you cannot see rows updated after snapshot creation, right?\n> \n> So phantom reads cannot occur, but we still are not truly\n> serializable.\n \nMy previous reply missed your point en... |
[
{
"msg_contents": "Greg Stark wrote:\n \n> The unsolved problems that have been raised are:\n> [legion]\n \nYeah, that's why this is a two to four year project. And I would\npoint out that if there *wasn't* a performance hit in serializable\nmode, none of the other isolation levels would exist. These less\nr... |
[
{
"msg_contents": "Attached patch is the part of the win64 patch that changes Datum to be\nuintptr_t, and associated changes, with only very minor changes from\nme. It also includes autoconf tests that I tricked Bruce into fixing\nfor me :-)\n\nComments?\n\nUnless there are objections, I'll go ahead and apply t... |
[
{
"msg_contents": "Greg Stark wrote:\n \n> Hm, this raises the issue that you'll have to figure out what\n> should happen if two different transactions are using different\n> isolation modes. Currently our two isolation modes only control\n> behaviour within your transaction so they co-exist perfectly fine.\n>... |
[
{
"msg_contents": "Robert Haas wrote:\n \n> It seems to me that the hard part of this problem is to describe\n> the general mechanism by which conflicts will be detected, with\n> specific references to the types of data structures that will be\n> used to hold that information.\n \nWell, the general approach to... |
[
{
"msg_contents": "Robert Haas wrote:\n \n> OK. I think it will become more clear whether the existing lock\n> data structures are adequate as you move into detailed design.\n \nI've gotten far enough in reviewing it to be pretty convinced that\nthey'll cover all the granularities I'm likely to want unless I g... |
[
{
"msg_contents": "Here's an updated version of the patch I posted here:\n\nhttp://archives.postgresql.org/pgsql-hackers/2009-12/msg01814.php\n\nVersion 2 has the following changes:\n\n1. It has been tested and debugged on Windows.\n2. As suggested by Robert Haas, I have renamed the BKI scripts to\ngenbki.pl an... |
[
{
"msg_contents": "I wrote:\n \n> It's just a little bit of a stretch to call SILOCKs locks, because\n> they don't actually block anything. They are used at various points\n> to see where a transaction is reading data which has been modified\n> by another transaction or vice versa. And they do need to be kept\n... |
[
{
"msg_contents": "Jeff Davis wrote:\n \n> I started a wiki page here:\n>\n> http://wiki.postgresql.org/wiki/Serializable\n \nI'll try to get that filled in with something useful over the\nweekend. I'm heading to a party soon, and may not be in shape to\nwork on it tomorrow.... ;-)\n \nHappy New Year, all!\n... |
[
{
"msg_contents": "Fellow Hackers,\n\nGiven this SQL:\n\n BEGIN;\n\n CREATE ROLE foo WITH NOLOGIN;\n\n CREATE ROLE foo_bar WITH LOGIN PASSWORD '***' INHERIT IN ROLE foo;\n\n CREATE ROLE foo_baz WITH LOGIN PASSWORD '***' INHERIT IN ROLE foo;\n\n CREATE ROLE foo_yow WITH LOGIN PASSWORD '***' INHERI... |
[
{
"msg_contents": "Greg Stark wrote:\n \n> aaah... I think I see where we've gone off track in previous\n> discussions...you think postgres keeps row level locks in a shared\n> memory data structure. It doesn't it stores all row level locks\n> *in* the tuple itself. It only stores the lock in memory briefly\n>... |
[
{
"msg_contents": "Robert Haas wrote:\n \n> What predicate locking? If you take ACCESS EXCLUSIVE locks on every\n> read, that should serialize all access to every table. Predicate\n> locking wouldn't do anything, because the table would be completely\n> inaccessible to all competing transactions.\n \nYeah, tha... |
[
{
"msg_contents": "Hi,\n\n\nEvery time i explain what is the fsync parameter for, the next thing i\nalways say is: \"never turn it off\", and now that we have\nsynchronous_commit there is no good reason for turn fsync off... so\nwhy are we still let it be in the postgresql.conf where it's so\nvulnerable to a mi... |
[
{
"msg_contents": "I'd like to apply the attached patch to the msvc build environment,\nwhich changes how config.pl is handled. (For the unenlightened, this\nis the replacement stuff we have for autoconf)\n\nToday, there is one config.pl. We ship it with a couple of defaults\n(which points to a very old install... |
[
{
"msg_contents": "I have adapted the win64 patches a bit, and now have a working build.\nAs in it runs the regression tests fine. However, I have well over a\nthousand warnings of the type:\nconversion from 'size_t' to 'int', possible loss of data\n\nMy first 5-6 checks of where these happen are all cases wher... |
[
{
"msg_contents": "The win64 port has showed that we have two sockets declared\nincorrectly. They are supposed to be declared as SOCKET on win32, but\nthey are declared as int. See attached patch.\n\nGiven that SOCKET is actually defined as int on win32 (no warnings or\nanything there, just on win64), I'm incli... |
[
{
"msg_contents": "\"Albe Laurenz\" wrote:\n \n> If I remember right, one necessity for the SIREAD lock technique\n> was that SIREAD locks taken by a transaction have to be kept after\n> the transaction has ended.\n \nCorrect. An SIREAD lock must be held until completion of the last\nserializable transaction ... |
[
{
"msg_contents": "There has been periodic discussion here about allowing psql to deal\nwith multiple sessions, or possibly creating another tool to allow\nthis sort of test. Is anyone working on this?\n \nIt's very soon going to be critical that I be able to test particular\ninterleavings of statements in par... |
[
{
"msg_contents": "Hello all,\n\nWith New Year !\n\nI realized my own gist index, and now I want to debug it :) I want to use\namdb program\nhttp://gist.cs.berkeley.edu/libgist-2.0/install.html\nBut I have problem with installing it.\nWhen I running \"gmake cmdline\"\nIt doesn't find iostream.h file :( and uses... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.