threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "This fixes bug #18456 [1]. Since we're in back-branch release freeze,\nI'll just park it for the moment. But I think we should shove it in\nonce the freeze lifts so it's in 17beta1.\n\n\t\t\tregards, tom lane\n\n[1] https://www.postgresql.org/message-id/18456-82d3d70134aefd28%40postgresql.o... |
[
{
"msg_contents": "Dear PostgreSQL Hackers,\n\nI hope this email finds you well. As I delve into the codebase, I've\nencountered some challenges understanding how routes are implemented within\nthe application.\n\nAs I navigate through the codebase, I've encountered some challenges\nunderstanding how routes are... |
[
{
"msg_contents": "Hi hackers,\n\nMore or less by chance, I stumbled on a Security Technical Implementation\nGuide (STIG, promulgated by the US Dept. of Defense, Defense Information\nSystems Agency) for PostgreSQL (specific to PG 9.x, so a bit dated).\n\nThere is a rule in the STIG that pertains to PLs, and see... |
[
{
"msg_contents": "Hello hackers,\n\nWhile trying to catch a sporadic regression test failure, I've discovered\nthat tests tsearch and advisory_lock in the parallel_schedule's group:\ntest: select_views ... tsearch ... advisory_lock indirect_toast equivclass\n\nmight fail, depending on timing, because the test... |
[
{
"msg_contents": "On 2024-04-03 Wn 04:21, Andrew Dunstan\n> I don't think a cast that doesn't cater for all the forms json can take\nis going to work very well. At the very least you would need to error out\nin cases you didn't want to cover, and have tests for all of > > those\nerrors. But the above is only a... |
[
{
"msg_contents": "Greetings,\n\nI've been asked by the debezium developers if it is possible to include\nxid8 in the logical replication protocol. Are there any previous threads on\nthis topic?\n\nAny reason why we wouldn't include the epoch ?\n\nDave Cramer\n\nGreetings,I've been asked by the debezium develop... |
[
{
"msg_contents": "One-line Summary: There are use cases for long table names so people might\nuse Oracle and MS SQL Server because Postgres does not support table names\nlonger than 63 characters, so the max character limit should be increased\nto 128 or higher.\n\nBusiness Use-case: I want to create a table\n... |
[
{
"msg_contents": "In March, I noticed that a backend got stuck overnight doing:\n\nbackend_start | 2024-03-27 22:34:12.774195-07\nxact_start | 2024-03-27 22:34:39.741933-07\nquery_start | 2024-03-27 22:34:41.997276-07\nstate_change | 2024-03-27 22:34:41.997307-07\nwait_event_type | IO\nwait_... |
[
{
"msg_contents": "Currently, it is pretty easy to subvert the restrictions imposed\nby row-level security and security_barrier views. All you have to\nto is use EXPLAIN (ANALYZE) and see how many rows were filtered\nout by the RLS policy or the view condition.\n\nThis is not considered a security bug (I asked... |
[
{
"msg_contents": "Hi,\r\n\r\nPlease find the draft of the 2024-05-09 release announcement.\r\n\r\nPlease review for corrections and any omissions that should be called \r\nout as part of this release.\r\n\r\nPlease provide feedback no later (and preferably sooner) than 2024-05-09 \r\n12:00 UTC.\r\n\r\nThanks,\... |
[
{
"msg_contents": "Instead of needing to be explicit, we can just iterate the\npgstat_kind_infos array to find the memory locations to read into.\n\nThis was originally thought of by Andres in\n5891c7a8ed8f2d3d577e7eea34dacff12d7b6bbd.\n\nNot a fix, per se, but it does remove a comment. Perhaps the discussion \... |
[
{
"msg_contents": "Hi PostgreSQL Community,\n\nI have been working on partitioned tables recently, and I have noticed\nsomething that doesn't seem correct with the EXPLAIN output of an\nupdate/delete query with a returning list.\n\nFor example, consider two partitioned tables, \"t1\" and \"t2,\" with\npartition... |
[
{
"msg_contents": "Hi all,\n\nRecently I dealt with a server where PAM had hung a connection\nindefinitely, suppressing our authentication timeout and preventing a\nclean shutdown. Worse, the xmin that was pinned by the opening\ntransaction cascaded to replicas and started messing things up\ndownstream.\n\nThe ... |
[
{
"msg_contents": "Hi PostgreSQL Community,\n\nI would like to bring to your attention an observation regarding the\nplanner's behavior for foreign table update/delete operations. It appears\nthat the planner adds rowmarks (ROW_MARK_COPY) for non-target tables, which\nI believe is unnecessary when using the pos... |
[
{
"msg_contents": "Robert pointed out [1] that the planner fails if we have $SUBJECT,\nbecause tidpath.c can seize on the RLS-derived ctid constraint\ninstead of the CurrentOfExpr. Since the executor can only handle\nCurrentOfExpr in a TidScan's tidquals, that leads to a confusing\nruntime error.\n\nHere's a p... |
[
{
"msg_contents": "hi,\n\nSELECT table_name, column_name, is_updatable\n FROM information_schema.columns\n WHERE table_name LIKE E'r_\\\\_view%'\n ORDER BY table_name, ordinal_position;\n\nat d1d286d83c0eed695910cb20d970ea9bea2e5001,\nthis query in src/test/regress/sql/updatable_views.sql\nmakes regress tests ... |
[
{
"msg_contents": "pg_dump -Fc |pg_restore -l -N schema:\n\n| 2; 3079 18187 EXTENSION - pg_buffercache \n\nWithout -N schema also shows:\n\n| 2562; 0 0 COMMENT - EXTENSION pg_buffercache \n\nI mean literal s-c-h-e-m-a, but I suppose anything else will work the\nsame.\n\nBTW, I noticed that pg_restore -v shows t... |
[
{
"msg_contents": "Hi,\n\nNazir Bilal Yavuz <byavuz81(at)gmail(dot)com> wrote:\n\n>Any kind of feedback would be appreciated.\n\nI know it's coming from copy-and-paste, but\nI believe the flags could be:\n- dstfd = OpenTransientFile(tofile, O_RDWR | O_CREAT | O_EXCL | PG_BINARY);\n+ dstfd = OpenTransientFile(to... |
[
{
"msg_contents": "In this commit:\n\n\tcommit 34768ee3616\n\tAuthor: Peter Eisentraut <peter@eisentraut.org>\n\tDate: Sun Mar 24 07:37:13 2024 +0100\n\t\n\t Add temporal FOREIGN KEY contraints\n\t\n\t Add PERIOD clause to foreign key constraint definitions. This is\n\t supported for range and multi... |
[
{
"msg_contents": "Hi,\n\nAs you may know, aggregates like SELECT MIN(unique1) FROM tenk1; are\nrewritten as SELECT unique1 FROM tenk1 ORDER BY unique1 USING < LIMIT\n1; by using the optional sortop field in the aggregator.\nHowever, this optimization is disabled for clauses that in itself have\nan ORDER BY cla... |
[
{
"msg_contents": "I happened to notice that the comment for AlterObjectNamespace_oid\nclaims that\n\n * ... it doesn't have to deal with certain special cases\n * such as not wanting to process array types --- those should never\n * be direct members of an extension anyway.\n\nThis struck me as probably broken... |
[
{
"msg_contents": "Fix overread in JSON parsing errors for incomplete byte sequences\n\njson_lex_string() relies on pg_encoding_mblen_bounded() to point to the\nend of a JSON string when generating an error message, and the input it\nuses is not guaranteed to be null-terminated.\n\nIt was possible to walk off t... |
[
{
"msg_contents": "Hello hackers,\n\nLooking at a recent failure on the buildfarm:\nhttps://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=morepork&dt=2024-04-30%2020%3A48%3A34\n\n# poll_query_until timed out executing this query:\n# SELECT archived_count FROM pg_stat_archiver\n# expecting this output:\n# 1\n#... |
[
{
"msg_contents": "I have committed the first draft of the PG 17 release notes; you can\nsee the results here:\n\n\thttps://momjian.us/pgsql_docs/release-17.html\n\nIt will be improved until the final release. The item count is 188,\nwhich is similar to recent releases:\n\n\trelease-10: 189\n\trelease-11: 1... |
[
{
"msg_contents": "In [0] I had noticed that we have no automated verification that global \nvariables are declared in header files. (For global functions, we have \nthis through -Wmissing-prototypes.) As I mentioned there, I discovered \nthe Clang compiler option -Wmissing-variable-declarations, which does \... |
[
{
"msg_contents": "Hi everyone, I just installed Postgres and pg_tle extension as I was\nlooking to contribute to pg_tle.\n\nSomehow, I am unable to update the shared_preload_libraries. It feels like\nALTER has happened but the SPL value is not updated:\n\n> test=# show shared_preload_libraries;\n> shared_prel... |
[
{
"msg_contents": "Hi hackers,\n\nI detected two problems about ECPG.\nI show my opinion. Please comment.\nIf it's correct, I will prepare a patch.\nThank you.\n\n1.\nIt is indefinite what PGTYPEStimestamp_from_asc() returns in error.\nThe following is written in document(36.6.8. Special Constants of pgtypeslib... |
[
{
"msg_contents": "Some findings\n\n\n1.\n\n>>Remove adminpack contrib extension (Daniel Gustafsson)\n\n>>This was used by non-end-of-life pgAdmin III.\n\nPerhaps you mean now-end-of-life (s/non/now/)\n\n2.\n>>All specification of partitioned table access methods (Justin Pryzby, >>Soumyadeep Chakraborty, Michae... |
[
{
"msg_contents": "Greetings,\n\nThe JDBC driver is currently keeping a per connection cache of types in the\ndriver. We are seeing cases where the number of columns is quite high. In\none case Prevent fetchFieldMetaData() from being run when unnecessary. ·\nIssue #3241 · pgjdbc/pgjdbc (github.com)\n<https://gi... |
[
{
"msg_contents": "Hi hackers,\n\nAfter several refactoring iterations, auxiliary processes are no\nlonger initialized from the bootstrapper. I think using the\nInitProcessing mode for initializing auxiliary processes is more\nappropriate.\n\nBest Regards,\nXing",
"msg_date": "Thu, 9 May 2024 21:12:45 +0800... |
[
{
"msg_contents": "$subject\n\nMake has one:\nhttps://www.postgresql.org/docs/current/docguide-build.html#DOCGUIDE-BUILD-SYNTAX-CHECK\n\nThis needs updating:\nhttps://www.postgresql.org/docs/current/docguide-build-meson.html\n\nI've been using \"ninja html\" which isn't shown here. Also, as a sanity\ncheck, ru... |
[
{
"msg_contents": "Hi,\n\nJust a few reminders about the open items list at\nhttps://wiki.postgresql.org/wiki/PostgreSQL_17_Open_Items --\n\n- Please don't add issues to this list unless they are the result of\ndevelopment done during this release cycle. This is not a\ngeneral-purpose bug tracker.\n\n- The owne... |
[
{
"msg_contents": "As we know, the deadlock error message isn't the most friendly one. All the client gets back is process PIDs, transaction IDs, and lock types. You have to check the server log to retrieve lock details. This is tedious.\n\nIn one of my apps I even added a deadlock exception handler on the app ... |
[
{
"msg_contents": "Hi,\n\nAnalyze logs within autovacuum uses specific variables\nVacuumPage{Hit,Miss,Dirty} to track the buffer usage count. However,\npgBufferUsage already provides block usage tracking and handles more cases\n(temporary tables, parallel workers...).\n\nThose variables were only used in two pl... |
[
{
"msg_contents": "Hello hackers,\n\nI've investigated a recent buildfarm failure:\nhttps://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=dodo&dt=2024-05-02%2006%3A40%3A36\n\nwhere the test failed due to a CRC error:\n2024-05-02 17:08:18.401 ACST [3406:7] LOG: incorrect resource manager data checksum in reco... |
[
{
"msg_contents": "SNI was brought up the discussions around the ALPN work, and I have had asks\nfor it off-list, so I decided to dust off an old patch I started around the\ntime we got client-side SNI support but never finished (until now). Since\nthere is discussion and thinking around how we handle SSL righ... |
[
{
"msg_contents": "Per src/tools/RELEASE_CHANGES, we still have some routine\ntasks to finish before beta1:\n\n* Run mechanical code beautification tools:\n pgindent, pgperltidy, and \"make reformat-dat-files\"\n (complete steps from src/tools/pgindent/README)\n\n* Renumber any manually-assigned OIDs between ... |
[
{
"msg_contents": "For example, 'i'::citext = 'İ'::citext fails to be true.\n\nIt must now be using UTF-8 (unlike, say, Drongo) and non-C ctype,\ngiven that the test isn't skipped. This isn't the first time that\nwe've noticed that Windows doesn't seem to know about İ→i (see [1]),\nbut I don't think anyone has... |
[
{
"msg_contents": "Hello Everyone!\n\nIs there any chance to get some kind of a result set sifting mechanism in\nPostgres?\n\nWhat I am looking for is a way to get for example: \"nulls last\" in a result\nset, without having to call \"order by\" or having to use UNION ALL, and if\npossible to get this in a sing... |
[
{
"msg_contents": "I spent some time looking into the performance complaint at [1],\nwhich for the sake of self-containedness is\n\nCREATE TABLE t(a int, b int);\n\nINSERT INTO t(a, b)\nSELECT\n (random() * 123456)::int AS a,\n (random() * 123456)::int AS b\nFROM\n generate_series(1, 12345678);\n\nCREA... |
[
{
"msg_contents": "I just joined the mailing list and I don't know how to respond to old messages. However, I have a few suggestions on the upcoming TLS and ALPN changes.\n\nTL;DR\n\nPrefer TLS over SSLRequest or plaintext (from the start)\n\n- ?sslmode=default # try tls, then sslrequest, then plaintext\n- ?ss... |
[
{
"msg_contents": "After sending out my v18 patches:\nhttps://www.postgresql.org/message-id/20240511.162307.2246647987352188848.t-ishii%40sranhm.sra.co.jp\n\nCFbot complains that the patch was broken:\nhttp://cfbot.cputube.org/patch_48_4460.log\n\n=== Applying patches on top of PostgreSQL commit ID 31e8f4e619d9... |
[
{
"msg_contents": "XLogReadBufferForRedoExtended() precedes RestoreBlockImage() with\nRBM_ZERO_AND_LOCK. Per src/backend/storage/buffer/README:\n\n Once one has determined that a tuple is interesting (visible to the current\n transaction) one may drop the content lock, yet continue to access the\n tuple's d... |
[
{
"msg_contents": "Hello!\nI have created a patch to allow additional commas between columns, and \nat the end of the SELECT clause.\n\nMotivation:\nCommas of this type are allowed in many programming languages, in some \nit is even recommended to use them at the ends of lists or objects. A \nnew generation of ... |
[
{
"msg_contents": "The sepgsql tests have not been integrated into the Meson build system \nyet. I propose to fix that here.\n\nOne problem there was that the tests use a very custom construction \nwhere a top-level shell script internally calls make. I have converted \nthis to a TAP script that does the prel... |
[
{
"msg_contents": "I noticed that the reference pages for initdb and pg_ctl claim in the \nEnvironment section that libpq variables are used, which does not seem \ncorrect to me. I think this was accidentally copied when this blurb was \nadded to other pages.\n\nWhile I was checking around that, I also noticed... |
[
{
"msg_contents": "hi.\n\nexplain(analyze, format json, serialize, memory, costs off, Timing\noff) select * from tenk1;\n QUERY PLAN\n---------------------------------\n [\n {\n \"Plan\": {\n \"Node Type\": \"Seq Scan\",\n \"Parallel Aware\": false,\n \"Async Capable\": false,... |
[
{
"msg_contents": "In master, if you look at ExecHashGetHashValue() in nodeHash.c, you\ncan see that it calls ExecEvalExpr() and then manually calls the hash\nfunction on the returned value. This process is repeated once for each\nhash key. This is inefficient for a few reasons:\n\n1) ExecEvalExpr() will only d... |
[
{
"msg_contents": "Hi,\n\nBookworm versions of the Debian CI images are available now [0]. The\npatches to use these images are attached.\n\n'v1-0001-Upgrade-Debian-CI-images-to-Bookworm_REL_16+.patch' patch can\nbe applied to both upstream and REL_16 and all of the tasks finish\nsuccessfully.\n\n'v1-0001-Upgra... |
[
{
"msg_contents": "Hi,\n\nMy collegue Konstantin Knizhnik pointed out that we fail to mark pages\nwith a non-standard page layout with page_std=false in\nRelationCopyStorageUsingBuffer when we WAL-log them. This causes us to\ninterpret the registered buffer as a standard buffer, and omit the\nhole in the page, ... |
[
{
"msg_contents": "Hi team!\n\nFirst, i want to thank you for having your hands in this. You are doing a\nfantastic and blessing job. Bless to you all!\n\nI have a special need i want to comment to you. This is not a bug, is a\nneed i have and i write here for been redirected where needed.\n\nI have to make a d... |
[
{
"msg_contents": "Hello,\n\nIn light of multiple threads [1-6] discussing sorting improvements, I'd like to consolidate the old (+some new) ideas as a starting point.\nIt might make sense to brain storm on a few of these ideas and maybe even identify some that are worth implementing and testing.\n\n1. Simple a... |
[
{
"msg_contents": "Hi,\n\nIt can be very useful to look at the log messages emitted by a larger number\nof postgres instances to see if anything unusual is happening. E.g. checking\nwhether there are an increased number of internal, IO, corruption errors (and\nLOGs too, because we emit plenty bad things as LOG)... |
[
{
"msg_contents": "I noticed that we (kind of) accept underscores in positional parameters.\nFor example, this works:\n\n => PREPARE p1 AS SELECT $1_2;\n PREPARE\n => EXECUTE p1 (123);\n ?column?\n ----------\n 123\n (1 row)\n\nParameter $1_2 is taken as $1 because in rule {param} in scan... |
[
{
"msg_contents": "hi.\n\nwhile reading this[1],\n<< More information about partial indexes can be found in [ston89b],\n[olson93], and [seshadri95].\nI googled around, still cannot find [olson93] related pdf or html link.\n\n\nin [2],\nI found out\n[ong90] “A Unified Framework for Version Modeling Using Product... |
[
{
"msg_contents": "Hi hackers,\n\nWhile resuming the work on refilenode stats (mentioned in [1] but did not share\nthe patch yet), I realized that my current POC patch is buggy enough to produce\nthings like:\n\n024-05-14 09:51:14.783 UTC [1788714] FATAL: can only drop stats once\n\nWhile the CONTEXT provides ... |
[
{
"msg_contents": "Hello, hackers!\r\n\r\nRecently I've been building postgres with different cflags and cppflags.\r\nAnd suddenly on REL_15_STABLE, REL_16_STABLE and master\r\nI faced a failure of a src/test/subscription/t/029_on_error.pl test when\r\n CPPFLAGS=\"-DWAL_DEBUG\"\r\nand\r\n printf \"wal... |
[
{
"msg_contents": "plpgsql fails to parse 1_000..1_000 as 1000..1000 in FOR loops:\n\n DO $$\n DECLARE\n i int;\n BEGIN\n FOR i IN 1_000..1_000 LOOP\n END LOOP;\n END $$;\n\n ERROR: syntax error at or near \"1_000.\"\n LINE 5: FOR i IN 1_000..1_000 LOOP\n\nThe scan.l ... |
[
{
"msg_contents": "This page has 3 items that are between parentheses, there is an explanation\nwhy they are used this way ?\n\nhttps://www.postgresql.org/docs/devel/functions-json.html#FUNCTIONS-SQLJSON-TABLE\n\n\nEach syntax element is described below in more detail.\n*context_item*, *path_expression* [ AS *j... |
[
{
"msg_contents": "While looking at pg_dump performance today I noticed that pg_dump fails to\nclear query results in binary_upgrade_set_pg_class_oids during binary upgrade\nmode. 9a974cbcba00 moved the query to the outer block, but left the PQclear\nand query buffer destruction in the is_index conditional, ma... |
[
{
"msg_contents": "While seeing changes and new features of\nhttps://www.postgresql.org/docs/devel/release-17.html\nI see that there are too little links to other DOC pages, which would be\nuseful.\n\nThere are links to\n\"logical-replication\", \"sql-merge\", \"plpgsql\", \"libpq\" and\n\"pgstatstatements\"\n\... |
[
{
"msg_contents": "Prompted by an off-list bugreport of pg_upgrade hanging (which turned out to be\nslow enough to be perceived to hang) for large schemas I had a look at pg_dump\nperformance during --binary-upgrade mode today. My initial take was to write\nmore or less exactly what Nathan did in [0], only to ... |
[
{
"msg_contents": "Hi hackers,\n\nThere are lots of subscription options listed on the CREATE\nSUBSCRIPTION page [1].\n\nAlthough these boolean options are capable of accepting different\nvalues like \"1|0\", \"on|off\", \"true|false\", here they are all described\nonly using values \"true|false\".\n\n~\n\nIMO ... |
[
{
"msg_contents": "Hi,\r\n\r\nAttached is a copy of the PostgreSQL 17 Beta 1 release announcement \r\ndraft. This contains a user-facing summary of some of the features that \r\nwill be available in the Beta, as well as a call to test. I've made an \r\neffort to group them logically around the different workflo... |
[
{
"msg_contents": "Hi hackers!\n\nStHighload conference will be held on June 24-25[0]. I’m planning to do “Pre-Commitfest Party” there.\n\nThe idea is to help promote patches among potential reviewers. And start working with the very beginning of PG18 development cycle.\nGood patch review of a valuable feature ... |
[
{
"msg_contents": "When writing a new SSL test for another patch it struck me that the SSL tests\nare doing configuration management without using the test framework API's. The\nattached patches cleans this up, no testcases are altered as part of this.\n\n0001 makes the test for PG_TEST_EXTRA a top-level if st... |
[
{
"msg_contents": "I wonder if we can avoid making MERGE_ACTION a keyword.\n\nI think we could parse it initially as a function and then transform it \nto a more special node later. In the attached patch, I'm doing this in \nparse analysis. We could try to do it even later and actually execute \nit as a funct... |
[
{
"msg_contents": "Hi,\n\nThe original intent of CommitFests, and of commitfest.postgresql.org\nby extension, was to provide a place where patches could be registered\nto indicate that they needed to be reviewed, thus enabling patch\nauthors and patch reviewers to find each other in a reasonably\nefficient way.... |
[
{
"msg_contents": "Hi,\n\nI built a gist index to try to test a theory in some other thread. For that\nthe indexes need to cover a lot of entries. With gist creating the index took\na long time, which made me strace the index build process.\n\nWhich lead me to notice this:\n\n...\nopenat(AT_FDCWD, \"base/16462/... |
[
{
"msg_contents": "Hi,\n\nIn the subthread at [1] I needed to trigger multiple rounds of index vacuuming\nwithin one vacuum.\n\nIt turns out that with the new dead tuple implementation, that got actually\nsomewhat expensive. Particularly if all tuples on all pages get deleted, the\nrepresentation is just \"too ... |
[
{
"msg_contents": "A number of pg_upgrade steps require connecting to each database and\nrunning a query. When there are many databases, these steps are\nparticularly time-consuming, especially since this is done sequentially in\na single process. At a quick glance, I see the following such steps:\n\n\t* crea... |
[
{
"msg_contents": "Hackers,\n\nThis is greatly simplified implementation of the patch proposed in [1] \nand hopefully it addresses the concerns expressed there. Since the \nimplementation is quite different it seemed like a new thread was \nappropriate, especially since the old thread title would be very \nmisl... |
[
{
"msg_contents": "Hi Hackers,\n\nI have been playing with PG on the Windows platform recently. An annoying\nthing I faced is that a lot of Visual Studio's temp files kept appearing in\ngit changed files. Therefore, I am submitting this very trivial patch to\nignore these temp files.\n\nLooking forward to the P... |
[
{
"msg_contents": "hi.\n\nhttps://wiki.postgresql.org/wiki/Todo\nDates and Times[edit]\nAllow infinite intervals just like infinite timestamps\nhttps://www.postgresql.org/message-id/4EB095C8.1050703@agliodbs.com\n\nthis done at\nhttps://git.postgresql.org/cgit/postgresql.git/commit/?id=519fc1bd9e9d7b408903e44f5... |
[
{
"msg_contents": "Hi\r\n\r\nafter migration on PostgreSQL 16 I seen 3x times (about every week) broken\r\ntables on replica nodes. The query fails with error\r\n\r\nERROR: could not access status of transaction 1442871302\r\nDETAIL: Could not open file \"pg_xact/0560\": No such file or directory\r\n\r\nverif... |
[
{
"msg_contents": "(moving to a new thread)\n\nOn Thu, May 16, 2024 at 09:16:46PM -0500, Nathan Bossart wrote:\n> On Thu, May 16, 2024 at 04:37:10PM +0000, Imseih (AWS), Sami wrote:\n>> Also, Not sure if I am mistaken here, but the \"+ 5\" in the existing docs\n>> seems wrong.\n>> \n>> If it refers to NUM_AUXI... |
[
{
"msg_contents": "Hi,\n\nI've been looking at GiST to see if there could be a good way to do\nparallel builds - and there might be, if the opclass supports sorted\nbuilds, because then we could parallelize the sort.\n\nBut then I noticed we support this mode only for point_ops, because\nthat's the only opclass... |
[
{
"msg_contents": "Hi!\n\nIn a thread about sorting comparators[0] Andres noted that we have infrastructure to help compiler optimize sorting. PFA attached PoC implementation. I've checked that it indeed works on the benchmark from that thread.\n\npostgres=# CREATE TABLE arrays_to_sort AS\n SELECT array_shuff... |
[
{
"msg_contents": "When deploying RLS, I was surprised to find that certain queries which used \nonly builtin indexes and operators had dramatically different query plans when \na policy is applied. In my case, the query `tsvector @@ tsquery` over a GIN\nindex was no longer able to use that index. I was able to... |
[
{
"msg_contents": "I noticed that PlannedStmt.hasReturning and hasModifyingCTE have an\noutdated comment now that MERGE supports RETURNING (per commit\nc649fa24a)\n\ni.e. these two:\n\n> bool hasReturning; /* is it insert|update|delete RETURNING? */\n\n> bool hasModifyingCTE; /* has insert|update|delete in WITH... |
[
{
"msg_contents": "\npostgresql 17 (testing) SLES 15.5 missing \"repodata\" so i can´t add \nrepo to OpenSUSE...\nPlease add... thank you..\n\nurl:\nhttps://download.postgresql.org/pub/repos/zypp/testing/17/suse/sles-15.5-x86_64/\n\n\n\n",
"msg_date": "Sun, 19 May 2024 14:03:42 +0200",
"msg_from": "... |
[
{
"msg_contents": "I'm fairly disturbed about the readiness of pg_createsubscriber.\nThe 040_pg_createsubscriber.pl TAP test is moderately unstable\nin the buildfarm [1], and there are various unaddressed complaints\nat the end of the patch thread (pretty much everything below [2]).\nI guess this is good enough... |
[
{
"msg_contents": "Hi all,\n\nI have a couple of extra toys for injection points in my bucket that\nI'd like to propose for integration in v18, based on some feedback I\nhave received:\n1) Preload an injection point into the backend-level cache without\nrunning it. This has come up because an injection point r... |
[
{
"msg_contents": "This patch converts the compile-time settings\n\n COPY_PARSE_PLAN_TREES\n WRITE_READ_PARSE_PLAN_TREES\n RAW_EXPRESSION_COVERAGE_TEST\n\ninto run-time parameters\n\n debug_copy_parse_plan_trees\n debug_write_read_parse_plan_trees\n debug_raw_expression_coverage_test\n\n... |
[
{
"msg_contents": "I noticed that there are slightly inconsistent messages regarding\nquoting policies.\n\n> This happens if you temporarily set \"wal_level=minimal\" on the server.\n> WAL generated with \"full_page_writes=off\" was replayed during online backup\n \n> pg_log_standby_snapshot() can only be used ... |
[
{
"msg_contents": "Hi\n\nI'm working on updating the build of PostgreSQL that pgAdmin uses in its\nWindows installers to use Meson ready for the v17 release. I'm using Visual\nStudio 2022, on Windows Server 2022.\n\nI've been unable to persuade Meson to detect zlib, whilst OpenSSL seems to\nbe fine.\n\nThe depe... |
[
{
"msg_contents": "Hey,\n\nI'm trying to read a timestamp column as EPOCH.\nMy query is as follows.\n```\nSELECT EXTRACT(EPOCH FROM timestamp_column) FROM table;\n\ncolumn\n----------\n\n1716213097.86486\n```\nWhen running in the console this query gives valid epoch output which\nappears to be of type double.\n... |
[
{
"msg_contents": "The following documentation comment has been logged on the website:\n\nPage: https://www.postgresql.org/docs/16/logical-replication-col-lists.html\nDescription:\n\nThe documentation on this page mentions:\r\n\r\n\"If no column list is specified, any columns added later are automatically\nrepl... |
[
{
"msg_contents": "Hello hackers,\n\nPlease look at a bunch of unused variables and a couple of other defects\nI found in the perl code, maybe you'll find them worth fixing:\ncontrib/amcheck/t/001_verify_heapam.pl\n$result # unused since introduction in 866e24d47\nunused sub:\nget_toast_for # not used since 860... |
[
{
"msg_contents": "Many years ago we in effect moved maintenance of the typedefs list for \npgindent into the buildfarm client. The reason was that there were a \nnumber of typedefs that were platform dependent, so we wanted to have \ncoverage across a number of platforms to get a comprehensive list.\n\nLately,... |
[
{
"msg_contents": "Earlier today in [1], a bug was reported regarding a problem with the\ncode added in 66c0185a3 where I'd failed to handle the case correctly\nwhere the UNION's targetlist has columns which are not sortable. For\npg_class, that's relfrozenxid, relminmxid and relacl.\n\nThe most minimal reprod... |
[
{
"msg_contents": "hi.\n\nhttps://www.postgresql.org/docs/current/functions-matching.html#FUNCTIONS-POSIX-REGEXP\n<<<start_quote\nThe regexp_replace function provides substitution of new text for\nsubstrings that match POSIX regular expression patterns. It has the\nsyntax regexp_replace(source, pattern, replace... |
[
{
"msg_contents": "Hello,\n\nApologies if this is the wrong place to make such a request.\n\nIt would be useful to have the ability to prevent clients from setting whatever isolation levels they choose. Specifically, it would be desirable to enforce SERIALIZABLE for all transactions since the serializable guara... |
[
{
"msg_contents": "Hi, all\n\nI want to add some columns of int(or Oid) array and declare GIN index for it in catalog when bootstrap.\n\nBut current catalogs all use btree, I tried to declare a GIN index but failed, ex:\npg_class.h\n```\nCATALOG(pg_class\n...\nInt32 \tmy_new_column[1] BKI_DEFAULT(_null_);\n...\... |
[
{
"msg_contents": "Hello Hackers,\n\nIf the user tries to open the relation in RangeVar and NoLock mode\ncalling *table_openrv(relation,\nNoLock), *it will internally call relation_openrv()-->relation_open().\nIn *relation_open()\n*we checking the Assert(lockmode >= NoLock && lockmode < MAX_LOCKMODES); ,\nhere ... |
[
{
"msg_contents": "Hi hackers,\n\nI'd submit an implementation of multi-key sort for review. Please see the\ncode as attachment. Thanks for your reponse in advance.\n\n\nOverview\n--------\n\nMKsort (multi-key sort) is an alternative of standard qsort algorithm,\nwhich has better performance for particular sort... |
[
{
"msg_contents": "Hi.\n\nPer Coverity.\n\n2. returned_null: SearchSysCacheAttName returns NULL (checked 20 out of 21\ntimes).\n3. var_assigned: Assigning: ptup = NULL return value from\nSearchSysCacheAttName.\n 964 ptup = SearchSysCacheAttName(relid, attname);\nCID 1545986: (#1 of 1): Dereferenc... |
[
{
"msg_contents": "Hi!\n\nI faced the issue, when the sorting node in the actual information \nshows a larger number of tuples than it actually is. And I can not \nunderstand why?\n\nI attached the dump file with my database and run this query that \nconsists underestimation and it works fine.\n\nUnfortunately... |
[
{
"msg_contents": "https://postgr.es/m/20240512232923.aa.nmisch@google.com wrote:\n> Separable, nontrivial things not fixed in the attached patch stack:\n> \n> - Inplace update uses transactional CacheInvalidateHeapTuple(). ROLLBACK of\n> CREATE INDEX wrongly discards the inval, leading to the relhasindex=t ... |
[
{
"msg_contents": "Currently the escape_json() function takes a cstring and char-by-char\nchecks each character in the string up to the NUL and adds the escape\nsequence if the character requires it.\n\nBecause this function requires a NUL terminated string, we're having\nto do a little more work in some places... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.