threads listlengths 1 2.99k |
|---|
[
{
"msg_contents": "Hi,\n\nI was looking for a pattern to destroy a hashtable (dynahash).allocated\nin TopMemoryContext\nI found one pattern : create_seq_hashtable uses TopMemoryContext\n memory context to create hash table. It calls hash_destroy in\n ResetSequenceCaches. hash_destroy will destroy the memory\nco... |
[
{
"msg_contents": "I checked the generated ecpg_config.h with make and meson, and the meson \none is missing\n\n#define HAVE_LONG_LONG_INT 1\n\nThis is obviously quite uninteresting, since that is required by C99. \nBut it would be more satisfactory if we didn't have discrepancies like \nthat. Note that we als... |
[
{
"msg_contents": "Hackers,\n I often use the ctrl-click on the link after getting help in psql. A\ngreat feature.\n\nChallenge, when there is no help, you don't get any link.\n\n My thought process is to add a default response that would take them to\nhttps://www.postgresql.org/search/?u=%2Fdocs%2F16%2F&q={... |
[
{
"msg_contents": "In the \"Differential code coverage between 16 and HEAD\" thread, Andres\npointed out that there wasn't test case coverage for\npg_combinebackup's code to handle files in tablespaces. I looked at\nadding that, and as nobody could possibly have predicted, found a bug.\n\nHere's a 2-patch serie... |
[
{
"msg_contents": "Hi,\n\nWe have a fair amount of code that uses non-constant function level static\nvariables for read-only data. Which makes little sense - it prevents the\ncompiler from understanding\n\na) that the data is read only and can thus be put into a segment that's shared\n between all invocation... |
[
{
"msg_contents": "While examining pg_upgrade on a cluster with many tables (created with the\ncommand in [0]), I noticed that a huge amount of pg_dump time goes towards\nthe binary_upgrade_set_pg_class_oids() function. This function executes a\nrather expensive query for a single row, and this function appear... |
[
{
"msg_contents": "On 18/04/2024 00:39, Andres Freund wrote:\n\n>We have a fair amount of code that uses non-constant function level static\n>variables for read-only data. Which makes little sense - it prevents the\n>compiler from understanding\n\n>a) that the data is read only and can thus be put into a segmen... |
[
{
"msg_contents": "Fix restore of not-null constraints with inheritance\n\nIn tables with primary keys, pg_dump creates tables with primary keys by\ninitially dumping them with throw-away not-null constraints (marked \"no\ninherit\" so that they don't create problems elsewhere), to later drop\nthem once the pri... |
[
{
"msg_contents": "Hello hackers,\n\nWhen using a server built with clang (15, 18) with sanitizers enabled,\nthe last query in the following script:\nSET parallel_setup_cost = 0;\nSET min_parallel_table_scan_size = 0;\n\nSELECT a::text INTO t FROM generate_series(1, 1000) a;\n\\timing on\nSELECT string_agg(a, '... |
[
{
"msg_contents": "Hi,\r\n\r\nPostgreSQL 17 Beta 1 is planned to be release on May 23, 2024. Please \r\ncontinue your hard work on closing out open items[1] ahead of the \r\nrelease and have the fixes targeted for the release committed by May 18, \r\n2024.\r\n\r\nThanks - it's very exciting that we're at this p... |
[
{
"msg_contents": "Here is a patch series that aims to clean up ecpg's preprocessor\ncode a little and fix the compile time problems we're seeing with\nlate-model clang [1]. I guess whether it's a cleanup is in the eye of\nthe beholder, but it definitely succeeds at fixing compile time: for\nme, the time neede... |
[
{
"msg_contents": "Hello\n\nThis page says that the `@` and `~` operators on various types can be\naccelerated by a GiST index.\n\nhttps://www.postgresql.org/docs/current/gist-builtin-opclasses.html\n\nThese operators have been listed in the file since it was created in 2014,\nbut if they exist then I don't kno... |
[
{
"msg_contents": "Hello!\n\nThere is a macro XLOG_CONTROL_FILE for control file name\ndefined in access/xlog_internal.h\nAnd there are some places in code where this macro is used\nlike here\nhttps://github.com/postgres/postgres/blob/84db9a0eb10dd1dbee6db509c0e427fa237177dc/src/bin/pg_resetwal/pg_resetwal.c#L5... |
[
{
"msg_contents": "Hi all,\n(Heikki in CC.)\n\nSince 91044ae4baea (require ALPN for direct SSL connections) and\nd39a49c1e459 (direct hanshake), direct SSL connections are supported\n(yeah!), still the thread where this has been discussed does not cover\nthe potential impact on HBA rules:\nhttps://www.postgresq... |
[
{
"msg_contents": "Hi, I have a use case where I want a particular database to not load a few\nmodules in shared_preload_libraries.\nI was wondering if there's any way to tweak the codebase to achieve this.\nOtherwise, can I block the modules' hooks/bgw from performing actions on my\nparticular database?\n\n-- ... |
[
{
"msg_contents": "Hi, hackers\n\nI see [1] has already implemented on login event trigger, why not implement\nthe logoff event trigger?\n\nMy friend Song Jinzhou and I try to implement the logoff event trigger, so\nattach it.\n\nHere is a problem with the regression test when using \\c to create a new\nsession... |
[
{
"msg_contents": "I just did a run of the regression test where this test was the last\none to finish by quite a lot. Key log entries:\n\n[13:35:48.583](0.039s) # initializing database system by copying initdb template\n...\n[13:35:52.397](0.108s) ok 5 - Check reset timestamp for\n'test_tab1_sub' is newer afte... |
[
{
"msg_contents": "Hi hackers,\n\nI noticed that the header file resowner_private.h is deprecated and no\nlonger useful after commit b8bff07[^1]. We should remove it.\n\n[^1]: https://github.com/postgres/postgres/commit/b8bff07daa85c837a2747b4d35cd5a27e73fb7b2\n\nBest Regards,\nXing",
"msg_date": "Sat, 20 A... |
[
{
"msg_contents": "Hello hackers,\n\nWhen playing with JSON_TABLE, I tried to replace tenk1 in regression tests\nwith a view based on JSON_TABLE, with the same content, and discovered\nthat for one sub-optimal query it's execution duration increased many-fold.\nWith the preparation script attached, I see the fo... |
[
{
"msg_contents": "Hi,\n\nWhile doing some testing with createdb, I noticed it only accepts\nfile_copy/wal_log as valid strategies, not FILE_COPY/WAL_LOG (which is\nwhat the docs say). The same thing applies to CREATE DATABASE.\n\nThe problem is that createdb() does the check using strcmp() which is\ncase-sensi... |
[
{
"msg_contents": "I came across an assert failure in _bt_preprocess_array_keys regarding\nthe sanity check on the datatype of the array elements. It can be\nreproduced with the query below.\n\ncreate table t (c int4range);\ncreate unique index on t (c);\n\nselect * from t where c in ('(1, 100]'::int4range, '(... |
[
{
"msg_contents": "hi.\nminor issue in guc.c.\n\nset work_mem to '1kB';\nERROR: 1 kB is outside the valid range for parameter \"work_mem\" (64\n.. 2147483647)\nshould it be\nERROR: 1 kB is outside the valid range for parameter \"work_mem\" (64\nkB .. 2147483647 kB)\n?\nsince the units for work_mem are { \"B\"... |
[
{
"msg_contents": "Hi,\n\nThis new thread is a follow-up of [1] and [2].\n\nProblem description:\n\nWe have occasionally observed objects having an orphaned dependency, the \nmost common case we have seen is functions not linked to any namespaces.\n\nExamples to produce such orphaned dependencies:\n\nScenario 1... |
[
{
"msg_contents": "Hi,\n\nWhile working on the 'reduce nodeToString output' patch, I noticed\nthat my IDE marked one field in the TidScanState node as 'unused'.\nAfter checking this seemed to be accurate, and I found a few more such\nfields in Node structs.\n\nPFA some patches that clean this up: 0001 is plain ... |
[
{
"msg_contents": "Hey,\n\nThe attached patch addresses four somewhat related aspects of the create\ntable reference page that bother me.\n\nThis got started with Bug# 15954 [1] (unlogged on a partitioned table\ndoesn't make sense) and I've added a paragraph under \"unlogged\" to address\nit.\n\nWhile doing tha... |
[
{
"msg_contents": "I used to do this step when I first started hacking on Postgres because\nthat's what it says to do, but I've only ever used the in-tree one for many\nyears now, and I'm not aware of any scenario where I might need to download\na new version from the buildfarm. I see that the in-tree copy was... |
[
{
"msg_contents": "hi.\n\n/*\n * clamp_row_est\n * Force a row-count estimate to a sane value.\n */\ndouble\nclamp_row_est(double nrows)\n{\n/*\n* Avoid infinite and NaN row estimates. Costs derived from such values\n* are going to be useless. Also force the estimate to be at least one\n* row, to make explain... |
[
{
"msg_contents": "Hi all,\n\nWhile analyzing the use of internal error codes in the code base, I've\nsome problems, and that's a mixed bag of:\n- Incorrect uses, for errors that can be triggered by users with\nvallid cases.\n- Expected error cases, wanted by the tests like corruption cases or\n just to keep s... |
[
{
"msg_contents": "Hi,\n\nI'm proposing a patch that making COPY format extendable:\nhttps://www.postgresql.org/message-id/20231204.153548.2126325458835528809.kou%40clear-code.com\nhttps://commitfest.postgresql.org/48/4681/\n\nIt's based on the discussion at:\nhttps://www.postgresql.org/message-id/flat/3741749.... |
[
{
"msg_contents": "Hi hackers,\n\nI proposal adding privileges test to improve\ntest coverage of pg_stat_statements.\n\n## test procedure\n\n./configure --enable-coverage --enable-tap-tests --with-llvm CFLAGS=-O0\nmake check-world\nmake coverage-html\n\n## coverage\n\nbefore Line Coverage 74.0 %(702/949 lines)\... |
[
{
"msg_contents": "Hi!\n\nI've been trying to introduce 64-bit transaction identifications to\nPostgres for quite a while [0]. All this implies,\nof course, an enormous amount of change that will have to take place in\nvarious modules. Consider this, the\npatch set become too big to be committed “at once”.\n\nT... |
[
{
"msg_contents": "Hi,\n\ndoc/src/sgml/monitoring.sgml seems to have a minor typo:\n\nIn pg_stat_database_conflicts section (around line 3621) we have:\n\n <para>\n Number of uses of logical slots in this database that have been\n canceled due to old snapshots or too low a <xref linkend=\"guc-w... |
[
{
"msg_contents": "Hi all,\r\n\r\nI would like to report an issue with the pg_trgm extension on cross-architecture replication scenarios. When an x86_64 standby server is replicating from an aarch64 primary server or vice versa, the gist_trgm_ops opclass returns different results on the primary and standby. Mas... |
[
{
"msg_contents": "Hey,\n\nI'm developing a postgres extension as a custom Table Interface method\ndefinition.\nWIthin the extension, I\"m planning to create two background processes using\n`fork()` that will process data in the background.\n\nAre there any recommendations / guidelines around creating backgroun... |
[
{
"msg_contents": "With one eye on the beta-release calendar, I thought it'd be a\ngood idea to test whether our tarball build script works for\nthe new plan where we'll use \"git archive\" instead of the\ntraditional process.\n\nIt doesn't.\n\nIt makes tarballs all right, but whatever commit ID you specify\nis... |
[
{
"msg_contents": "Hi,\n\nCurrently, ALTER DEFAULT PRIVILEGE doesn't support large objects,\nso if we want to allow users other than the owner to use the large\nobject, we need to grant a privilege on it every time a large object\nis created. One of our clients feels that this is annoying, so I would\nlike prop... |
[
{
"msg_contents": "Hello,\n\nA diff tool that would generate create, drop, alter, etc. commands from the\ndifferences between 2 specified schemes would be very useful. The diff\ncould even manage data, so there would be insert, delete update command\noutputs, although I think the schema diff management is much ... |
[
{
"msg_contents": "Hi,\n\nHi,\n\nWe can specify more than one privilege type in \n\"ALTER DEFAULT PRIVILEGES GRANT/REVOKE ON SCHEMAS\",\nfor example,\n\n ALTER DEFAULT PRIVILEGES GRANT USAGE,CREATE ON SCHEMAS TO PUBLIC;\n\nHowever, the syntax described in the documentation looks to\nbe allowing only one,\n\n G... |
[
{
"msg_contents": "hi.\n\nin make_pathkey_from_sortinfo\n\nequality_op = get_opfamily_member(opfamily,\n opcintype,\n opcintype,\n BTEqualStrategyNumber);\nif (!OidIsValid(equality_op)) /* shouldn't happen */\nelog(ERROR, \"missing operator %d(%u,%u) in opfamily %u\",\nBTEqualStrategyNumber, opcintype, opcin... |
[
{
"msg_contents": "Hi all,\n\nAs a recent poke on a thread of 2019 has reminded me, the current\nsituation of partitioned tables with unlogged is kind of weird:\nhttps://www.postgresql.org/message-id/15954-b61523bed4b110c4%40postgresql.org\n\nTo sum up the situation:\n- ALTER TABLE .. SET LOGGED/UNLOGGED does n... |
[
{
"msg_contents": "libpq's pqTraceOutputMessage() used to look like this:\n\n case 'Z': /* Ready For Query */\n pqTraceOutputZ(conn->Pfdebug, message, &logCursor);\n break;\n\nCommit f4b54e1ed98 introduced macros for protocol characters, so now\nit looks like this:\n\n case... |
[
{
"msg_contents": "In sort_inner_and_outer, we create mergejoin join paths by explicitly\nsorting both relations on each possible ordering of the available\nmergejoin clauses. However, if there are no available mergejoin\nclauses, we can skip this process entirely. It seems that this is a\nrelatively common s... |
[
{
"msg_contents": "Hi,\n\nI noticed that a permission check is performed in be_lo_put()\njust after returning inv_open(), but teh permission should be\nalready checked in inv_open(), so I think we can remove this\npart of codes. I attached a patch for this fix.\n\nRegards,\nYugo Nagata\n\n-- \nYugo NAGATA <naga... |
[
{
"msg_contents": "Yesterday Fedora 40 was released as GA available release.\n\n\nAccording to the project website (download for Linux/Fedora):\n\n\n\" the PostgreSQL project provides a repository<https://www.postgresql.org/download/linux/redhat/#yum> of packages of all supported versions for the most common di... |
[
{
"msg_contents": "Hello,\n\nI would like to suggest a new parameter, autovacuum_max_threshold, which \nwould set an upper limit on the number of tuples to delete/update/insert \nprior to vacuum/analyze.\n\nA good default might be 500000.\n\nThe idea would be to replace the following calculation :\n\nvacthresh ... |
[
{
"msg_contents": "Hello everyone, I hope you're doing well. Does anyone have a guide or know\nhow to perform an upgrade from PostgreSQL 13.12 to 13.14 on Linux? I've\nsearched in various places but haven't found any solid guides, and truth be\ntold, I'm a bit of a novice with PostgreSQL. Any help would be appr... |
[
{
"msg_contents": "Hi,\n\nCommit 91f2cae7a4e that introduced WALReadFromBuffers only used it for\nphysical walsenders. It can also be used in more places benefitting\nlogical walsenders, backends running pg_walinspect and logical\ndecoding functions if the WAL is available in WAL buffers. I'm\nattaching a 0001 ... |
[
{
"msg_contents": "Tomas Vondra pointed out to me a couple of mistakes that I made with\nregard to pg_combinebackup and tablespaces.\n\nOne is that I screwed up the long_options array by specifying\ntablespace-mapping as no_argument rather than required_argument. That\ndoesn't break the tests I just committed b... |
[
{
"msg_contents": "Hi\n\nHere:\n\n https://www.postgresql.org/docs/current/functions-range.html#MULTIRANGE-FUNCTIONS-TABLE\n\nthe description for \"lower(anymultirange)\":\n\n> (NULL if the multirange is empty has no lower bound).\n\nis missing \"or\" and should be:\n\n> (NULL if the multirange is empty or ... |
[
{
"msg_contents": "Hi,\n\nI'm curious about composite types in PostgreSQL. By default, when we\ncreate a composite type, it utilizes the \"record_in\" and \"record_out\"\nfunctions for input/output. Do you think it would be beneficial to\nexpand the syntax to allow users to specify custom input/output\nfunction... |
[
{
"msg_contents": "Hi\n\nyesterday, I had to fix strange issue on standby server\n\nThe query to freshly updated data fails\n\nselect * from seller_success_rate where create_time::date = '2024-04-23';\nERROR: 58P01: could not access status of transaction 1393466389\nDETAIL: Could not open file \"pg_xact/0530\... |
[
{
"msg_contents": "Hi,\n\nCurrently the launcher's latch is used for the following: a) worker\nprocess attach b) worker process exit and c) subscription creation.\nSince this same latch is used for multiple cases, the launcher process\nis not able to handle concurrent scenarios like: a) Launcher started a\nnew ... |
[
{
"msg_contents": "Hi hackers,\n\nAn IRC conversation just now made me notice that it would be handy to\nhave stable links for the descrpitions of the various COPY formats, per\nthe attached patch.\n\n- ilmari",
"msg_date": "Thu, 25 Apr 2024 12:23:18 +0100",
"msg_from": "=?utf-8?Q?Dagfinn_Ilmari_Manns=C... |
[
{
"msg_contents": "Hi!\n\nRecently, I'm stumbled in such an easy task as build Postgres with clang\nand\nsanitizer.\n\nUsually, I use autotools to build Postgres something like this:\n================================================================================\nSRC=\"../postgres\"\nTRG=\"/tmp\"\n\nLINUX_CON... |
[
{
"msg_contents": "Can anyone tell me why my index access method isn't seeing an order_by\nScanKey when there is a query with an ORDER BY clause that uses an operator\nthat the access method supports?\n\n(Apologies for the volume of code below. I just don't know what is\nimportant. This is all in Rust.)\n\n\nCR... |
[
{
"msg_contents": "Hi PostgreSQL Community,\nRecently I have been working on foreign servers regarding my project and\nwanted to add some extensions in server options to support query pushdown.\nFor this, suppose I had 20 extensions in the beginning I used ALTER SERVER\nsrv OPTIONS (ADD EXTENSIONS 'all 20 exten... |
[
{
"msg_contents": "The Core Team would like to extend our congratulations to Melanie \r\nPlageman and Richard Guo, who have accepted invitations to become our \r\nnewest PostgreSQL committers.\r\n\r\nPlease join us in wishing them much success and few reverts!\r\n\r\nThanks,\r\n\r\nJonathan",
"msg_date": "F... |
[
{
"msg_contents": "Hi,\n\nManually specifying tablespace mappings in pg_basebackup, especially in\nenvironments where tablespaces can come and go, or with incremental\nbackups, can be tedious and error-prone. I propose a solution using\npattern-based mapping to automate this process.\n\nSo rather than having to... |
[
{
"msg_contents": "hi.\n\nI found some minor issues related to the EXPLAIN command.\n\ncannot auto-complete with a white space.\nsrc8=# explain (analyze,b\n\ncan auto-complete:\nsrc8=# explain (analyze, b\n\nto make tab-complete work, comma, must be followed with a white space,\nnot sure why.\n--------------\ne... |
[
{
"msg_contents": "Hey,\n\nI\"m trying to read the rows of a table in chunks to process them in a\nbackground worker.\nI want to ensure that each row is processed only once.\n\nI was thinking of using the `SELECT * ... OFFSET {offset_size} LIMIT\n{limit_size}` functionality for this but I\"m running into issues... |
[
{
"msg_contents": "Hi all,\n\nAttached is a patch that fixes some overflow/underflow hazards in\n`timestamp_pl_interval`. The microseconds overflow could produce\nincorrect result. The month overflow would generally still result in an\nerror from the timestamp month field being too low, but it's still\nbetter t... |
[
{
"msg_contents": "Yesterday I noticed a failure on cirrus-ci for the 'Right Semi Join'\npatch. The failure can be found at [1], and it looks like:\n\n--- /tmp/cirrus-ci-build/src/test/regress/expected/prepared_xacts.out\n2024-04-27 00:41:25.831297000 +0000\n+++\n/tmp/cirrus-ci-build/build/testrun/regress-runn... |
[
{
"msg_contents": "hi.\n\nselect * from pg_input_error_info('42000000000', 'integer')\nselect message, detail from pg_input_error_info('1234.567', 'numeric(7,4)')\nI found above two examples at [0] crammed together.\n\n\n <para>\n <literal>select * from pg_input_error_info('42000000000',\n'integer'... |
[
{
"msg_contents": "This patch adds support for using LIKE with nondeterministic collations. \n So you can do things such as\n\n col LIKE 'foo%' COLLATE case_insensitive\n\nThis currently results in a \"not supported\" error. The reason for that \nis that when I first developed support for nondeterministic... |
[
{
"msg_contents": "I found two mistakes related to collation and/or ICU support in the \ndocumentation that should probably be fixed and backpatched. See \nattached patches.",
"msg_date": "Mon, 29 Apr 2024 09:05:46 +0200",
"msg_from": "Peter Eisentraut <peter@eisentraut.org>",
"msg_from_op": true,
... |
[
{
"msg_contents": "While the \"unique keys\" feature [1] is still under development, I'm thinking\nhow it could be used to enhance the removal of useless outer joins. Is\nsomething really bad about the 0002 patch attached?\n\nI recognize it may be weird that a join relation possibly produces non-join\npaths (e.... |
[
{
"msg_contents": "Here is a patch set to implement virtual generated columns.\n\nSome history first: The original development of generated columns was \ndiscussed in [0]. It started with virtual columns, then added stored \ncolumns. Before the release of PG12, it was decided that only stored \ncolumns were ... |
[
{
"msg_contents": "The attached patch tries to fix a corner case where attr_needed of an inner\nrelation of an OJ contains the join relid only because another,\nalready-removed OJ, needed some of its attributes. The unnecessary presence of\nthe join relid in attr_needed can prevent the planner from further join... |
[
{
"msg_contents": "Hi,\n\nPer Coverity.\n\nCID 1542943: (#1 of 1): Data race condition (MISSING_LOCK)\n3. missing_lock: Accessing slot->AH without holding lock signal_info_lock.\nElsewhere, ParallelSlot.AH is written to with signal_info_lock held 1 out\nof 1 times (1 of these accesses strongly imply that it is ... |
[
{
"msg_contents": "There's been a bunch of bugs, and discussion on the intended behavior of \nsslnegotiation and ALPN. This email summarizes the current status:\n\n## Status and loose ends for beta1\n\nAll reported bugs have now been fixed. We now enforce ALPN in all the \nright places. Please let me know if I ... |
[
{
"msg_contents": "I'm developing an index access method.\n\nSELECT *\nFROM foo\nWHERE col <=> constant\nORDER BY col <==> constant\nLIMIT 10;\n\nI'm successfully getting the WHERE and the ORDER BY clauses in my\nbeginscan() method. Is there any way to get the LIMIT (or OFFSET, for that\nmatter)?\n\nMy access m... |
[
{
"msg_contents": "Hi,\n\nWith TLS 1.3 and others there is possibly a security flaw using ALPN [1].\n\nIt seems to me that the ALPN protocol can be bypassed if the client does\nnot correctly inform the ClientHello header.\n\nSo, the suggestion is to check the ClientHello header in the server and\nterminate the ... |
[
{
"msg_contents": "Do we know, is it posted anywhere on the postgresql.org site what CVE's will be addressed in the next round up updates\nto Postgres which should come out next Thursday, May 9th, 2024?\n\nThanks, Mark\n\n\n\n\n\n\n\n\n\nDo we know, is it posted anywhere on the postgresql.org site what CVE’s wi... |
[
{
"msg_contents": "Hello\n\n \n\nWhen using ctid as a\nrestriction clause with lower and upper bounds, PostgreSQL's planner will use\nTID range scan plan to handle such query. This works and generally fine.\nHowever, if the ctid range covers a huge amount of data, the planner will not\nuse parallel worker to pe... |
[
{
"msg_contents": "Hello hackers,\n\nI noticed that the jsonpath date/time functions (.time() and timestamp(), et al.) don’t support some valid but special-case PostgreSQL values, notably `infinity`, `-infinity`, and, for times, '24:00:00`:\n\n❯ psql\npsql (17devel)\nType \"help\" for help.\n\ndavid=# select js... |
[
{
"msg_contents": "Hackers,\n\nWhile testing pgAudit against PG17 I noticed the following behavioral \nchange:\n\nCREATE TABLE public.test\n(\n\tid INT,\n\tname TEXT,\n\tdescription TEXT,\n\tCONSTRAINT test_pkey PRIMARY KEY (id)\n);\nNOTICE: AUDIT: SESSION,23,1,DDL,CREATE TABLE,TABLE,public.test,\"CREATE \nTAB... |
[
{
"msg_contents": "Hi all,\n\nWhen json_lex_string() hits certain types of invalid input, it calls\npg_encoding_mblen_bounded(), which assumes that its input is\nnull-terminated and calls strnlen(). But the JSON lexer is constructed\nwith an explicit string length, and we don't ensure that the string is\nnull-t... |
[
{
"msg_contents": "\nHi,\n\nI wanted to check my understanding of how control flows in a walsender doing logical replication. My understanding is that the (single) thread in each walsender process, in the simplest case, loops on:\n\n1. Pull a record out of the WAL.\n2. Pass it to the recorder buffer code, whic... |
[
{
"msg_contents": "Hi,\n\nWhen a client of our JSON parser registers semantic action callbacks,\nthe parser will allocate copies of the lexed tokens to pass into those\ncallbacks. The intent is for the callbacks to always take ownership of\nthe copied memory, and if they don't want it then they can pfree() it.\... |
[
{
"msg_contents": "Hello, I've recently joined the list on a tip from one of the maintainers\nof jdbc-postgres as I wanted to discuss an issue we've run into and find\nout if the fix we've worked out is the right thing to do, or if there is\nactually a bug that needs to be fixed.\n\nThe full details can be foun... |
[
{
"msg_contents": "If you create an unlogged sequence on a primary, pg_sequence_last_value()\nfor that sequence on a standby will error like so:\n\n\tpostgres=# select pg_sequence_last_value('test'::regclass);\n\tERROR: could not open file \"base/5/16388\": No such file or directory\n\nThis function is used by... |
[
{
"msg_contents": "Hi,\n\nOver in [1] it was rediscovered that our documentation assumes the reader\nis familiar with NULL. It seems worthwhile to provide both an introduction\nto the topic and an overview of how this special value gets handled\nthroughout the system.\n\nAttached is a very rough draft attempti... |
[
{
"msg_contents": "Hi,\n\nComparing the current SSE4.2 implementation of the CRC32C algorithm in Postgres, to an optimized AVX-512 algorithm [0] we observed significant gains. The result was a ~6.6X average multiplier of increased performance measured on 3 different Intel products. Details below. The AVX-512 al... |
[
{
"msg_contents": "Hi PostgreSQL Community,\nI'm currently delving into Postgres HLL (HyperLogLog) functionality and\nhave encountered an unexpected behavior while executing queries from the \"\ncumulative_add_sparse_edge.sql\n<https://github.com/citusdata/postgresql-hll/blob/master/sql/cumulative_add_sparse_ed... |
[
{
"msg_contents": "Hi hackers,\n\nIt seems the get_actual_variable_range function has a long history of\nfixes attempting to improve its worst-case behaviour, most recently in\n9c6ad5eaa95, which limited the number of heap page fetches to 100.\nThere's currently no limit on the number of index pages fetched.\n\... |
[
{
"msg_contents": "I tried to initialize a table with values for smallint columns.\n\nThe final goal is to get mask values for logical operations.\n\n\nThe insert failed with ERROR: smallint out of range.\n\n\nthe same occurs when using a simple select statement like:\n\nselect -32768::smallint;\nselect -214748... |
[
{
"msg_contents": "hi.\n\njust found out we can:\nexplain (verbose, verbose off, analyze on, analyze off, analyze on)\nselect count(*) from tenk1;\n\nsimilar to COPY, do we want to error out these redundant options?\n\n\n",
"msg_date": "Thu, 2 May 2024 21:16:42 +0800",
"msg_from": "jian he <jian.univers... |
[
{
"msg_contents": "Hi,\n\nIn PG17 we shall have parallel CREATE INDEX for BRIN indexes, and back\nwhen working on that I was thinking how difficult would it be to do\nsomething similar to do that for other index types, like GIN. I even had\nthat on my list of ideas to pitch to potential contributors, as I was\n... |
[
{
"msg_contents": "I thought that this might be a small quality of life improvement for \npeople scrolling through logs wondering which tranche name wasn't \nregistered.\n\n-- \nTristan Partin\nNeon (https://neon.tech)",
"msg_date": "Thu, 02 May 2024 11:23:06 -0500",
"msg_from": "\"Tristan Partin\" <tri... |
[
{
"msg_contents": "Sorry to have to ask for help here, but no amount of stepping through code\nis giving me the answer.\n\nI'm building an index access method which supports an ordering operator:\n\n CREATE OPERATOR pg_catalog.<<=>> (\n FUNCTION = rdb.rank_match,\n LEFTARG = record,\n RI... |
[
{
"msg_contents": "Hi Hackers,\n\nThere is a comment like below in src/include/libpq/libpq.h,\n\n /*\n * prototypes for functions in be-secure.c\n */\nextern PGDLLIMPORT char *ssl_library;\nextern PGDLLIMPORT char *ssl_cert_file;\n\n...\n\nHowever, 'ssl_library', 'ssl_cert_file' and the rest are global \np... |
[
{
"msg_contents": "\nHi,\n\npg_ugprade from v15 to v16 failed in an environment. Often we get a\nreasonable message, but this time it was a bit weird. First, error\nmessage:\n\n=====================================================================================\npg_restore: creating TYPE \"foobar._packagestopt... |
[
{
"msg_contents": "Similar to 'pg_dump --binary-upgrade' [0], we can speed up pg_dump with\nmany sequences by gathering the required information in a single query\ninstead of two queries per sequence. The attached patches are\nworks-in-progress, but here are the results I see on my machine for\n'pg_dump --sche... |
[
{
"msg_contents": "I'm trying to figure out why BufFileSize() Asserts that file->fileset\nisn't NULL, per 1a990b207.\n\nThe discussion for that commit is in [1], but I don't see any\nexplanation of the Assert in the discussion or commit message and\nthere's no comment explaining why it's there.\n\nThe code that... |
[
{
"msg_contents": "Hi\n\nI found https://github.com/vnmakarov/mir?tab=readme-ov-file\n\nIt should be significantly faster than llvm (compilation).\n\nRegards\n\nPavel\n\nHiI found https://github.com/vnmakarov/mir?tab=readme-ov-fileIt should be significantly faster than llvm (compilation).RegardsPavel",
"msg... |
[
{
"msg_contents": "(40af10b57 did this for tuplesort.c, this is the same, but for tuplestore.c)\n\nI was looking at the tuplestore.c code a few days ago and noticed that\nit allocates tuples in the memory context that tuplestore_begin_heap()\nis called in, which for nodeMaterial.c, is ExecutorState.\n\nI didn't... |
[
{
"msg_contents": "This is likely small potatoes compared to some of the other\npg_upgrade-related improvements I've proposed [0] [1] or plan to propose,\nbut this is easy enough, and I already wrote the patch, so here it is.\nAFAICT there's no reason to bother syncing these dump files to disk. If\nsomeone pul... |
[
{
"msg_contents": "See\n\nhttps://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=7155cc4a60e7bfc837233b2dea2563a2edc673fd\n\nAs usual, please send any corrections by Sunday.\n\n\t\t\tregards, tom lane\n\n\n",
"msg_date": "Fri, 03 May 2024 14:12:32 -0400",
"msg_from": "Tom Lane <tgl@sss.pgh.p... |
[
{
"msg_contents": "Dear pgsql-hackers,\n\nOne-line Summary:\nProposal to introduce the CREATE OR REPLACE syntax for EVENT TRIGGER in\nPostgreSQL.\n\nBusiness Use-case:\nCurrently, to modify an EVENT TRIGGER, one must drop and recreate it. This\nproposal aims to introduce a CREATE OR REPLACE syntax for EVENT TRI... |
[
{
"msg_contents": "Hi all,\n\nThere's a rare edge case in `alter table` that can prevent users from\ndropping a column as shown below\n\n # create table atacc1(a int, \"........pg.dropped.1........\" int);\n CREATE TABLE\n # alter table atacc1 drop column a;\n ERROR: duplicate key value violates un... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.