| From: |
| PWN via PostgreSQL Announce <announce-noreply-AT-postgresql.org> |
| To: |
| PostgreSQL Announce <pgsql-announce-AT-lists.postgresql.org> |
| Subject: |
| PostgreSQL Weekly News - November 8, 2020 |
| Date: |
| Mon, 09 Nov 2020 13:31:16 +0000 |
| Message-ID: |
| <160492867662.25505.9664044358928345339@wrigleys.postgresql.org> |
# PostgreSQL Weekly News - November 8, 2020
Congratulations to new core team members Andres Freund and Jonathan Katz!
[https://www.postgresql.org/developer/core/](https://www.postgresql.org/developer/core/)
Person of the week:
[https://postgresql.life/post/elein\_mustain/](https://postgresql.life/post/elein\_mustain/)
# PostgreSQL Product News
phpPgAdmin 7.13.1, a web-based administration tool for PostgreSQL, released.
[https://xzilla.net//blog/2020/Nov/phpPgAdmin-7-13-1-relea...](https://xzilla.net//blog/2020/Nov/phpPgAdmin-7-13-1-relea...)
Ajqvue Version 3.3, a java-based UI which supports PostgreSQL, released.
[http://ajqvue.com](http://ajqvue.com)
pg\_statement\_rollback, an extension that adds server side
transaction with rollback at statement level, released.
[https://github.com/lzlabs/pg\_statement\_rollback/releases/](https://github.com/lzlabs/pg\_statement\_rollback/releases/)
# PostgreSQL Jobs for November
[http://archives.postgresql.org/pgsql-jobs/2020-11/](http://archives.postgresql.org/pgsql-jobs/2020-11/)
# PostgreSQL in the News
Planet PostgreSQL: [http://planet.postgresql.org/](http://planet.postgresql.org/)
PostgreSQL Weekly News is brought to you this week by David Fetter
Submit news and announcements by Sunday at 3:00pm PST8PDT to david@fetter.org.
# Applied Patches
Tom Lane pushed:
- Fix two issues in TOAST decompression. pglz\_maximum\_compressed\_size()
potentially underestimated the amount of compressed data required to produce N
bytes of decompressed data; this is a fault in commit 11a078cf8. Separately
from that, pglz\_decompress() failed to protect itself against corrupt
compressed data, particularly off == 0 in a match tag. Commit c60e520f6
turned such a situation into an infinite loop, where before it'd just have
resulted in garbage output. The combination of these two bugs seems like it
may explain bug #16694 from Tom Vijlbrief, though it's impossible to be quite
sure without direct inspection of the failing session. (One needs to assume
that the pglz\_maximum\_compressed\_size() bug caused us to fail to fetch the
second byte of a match tag, and what happened to be there instead was a zero.
The reported infinite loop is hard to explain without off == 0, though.)
Aside from fixing the bugs, rewrite associated comments for more clarity.
Back-patch to v13 where both these commits landed. Discussion:
[https://postgr.es/m/16694-f107871e499ec114@postgresql.org](https://postgr.es/m/16694-f107871e499ec114@postgresql.org)
[https://git.postgresql.org/pg/commitdiff/dfc797730fc7a07c...](https://git.postgresql.org/pg/commitdiff/dfc797730fc7a07c...)
- Second thoughts on TOAST decompression. On detecting a corrupted match tag,
pglz\_decompress() should just summarily return -1. Breaking out of the loop,
as I did in dfc797730, doesn't quite guarantee that will happen. Also, we can
use unlikely() on that check, just in case it helps. Backpatch to v13, like
the previous patch.
[https://git.postgresql.org/pg/commitdiff/fd2997565c6f6683...](https://git.postgresql.org/pg/commitdiff/fd2997565c6f6683...)
- Rethink the generation rule for fmgroids.h macros. Traditionally, the names of
fmgroids.h macros for pg\_proc OIDs have been constructed from the prosrc
field. But sometimes the same C function underlies multiple pg\_proc entries,
forcing us to make an arbitrary choice of which OID to reference; the other
entries are then not namable via fmgroids.h. Moreover, we could not have
macros at all for pg\_proc entries that aren't for C-coded functions. Instead,
use the proname field, and append the proargtypes field (replacing
inter-argument spaces with underscores) if proname is not unique.
Special-casing unique entries such as F\_OIDEQ removes the need to change a lot
of code. Indeed, I can only find two places in the tree that need to be
adjusted; while this changes quite a few existing entries in fmgroids.h, few
of them are referenced from C code. With this patch, all entries in
pg\_proc.dat have macros in fmgroids.h. Discussion:
[https://postgr.es/m/472274.1604258384@sss.pgh.pa.us](https://postgr.es/m/472274.1604258384@sss.pgh.pa.us)
[https://git.postgresql.org/pg/commitdiff/8e1f37c07aafd4bb...](https://git.postgresql.org/pg/commitdiff/8e1f37c07aafd4bb...)
- Remove special checks for pg\_rewrite.ev\_qual and ev\_action being NULL.
make\_ruledef() and make\_viewdef() were coded to cope with possible null-ness
of these columns, but they've been marked BKI\_FORCE\_NOT\_NULL for some time.
So there's not really any need to do more than what we do for the other
columns of pg\_rewrite, i.e. just Assert that we got non-null results. (There
is a school of thought that says Asserts aren't the thing to do to check for
corrupt data, but surely here is not the place to start if we want such a
policy.) Also, remove long-dead-if-indeed-it-ever-wasn't-dead handling of an
empty actions list in make\_ruledef(). That's an error case and should be
treated as such. (DO INSTEAD NOTHING is represented by a CMD\_NOTHING Query,
not an empty list; cf transformRuleStmt.) Kyotaro Horiguchi, some changes by
me Discussion:
[https://postgr.es/m/CAEudQApoA=tMTic6xEPYP\_hsNZ8XtToVThK...](https://postgr.es/m/CAEudQApoA=tMTic6xEPYP\_hsNZ8XtToVThK...)
[https://git.postgresql.org/pg/commitdiff/e1339bfc7a2fd462...](https://git.postgresql.org/pg/commitdiff/e1339bfc7a2fd462...)
- Fix unportable use of getnameinfo() in pg\_hba\_file\_rules view. fill\_hba\_line()
thought it could get away with passing sizeof(struct sockaddr\_storage) rather
than the actual addrlen previously returned by getaddrinfo(). While that
appears to work on many platforms, it does not work on FreeBSD 11: you get
back a failure, which leads to the view showing NULL for the address and
netmask columns in all rows. The POSIX spec for getnameinfo() is pretty
clearly on FreeBSD's side here: you should pass the actual address length. So
it seems plausible that there are other platforms where this coding also
fails, and we just hadn't noticed. Also, IMO the fact that getnameinfo()
failure leads to a NULL output is pretty bogus in itself. Our
pg\_getnameinfo\_all() wrapper is careful to emit "???" on failure, and we
should use that in such cases. NULL should only be emitted in rows that don't
have IP addresses. Per bug #16695 from Peter Vandivier. Back-patch to v10
where this code was added. Discussion:
[https://postgr.es/m/16695-a665558e2f630be7@postgresql.org](https://postgr.es/m/16695-a665558e2f630be7@postgresql.org)
[https://git.postgresql.org/pg/commitdiff/0a4b34031279d938...](https://git.postgresql.org/pg/commitdiff/0a4b34031279d938...)
- Allow users with BYPASSRLS to alter their own passwords. The intention in
commit 491c029db was to require superuserness to change the BYPASSRLS
property, but the actual effect of the coding in AlterRole() was to require
superuserness to change anything at all about a BYPASSRLS role. Other
properties of a BYPASSRLS role should be changeable under the same rules as
for a normal role, though. Fix that, and also take care of some documentation
omissions related to BYPASSRLS and REPLICATION role properties. Tom Lane and
Stephen Frost, per bug report from Wolfgang Walther. Back-patch to all
supported branches. Discussion:
[https://postgr.es/m/a5548a9f-89ee-3167-129d-162b5985fcf8@...](https://postgr.es/m/a5548a9f-89ee-3167-129d-162b5985fcf8@...)
[https://git.postgresql.org/pg/commitdiff/d907bd0543aa63e5...](https://git.postgresql.org/pg/commitdiff/d907bd0543aa63e5...)
- Improve error messages around REPLICATION and BYPASSRLS properties. Clarify
wording as per suggestion from Wolfgang Walther. No back-patch; this doesn't
seem worth thrashing translatable strings in the back branches. Tom Lane and
Stephen Frost Discussion:
[https://postgr.es/m/a5548a9f-89ee-3167-129d-162b5985fcf8@...](https://postgr.es/m/a5548a9f-89ee-3167-129d-162b5985fcf8@...)
[https://git.postgresql.org/pg/commitdiff/17fb60387ce3fdc2...](https://git.postgresql.org/pg/commitdiff/17fb60387ce3fdc2...)
- Guard against core dump from uninitialized subplan. If the planner erroneously
puts a non-parallel-safe SubPlan into a parallelized portion of the query
tree, nodeSubplan.c will fail in the worker processes because it finds a null
in es\_subplanstates, which it's unable to cope with. It seems worth a
test-and-elog to make that an error case rather than a core dump case. This
probably should have been included in commit 16ebab688, which was responsible
for allowing nulls to appear in es\_subplanstates to begin with. So,
back-patch to v10 where that came in. Discussion:
[https://postgr.es/m/924226.1604422326@sss.pgh.pa.us](https://postgr.es/m/924226.1604422326@sss.pgh.pa.us)
[https://git.postgresql.org/pg/commitdiff/92f87182f2c617fd...](https://git.postgresql.org/pg/commitdiff/92f87182f2c617fd...)
- Remove useless entries for aggregate functions from fmgrtab.c. Gen\_fmgrtab.pl
treated aggregate functions the same as other built-in functions, which is
wasteful because there is no real need to have entries for them in the
fmgr\_builtins[] table. Suppressing those entries saves about 3KB in the
compiled table on my machine; which is not a lot but it's not nothing either,
considering that that table is pretty "hot". The only outside code change
needed is that ExecInitWindowAgg() can't be allowed to call fmgr\_info\_cxt() on
a plain aggregate function. But that saves a few cycles anyway. Having done
that, the aggregate\_dummy() function is unreferenced and might as well be
dropped. Using "aggregate\_dummy" as the prosrc value for an aggregate is now
just a documentation convention not something that matters. There was some
discussion of using NULL instead to save a few bytes in pg\_proc, but we'd have
to remove prosrc's BKI\_FORCE\_NOT\_NULL marking which doesn't seem a great idea.
Anyway, it's possible there's client-side code that expects to see
"aggregate\_dummy" there, so I'm loath to change it without a strong reason.
Discussion:
[https://postgr.es/m/533989.1604263665@sss.pgh.pa.us](https://postgr.es/m/533989.1604263665@sss.pgh.pa.us)
[https://git.postgresql.org/pg/commitdiff/f21636e5d5b8394e...](https://git.postgresql.org/pg/commitdiff/f21636e5d5b8394e...)
- Improve our ability to regurgitate SQL-syntax function calls. The SQL spec
calls out nonstandard syntax for certain function calls, for example
substring() with numeric position info is supposed to be spelled
"SUBSTRING(string FROM start FOR count)". We accept many of these things, but
up to now would not print them in the same format, instead simplifying down to
"substring"(string, start, count). That's long annoyed me because it creates
an interoperability problem: we're gratuitously injecting Postgres-specific
syntax into what might otherwise be a perfectly spec-compliant view
definition. However, the real reason for addressing it right now is to support
a planned change in the semantics of EXTRACT() a/k/a date\_part(). When we
switch that to returning numeric, we'll have the parser translate EXTRACT() to
some new function name (might as well be "extract" if you ask me) and then
teach ruleutils.c to reverse-list that per SQL spec. In this way existing
calls to date\_part() will continue to have the old semantics. To implement
this, invent a new CoercionForm value COERCE\_SQL\_SYNTAX, and make the parser
insert that rather than COERCE\_EXPLICIT\_CALL when the input has SQL-spec
decoration. (But if the input has the form of a plain function call, continue
to mark it COERCE\_EXPLICIT\_CALL, even if it's calling one of these functions.)
Then ruleutils.c recognizes COERCE\_SQL\_SYNTAX as a cue to emit SQL call
syntax. It can know which decoration to emit using hard-wired knowledge about
the functions that could be called this way. (While this solution isn't
extensible without manual additions, neither is the grammar, so this doesn't
seem unmaintainable.) Notice that this solution will reverse-list a function
call with SQL decoration only if it was entered that way; so dump-and-reload
will not by itself produce any changes in the appearance of views. This
requires adding a CoercionForm field to struct FuncCall. (I couldn't resist
the temptation to rearrange that struct's field order a tad while I was at
it.) FuncCall doesn't appear in stored rules, so that change isn't a reason
for a catversion bump, but I did one anyway because the new enum value for
CoercionForm fields could confuse old backend code. Possible future work: \*
Perhaps CoercionForm should now be renamed to DisplayForm, or something like
that, to reflect its more general meaning. This'd require touching a couple
hundred places, so it's not clear it's worth the code churn. \* The
SQLValueFunction node type, which was invented partly for the same goal of
improving SQL-compatibility of view output, could perhaps be replaced with
regular function calls marked with COERCE\_SQL\_SYNTAX. It's unclear if this
would be a net code savings, however. Discussion:
[https://postgr.es/m/42b73d2d-da12-ba9f-570a-420e0cce19d9@...](https://postgr.es/m/42b73d2d-da12-ba9f-570a-420e0cce19d9@...)
[https://git.postgresql.org/pg/commitdiff/40c24bfef92530bd...](https://git.postgresql.org/pg/commitdiff/40c24bfef92530bd...)
- Declare lead() and lag() using anycompatible not anyelement. This allows use
of a "default" expression that doesn't slavishly match the data column's type.
Formerly you got something like "function lag(numeric, integer, integer) does
not exist", which is not just unhelpful but actively misleading. The SQL spec
suggests that the default should be coerced to the data column's type, but
this implementation instead chooses the common supertype, which seems at least
as reasonable. (Note: I took the opportunity to run "make reformat-dat-files"
on pg\_proc.dat, so this commit includes some cosmetic changes to
recently-added entries that aren't related to lead/lag.) Vik Fearing
Discussion:
[https://postgr.es/m/77675130-89da-dab1-51dd-492c93dcf5d1@...](https://postgr.es/m/77675130-89da-dab1-51dd-492c93dcf5d1@...)
[https://git.postgresql.org/pg/commitdiff/5c292e6b90433c76...](https://git.postgresql.org/pg/commitdiff/5c292e6b90433c76...)
- Declare assorted array functions using anycompatible not anyelement. Convert
array\_append, array\_prepend, array\_cat, array\_position, array\_positions,
array\_remove, array\_replace, and width\_bucket to use anycompatiblearray. This
is a simple extension of commit 5c292e6b9 to hit some other places where
there's a pretty obvious gain in usability from doing so. Ideally we'd also
modify other functions taking multiple old-style polymorphic arguments. But
most of the remainder are tied into one or more operator classes, making any
such change a much larger can of worms than I desire to open right now.
Discussion:
[https://postgr.es/m/77675130-89da-dab1-51dd-492c93dcf5d1@...](https://postgr.es/m/77675130-89da-dab1-51dd-492c93dcf5d1@...)
[https://git.postgresql.org/pg/commitdiff/9e38c2bb5093ceb0...](https://git.postgresql.org/pg/commitdiff/9e38c2bb5093ceb0...)
- Remove underflow error in float division with infinite divisor. float4\_div and
float8\_div correctly produced zero for zero divided by infinity, but threw an
underflow error for nonzero finite values divided by infinity. This seems
wrong; at the very least it's inconsistent with the behavior recently
implemented for numeric infinities. Remove the error and allow zero to be
returned. This patch also removes a useless isinf() test from the overflow
checks in these functions (non-Inf divided by Inf can't produce Inf).
Extracted from a larger patch; this seems significant outside the context of
geometric operators, so it deserves its own commit. Kyotaro Horiguchi
Discussion:
[https://postgr.es/m/CAGf+fX70rWFOk5cd00uMfa\_\_0yP+vtQg5c...](https://postgr.es/m/CAGf+fX70rWFOk5cd00uMfa\_\_0yP+vtQg5c...)
[https://git.postgresql.org/pg/commitdiff/fac83dbd6fe1ac3d...](https://git.postgresql.org/pg/commitdiff/fac83dbd6fe1ac3d...)
- Don't throw an error for LOCK TABLE on a self-referential view. LOCK TABLE has
complained about "infinite recursion" when applied to a self-referential view,
ever since we made it recurse into views in v11. However, that breaks
pg\_dump's new assumption that it's okay to lock every relation. There doesn't
seem to be any good reason to throw an error: if we just abandon the
recursion, we've still satisfied the requirement of locking every referenced
relation. Per bug #16703 from Andrew Bille (via Alexander Lakhin).
Discussion:
[https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org](https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org)
[https://git.postgresql.org/pg/commitdiff/5b7bfc39726ff9f6...](https://git.postgresql.org/pg/commitdiff/5b7bfc39726ff9f6...)
- Revert "pg\_dump: Lock all relations, not just plain tables". Revert 403a3d91c,
as well as the followup fix 7f4235032, in all branches. We need to think a
bit harder about what the behavior of LOCK TABLE on views should be, and
there's no time for that before next week's releases. We'll take another
crack at this later. Discussion:
[https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org](https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org)
[https://git.postgresql.org/pg/commitdiff/d3adaabaf7d555ec...](https://git.postgresql.org/pg/commitdiff/d3adaabaf7d555ec...)
- Revert "Accept relations of any kind in LOCK TABLE". Revert 59ab4ac32, as well
as the followup fix 33862cb9c, in all branches. We need to think a bit harder
about what the behavior of LOCK TABLE on views should be, and there's no time
for that before next week's releases. We'll take another crack at this later.
Discussion:
[https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org](https://postgr.es/m/16703-e348f58aab3cf6cc@postgresql.org)
[https://git.postgresql.org/pg/commitdiff/eeda7f6338095701...](https://git.postgresql.org/pg/commitdiff/eeda7f6338095701...)
- Fix ecpg's mishandling of B'...' and X'...' literals. These were broken in
multiple ways: \* The xbstart and xhstart lexer actions neglected to set
"state\_before\_str\_start" before transitioning to the xb/xh states, thus
possibly resulting in "internal error: unreachable state" later. \* The test
for valid string contents at the end of xb state was flat out wrong, as it
accounted incorrectly for the "b" prefix that the xbstart action had injected.
Meanwhile, the xh state had no such check at all. \* The generated literal
value failed to include any quote marks. \* The grammar did the wrong thing
anyway, typically ignoring the literal value and emitting something else,
since BCONST and XCONST tokens were handled randomly differently from SCONST
tokens. The first of these problems is evidently an oversight in commit
7f380c59f, but the others seem to be very ancient. The lack of complaints
shows that ECPG users aren't using these syntaxes much (although I do vaguely
remember one previous complaint). As written, this patch is dependent on
7f380c59f, so it can't go back further than v13. Given the shortage of
complaints, I'm not excited about adapting the patch to prior branches.
Report and patch by Shenhao Wang (test case adjusted by me) Discussion:
[https://postgr.es/m/d6402f1bacb74ecba22ef715dbba17fd@G08C...](https://postgr.es/m/d6402f1bacb74ecba22ef715dbba17fd@G08C...)
[https://git.postgresql.org/pg/commitdiff/1e3868ab3bef5cfa...](https://git.postgresql.org/pg/commitdiff/1e3868ab3bef5cfa...)
- Avoid re-using output variables in new ecpg test case. The buildfarm thinks
this leads to memory stomps, though annoyingly I can't duplicate that here.
The existing code in strings.pgc is doing something that doesn't seem to be
sanctioned at all really by the documentation, but I'm disinclined to try to
make that nicer right now. Let's just declare some more output variables in
hopes of working around it.
[https://git.postgresql.org/pg/commitdiff/eed4356fad84b0fd...](https://git.postgresql.org/pg/commitdiff/eed4356fad84b0fd...)
David Rowley pushed:
- Allow run-time pruning on nested Append/MergeAppend nodes. Previously we only
tagged on the required information to allow the executor to perform run-time
partition pruning for Append/MergeAppend nodes belonging to base relations.
It was thought that nested Append/MergeAppend nodes were just about always
pulled up into the top-level Append/MergeAppend and that making the run-time
pruning info for any sub Append/MergeAppend nodes was a waste of time.
However, that was likely badly thought through. Some examples of cases we're
unable to pullup nested Append/MergeAppends are: 1) Parallel Append nodes with
a mix of parallel and non-parallel paths into a Parallel Append. 2) When
planning an ordered Append scan a sub-partition which is unordered may require
a nested MergeAppend path to ensure sub-partitions don't mix up the order of
tuples being fed into the top-level Append. Unfortunately, it was not just as
simple as removing the lines in createplan.c which were purposefully not
building the run-time pruning info for anything but RELOPT\_BASEREL relations.
The code in add\_paths\_to\_append\_rel() was far too sloppy about which
partitioned\_rels it included for the Append/MergeAppend paths. The original
code there would always assume accumulate\_append\_subpath() would pull each
sub-Append and sub-MergeAppend path into the top-level path. While it does
not appear that there were any actual bugs caused by having the additional
partitioned table RT indexes recorded, what it did mean is that later in
planning, when we built the run-time pruning info that we wasted effort and
built PartitionedRelPruneInfos for partitioned tables that we had no subpaths
for the executor to run-time prune. Here we tighten that up so that
partitioned\_rels only ever contains the RT index for partitioned tables which
actually have subpaths in the given Append/MergeAppend. We can now Assert
that every PartitionedRelPruneInfo has a non-empty present\_parts. That should
allow us to catch any weird corner cases that have been missed. In passing,
it seems there is no longer a good reason to have the AppendPath and
MergeAppendPath's partitioned\_rel fields a List of IntList. We can simply have
a List of Relids instead. This is more compact in memory and faster to add
new members to. We still know which is the root level partition as these
always have a lower relid than their children. Previously this field was used
for more things, but run-time partition pruning now remains the only user of
it and it has no need for a List of IntLists. Here we also get rid of the
RelOptInfo partitioned\_child\_rels field. This is what was previously used to
(sometimes incorrectly) set the Append/MergeAppend path's partitioned\_rels
field. That was the only usage of that field, so we can happily just remove
it. I also couldn't resist changing some nearby code to make use of the newly
added for\_each\_from macro so we can skip the first element in the list without
checking if the current item was the first one on each iteration. A bug
report from Andreas Kretschmer prompted all this work, however, after some
consideration, I'm not personally classing this as a bug fix. So no backpatch.
In Andreas' test case, it just wasn't that clear that there was a nested
Append since the top-level Append just had a single sub-path which was pulled
up a level, per 8edd0e794. Author: David Rowley Reviewed-by: Amit Langote
Discussion:
[https://postgr.es/m/flat/CAApHDvqSchs%2BubdybcfFaSPB%2B%2...](https://postgr.es/m/flat/CAApHDvqSchs%2BubdybcfFaSPB%2B%2...)
[https://git.postgresql.org/pg/commitdiff/a929e17e5a8c9b75...](https://git.postgresql.org/pg/commitdiff/a929e17e5a8c9b75...)
- Fix unstable partition\_prune regression tests. This was broken recently by
a929e17e5. I'd failed to remember that parallel tests should have their
EXPLAIN output run through the explain\_parallel\_append function so that the
output is stable when parallel workers fail to start. fairywren was first to
notice. Reported-by: Michael Paquier Discussion:
[https://postgr.es/m/20201102062951.GB15770@paquier.xyz](https://postgr.es/m/20201102062951.GB15770@paquier.xyz)
[https://git.postgresql.org/pg/commitdiff/90d8f1b1826ce076...](https://git.postgresql.org/pg/commitdiff/90d8f1b1826ce076...)
Amit Kapila pushed:
- Use Enum for top level logical replication message types. Logical replication
protocol uses a single byte character to identify a message type in logical
replication protocol. The code uses string literals for the same. Use Enum so
that 1. All the string literals used can be found at a single place. This
makes it easy to add more types without the risk of conflicts. 2. It's easy
to locate the code handling a given message type. 3. When used with switch
statements, it is easy to identify the missing cases using -Wswitch. Author:
Ashutosh Bapat Reviewed-by: Kyotaro Horiguchi, Andres Freund, Peter Smith and
Amit Kapila Discussion:
[https://postgr.es/m/CAExHW5uPzQ7L0oAd\_ENyvaiYMOPgkrAoJpE...](https://postgr.es/m/CAExHW5uPzQ7L0oAd\_ENyvaiYMOPgkrAoJpE...)
[https://git.postgresql.org/pg/commitdiff/644f0d7cc9c2cb27...](https://git.postgresql.org/pg/commitdiff/644f0d7cc9c2cb27...)
- Fix typos. Author: Hou Zhijie Discussion:
[https://postgr.es/m/855a9421839d402b8b351d273c89a8f8@G08C...](https://postgr.es/m/855a9421839d402b8b351d273c89a8f8@G08C...)
[https://git.postgresql.org/pg/commitdiff/8c2d8f6cc4848cf9...](https://git.postgresql.org/pg/commitdiff/8c2d8f6cc4848cf9...)
- Use strlcpy instead of memcpy for copying the slot name in pgstat.c. There is
no outright bug here but it is better to be consistent with the usage at other
places in the same file. In the passing, fix a wrong assertion in
pgstat\_recv\_replslot. Author: Kyotaro Horiguchi Reviewed-by: Sawada Masahiko
and Amit Kapila Discussion:
[https://postgr.es/m/20201104.175523.1704166915688949637.h...](https://postgr.es/m/20201104.175523.1704166915688949637.h...)
[https://git.postgresql.org/pg/commitdiff/4f841ce3f7f4d429...](https://git.postgresql.org/pg/commitdiff/4f841ce3f7f4d429...)
Michaël Paquier pushed:
- Fix some grammar and typos in comments and docs. The documentation fixes are
backpatched down to where they apply. Author: Justin Pryzby Discussion:
[https://postgr.es/m/20201031020801.GD3080@telsasoft.com](https://postgr.es/m/20201031020801.GD3080@telsasoft.com) Backpatch-through: 9.6
[https://git.postgresql.org/pg/commitdiff/8a15e735be00f156...](https://git.postgresql.org/pg/commitdiff/8a15e735be00f156...)
- Revert pg\_relation\_check\_pages(). This reverts the following set of commits,
following complaints about the lack of portability of the central part of the
code in bufmgr.c as well as the use of partition mapping locks during page
reads: c780a7a9 f2b88396 b787d4ce ce7f772c 60a51c6b Per discussion with
Andres Freund, Robert Haas and myself. Bump catalog version. Discussion:
[https://postgr.es/m/20201029181729.2nrub47u7yqncsv7@alap3...](https://postgr.es/m/20201029181729.2nrub47u7yqncsv7@alap3...)
[https://git.postgresql.org/pg/commitdiff/e152506adef4bc50...](https://git.postgresql.org/pg/commitdiff/e152506adef4bc50...)
- Fix minor issues with new unicode {de,re}composition code. The table
generation script would incorrectly complain in the recomposition sorting when
matching code points. This would not have caused the generation of an
incorrect table. Note that this condition is not reachable yet, but could
have been reached with future updates. pg\_bswap.h does not need to be
included in the frontend.x Author: John Naylor Discussion:
[https://postgr.es/m/CAFBsxsGWmExpvv=61vtDKCs7+kBbhkwBDL2P...](https://postgr.es/m/CAFBsxsGWmExpvv=61vtDKCs7+kBbhkwBDL2P...)
[https://git.postgresql.org/pg/commitdiff/ceaeac54f75f0117...](https://git.postgresql.org/pg/commitdiff/ceaeac54f75f0117...)
- Add GUC\_LIST\_INPUT and GUC\_LIST\_QUOTE to unix\_socket\_directories. This should
have been done in the initial commit that made unix\_socket\_directories a list
as of c9b0cbe. This change allows to support correctly the case of ALTER
SYSTEM, where it is possible to specify multiple paths as a list, like the
following pattern where flattening is applied to each item: ALTER SYSTEM SET
unix\_socket\_directories = '/path1', '/path2'; Any parameters specified in
postgresql.conf are parsed the same way, so there is no compatibility change.
pg\_dump has a hardcoded list of parameters marked with GUC\_LIST\_QUOTE, that
gets its routine update. These are reordered alphabetically for clarity.
Author: Ian Lawrence Barwick Reviewed-by: Peter Eisentraunt, Tom Lane, Michael
Paquier Discussion:
[https://postgr.es/m/CAB8KJ=iMOtNY6\_sUwV=LQVCJ2zgYHBDyNzV...](https://postgr.es/m/CAB8KJ=iMOtNY6\_sUwV=LQVCJ2zgYHBDyNzV...)
[https://git.postgresql.org/pg/commitdiff/a05dbf477b0ef173...](https://git.postgresql.org/pg/commitdiff/a05dbf477b0ef173...)
Heikki Linnakangas pushed:
- doc: Mention UNION/ORDER BY etc. keywords in section headers. Most of the
section and sub-section headers in the Queries chapter have the keywords
literally stated, but neither "Sorting Rows" nor "Combining Rows" did. There's
no rule that they must be, but it seems like a good practice. The keywords
will ring a bell to anyone with with even a little bit of SQL experience.
David G. Johnston, per suggestion by bilge@scriptfusion.com Discussion:
[https://www.postgresql.org/message-id/159981394174.31338....](https://www.postgresql.org/message-id/159981394174.31338....)
[https://git.postgresql.org/pg/commitdiff/8ef2a5afdf8ec9e4...](https://git.postgresql.org/pg/commitdiff/8ef2a5afdf8ec9e4...)
- pg\_rewind: Move syncTargetDirectory() to file\_ops.c. For consistency. All the
other low-level functions that operate on the target directory are in
file\_ops.c. Reviewed-by: Michael Paquier Discussion:
[https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...](https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...)
[https://git.postgresql.org/pg/commitdiff/ffb4e27e9c5ea87f...](https://git.postgresql.org/pg/commitdiff/ffb4e27e9c5ea87f...)
- Refactor pg\_rewind for more clear decision making. Deciding what to do with
each file is now a separate step after all the necessary information has been
gathered. It is more clear that way. Previously, the decision-making was
divided between process\_source\_file() and process\_target\_file(), and it was a
bit hard to piece together what the overall rules were. Reviewed-by: Kyotaro
Horiguchi, Soumyadeep Chakraborty Discussion:
[https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...](https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...)
[https://git.postgresql.org/pg/commitdiff/eb00f1d4bf96bdba...](https://git.postgresql.org/pg/commitdiff/eb00f1d4bf96bdba...)
- pg\_rewind: Replace the hybrid list+array data structure with simplehash. Now
that simplehash can be used in frontend code, let's make use of it.
Reviewed-by: Kyotaro Horiguchi, Soumyadeep Chakraborty Discussion:
[https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...](https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...)
[https://git.postgresql.org/pg/commitdiff/f81e97d0475cd4bc...](https://git.postgresql.org/pg/commitdiff/f81e97d0475cd4bc...)
- pg\_rewind: Refactor the abstraction to fetch from local/libpq source. This
makes the abstraction of a "source" server more clear, by introducing a common
abstract class, borrowing the object-oriented programming term, that
represents all the operations that can be done on the source server. There are
two implementations of it, one for fetching via libpq, and another to fetch
from a local directory. This adds some code, but makes it easier to understand
what's going on. The copy\_executeFileMap() and libpq\_executeFileMap()
functions contained basically the same logic, just calling different functions
to fetch the source files. Refactor so that the common logic is in one place,
in a new function called perform\_rewind(). Reviewed-by: Kyotaro Horiguchi,
Soumyadeep Chakraborty Discussion:
[https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...](https://www.postgresql.org/message-id/0c5b3783-af52-3ee5-...)
[https://git.postgresql.org/pg/commitdiff/37d2ff38031262a1...](https://git.postgresql.org/pg/commitdiff/37d2ff38031262a1...)
Thomas Munro pushed:
- Add pg\_depend.refobjversion. Provide a place for the version of referenced
database objects to be recorded. A follow-up commit will use this to record
dependencies on collation versions for indexes, but similar ideas for other
kinds of objects have also been mooted. Author: Thomas Munro
thomas.munro [AT] gmail.com Reviewed-by: Julien Rouhaud rjuju123 [AT] gmail.com
Reviewed-by: Peter Eisentraut peter.eisentraut [AT] 2ndquadrant.com Discussion:
[https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...](https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...)
[https://git.postgresql.org/pg/commitdiff/cd6f479e79f3a33e...](https://git.postgresql.org/pg/commitdiff/cd6f479e79f3a33e...)
- Remove pg\_collation.collversion. This model couldn't be extended to cover the
default collation, and didn't have any information about the affected database
objects when the version changed. Remove, in preparation for a follow-up
commit that will add a new mechanism. Author: Thomas Munro
thomas.munro [AT] gmail.com Reviewed-by: Julien Rouhaud rjuju123 [AT] gmail.com
Reviewed-by: Peter Eisentraut peter.eisentraut [AT] 2ndquadrant.com Discussion:
[https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...](https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...)
[https://git.postgresql.org/pg/commitdiff/7d1297df0830725d...](https://git.postgresql.org/pg/commitdiff/7d1297df0830725d...)
- Track collation versions for indexes. Record the current version of dependent
collations in pg\_depend when creating or rebuilding an index. When accessing
the index later, warn that the index may be corrupted if the current version
doesn't match. Thanks to Douglas Doole, Peter Eisentraut, Christoph Berg,
Laurenz Albe, Michael Paquier, Robert Haas, Tom Lane and others for very
helpful discussion. Author: Thomas Munro thomas.munro [AT] gmail.com Author:
Julien Rouhaud rjuju123 [AT] gmail.com Reviewed-by: Peter Eisentraut
peter.eisentraut [AT] 2ndquadrant.com (earlier versions) Discussion:
[https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...](https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...)
[https://git.postgresql.org/pg/commitdiff/257836a75585934c...](https://git.postgresql.org/pg/commitdiff/257836a75585934c...)
- Tolerate version lookup failure for old style Windows locale names. Accept
that we can't get versions for such locale names for now. Users will need to
specify the newer language tag format to enable the collation versioning
feature. It's not clear that we can do automatic conversion from the old
style to the new style reliably enough for this purpose. Unfortunately, this
means that collation versioning probably won't work for the default collation
unless you provide something like en-US at initdb or CREATE DATABASE time
(though, for reasons not yet understood, it does seem to work on some
systems). It'd be nice to find a better solution, or document this quirk if
we settle on it, but this should unbreak the 3 failing build farm animals in
the meantime. Reviewed-by: David Rowley dgrowleyml [AT] gmail.com Reviewed-by:
Tom Lane tgl [AT] sss.pgh.pa.us Discussion:
[https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...](https://postgr.es/m/CAEepm%3D0uEQCpfq\_%2BLYFBdArCe4Ot98t...)
[https://git.postgresql.org/pg/commitdiff/9f12a3b95dd56c89...](https://git.postgresql.org/pg/commitdiff/9f12a3b95dd56c89...)
- Fix unlinking of SLRU segments. Commit dee663f7 intended to drop any queued up
fsync requests before unlinking segment files, but missed a code path. Fix,
by centralizing the forget-and-unlink code into a single function.
Reported-by: Tomas Vondra tomas.vondra [AT] 2ndquadrant.com Discussion:
[https://postgr.es/m/20201104013205.icogbi773przyny5%40dev...](https://postgr.es/m/20201104013205.icogbi773przyny5%40dev...)
[https://git.postgresql.org/pg/commitdiff/c732c3f8c122009d...](https://git.postgresql.org/pg/commitdiff/c732c3f8c122009d...)
- Fix assertion in collation version lookup. Commit 257836a7 included an
assertion that a version lookup routine is not trying to look up "C" or
"POSIX", but that case is reachable with the user-facing SQL function
pg\_collation\_actual\_version(). Remove the assertion.
[https://git.postgresql.org/pg/commitdiff/d50e3b1f8d763d51...](https://git.postgresql.org/pg/commitdiff/d50e3b1f8d763d51...)
Magnus Hagander pushed:
- Clarify temporary table name shadowing in CREATE TABLE docs. Author: David
Johnston
[https://git.postgresql.org/pg/commitdiff/5b3dca096f8e504b...](https://git.postgresql.org/pg/commitdiff/5b3dca096f8e504b...)
- Use be\_tls\_\* API for SSL information in sslinfo. sslinfo was passing the
Port->ssl member directly to OpenSSL in order to extract information regarding
the connection. This breaks the API provided by the backend TLS
implementation, as well as duplicates code for no benefit. Rewrite to make use
of the backend API as much as possible. Author: Daniel Gustafsson
daniel [AT] yesql.se
[https://git.postgresql.org/pg/commitdiff/5d1833f414973595...](https://git.postgresql.org/pg/commitdiff/5d1833f414973595...)
- Improve error handling in backend OpenSSL implementation. Commit d94c36a45ab
introduced error handling to sslinfo to handle OpenSSL errors gracefully. This
ports this errorhandling to the backend TLS implementation. Author: Daniel
Gustafsson daniel [AT] yesql.se
[https://git.postgresql.org/pg/commitdiff/13cfa02f77936895...](https://git.postgresql.org/pg/commitdiff/13cfa02f77936895...)
- Use the non-deprecated TG\_TABLE\_MAME in test trigger. Commit 3a9ae3d2068 (back
in 2006) deprecated TG\_RELNAME in favor of TG\_TABLE\_NAME, but the existing
usage in test cases has remained till today. Change to use TG\_TABLE\_NAME
instead (TG\_RELNAME is still covered by a test case).
[https://git.postgresql.org/pg/commitdiff/44a184cb686866b1...](https://git.postgresql.org/pg/commitdiff/44a184cb686866b1...)
- Add pg\_strong\_random\_init function to initialize random number generator.
Currently only OpenSSL requires this initialization, but in the future other
SSL implementations are likely to need it as well. Abstracting this
functionality out into a separate function makes this cleaner and more clear,
and also removes the dependency on OpenSSL headers from fork\_process.c.
OpenSSL is special in that we need to initialize this random number generator
even if we're not going to use it directly, until we drop support for
everything prior to OpenSSL 1.1.1. (And of course also if we actually use it).
All other implementations are left empty at this time, but more are expected
to be added in the future. Author: Daniel Gustafsson daniel [AT] yesql.se,
Michael Paquier michael [AT] paquier.xyz Reviewed-By: Magnus Hagander
magnus [AT] hagander.net Discussion:
[https://postgr.es/m/F6291C3C-747C-4C93-BCE0-28BB420B1FF5@...](https://postgr.es/m/F6291C3C-747C-4C93-BCE0-28BB420B1FF5@...)
[https://git.postgresql.org/pg/commitdiff/5ee180a3947060b9...](https://git.postgresql.org/pg/commitdiff/5ee180a3947060b9...)
Peter Eisentraut pushed:
- Use PG\_GETARG\_TRANSACTIONID where appropriate. Some places were using
PG\_GETARG\_UINT32 where PG\_GETARG\_TRANSACTIONID would be more appropriate. (Of
course, they are the same internally, so there is no externally visible
effect.) To do that, export PG\_GETARG\_TRANSACTIONID outside of xid.c. We
also export PG\_RETURN\_TRANSACTIONID for symmetry, even though there are
currently no external users. Author: Ashutosh Bapat
ashutosh.bapat [AT] 2ndquadrant.com Discussion:
[https://www.postgresql.org/message-id/flat/d8f6bdd536df40...](https://www.postgresql.org/message-id/flat/d8f6bdd536df40...)
[https://git.postgresql.org/pg/commitdiff/dd26a0ad760b1123...](https://git.postgresql.org/pg/commitdiff/dd26a0ad760b1123...)
- Remove use of deprecated containment operators in tests. Switch @ to `<@` and ~
to `@>` in gist-related tests. The old operator names have been deprecated and
will eventually (possibly soon) be removed. Author: Justin Pryzby
pryzby [AT] telsasoft.com Discussion:
[https://www.postgresql.org/message-id/flat/20201027032511...](https://www.postgresql.org/message-id/flat/20201027032511...)
[https://git.postgresql.org/pg/commitdiff/3165426e54df02a6...](https://git.postgresql.org/pg/commitdiff/3165426e54df02a6...)
- Remove deprecated containment operators for built-in types. Remove old
containment operators @ and ~ for built-in geometry data types. These have
been deprecated; use `<@` and `@>` instead. (Some contrib modules still contain
the same deprecated operators. That will be dealt with separately.) Author:
Justin Pryzby pryzby [AT] telsasoft.com Discussion:
[https://www.postgresql.org/message-id/flat/20201027032511...](https://www.postgresql.org/message-id/flat/20201027032511...)
[https://git.postgresql.org/pg/commitdiff/2f70fdb0644c32c4...](https://git.postgresql.org/pg/commitdiff/2f70fdb0644c32c4...)
- Disallow ALTER TABLE ONLY / DROP EXPRESSION. The current implementation cannot
handle this correctly, so just forbid it for now. GENERATED clauses must be
attached to the column definition and cannot be added later like DEFAULT, so
if a child table has a generation expression that the parent does not have,
the child column will necessarily be an attlocal column. So to implement
ALTER TABLE ONLY / DROP EXPRESSION, we'd need extra code to update attislocal
of the direct child tables, somewhat similar to how DROP COLUMN does it, so
that the resulting state can be properly dumped and restored. Discussion:
[https://www.postgresql.org/message-id/flat/15830.15754688...](https://www.postgresql.org/message-id/flat/15830.15754688...)
[https://git.postgresql.org/pg/commitdiff/bf797a8d9768239f...](https://git.postgresql.org/pg/commitdiff/bf797a8d9768239f...)
- Enable hash partitioning of text arrays. hash\_array\_extended() needs to pass
PG\_GET\_COLLATION() to the hash function of the element type. Otherwise, the
hash function of a collation-aware data type such as text will error out,
since the introduction of nondeterministic collation made hash functions
require a collation, too. The consequence of this is that before this change,
hash partitioning using an array over text in the partition key would not
work. Reviewed-by: Heikki Linnakangas hlinnaka [AT] iki.fi Reviewed-by: Tom Lane
tgl [AT] sss.pgh.pa.us Reviewed-by: Michael Paquier michael [AT] paquier.xyz
Discussion:
[https://www.postgresql.org/message-id/flat/32c1fdae-95c6-...](https://www.postgresql.org/message-id/flat/32c1fdae-95c6-...)
[https://git.postgresql.org/pg/commitdiff/560564d3addcb6c5...](https://git.postgresql.org/pg/commitdiff/560564d3addcb6c5...)
- Move catalog toast table declarations. Move the system catalog toast table
declarations from catalog/toasting.h to the respective parent tables'
catalog/pg\_\*.h files. The original reason for having it split was that the
old genbki system produced the output in the order of the catalog files it
read, so all the toasting stuff needed to come separately. But this is no
longer the case, and keeping it together makes more sense. Reviewed-by: John
Naylor john.naylor [AT] enterprisedb.com Discussion:
[https://www.postgresql.org/message-id/flat/c7cc82d6-f976-...](https://www.postgresql.org/message-id/flat/c7cc82d6-f976-...)
[https://git.postgresql.org/pg/commitdiff/b4c9695e79f79d39...](https://git.postgresql.org/pg/commitdiff/b4c9695e79f79d39...)
- Move catalog index declarations. Move the system catalog index declarations
from catalog/indexing.h to the respective parent tables' catalog/pg\_\*.h files.
The original reason for having it split was that the old genbki system
produced the output in the order of the catalog files it read, so all the
indexing stuff needed to come separately. But this is no longer the case, and
keeping it together makes more sense. Reviewed-by: John Naylor
john.naylor [AT] enterprisedb.com Discussion:
[https://www.postgresql.org/message-id/flat/c7cc82d6-f976-...](https://www.postgresql.org/message-id/flat/c7cc82d6-f976-...)
[https://git.postgresql.org/pg/commitdiff/bdc4edbea6fc847f...](https://git.postgresql.org/pg/commitdiff/bdc4edbea6fc847f...)
- Fix redundant error messages in client tools. A few client tools duplicate
error messages already provided by libpq. Discussion:
[https://www.postgresql.org/message-id/flat/3e937641-88a1-...](https://www.postgresql.org/message-id/flat/3e937641-88a1-...)
[https://git.postgresql.org/pg/commitdiff/6be725e701611660...](https://git.postgresql.org/pg/commitdiff/6be725e701611660...)
- Fix test for error message change. fix for
6be725e701611660b36642de9ff1d665a1eb24f5
[https://git.postgresql.org/pg/commitdiff/8cff66d309b90e03...](https://git.postgresql.org/pg/commitdiff/8cff66d309b90e03...)
Tomáš Vondra pushed:
- Fix get\_useful\_pathkeys\_for\_relation for volatile expressions. When
considering Incremental Sort below a Gather Merge, we need to be a bit more
careful when matching pathkeys to EC members. It's not enough to find a member
whose Vars are all in the current relation's target; volatile expressions in
particular need to be contained in the target, otherwise it's too early to use
the pathkey. Reported-by: Jaime Casanova Author: James Coleman Reviewed-by:
Tomas Vondra Backpatch-through: 13, where the incremental sort code was added
Discussion:
[https://postgr.es/m/CAJGNTeNaxpXgBVcRhJX%2B2vSbq%2BF2kJqG...](https://postgr.es/m/CAJGNTeNaxpXgBVcRhJX%2B2vSbq%2BF2kJqG...)
[https://git.postgresql.org/pg/commitdiff/ebb7ae839d033d0f...](https://git.postgresql.org/pg/commitdiff/ebb7ae839d033d0f...)
- Use INT64\_FORMAT to print int64 variables in sort debug. Commit 6ee3b5fb99
cleaned up most of the long/int64 confusion related to incremental sort, but
the sort debug messages were still using %ld for int64 variables. So fix that.
Author: Haiying Tang Backpatch-through: 13, where the incremental sort code
was added Discussion:
[https://postgr.es/m/4250be9d350c4992abb722a76e288aef%40G0...](https://postgr.es/m/4250be9d350c4992abb722a76e288aef%40G0...)
[https://git.postgresql.org/pg/commitdiff/90851d1d26f54ccb...](https://git.postgresql.org/pg/commitdiff/90851d1d26f54ccb...)
- Properly detoast data in brin\_form\_tuple. brin\_form\_tuple failed to consider
the values may be toasted, inserting the toast pointer into the index. This
may easily result in index corruption, as the toast data may be deleted and
cleaned up by vacuum. The cleanup however does not care about indexes, leaving
invalid toast pointers behind, which triggers errors like this: ERROR:
missing chunk number 0 for toast value 16433 in pg\_toast\_16426 A less severe
consequence are inconsistent failures due to the index row being too large,
depending on whether brin\_form\_tuple operated on plain or toasted version of
the row. For example CREATE TABLE t (val TEXT); INSERT INTO t VALUES
('... long value ...') CREATE INDEX idx ON t USING brin (val); would
likely succeed, as the row would likely include toast pointer. Switching the
order of INSERT and CREATE INDEX would likely fail: ERROR: index row
size 8712 exceeds maximum 8152 for index "idx" because this happens before
the row values are toasted. The bug exists since PostgreSQL 9.5 where BRIN
indexes were introduced. So backpatch all the way back. Author: Tomas Vondra
Reviewed-by: Alvaro Herrera Backpatch-through: 9.5 Discussion:
[https://postgr.es/m/20201001184133.oq5uq75sb45pu3aw@devel...](https://postgr.es/m/20201001184133.oq5uq75sb45pu3aw@devel...) Discussion:
[https://postgr.es/m/20201104010544.zexj52mlldagzowv%40dev...](https://postgr.es/m/20201104010544.zexj52mlldagzowv%40dev...)
[https://git.postgresql.org/pg/commitdiff/7577dd84807a808f...](https://git.postgresql.org/pg/commitdiff/7577dd84807a808f...)
Fujii Masao pushed:
- Use standard SIGHUP handler in syslogger. Commit 1e53fe0e70 changed background
processes so that they use standard SIGHUP handler. Like that, this commit
makes syslogger use standard SIGHUP handler to simplify the code. Author:
Bharath Rupireddy Reviewed-by: Fujii Masao Discussion:
[https://postgr.es/m/CALj2ACXPorUqePswDtOeM\_s82v9RW32E1fY...](https://postgr.es/m/CALj2ACXPorUqePswDtOeM\_s82v9RW32E1fY...)
[https://git.postgresql.org/pg/commitdiff/02d332297f9bfe63...](https://git.postgresql.org/pg/commitdiff/02d332297f9bfe63...)
- Get rid of the dedicated latch for signaling the startup process. This commit
gets rid of the dedicated latch for signaling the startup process in favor of
using its procLatch, since that comports better with possible generic signal
handlers using that latch. Commit 1e53fe0e70 changed background processes so
that they use standard SIGHUP handler. Like that, this commit also makes the
startup process use standard SIGHUP handler to simplify the code. Author:
Fujii Masao Reviewed-by: Bharath Rupireddy, Michael Paquier Discussion:
[https://postgr.es/m/CALj2ACXPorUqePswDtOeM\_s82v9RW32E1fY...](https://postgr.es/m/CALj2ACXPorUqePswDtOeM\_s82v9RW32E1fY...)
[https://git.postgresql.org/pg/commitdiff/ac22929a2613e122...](https://git.postgresql.org/pg/commitdiff/ac22929a2613e122...)
- Fix segmentation fault that commit ac22929a26 caused. Commit ac22929a26
changed recoveryWakeupLatch so that it's reset to NULL at the end of recovery.
This change could cause a segmentation fault in the buildfarm member 'elver'.
Previously the latch was reset to NULL after calling ShutdownWalRcv(). But
there could be a window between ShutdownWalRcv() and the actual exit of
walreceiver. If walreceiver set the latch during that window, the segmentation
fault could happen. To fix the issue, this commit changes walreceiver so that
it sets the latch only when the latch has not been reset to NULL yet. Author:
Fujii Masao Discussion:
[https://postgr.es/m/5c1f8a85-747c-7bf9-241e-dd467d8a3586@...](https://postgr.es/m/5c1f8a85-747c-7bf9-241e-dd467d8a3586@...)
[https://git.postgresql.org/pg/commitdiff/113d3591b859fb8d...](https://git.postgresql.org/pg/commitdiff/113d3591b859fb8d...)
- pg\_prewarm: make autoprewarm leader use standard SIGHUP and SIGTERM handlers.
Commit 1e53fe0e70 changed background processes so that they use standard
SIGHUP handler. Like that, this commit makes autoprewarm leader process also
use standard SIGHUP and SIGTERM handlers, to simplify the code. Author:
Bharath Rupireddy Reviewed-by: Kyotaro Horiguchi, Fujii Masao Discussion:
[https://postgr.es/m/CALj2ACXPorUqePswDtOeM\_s82v9RW32E1fY...](https://postgr.es/m/CALj2ACXPorUqePswDtOeM\_s82v9RW32E1fY...)
[https://git.postgresql.org/pg/commitdiff/53f614f1302d56e5...](https://git.postgresql.org/pg/commitdiff/53f614f1302d56e5...)
Peter Geoghegan pushed:
- Fix nbtree cleanup-only VACUUM stats inaccuracies. Logic for counting heap
TIDs from posting list tuples (added by commit 0d861bbb) was faulty. It
didn't count any TIDs/index tuples in the event of no callback being set.
This meant that we incorrectly counted no index tuples in clean-up only
VACUUMs, which could lead to pg\_class.reltuples being spuriously set to 0 in
affected indexes. To fix, go back to counting items from the page in cases
where there is no callback. This approach isn't very accurate, but it works
well enough in practice while avoiding the expense of accessing every index
tuple during cleanup-only VACUUMs. Author: Peter Geoghegan pg [AT] bowt.ie
Reported-By: Jehan-Guillaume de Rorthais jgdr [AT] dalibo.com
[https://postgr.es/m/20201023174451.69e358f1@firost](https://postgr.es/m/20201023174451.69e358f1@firost) Backpatch: 13-, where
nbtree deduplication was introduced
[https://git.postgresql.org/pg/commitdiff/48e1291342dd7771...](https://git.postgresql.org/pg/commitdiff/48e1291342dd7771...)
- Fix wal\_consistency\_checking nbtree bug. wal\_consistency\_checking indicated an
inconsistency in certain cases involving nbtree page deletion. The underlying
issue is that there was a minor difference between the page image produced
after a REDO routine ran and the corresponding page image following original
execution. This harmless inconsistency has been around forever. We more or
less expect total consistency among even deleted nbtree pages these days,
though, so this won't do anymore. To fix, tweak the REDO routine to match
original execution. Oversight in commit f47b5e13.
[https://git.postgresql.org/pg/commitdiff/efc5dcfd8ad4e1df...](https://git.postgresql.org/pg/commitdiff/efc5dcfd8ad4e1df...)
- Improve nbtree README's LP\_DEAD section. The description of how LP\_DEAD bit
setting by index scans works following commit 2ed5b87f was rather unclear.
Clean that up a bit. Also refer to LP\_DEAD bit setting within
\_bt\_check\_unique() at the start of the same section. This mechanism may
actually be more important than the generic kill\_prior\_tuple mechanism that
the section focuses on, so it at least deserves to be mentioned in passing.
[https://git.postgresql.org/pg/commitdiff/5a2f154a2ecaf545...](https://git.postgresql.org/pg/commitdiff/5a2f154a2ecaf545...)
Álvaro Herrera pushed:
- Plug memory leak in index\_get\_partition. The list of indexes was being leaked
when asked for an index that doesn't have an index partition in the table
partition. Not a common case admittedly --and in most cases where it occurs,
caller throws an error anyway-- but worth fixing for cleanliness and in case
any third-party code is calling this function. While at it, remove use of
lfirst\_oid() to obtain a value we already have. Author: Justin Pryzby
pryzby [AT] telsasoft.com Reviewed-by: Michael Paquier michael [AT] paquier.xyz
Discussion:
[https://postgr.es/m/20201105203606.GF22691@telsasoft.com](https://postgr.es/m/20201105203606.GF22691@telsasoft.com)
[https://git.postgresql.org/pg/commitdiff/623644f02cbde7ad...](https://git.postgresql.org/pg/commitdiff/623644f02cbde7ad...)
- Message style improvements. \* Avoid pointlessly highlighting that an index
vacuum was executed by a parallel worker; user doesn't care. \* Don't give
the impression that a non-concurrent reindex of an invalid index on a TOAST
table would work, because it wouldn't. \* Add a "translator:" comment for a
mysterious message. Discussion:
[https://postgr.es/m/20201107034943.GA16596@alvherre.pgsql](https://postgr.es/m/20201107034943.GA16596@alvherre.pgsql) Reviewed-by: Michael
Paquier michael [AT] paquier.xyz
[https://git.postgresql.org/pg/commitdiff/52eec1c53aa6a7df...](https://git.postgresql.org/pg/commitdiff/52eec1c53aa6a7df...)
# Pending Patches
Nikhil Benesch sent in a patch to support negative indexes in the split\_part()
function, those counting from the end of the array instead of the start.
Justin Pryzby sent in another revision of a patch to refactor CIC to rely on
REINDEX CONCURRENTLY, refactor same to allow reindexing all index partitions at
once, and make ReindexPartitions() set indisvalid so things that come by later
can see they're ready to go.
Magnus Hagander sent in another revision of a patch to remove the obsolete
analyze\_new\_cluster.sh script and things that know about it from pg\_upgrade.
Anastasia Lubennikova sent in another revision of a patch to teach COPY FREEZE
to set PD\_ALL\_VISIBLE and visibility map bits.
David G. Johnston sent in another revision of a patch to clarify the fact that
signal functions have no feedback.
Heikki Linnakangas sent in four revisions of a patch to split copy.c into
copyto.c and copyfrom.c, and further split copyfrom.c into copyfrom.c and
copyfromparse.c. This will make working on the usually independent
functionalities of the split files more convenient and easier to read.
Álvaro Herrera sent in another revision of a patch to add batch/pipelining
support to libpq.
Pavel Stěhule sent in a patch to reduce overhead of execution of CALL statement
in no atomic mode from PL/pgSQL.
Magnus Hagander sent in two revisions of a patch to remove the obsolete -O
switch for postgres, which once allowed passing options to each server process.
Kyotaro HORIGUCHI sent in two more revisions of a patch to fix a dereference
before NULL check in src/backend/storage/ipc/latch.c.
David Rowley sent in two revisions of a patch to reduce the number of special
cases to build contrib modules on windows.
Konstantin Knizhnik sent in three more revisions of a patch to add custom
compression to libpq.
Fabien COELHO sent in another revision of a patch to pgbench to add an option
which delays queries until connections are established.
Thomas Munro and David Rowley traded patches to implement collation versioning.
Jinbao Chen sent in a patch to add a new table am 'tid\_visible' to test the
visibility of a tid.
Peter Geoghegan sent in another revision of a patch to add delete deduplication
to nbtree.
Stephen Frost sent in two more revisions of a patch to use pre-fetching for
ANALYZE.
Tomáš Vondra sent in another revision of a patch to use INT64\_FORMAT to print
int64 variables in sort debug.
Bharath Rupireddy sent in another revision of a patch to use multi Inserts in
Create Table As.
Amit Langote sent in another revision of a patch to call BeginDirectModify from
ExecInitModifyTable, and initialize result relation information lazily. This
work builds infrastructure that will later be used to make writes on foreign
tables more efficient.
Vigneshwaran C sent in two more revisions of a patch to improve the connection
authorization message for GSS authenticated/encrypted connections by adding a
log message to include GSS authentication, encryption, and principal
information. This message will help the user to know whether GSS authentication
or encryption was used and which GSS principal was used.
Tomáš Vondra sent in three more revisions of a patch to implement BRIN
multi-range indexes.
Álvaro Herrera sent in another revision of a patch to implement ALTER TABLE ...
DETACH CONCURRENTLY.
Tsutomu Yamada sent in another revision of a patch to add \\dX, which lists
extended statistics, to psql.
Pavel Borisov sent in two more revisions of a patch to deprecate and replace `<^`
and `>^` operators for points.
Melanie Plageman sent in another revision of a patch to support parallel FULL
JOIN and RIGHT JOIN.
Kyotaro HORIGUCHI sent in two more revisions of a patch to use shared memory
instead of files for storage in the stats collector.
Ajin Cherian and Peter Smith traded patches to add logical decoding of two-phase
transactions.
Kirk Jamison sent in another revision of a patch to make
DropRelFileNodeBuffers() more efficient during recovery by avoiding scanning the
whole buffer pool when the relation is small enough, or the the total number of
blocks to be invalidated is below the threshold of full scanning.
Daniel Gustafsson sent in two more revisions of a patch to support NSS as a TLS
backend for libpq.
Takamichi Osumi sent in three more revisions of a patch to implement CREATE OR
REPLACE TRIGGER.
Fujii Masao sent in another revision of a patch to use standard SIGHUP and
SIGTERM handlers in the autoprewarm process.
Justin Pryzby sent in another revision of a patch to Implement CLUSTER of
partitioned table. This requires either specification of a partitioned index on
which to cluster, or that an partitioned index was previously set clustered.
Kyotaro HORIGUCHI sent in another revision of a patch to fix NaN handling in
some geometric operators and functions.
Michaël Paquier sent in another revision of a patch to rework the SHA2 APIs,
switch sha2\_openssl.c to use EVP, and make pgcrypto use the in-core resowner
facility for EVP.
Yuzuko Hosoya sent in another revision of a patch to fix some infelicities
between autovacuum and partitioned tables.
Justin Pryzby sent in another revision of a patch to make pg\_ls\_\* show
directories and shared filesets.
Seino Yuki sent in another revision of a patch to enable pg\_stat\_statements to
track rows processed by REFRESH MATERIALIZED VIEW.
Kyotaro HORIGUCHI sent in two more revisions of a patch to implement CatCache
expiration.
David Pirotte sent in another revision of a patch to add logical decoding
messages to pgoutput.
Masahiko Sawada sent in two more revisions of a patch to implement a transaction
manager for foreign transactions.
Masahiro Ikeda sent in another revision of a patch to add statistics to the
pg\_stat\_wal view.
Justin Pryzby sent in a patch atop the one for incremental view maintenance
patch which fixes some documentation.
Michaël Paquier sent in a patch to refactor the MD5 implementations to be just
one, and switch to EVP for OpenSSL.
Justin Pryzby sent in another revision of a patch to clarify the computation of
min/max IO and specifically the double use and effect of correlation, avoid
re-using the "pages\_fetched" variable, and use the correlation statistic in
costing bitmap scans as for an index scan.
Sergei Kornilov sent in another revision of a patch to allow some recovery
parameters to be changed with reload.
Marina Polyakova sent in two revisions of a patch to fix a bug that manifested
as pgbench no longer supporting a large number of client connections on Windows.
Andrey Borodin sent in another revision of a patch to add Sortsupport for
sorting GiST build for gist\_btree types.
Jürgen Purtz and Erik Rijkers traded patches to add an architecture chapter to
the tutorial.
Dilip Kumar sent in another revision of a patch to implement custom table
compression methods.
Tomáš Vondra sent in a patch to remove some duplicate code from
brin\_memtuple\_initialize.
