Skip to content

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also compare across forks.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also compare across forks.
Commits on Sep 23, 2013
@hlinnaka hlinnaka Fix two timeline handling bugs in pg_receivexlog.
When a timeline history file is fetched from server, it is initially created
with a temporary file name, and renamed to place. However, the temporary
file name was constructed using an uninitialized buffer. Usually that meant
that the file was created in current directory instead of the target, which
usually goes unnoticed, but if the target is on a different filesystem than
the current dir, the rename() would fail. Fix that.

The second issue is that pg_receivexlog would not take .partial files into
account when determining when scanning the target directory for existing
WAL files. If the timeline has switched in the server several times in the
last WAL segment, and pg_receivexlog is restarted, it would choose a too
old starting point. That's not a problem as long as the old WAL segment
exists in the server and can be streamed over, but will cause a failure if
it's not.

Backpatch to 9.3, where this timeline handling code was written.

Analysed by Andrew Gierth, bug #8453, based on a bug report on IRC.
@sfrost sfrost Fix SSL deadlock risk in libpq
In libpq, we set up and pass to OpenSSL callback routines to handle
locking.  When we run out of SSL connections, we try to clean things
up by de-registering the hooks.  Unfortunately, we had a few calls
into the OpenSSL library after these hooks were de-registered during
SSL cleanup which lead to deadlocking.  This moves the thread callback
cleanup to be after all SSL-cleanup related OpenSSL library calls.
I've been unable to reproduce the deadlock with this fix.

In passing, also move the close_SSL call to be after unlocking our
ssl_config mutex when in a failure state.  While it looks pretty
unlikely to be an issue, it could have resulted in deadlocks if we
ended up in this code path due to something other than SSL_new
failing.  Thanks to Heikki for pointing this out.

Back-patch to all supported versions; note that the close_SSL issue
only goes back to 9.0, so that hunk isn't included in the 8.4 patch.

Initially found and reported by Vesa-Matti J Kari; many thanks to
both Heikki and Andres for their help running down the specific
issue and reviewing the patch.
@bmomjian bmomjian pg_upgrade: fix C comment typo f7cf5fa
@bmomjian bmomjian pg_upgrade: more C comment fixes ff2a1f5
Robert Haas Don't allow system columns in CHECK constraints, except tableoid.
Previously, arbitray system columns could be mentioned in table
constraints, but they were not correctly checked at runtime, because
the values weren't actually set correctly in the tuple.  Since it
seems easy enough to initialize the table OID properly, do that,
and continue allowing that column, but disallow the rest unless and
until someone figures out a way to make them work properly.

No back-patch, because this doesn't seem important enough to take the
risk of destabilizing the back branches.  In fact, this will pose a
dump-and-reload hazard for those upgrading from previous versions:
constraints that were accepted before but were not correctly enforced
will now either be enforced correctly or not accepted at all.  Either
could result in restore failures, but in practice I think very few
users will notice the difference, since the use case is pretty
marginal anyway and few users will be relying on features that have
not historically worked.

Amit Kapila, reviewed by Rushabh Lathia, with doc changes by me.
Robert Haas doc: Clarify that file_fdw options require values.
Mike Blackwell and Robert Haas
@nmisch nmisch pgbench: Tweak documentation.
@nmisch nmisch Use @libdir@ in both of regress/{input,output}/security_label.source
Though @libdir@ almost always matches @abs_builddir@ in this context,
the test could only fail if they differed.  Back-patch to 9.1, where the
test was introduced.

Hamid Quddus Akhtar
Commits on Sep 24, 2013
@alvherre alvherre Fix pgindent comment breakage b2fc4d6
Commits on Sep 25, 2013
@hlinnaka hlinnaka Plug memory leak in range_cmp function.
B-tree operators are not allowed to leak memory into the current memory
context. Range_cmp leaked detoasted copies of the arguments. That caused
a quick out-of-memory error when creating an index on a range column.

Reported by Marian Krucina, bug #8468.
Commits on Sep 26, 2013
@nmisch nmisch pgbench: Correct for bias in --rate schedule generation.
Previous code gave a mean delay 0.44% below target.  This change also
has the effect of increasing the maximum possible delay.

@hlinnaka hlinnaka Fix spurious warning after vacuuming a page on a table with no indexes.
There is a rare race condition, when a transaction that inserted a tuple
aborts while vacuum is processing the page containing the inserted tuple.
Vacuum prunes the page first, which normally removes any dead tuples, but
if the inserting transaction aborts right after that, the loop after
pruning will see a dead tuple and remove it instead. That's OK, but if the
page is on a table with no indexes, and the page becomes completely empty
after removing the dead tuple (or tuples) on it, it will be immediately
marked as all-visible. That's OK, but the sanity check in vacuum would
throw a warning because it thinks that the page contains dead tuples and
was nevertheless marked as all-visible, even though it just vacuumed away
the dead tuples and so it doesn't actually contain any.

Spotted this while reading the code. It's difficult to hit the race
condition otherwise, but can be done by putting a breakpoint after the
heap_page_prune() call.

Backpatch all the way to 8.4, where this code first appeared.
@adunstan adunstan Fix erroneous statements about multiply specified JSON columns.
The behaviour in json_populate_record() and json_populate_recordset()
was changed during development but the docs were not.
Robert Haas Allow printf-style padding specifications in log_line_prefix.
David Rowley, after a suggestion from Heikki Linnakangas.  Reviewed by
Albe Laurenz, and further edited by me.
Commits on Sep 29, 2013
@adunstan adunstan Ensure installation dirs are built before contents are installed.
Cédric Villemain
@adunstan adunstan Use a new hstore extension version for added json functions.
This should have been done when the json functionality was added to
hstore in 9.3.0. To handle this correctly, the upgrade script therefore
uses conditional logic by using plpgsql in a DO statement to add the two
new functions and the new cast. If hstore_to_json_loose is detected as
already present and dependent on the hstore extension nothing is done.
This will require that the database be loaded with plpgsql.

People who have installed the earlier and spurious 1.1 version of hstore
will need to do:


to pick up the new functions properly.
Commits on Sep 30, 2013
@adunstan adunstan Fix makefile broken by hstore fix. 42bf7fc
Fujii Masao Correct comment of pgbench "filler" columns.
Pavan Deolasee
@hlinnaka hlinnaka Fix snapshot leak if lo_open called on non-existent object.
lo_open registers the currently active snapshot, and checks if the
large object exists after that. Normally, snapshots registered by lo_open
are unregistered at end of transaction when the lo descriptor is closed, but
if we error out before the lo descriptor is added to the list of open
descriptors, it is leaked. Fix by moving the snapshot registration to after
checking if the large object exists.

Reported by Pavel Stehule. Backpatch to 8.4. The snapshot registration
system was introduced in 8.4, so prior versions are not affected (and not
supported, anyway).
@hlinnaka hlinnaka In bms_add_member(), use repalloc() if the bms needs to be enlarged.
Previously bms_add_member() would palloc a whole-new copy of the existing
set, copy the words, and pfree the old one. repalloc() is potentially much
faster, and more importantly, this is less surprising if CurrentMemoryContext
is not the same as the context the old set is in. bms_add_member() still
allocates a new bitmapset in CurrentMemoryContext if NULL is passed as
argument, but that is a lot less likely to induce bugs.

Nicholas White.
@adunstan adunstan Ensure installation dirs are built before contents are installed (v2)
Push dependency on installdirs down to individual targets.

Christoph Berg
@adunstan adunstan Add missing condition for pg_depend in hstore migration script.
Error noted by Andres Freund.
Commits on Oct 01, 2013
@alvherre alvherre Remove broken PGXS code for pg_xlogdump
With the PGXS boilerplate in place, pg_xlogdump currently fails with an
ominous error message that certain targets cannot be built because
certain files do not exist.  Remove that and instead throw a quick error
message alerting the user of the actual problem, which should be easier
to diagnose that the statu quo.

Andres Freund
@alvherre alvherre Add WaitForLockers in lmgr, refactoring index.c code
This is in support of a future REINDEX CONCURRENTLY feature.

Michael Paquier
@bmomjian bmomjian Adjust C comments that would be wrap-able. d50f281
Commits on Oct 02, 2013
@mhagander mhagander Fix copy/paste error 542131c
@bmomjian bmomjian doc: fix hstore_to_json_loose() doc wording eb5d87e
Commits on Oct 03, 2013
@petere petere psql: Set up cancel handler later
The cancel handler was uselessly set up even before the first connection
was opened.  By setting it up afterwards, the user can use Ctrl+C to
abort psql if the initial connection attempt hangs.

Reviewed-by: Dean Rasheed <>
Reviewed-by: Ryan Kelly <>
@petere petere doc: Correct psycopg URL 627882d
@hlinnaka hlinnaka Minor GIN code refactoring.
It makes for cleaner code to have separate Get/Add functions for PostingItems
and ItemPointers.  A few callsites that have to deal with both types need to
be duplicated because of this, but all the callers have to know which one
they're dealing with anyway. Overall, this reduces the amount of casting

Extracted from Alexander Korotkov's larger patch to change the data page
Robert Haas psql: Make \pset without arguments show all settings.
Gilles Darold, reviewed by Pavel Stehule
Robert Haas Add DISCARD SEQUENCES command.
DISCARD ALL will now discard cached sequence information, as well.

Fabrízio de Royes Mello, reviewed by Zoltán Böszörményi, with some
further tweaks by me.
Commits on Oct 04, 2013
Robert Haas Fix silly thinko in ResetSequenceCaches.
Report from Kevin Hale Boyes.
Robert Haas doc: Add missing words to bgworker docs.
Maciek Sakrejda
@bmomjian bmomjian Issue error on SET outside transaction block in some cases
Issue error for SET LOCAL/CONSTRAINTS/TRANSACTION outside a transaction
block, as they have no effect.

Per suggestion from Morten Hustveit
@alvherre alvherre isolationtester: Allow tuples to be returned in more places
Previously, isolationtester would forbid returning tuples in
session-specific teardown (but not global teardown), as well as in
global setup.  Allow these places to return tuples, too.
@alvherre alvherre Make some isolationtester specs more complete
Also, make sure they pass on all transaction isolation levels.
@alvherre alvherre add multixact-no-deadlock to schedule 1310d4c
Commits on Oct 05, 2013
@bmomjian bmomjian pg_upgrade doc: link mode additions
Mention that link mode uses less disk space, and uses junction points on

Backpatch to 9.3.
@nmisch nmisch pgbench: Remove stray use of "float" math.
Oversight in commit 4a87f30.

@nmisch nmisch pgbench: Elaborate latency reporting.
Isolate transaction latency (elapsed time between submitting first
command and receiving response to last command) from client-side delays
pertaining to the --rate schedule.  Under --rate, report schedule lag as
defined in the documentation.  Report latency standard deviation
whenever we collect the measurements to do so.  All of these changes
affect --progress messages and the final report.

Fabien COELHO, reviewed by Pavel Stehule.
Commits on Oct 06, 2013
@alvherre alvherre Fix various bugs in postmaster SIGKILL processing
Clamp the minimum sleep time during immediate shutdown or crash to a
minimum of zero, not a maximum of one second.  The previous code could
result in a negative sleep time, leading to failure in select() calls.

Also, on crash recovery, reset AbortStartTime as soon as SIGKILL is sent
or abort processing has commenced instead of waiting until the startup
process completes.  Reset AbortStartTime as soon as SIGKILL is sent,
too, to avoid doing that repeatedly.

Per trouble report from Jeff Janes on

Author: MauMau
@nmisch nmisch pgbench: Comment on thread timing hazards.
Reviewed by Fabien COELHO.
Commits on Oct 07, 2013
@adunstan adunstan Document support for VPATH builds of extensions.
Cédric Villemain and me.
@kgrittn kgrittn Eliminate xmin from hash tag for predicate locks on heap tuples.
If a tuple was frozen while its predicate locks mattered,
read-write dependencies could be missed, resulting in failure to
detect conflicts which could lead to anomalies in committed
serializable transactions.

This field was added to the tag when we still thought that it was
necessary to carry locks forward to a new version of an updated
row.  That was later proven to be unnecessary, which allowed
simplification of the code, but elimination of xmin from the tag
was missed at the time.

Per report and analysis by Heikki Linnakangas.
Backpatch to 9.1.
Robert Haas plpgsql: Add new option print_strict_params.
This option provides more detailed error messages when STRICT is used
and the number of rows returned is not one.

Marko Tiikkaja, reviewed by Ian Lawrence Barwick
Robert Haas Make DISCARD SEQUENCES also discard the last used sequence.
Otherwise, we access already-freed memory.  Oops.

Report by Michael Paquier.  Fix by me.
@petere petere Translation updates 0b109c8
@hlinnaka hlinnaka Fix bugs in SSI tuple locking.
1. In heap_hot_search_buffer(), the PredicateLockTuple() call is passed
wrong offset number. heapTuple->t_self is set to the tid of the first
tuple in the chain that's visited, not the one actually being read.

2. CheckForSerializableConflictIn() uses the tuple's t_ctid field
instead of t_self to check for exiting predicate locks on the tuple. If
the tuple was updated, but the updater rolled back, t_ctid points to the
aborted dead tuple.

Reported by Hannu Krosing. Backpatch to 9.1.
@hlinnaka hlinnaka TYPEALIGN doesn't work on int64 on 32-bit platforms.
The TYPEALIGN macro, and the related ones like MAXALIGN, don't work with
values larger than intptr_t, because TYPEALIGN casts the argument to
intptr_t to do the arithmetic. That's not a problem when dealing with
pointers or lengths or offsets related to pointers, but the XLogInsert
scaling patch added a call to MAXALIGN with an XLogRecPtr argument.

To fix, add wider variants of the macros, called TYPEALIGN64 and MAXALIGN64,
which are just like the existing variants but work with uint64 instead of

Report and patch by David Rowley, analysis by Andres Freund.
Commits on Oct 08, 2013
@bmomjian bmomjian docs: update release notes for 8.4.18, 9.0.14, 9.1.10, 9.2.5, 9.3.1 1732ce4
@bmomjian bmomjian Update instructions on creating minor release notes. a288324
@bmomjian bmomjian Additional instructions on minor release note creation. d29a031
@bmomjian bmomjian Auto-tune effective_cache size to be 4x shared buffers ee1e566
@bmomjian bmomjian unaccent: mark unaccent() functions as immutable
Suggestion from Pavel Stehule
@bmomjian bmomjian docs: clarify references to md5 hash and md5 crypt in pgcrypto docs
Suggestion from Richard Neill
@bmomjian bmomjian Update postgres.conf.sample for effective_cache_size's new default 6648775
@kgrittn kgrittn Allow drop-index-concurrently-1 test to run at any isolation level.
It previously reported failure at REPEATABLE READ and SERIALIZABLE
transaction isolation levels for make installcheck.
@bmomjian bmomjian doc: Fix typo in effective_cache_size patch 3e5a9ee
@bmomjian bmomjian Fix C comment in check_effective_cache_size() bf46524
Commits on Oct 09, 2013
@bmomjian bmomjian Move new effective_cache_size function
Previously set_default_effective_cache_size() could not handle fork,
non-fork, and bootstrap cases.
@bmomjian bmomjian Again move function where we set effective_cache_size's default 6b82f78
@bmomjian bmomjian Adjust the effective_cache_size default for standalone backends 96dfa6e
@bmomjian bmomjian Centralize effective_cache_size default setting 0c6b675
@bmomjian bmomjian doc: fix typo in release notes
Backpatch through 8.4

Per suggestion by Amit Langote
@kgrittn kgrittn Add record_image_ops opclass for matview concurrent refresh.
containing a column of a type without a default btree operator
class.  It also did not produce results consistent with a non-
concurrent REFRESH or a normal view if any column was of a type
which allowed user-visible differences between values which
compared as equal according to the type's default btree opclass.
Concurrent matview refresh was modified to use the new operators
to solve these problems.

Documentation was added for record comparison, both for the
default btree operator class for record, and the newly added
operators.  Regression tests now check for proper behavior both
for a matview with a box column and a matview containing a citext

Reviewed by Steve Singer, who suggested some of the doc language.
Commits on Oct 10, 2013
Robert Haas Allow dynamic allocation of shared memory segments.
Patch by myself and Amit Kapila.  Design help from Noah Misch.  Review
by Andres Freund.
@petere petere Revive line type
Change the input/output format to {A,B,C}, to match the internal

Complete the implementations of line_in, line_out, line_recv, line_send.
Remove comments and error messages about the line type not being
implemented.  Add regression tests for existing line operators and

Reviewed-by: rui hua <>
Reviewed-by: Álvaro Herrera <>
Reviewed-by: Jeevan Chalke <>
@petere petere pg_upgrade: Split off pg_fatal() from pg_log()
This allows decorating pg_fatal() with noreturn compiler hints, leading
to better diagnostics.

Reviewed-by: Marko Tiikkaja <>
Robert Haas Fix incorrect use of shm_unlink where unlink should be used.
Per buildfarm.
@adunstan adunstan json_typeof function.
Andrew Tipton.
@kgrittn kgrittn Fix bug in record_image_ops on big endian machines.
The buildfarm pointed out the problem.

Fix based on suggestion by Robert Haas.
Robert Haas initdb: Select working dynamic shared memory implementation.
If POSIX shared memory is supported and works, we prefer it.
Otherwise, we prefer System V, except on Windows, where we use
the implementation specific to that platform.
@petere petere Update regression tests for line type patch
Erroneously omitted in 261c7d4
Commits on Oct 11, 2013
@petere petere Replace duplicate_oids with Perl implementation
It is more portable, more robust, and more readable.

From: Andrew Dunstan <>
@petere petere Remove maintainer-check target, fold into normal build
make maintainer-check was obscure and rarely called in practice, and
many breakages were missed.  Fold everything that make maintainer-check
used to do into the normal build.  Specifically:

- Call duplicate_oids when is called.

- Check for tabs in SGML files when the documentation is built.

- Run msgfmt with the -c option during the regular build.  Add an
  additional configure check to see whether we are using the GNU
  version.  (make maintainer-check probably used to fail with non-GNU

Keep maintainer-check as around as phony target for the time being in
case anyone is calling it.  But it won't do anything anymore.
@petere petere doc: Handle additional character entities for SGML/XML conversion 92c2d2b
@petere petere doc: Fix table column number declaration 8238d1e
@petere petere doc: Move check-tabs target into html target
The previous plan of having the check-tabs target a prerequisite of
"all" and "distprep" caused make distcheck to fail because make -q
distprep would never be satisfied.  Put check-tabs into the html target
instead, so it is only called when a build actually happens.
@alvherre alvherre Rework SSL renegotiation code
The existing renegotiation code was home for several bugs: it might
erroneously report that renegotiation had failed; it might try to
execute another renegotiation while the previous one was pending; it
failed to terminate the connection if the renegotiation never actually
took place; if a renegotiation was started, the byte count was reset,
even if the renegotiation wasn't completed (this isn't good from a
security perspective because it means continuing to use a session that
should be considered compromised due to volume of data transferred.)

The new code is structured to avoid these pitfalls: renegotiation is
started a little earlier than the limit has expired; the handshake
sequence is retried until it has actually returned successfully, and no
more than that, but if it fails too many times, the connection is
closed.  The byte count is reset only when the renegotiation has
succeeded, and if the renegotiation byte count limit expires, the
connection is terminated.

This commit only touches the master branch, because some of the changes
are controversial.  If everything goes well, a back-patch might be

Per discussion started by message
@alvherre alvherre Use $(PERL) to invoke duplicate_oids
Per buildfarm failure reported by smilodon
@kgrittn kgrittn Fix several possibly non-portable gaffs in record_image_ops.
Sparc machines in the buildfarm were made happy by the previous
fix, but PowerPC machines still are still failing.  Hopefully this
will cure that.
Commits on Oct 13, 2013
@petere petere Tweak "line" test to avoid platform-specific floating-point output a53dee4
@petere petere Add use of asprintf()
Add asprintf(), pg_asprintf(), and psprintf() to simplify string
allocation and composition.  Replacement implementations taken from

Reviewed-by: Álvaro Herrera <>
Reviewed-by: Asif Naeem <>
Commits on Oct 14, 2013
@petere petere Attempt to fix MSVC build for asprintf addition d3d3975
@petere petere Add asprintf.c
Forgotten in 5b6d08c
@petere petere Translation updates to fix build failures
Now that msgfmt is run with -c by default, older versions of gettext are
complaining about the PO headers Last-Translator and Language-Team
still having their default values.  Newer gettext versions fail to catch
this because of a bug (, which is
why this hasn't been noticed before.

Copy updated versions of affected translation files from the
pgtranslations repository, were those files have been fixed.
Robert Haas Fix details missed by dynamic shared memory patch.
Additional documentation update, and a comment fix.

Both issues reported by Amit Kapila.
Robert Haas In dsm_impl_windows, don't error out when the segment already exists.
This is the behavior of the other implementations, and the behavior
expected by the callers of this function.

Amit Kapila
Commits on Oct 15, 2013
@bmomjian bmomjian docs: correct 9.1 and 9.2 release note mention of timeline switch fix
Backpatch through 9.1.

KONDO Mitsumasa
Commits on Oct 16, 2013
@petere petere doc: Enable book index in XSLT builds
The XSLT toolchain requires an empty <index> element where the index is
supposed to appear.  Add that with conditionals to hide it from the
DSSSL build.
@petere petere Switch order of -lpgport and -lpgcommon
Conceptually, libpgcommon can depend on libpgport, but not the other way
around.  In the past, this might not have mattered, but it's needed now
for asprintf.
Robert Haas initdb: Suppress dynamic shared memory when probing for max_connections.
This might not be the right long-term solution here, but it will
hopefully turn the buildfarm green again.

Oversight noted by Andres Freund
@bmomjian bmomjian Allow 5+ digit years for non-ISO timestamp/date strings, where approp…

Report from Haribabu Kommi
Commits on Oct 17, 2013
@alvherre alvherre Silence compiler warning when SSL not in use
Per Jaime Casanova and Vik Fearing
@adunstan adunstan Return valid json when converting an empty hstore.
Oskari Saarenmaa.
Robert Haas Remove spinlock support for SINIX, Sun3, and NS32K.
All of these platforms are very much obsolete.

As far as I can determine, the last version of SINIX, later renamed
Reliant, occurred some time between 2002 and 2005.

The last release of SunOS that would run on a sun3 was released in
November of 1991; the last release of OpenBSD which supported that
platform was in 2001.  The highest clock speed of any processor in
the family was 25MHz.

The NS32K (national semiconductor 320xx) architecture was retired
in 1990.

Support can be re-added if a maintainer emerges for any of these
platforms, but it seems unlikely.

Reviewed by Andres Freund.
Commits on Oct 18, 2013
@petere petere doc: Configure TOC generation in XSLT HTML build
The default table of contents in the XSLT HTML build is much too big and
deep.  Configure it to look more like the one that is currently being
produced by the DSSSL build.
@petere petere Switch dependency order of libpgcommon and libpgport
Continuing 63f32f3, libpgcommon should
depend on libpgport, but not vice versa.  But wait_result_to_str() in
wait_error.c depends on pstrdup() in libpgcommon.  So move exec.c and
wait_error.c from libpgport to libpgcommon.  Also switch the link order
in the place that's actually used by the failing ecpg builds.

The function declarations have been left in port.h for now.  That should
perhaps be separated sometime.
Robert Haas Remove IRIX port.
Development of IRIX has been discontinued, and support is scheduled
to end in December of 2013.  Therefore, there will be no supported
versions of this operating system by the time PostgreSQL 9.4 is
released.  Furthermore, we have no maintainer for this platform.
@petere petere Switch order of libpgport and libpgcommon in MSVC build as well 3c093b4
@petere petere Add libpgport to isolationtester on MSVC
From: Asif Naeem <>
@petere petere Fix for lack of va_copy() on certain Windows versions
Based-on-patch-by: David Rowley <>
Robert Haas Provide a reliable mechanism for terminating a background worker.
Although previously-introduced APIs allow the process that registers a
background worker to obtain the worker's PID, there's no way to prevent
a worker that is not currently running from being restarted.  This
patch introduces a new API TerminateBackgroundWorker() that prevents
the background worker from being restarted, terminates it if it is
currently running, and causes it to be unregistered if or when it is
not running.

Patch by me.  Review by Michael Paquier and KaiGai Kohei.
Robert Haas Allow only some columns of a view to be auto-updateable.
Previously, unless all columns were auto-updateable, we wouldn't
inserts, updates, or deletes, or at least not without a rule or trigger;
now, we'll allow inserts and updates that target only the auto-updateable
columns, and deletes even if there are no auto-updateable columns at
all provided the view definition is otherwise suitable.

Dean Rasheed, reviewed by Marko Tiikkaja
Commits on Oct 19, 2013
@petere petere Move pgfnames() from libpgport to libpgcommon
It requires pstrdup() from libpgcommon.
@petere petere Move rmtree() from libpgport to libpgcommon
It requires pgfnames() from libpgcommon.
@petere petere Add *.pot to .gitignore 07bf703
@petere petere Add libpgcommon to backend gettext source files
This ought to have been done when libpgcommon was split off from
Commits on Oct 21, 2013
@nmisch nmisch Consistently use unsigned arithmetic for alignment calculations.
This avoids an assumption about the signed number representation.  It is
anticipated to have no functional changes on supported configurations;
many two's complement assumptions remain elsewhere.

Per a suggestion from Andres Freund.
@petere petere doc: Improve setup for documentation building with FOP
Add a makefile rule for building PDFs with FOP.  Two new build targets
in doc/src/sgml are postgres-A4-fop.pdf and postgres-US-fop.pdf.

Run .fo output through xmllint for reformatting, so that errors are
easier to find.  (The default output has hardly any line breaks, so you
might be looking for an error in column 20000.)

Set some XSLT parameters to optimize for building with FOP.

Remove some redundant or somewhat useless chapterinfo/author
information, because it renders strangely with the FO stylesheet.

Reviewed-by: Álvaro Herrera <>
@hlinnaka hlinnaka Extend cube on-disk format to pack points more tightly.
If the lower left and upper right corners of a cube are the same, set a
flag in the cube header, and only store one copy of the coordinates. That
cuts the on-disk size into half for the common case that the cube datatype
is used to represent points rather than boxes.

The new format is backwards-compatible with the old one, so pg_upgrade
still works. However, to get the space savings, the data needs to be
rewritten. A simple VACUUM FULL or REINDEX is not enough, as the old
Datums will just be moved to the new heap/index as is. A pg_dump and
reload, or something similar like casting to text and back, will do the

This patch deliberately doesn't update all the alternative expected output
files, as I don't have access to machines that produce those outputs. I'm
not sure if they are still relevant, but if they are, the buildfarm will
tell us and produce the diff required to fix it. If none of the buildfarm
animals need them, they should be removed altogether.

Patch by Stas Kelvich.
Commits on Oct 22, 2013
@tglsfdc tglsfdc Fix blatantly broken record_image_cmp() logic for pass-by-value fields.
Doesn't anybody here pay attention to compiler warnings?
@hlinnaka hlinnaka Adjust cube.out expected output for new test queries.
Previous commit modified the test case, but I didn't update cube.out
expected output file in previous commit because it was not needed by the
platforms I have easy access to. Buildfarm animal 'dugong', running
"Debian 4.0 icc 10.1.011 ia64", has now gone red because of that, so update
it now.

Also adjust cube_3.out. According to git history, it was added to support
64-bit MinGW. There is no such animal in the buildfarm, so I'm doing this
blindly, but it was added quite recently so maybe someone still cares.
@petere petere Make use of psprintf() in recent changes 586a8fc