[COMMITTERS] pgsql: Unbreak comments on composite type attributes.

2010-10-19 Thread Robert Haas
Unbreak comments on composite type attributes. Report and diagnosis by Peter Eisentraut. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=262c1a42dcd36831a964099659026fb9cec0eecc Modified Files -- src/backend/catalog/objectaddress

[COMMITTERS] pgsql: Add levenshtein_less_equal, optimized version for small distance

2010-10-19 Thread Robert Haas
Add levenshtein_less_equal, optimized version for small distances. Alexander Korotkov, heavily revised by me. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=604ab08145d0dde91279c430e08d570717c12b30 Modified Files -- contrib/fuzz

[COMMITTERS] pgsql: Add mention of using tools/fsync to test fsync methods. Restruct

2010-10-19 Thread Bruce Momjian
Add mention of using tools/fsync to test fsync methods. Restructure recent wal_sync_method doc paragraph to be clearer. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=f75d6a1b19ec334a9b1455c4b8030a09d582581d Modified Files -- doc

[COMMITTERS] pgsql: Add mention of using tools/fsync to test fsync methods. Restruct

2010-10-19 Thread Bruce Momjian
Add mention of using tools/fsync to test fsync methods. Restructure recent wal_sync_method doc paragraph to be clearer. Branch -- REL9_0_STABLE Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=6f609199df2c8d38ed0391180a2ff467333ae57c Modified Files ---

[COMMITTERS] pgsql: Add mention of using tools/fsync to test fsync methods. Restruct

2010-10-19 Thread Bruce Momjian
Add mention of using tools/fsync to test fsync methods. Restructure recent wal_sync_method doc paragraph to be clearer. Branch -- REL8_3_STABLE Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=8a691873252df085544312e0e0ef2d842c457e88 Modified Files ---

Re: [COMMITTERS] pgsql: Add mention of using tools/fsync to test fsync methods. Restruct

2010-10-19 Thread Tom Lane
Bruce Momjian writes: > Add mention of using tools/fsync to test fsync methods. Restructure > recent wal_sync_method doc paragraph to be clearer. I see pushes of this in master, 9.0, 8.3 ... forget something? regards, tom lane -- Sent via pgsql-committers mailing list (

[COMMITTERS] pgsql: Add mention of using tools/fsync to test fsync methods. Restruct

2010-10-19 Thread Bruce Momjian
Add mention of using tools/fsync to test fsync methods. Restructure recent wal_sync_method doc paragraph to be clearer. Branch -- REL8_4_STABLE Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=d94d0a8c83715cb403267f0c89068781b706ddff Modified Files ---

Re: [COMMITTERS] pgsql: Add mention of using tools/fsync to test fsync methods. Restruct

2010-10-19 Thread Bruce Momjian
Tom Lane wrote: > Bruce Momjian writes: > > Add mention of using tools/fsync to test fsync methods. Restructure > > recent wal_sync_method doc paragraph to be clearer. > > I see pushes of this in master, 9.0, 8.3 ... forget something? Yes, thanks, applied to 8.4. This was my first pre-9.0 git c

[COMMITTERS] pgsql: Add removal of PG_VERSION to optional old cluster deletion scrip

2010-10-19 Thread Bruce Momjian
Add removal of PG_VERSION to optional old cluster deletion script. Backpatch to 9.0.X. Branch -- REL9_0_STABLE Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=80a65c05777ef2b05dfe3a276568cad438fb804d Modified Files -- contrib/pg_upgrade/check.c

[COMMITTERS] pgsql: Add removal of PG_VERSION to optional old cluster deletion scrip

2010-10-19 Thread Bruce Momjian
Add removal of PG_VERSION to optional old cluster deletion script. Backpatch to 9.0.X. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=51eeccd6cc2e0fcb95112e0e4a7a20d0f95c181a Modified Files -- contrib/pg_upgrade/check.c |

[COMMITTERS] pgsql: In pg_upgrade, rename SHELL_EXT to SCRIPT_EXT, for clarity.

2010-10-19 Thread Bruce Momjian
In pg_upgrade, rename SHELL_EXT to SCRIPT_EXT, for clarity. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=722d5beeb266ae83f548fc3953df700a71f30134 Modified Files -- contrib/pg_upgrade/check.c |2 +- contrib/pg_upgrade/pg

[COMMITTERS] pgsql: Fix incorrect generation of whole-row variables in planner.

2010-10-19 Thread Tom Lane
Fix incorrect generation of whole-row variables in planner. A couple of places in the planner need to generate whole-row Vars, and were cutting corners by setting vartype = RECORDOID in the Vars, even in cases where there's an identifiable named composite type for the RTE being referenced. While

[COMMITTERS] pgsql: Fix incorrect generation of whole-row variables in planner.

2010-10-19 Thread Tom Lane
Fix incorrect generation of whole-row variables in planner. A couple of places in the planner need to generate whole-row Vars, and were cutting corners by setting vartype = RECORDOID in the Vars, even in cases where there's an identifiable named composite type for the RTE being referenced. While

[COMMITTERS] pgsql: Restructure the pg_upgrade code to use several global structures

2010-10-19 Thread Bruce Momjian
Restructure the pg_upgrade code to use several global structures rather than packing everything into 'ctx' and passing that to every function. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=e13f7e9a711c9117e438516537066bbe61df6455 Modified F

[COMMITTERS] pgsql: Pgindent run on pg_upgrade source after restructuring.

2010-10-19 Thread Bruce Momjian
Pgindent run on pg_upgrade source after restructuring. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=07456b45e6359e8a15807e3f71c0f2e6a76fa4e9 Modified Files -- contrib/pg_upgrade/check.c | 28 ++--

[COMMITTERS] pgsql: In pg_upgrade, use cached copy of directory listing, rather than

2010-10-19 Thread Bruce Momjian
In pg_upgrade, use cached copy of directory listing, rather than calling scandir() with a pattern for every table. Optimization after report of pg_upgrade slowness with 150k tables. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=ad06db2610f4

[COMMITTERS] pgsql: Cache most recent relfilenode lookups, for speed, after report o

2010-10-19 Thread Bruce Momjian
Cache most recent relfilenode lookups, for speed, after report of pg_upgrade slowness for 150k tables. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=333b4a45f1d984bd7f32aecb1ba551feed088ec4 Modified Files -- contrib/pg_upgrade/i

[COMMITTERS] pgsql: Fix for new pg_upgrade cache code.

2010-10-19 Thread Bruce Momjian
Fix for new pg_upgrade cache code. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=1c72614be710127b198411f49301a56453e3abc4 Modified Files -- contrib/pg_upgrade/info.c |3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)

[COMMITTERS] pgsql: Update storage.sgml to describe the 9.0 tablespace directory lay

2010-10-19 Thread Tom Lane
Update storage.sgml to describe the 9.0 tablespace directory layout. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=c33bfb8b9b0d955b813d99e8f9f5c62c31f7387c Modified Files -- doc/src/sgml/storage.sgml | 22 -

[COMMITTERS] pgsql: Update storage.sgml to describe the 9.0 tablespace directory lay

2010-10-19 Thread Tom Lane
Update storage.sgml to describe the 9.0 tablespace directory layout. Branch -- REL9_0_STABLE Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=486187ae847b203d0c9e0d1c1b482104c659e557 Modified Files -- doc/src/sgml/storage.sgml | 22 ++

[COMMITTERS] pgsql: Further speed up pg_upgrade lookups.

2010-10-19 Thread Bruce Momjian
Further speed up pg_upgrade lookups. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=3325c9bddb332487914c5c96d838286350044814 Modified Files -- contrib/pg_upgrade/info.c | 39 --- contrib

[COMMITTERS] pgsql: Rename pg_upgrade 'log' to 'log_opts', to avoid platform naming

2010-10-19 Thread Bruce Momjian
Rename pg_upgrade 'log' to 'log_opts', to avoid platform naming conflict. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=691a67b922f2e8374356215228700958165e7d44 Modified Files -- contrib/pg_upgrade/check.c |2 +- contri

[COMMITTERS] pgsql: Add some caveats to the contrib/isn docs.

2010-10-19 Thread Robert Haas
Add some caveats to the contrib/isn docs. Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=4b6623a7e8f0f0116ff68ebfd4c98830fe59419d Modified Files -- doc/src/sgml/isn.sgml | 11 +-- 1 files changed, 9 insertions(+), 2 dele

[COMMITTERS] pgsql: Updates to contrib/isn ISBN tables.

2010-10-19 Thread Robert Haas
Updates to contrib/isn ISBN tables. Jan Otto, reviewed by Peter Geoghegan Branch -- master Details --- http://git.postgresql.org/gitweb?p=postgresql.git;a=commitdiff;h=6d1af7b2180719102a907bd3e35d218b43e76ad1 Modified Files -- contrib/isn/ISBN.h | 375 ++

[COMMITTERS] pgsql: Fix ecpg test building process to not generate *.dSYM junk on Ma

2010-10-19 Thread Tom Lane
Fix ecpg test building process to not generate *.dSYM junk on Macs. The trick is to not try to build executables directly from .c files, but to always build the intermediate .o files. For obscure reasons, Darwin's version of gcc will leave debug cruft behind in the first case but not the second.

[COMMITTERS] pgsql: Fix ecpg test building process to not generate *.dSYM junk on Ma

2010-10-19 Thread Tom Lane
Fix ecpg test building process to not generate *.dSYM junk on Macs. The trick is to not try to build executables directly from .c files, but to always build the intermediate .o files. For obscure reasons, Darwin's version of gcc will leave debug cruft behind in the first case but not the second.

[COMMITTERS] pgsql: Fix ecpg test building process to not generate *.dSYM junk on Ma

2010-10-19 Thread Tom Lane
Fix ecpg test building process to not generate *.dSYM junk on Macs. The trick is to not try to build executables directly from .c files, but to always build the intermediate .o files. For obscure reasons, Darwin's version of gcc will leave debug cruft behind in the first case but not the second.

[COMMITTERS] pgsql: Fix ecpg test building process to not generate *.dSYM junk on Ma

2010-10-19 Thread Tom Lane
Fix ecpg test building process to not generate *.dSYM junk on Macs. The trick is to not try to build executables directly from .c files, but to always build the intermediate .o files. For obscure reasons, Darwin's version of gcc will leave debug cruft behind in the first case but not the second.

[COMMITTERS] pgsql: Fix ecpg test building process to not generate *.dSYM junk on Ma

2010-10-19 Thread Tom Lane
Fix ecpg test building process to not generate *.dSYM junk on Macs. The trick is to not try to build executables directly from .c files, but to always build the intermediate .o files. For obscure reasons, Darwin's version of gcc will leave debug cruft behind in the first case but not the second.