|
| 1 | +2001-08-16 11:21 momjian |
| 2 | + |
| 3 | +* src/interfaces/python/pgmodule.c: This patch fixes the well-known |
| 4 | +but unfixed bug that fetchone() always returns the first result in |
| 5 | +the DB-API compliant wrapper. It turned out that the bug was way |
| 6 | +down in the C code. |
| 7 | + |
| 8 | +Gerhard Häring |
| 9 | + |
| 10 | +2001-08-13 16:09 tgl |
| 11 | + |
| 12 | +* configure: Hmm, looks like Bruce's auto-run-autoconf script |
| 13 | +doesn't work for old branches... |
| 14 | + |
| 15 | +2001-08-13 15:54 tgl |
| 16 | + |
| 17 | +* src/backend/executor/nodeHash.c: Make hashjoin give the right |
| 18 | +answer with toasted input data. |
| 19 | + |
| 20 | +2001-08-09 15:22 inoue |
| 21 | + |
| 22 | +* src/: backend/access/heap/heapam.c, backend/executor/execMain.c, |
| 23 | +backend/utils/adt/tid.c, include/access/heapam.h: fix my old fault. |
| 24 | + |
| 25 | +2001-08-09 12:39 momjian |
| 26 | + |
| 27 | +* configure.in, register.txt, doc/bug.template, |
| 28 | +doc/src/sgml/version.sgml, src/include/config.h.win32, |
| 29 | +src/interfaces/libpq/libpq.rc: Update branding for 7.1.3, no |
| 30 | +HISTORY yet. |
| 31 | + |
| 32 | +2001-08-09 09:52 tgl |
| 33 | + |
| 34 | +* src/tutorial/Makefile: Fix Cygwin build, per Jason Tishler. |
| 35 | + |
| 36 | +2001-08-08 18:32 tgl |
| 37 | + |
| 38 | +* src/backend/commands/copy.c: Back-patch fix to disallow COPY |
| 39 | +TO/FROM a view (or anything else that's not a plain relation). |
| 40 | + |
| 41 | +2001-08-08 18:25 tgl |
| 42 | + |
| 43 | +* src/backend/utils/init/miscinit.c: Back-patch fix to prevent |
| 44 | +infinite loop when $PGDATA is not writable. |
| 45 | + |
| 46 | +2001-08-07 14:36 momjian |
| 47 | + |
| 48 | +* src/: backend/port/beos/support.c, backend/port/dynloader/beos.c, |
| 49 | +include/port/beos.h: Commit BEOS patch to 7.1.X. |
| 50 | + |
| 51 | +2001-08-03 16:14 tgl |
| 52 | + |
| 53 | +* src/bin/pg_dump/: pg_dump.c, pg_dump.h: Back-patch fixes for |
| 54 | +dumping user-defined types and dumping comments on views. |
| 55 | + |
| 56 | +2001-07-31 14:39 tgl |
| 57 | + |
| 58 | +* src/: backend/optimizer/path/allpaths.c, |
| 59 | +backend/optimizer/util/clauses.c, backend/utils/adt/ruleutils.c, |
| 60 | +include/optimizer/clauses.h: Fix optimizer to not try to push WHERE |
| 61 | +clauses down into a sub-SELECT that has a DISTINCT ON clause, per |
| 62 | +bug report from Anthony Wood. While at it, improve the |
| 63 | +DISTINCT-ON-clause recognizer routine to not be fooled by out- |
| 64 | +of-order DISTINCT lists. Also, back-patch earlier fix to not push |
| 65 | +down into sub-SELECT with LIMIT. |
| 66 | + |
| 67 | +2001-07-29 18:12 tgl |
| 68 | + |
| 69 | +* src/bin/pg_dump/pg_dump.c: Arrange for GRANT/REVOKE on a view to |
| 70 | +be dumped at the right time, namely after the view definition |
| 71 | +rather than before it.Bug introduced in 7.1 by changes to dump |
| 72 | +stuff in OID ordering. |
| 73 | + |
| 74 | +2001-07-05 22:13 ishii |
| 75 | + |
| 76 | +* doc/src/sgml/backup.sgml: Fix typo. createdb -t --> createdb -T |
| 77 | + |
| 78 | +2001-07-02 15:34 tgl |
| 79 | + |
| 80 | +* src/test/regress/expected/: abstime-solaris-1947.out, |
| 81 | +abstime.out: In any case, it seems the REL7_1 branch needs the |
| 82 | +update too... |
| 83 | + |
| 84 | +2001-06-29 12:34 tgl |
| 85 | + |
| 86 | +* src/backend/commands/vacuum.c: Fix longstanding error in VACUUM: |
| 87 | +sometimes would examine a buffer page after writing/unpinning it. |
| 88 | +An actual failure is unlikely, unless the system is tremendously |
| 89 | +short of buffers ... but a bug is a bug. |
| 90 | + |
| 91 | +2001-06-12 21:02 tgl |
| 92 | + |
| 93 | +* src/pl/plpgsql/src/pl_exec.c: Back-patch fix for attempt to pfree |
| 94 | +a value that's not palloc'd (it's a field of a tuple).I see Jan |
| 95 | +has already fixed this in current sources, but 7.1.* is pretty |
| 96 | +badly broken here. |
| 97 | + |
| 98 | +2001-06-12 14:54 tgl |
| 99 | + |
| 100 | +* src/backend/rewrite/rewriteHandler.c: Repair problem with |
| 101 | +multi-action rules in combination with any nontrivial manipulation |
| 102 | +of rtable/jointree by planner.Rewriter was generating actions |
| 103 | +that shared rtable/jointree substructure, which caused havoc when |
| 104 | +planner got to the later actions that it'd already mucked up. |
| 105 | + |
| 106 | +2001-06-06 13:18 tgl |
| 107 | + |
| 108 | +* src/backend/access/transam/xlog.c: Back-patch change to not keep |
| 109 | +WAL segments just for UNDO information. |
| 110 | + |
| 111 | +2001-05-31 17:50 momjian |
| 112 | + |
| 113 | +* doc/src/sgml/release.sgml: Forgot SGML section section id tag for |
| 114 | +7.1. |
| 115 | + |
| 116 | +2001-05-31 13:33 tgl |
| 117 | + |
| 118 | +* src/backend/utils/adt/ri_triggers.c: RI triggers would fail for |
| 119 | +datatypes using old-style equal function, because cached fmgr info |
| 120 | +contained reference to a shorter-lived data structure.Also guard |
| 121 | +against possibility that fmgr_info could fail, leaving an |
| 122 | +incomplete entry present in the hash table. |
| 123 | + |
| 124 | +2001-05-27 21:01 ishii |
| 125 | + |
| 126 | +* src/backend/utils/mb/conv.c: Fix a message error in utf_to_local |
| 127 | + |
| 128 | +2001-05-22 10:46 thomas |
| 129 | + |
| 130 | +* src/interfaces/jdbc/org/postgresql/jdbc2/ResultSet.java: Patch |
| 131 | +from Barry Lind to correctly decode time zones in timestamp |
| 132 | +results. Without patch, the time zone field is ignored and the |
| 133 | +returned time is not correct. Already applied to the |
| 134 | +development tree... |
| 135 | + |
| 136 | +2001-05-18 12:00 tgl |
| 137 | + |
| 138 | +* src/include/catalog/pg_aggregate.h: Correct recently-broken |
| 139 | +avg(interval) definition. We can't force an initdb to fix this in |
| 140 | +7.1 installations, but it seems better to be shipping a correct |
| 141 | +entry than a wrong one. |
| 142 | + |