robocop
d112cc5a2b
increment build number
2008-10-21 06:17:04 +00:00
robocop
d50c65a3d9
Close some of my favorite B.O.'s since FB1. To be reviewed by Vlad.
2008-10-21 06:17:03 +00:00
robocop
3d97a35be5
increment build number
2008-10-21 04:25:50 +00:00
robocop
43ee87b07d
Misc.
2008-10-21 04:25:49 +00:00
asfernandes
1c1dd9b8e5
increment build number
2008-10-20 23:46:47 +00:00
asfernandes
ad716e38d9
Misc
2008-10-20 23:46:46 +00:00
asfernandes
6a9ab44d35
increment build number
2008-10-20 23:46:45 +00:00
dimitr
6ffe2391c6
increment build number
2008-10-20 14:30:21 +00:00
dimitr
fa6eaff430
An attempt to fix non-HAVE_MMAP builds. To be refactored by Alex later.
2008-10-20 14:30:20 +00:00
asfernandes
23cfefbfab
increment build number
2008-10-20 13:46:34 +00:00
asfernandes
b3df17bf85
increment build number
2008-10-20 13:19:34 +00:00
asfernandes
608fa01368
Patch by Dmitry Kovalenko for CORE-1596
2008-10-20 13:19:33 +00:00
dimitr
d45449ef72
increment build number
2008-10-20 11:12:02 +00:00
dimitr
f18f99ff32
Cleanup.
2008-10-20 11:12:01 +00:00
dimitr
6c4cf0f97c
increment build number
2008-10-20 10:43:51 +00:00
dimitr
7729fe8622
1) Let's avoid a deep recursion for the IN (<list>) predicate. At least on the parser side.
...
2) Fixed the IN elements limit to be 1500 (as documented) instead of 1499.
2008-10-20 10:43:50 +00:00
dimitr
a305538565
increment build number
2008-10-20 04:55:33 +00:00
dimitr
f8b294253c
increment build number
2008-10-20 04:32:43 +00:00
dimitr
2fa05ed659
Update.
2008-10-20 04:32:42 +00:00
dimitr
bc19d82941
increment build number
2008-10-19 12:05:37 +00:00
dimitr
a58aaafdf2
Fixed CORE-2132: Indexed retrieval cannot be chosen if a stored procedure is used inside the comparison predicate.
2008-10-19 12:05:36 +00:00
dimitr
b703879da6
increment build number
2008-10-19 11:41:45 +00:00
dimitr
715bf9b999
increment build number
2008-10-19 11:41:44 +00:00
dimitr
8e72ece026
Uncouple Mutex and RefMutex. This resolves the circular dependency: "AtomicCounter->Mutex->RefCounted->AtomicCounter" and thus enables all non-x86 POSIX builds.
2008-10-19 11:41:43 +00:00
dimitr
f764f68630
increment build number
2008-10-19 11:39:40 +00:00
dimitr
415bdc62aa
increment build number
2008-10-19 11:35:13 +00:00
asfernandes
e8c879f8bc
increment build number
2008-10-17 19:49:14 +00:00
asfernandes
99246d848c
Fixed CORE-2122 - Translation of large text BLOB between UNICODE_FSS (UTF8) and other charsets
2008-10-17 19:49:13 +00:00
hvlad
2f8312e234
increment build number
2008-10-16 13:10:42 +00:00
robocop
60d4b42518
Straighten some multi-line statements that can fit in less than 100 columns.
2008-10-16 11:30:10 +00:00
robocop
97e9a515cd
Fixed inconsistencies noticed by Adriano.
2008-10-16 10:15:02 +00:00
robocop
41a111e4c1
Misc.
2008-10-16 10:14:35 +00:00
robocop
96b399254e
Misc.
2008-10-16 08:52:33 +00:00
robocop
66f41f7986
Simplify the #ifdef.
2008-10-16 08:52:16 +00:00
robocop
bbc988bd9f
Correction.
2008-10-16 08:51:51 +00:00
dimitr
c4530625d9
Undo log shouldn't depend on the outer transaction.
2008-10-16 05:36:36 +00:00
asfernandes
262380af7b
Misc
2008-10-16 02:31:49 +00:00
hvlad
5805e37ba4
Implement new feature CORE-2054 : Ability to query contents database file pages via ISC API interface in order to get low-level information for debug and deep analysis purposes
2008-10-15 10:04:07 +00:00
alexpeshkoff
0c021b27b5
Avoid code duplication as suggested by Claudio
2008-10-14 14:07:10 +00:00
hvlad
ea9555d917
Fixed bug CORE-2018 : Only one client can access a readonly database
...
Missed parts, sorry
2008-10-14 13:15:25 +00:00
dimitr
5ac621c954
Let's not hide the line number of a particular bugcheck call behind a dummy function.
2008-10-14 13:09:48 +00:00
dimitr
e0db5f3ecd
Splitted the undo log into two parts: in-memory one for initial undo items (but they're much more dense now) and on-disk (cached up to TempSpaceLimit) one for full record versions (used in update-in-place).
...
This dramatically reduces the memory usage for bulk updates, thus solving CORE-1575 (serious memory bug on multiple update a table in a single transaction) and CORE-1477 (Avoiding excessive memory consumption by undo-log after series of updates) and predicting out-of-memory conditions that could often lead to record-level corruptions.
2008-10-14 10:32:54 +00:00
dimitr
1a943d1990
Splitted the undo log into two parts: in-memory one for initial undo items (but they're much more dense now) and on-disk (cached up to TempSpaceLimit) one for full record versions (used in update-in-place).
...
This dramatically reduces the memory usage for bulk updates, thus solving CORE-1575 (serious memory bug on multiple update a table in a single transaction) and CORE-1477 (Avoiding excessive memory consumption by undo-log after series of updates) and predicting out-of-memory conditions that could often lead to record-level corruptions.
2008-10-14 10:07:37 +00:00
roman-simakov
8edd7733d4
Fixed authentication bug with sysdba login written in lowercase letters.
2008-10-14 10:05:10 +00:00
hvlad
466f57189f
Correction
2008-10-14 09:51:21 +00:00
hvlad
a3d266ba83
Fixed bug CORE-2018 : Only one client can access a readonly database
2008-10-14 09:10:36 +00:00
dimitr
230071848e
Fixed the Windows build.
2008-10-13 19:21:43 +00:00
alexpeshkoff
065c4b0430
Disconnect all ports after shutdown of working threads
2008-10-13 16:04:15 +00:00
alexpeshkoff
1483d0e465
Need to keep reference count here too
2008-10-13 12:19:26 +00:00
dimitr
e6dc2d8f39
Fixed CORE-2115: Query plan is missing for the long query.
...
Some cleanup and misc code refactoring.
2008-10-13 07:23:19 +00:00
