Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
100 events occurring around 621824092d443425.
|
2010-08-02
| ||
| 14:32 | Experimental refactoring of the Pager object state. This version is surely buggy. (check-in: 03a24051 user: dan tags: experimental) | |
| 10:59 | Modify wal2.test to disable tests requiring TCL 8.5 if the test harness is compiled using TCL 8.4. (check-in: 016486c7 user: drh tags: trunk) | |
| 10:47 | In shared-cache mode, do not allow one connection to checkpoint a database while a second connection is reading or writing the same shared-cache. (check-in: e75b52d1 user: dan tags: trunk) | |
|
2010-08-01
| ||
| 22:41 | Fix a typo in an error message of the TCL interface. (check-in: 8eadd7b8 user: drh tags: trunk) | |
| 22:25 | • Fixed ticket [168d0f71]: Segfault while reading database with a corrupt index plus 3 other changes (artifact: 184435bf user: drh) ... 1 similar event omitted. | |
|
2010-07-31
| ||
| 20:55 | • Fixed ticket [ce7c133e]: Foreign key constraint fails when it should succeed. (artifact: 7b3f9b14 user: drh) ... 3 similar events omitted. | |
|
2010-07-30
| ||
| 18:40 | If a database becomes corrupted such that an index is out of sync with its table, make sure the corruption is detected and reported back. Do not assume that indices always contain rowids for valid table rows. (check-in: 83395a3d user: drh tags: trunk) | |
| 16:54 | Fix typos in comments. No code or logic changes. (check-in: 15c294ca user: drh tags: trunk) | |
| 16:44 | Remove a test from sqlite3PagerCommitPhaseOne which is now always true after the multi-file transaction fix. (check-in: 6758f6a8 user: drh tags: trunk) | |
| 15:43 | Remove variable Pager.needSync, which was almost completely unused. (check-in: 347f22a5 user: dan tags: experimental) | |
| 14:39 | Merge trunk changes into experimental again. (check-in: 87e0f4e1 user: dan tags: experimental) | |
| 11:37 | • Fixed ticket [f3e5abed]: Problems retrying COMMIT after SQLITE_BUSY on multi-file transaction plus 2 other changes (artifact: d07c2800 user: drh) | |
| 11:31 | Changes to the comments describing the Pager.setMaster variable in pager.c. Add an assert() statement to verify that two master journal pointers are not written to a single journal file. (check-in: ad78ccac user: dan tags: trunk) | |
| 11:20 | Cherry-pick the multi-file transaction fix for ticket [f3e5abed55] out of the experimental branch. (check-in: 40f7f0a5 user: drh tags: trunk) | |
| 10:09 | Add the test cases for bug [f3e5abed55]. (check-in: cbbaf8e6 user: dan tags: experimental) | |
| 10:02 | Make sure a connection has an exclusive lock on all database files involved in a multi-file transaction before writing the master-journal pointer into any journal files. Fix for [f3e5abed55]. (check-in: 50c0f220 user: dan tags: experimental) | |
| 07:26 | Merge further trunk changes into experimental branch. (check-in: fb847d70 user: dan tags: experimental) | |
| 05:06 | Add tests to check that the ICU regexp() function can only be called with exactly two arguments. (check-in: 451d9657 user: dan tags: trunk) | |
| 00:31 | Updates to the README.txt file for ICU. (check-in: d37443d3 user: drh tags: trunk) | |
|
2010-07-29
| ||
| 19:02 | • Add attachment tkt-f3e5abed55.test to ticket [f3e5abed] (artifact: aee9538c user: dan) | |
| 18:59 | • New ticket [f3e5abed] Problems retrying COMMIT after SQLITE_BUSY on multi-file transaction. (artifact: fd92e3d4 user: dan) | |
| 10:13 | Unconditionally include the pTmpSpace size in the estimate of the pager memory used. (check-in: 4b97f864 user: drh tags: trunk) | |
| 10:08 | • Fixed ticket [c43940c4]: sqlite3_profile gives milliseconds (was nanoseconds in earlier releases) plus 1 other change (artifact: 057fea73 user: drh) | |
| 10:07 | Change the profile timer units back to nanoseconds and update the sqlite3_profile() documentation. Ticket [c43940c49b74c70a69] (check-in: 7783b98a user: drh tags: trunk) | |
| 04:58 | • Ticket [c43940c4] sqlite3_profile gives milliseconds (was nanoseconds in earlier releases) status still Open with 3 other changes (artifact: 5ea7d43a user: rogerb) ... 1 similar event omitted. | |
| 01:55 | • Fixed ticket [ce7c133e]: Foreign key constraint fails when it should succeed. plus 1 other change (artifact: b3a12114 user: drh) ... 1 similar event omitted. | |
| 01:50 | Correct handling of compound foreign key constraints that include the integer primary key as one of the columns. Ticket [ce7c133ea6cc9ccdc1] (check-in: 53902f7d user: drh tags: trunk) | |
|
2010-07-28
| ||
| 20:57 | • New ticket [ce7c133e] Foreign key constraint fails when it should succeed.. (artifact: 4202cdf3 user: drh) | |
| 19:17 | Get SQLITE_OMIT_VIRTUALTABLE working again after being broken by recent changes. (check-in: 33b1e862 user: drh tags: trunk) | |
| 18:51 | Fix the dbstatus.test script so that it works correctly on 64-bit machines. (check-in: a3401d9e user: drh tags: trunk) | |
| 18:35 | Merge trunk changes into experimental branch. (check-in: aa819001 user: dan tags: experimental) | |
| 17:36 | Improve the accuracy of the Pager heap usage estimate. (check-in: ae89777e user: drh tags: trunk) | |
| 17:16 | Adjust the shell test script "shell4.test" to account for changes in the ".stat on" display format from the previous checkin. (check-in: 2f2fa7dd user: drh tags: trunk) | |
| 17:01 | Record the pcache allocation size statistics even for pcache overflow allocations. Adjust the wording on one of the stat output lines in the shell. (check-in: f9adf66a user: drh tags: trunk) | |
| 16:05 | Modify CLI to optionally display "stats". (check-in: 419ce0ed user: shaneh tags: trunk) | |
| 15:52 | Lookaside memory is not used to store schemas. Change the SQLITE_DBSATUS_SCHEMA_USED documentation to reflect this fact. (check-in: 07abfd52 user: drh tags: trunk) | |
| 15:49 | Add documentation for the SQLITE_STATUS_MALLOC_COUNT parameter to sqlite3_status(). (check-in: dbfbdb60 user: drh tags: trunk) | |
| 15:10 | Fix errors in wal3.test caused by recent modifications. (check-in: 0714aecc user: dan tags: experimental) | |
| 14:26 | Support FCNTL_CHUNK_SIZE on windows too. (check-in: a038688c user: dan tags: experimental) | |
| 02:53 | • Fixed ticket [8011086c]: Query planner creates unnecessary and expensive automatic index plus 2 other changes (artifact: b5dd6bce user: drh) | |
| 02:53 | Do not allow automatic indices for the RHS of IN expressions which are not correlated subqueries. Ticket [8011086c85c6c404014c9] (check-in: 267492d3 user: drh tags: trunk) | |
| 00:33 | • New ticket [8011086c] Query planner creates unnecessary and expensive automatic index. (artifact: 333f2a70 user: drh) | |
|
2010-07-27
| ||
| 18:37 | • Edit [621824092d443425|62182409]: Change branch background color to "#c0ffc0". Timestamp 2010-07-27 18:36:38. (artifact: 479f2044 user: dan) | |
| 18:37 | • Edit [7cf0e851d4c5e826|7cf0e851]: Change background color to "#c0ffc0". Timestamp 2010-07-27 18:34:16. (artifact: d94e1320 user: dan) | |
| 18:36 | Merge trunk changes into experimental branch. (check-in: 62182409 user: dan tags: experimental) | |
| 18:34 | Add experimental unix-only file-control to grow and truncate the database file by a configurable chunk size. (check-in: 7cf0e851 user: dan tags: experimental) | |
| 16:42 | Update the makefiles to prefer TCL version 8.5 instead of 8.4. (check-in: 8118de2a user: drh tags: trunk) | |
|
2010-07-26
| ||
| 19:09 | Fix a bug in the SQLITE_STATUS_MALLOC_COUNT counter. Add an ALWAYS() around a condition in the SQLITE_DBSTATUS_SCHEMA_USED logic that is always true. (check-in: 6df081ad user: drh tags: trunk) | |
| 18:43 | Add the SQLITE_STATUS_MALLOC_COUNT option for sqlite3_status(). (check-in: 8f8e442b user: drh tags: trunk) | |
| 16:40 | Increase the version number to 3.7.1 due to the API enhancement of the previous check-in. (check-in: 1d7571e4 user: drh tags: trunk) | |
| 16:25 | • Edit [548bf3f7d7b962d3|548bf3f7]: Mark "Closed". Edit check-in comment. Timestamp 2010-07-25 02:39:07. (artifact: 40b37362 user: drh) | |
| 16:24 | Add support for SQLITE_DBSTATUS_SCHEMA_USED and SQLITE_DBSTATUS_STMT_USED to the trunk. (check-in: 008368b2 user: drh tags: trunk) | |
| 15:57 | • Edit [e327ef37faec52ce|e327ef37]: Edit check-in comment. Timestamp 2010-07-26 15:57:02. (artifact: b93607ad user: drh) | |
| 15:57 | When calculating schema memory, use the actual allocated size of hash elements, not sizeof(HashElem). Also fix a bug in dbstatus.test. (Closed-Leaf check-in: e327ef37 user: dan tags: experimental) | |
| 14:47 | Add virtual table test cases to dbstatus.test. (check-in: 72b84d06 user: dan tags: experimental) | |
| 14:20 | Further fixes to the P4_MEM size measurement logic. (check-in: 934cda29 user: drh tags: experimental) | |
| 13:58 | Fix an issue with P4_MEM and the schema size measurement logic. Also fix a compiler warning. (check-in: 8166f338 user: drh tags: experimental) | |
| 12:38 | Make sure sqlite3_free() is not called for P4_MPRINTF during a size measurement. (check-in: 56f11f88 user: drh tags: experimental) | |
| 12:36 | Remove an assert() that uses a variable removed earlier. (check-in: c04907e6 user: dan tags: experimental) | |
| 12:05 | Change the way SubProgram objects are deleted so that the code is the same for deletion and measurement. (check-in: 00e55102 user: dan tags: experimental) | |
| 11:59 | Update comments to better documentation the new memory measurement functions. (check-in: 620bad03 user: drh tags: experimental) | |
| 11:08 | • Edit [629e38a8c9e31111|629e38a8]: Change branch background color to "#c0ffc0". Edit check-in comment. Timestamp 2010-07-26 11:07:21. (artifact: fbca0b4e user: drh) | |
| 11:07 | Make sure all memory from sqlite3DbMalloc() is freed by sqlite3DbFree() and all memory from sqlite3_malloc() is freed by sqlite3_free(). (check-in: 629e38a8 user: drh tags: experimental) | |
|
2010-07-25
| ||
| 02:39 | Fixes to prior checkins so that they compile and run even if SQLITE_MEMDEBUG is not defined. (Closed-Leaf check-in: 548bf3f7 user: drh tags: malloc-enhancement) | |
| 02:12 | Further examples of using automatic deallocation to replace "delete" methods. (check-in: da2f62c5 user: drh tags: malloc-enhancement) | |
|
2010-07-24
| ||
| 19:08 | Additional malloc sanity changes. Use sqlite3MemLink() on Index.zColAff and Table.zColAff as a proof of concept. (check-in: e5ecb159 user: drh tags: malloc-enhancement) | |
| 18:25 | Add infrastructure to support a hierarchy of memory allocations with automatic deallocation of substructure. (check-in: 48ef221c user: drh tags: malloc-enhancement) | |
| 16:34 | Make sure all memory from sqlite3DbMalloc() is freed by sqlite3DbFree() and all memory from sqlite3_malloc() is freed by sqlite3_free(). (check-in: ac1f37a6 user: drh tags: malloc-enhancement) | |
| 11:29 | • Edit [9aa30342f4de4eff|9aa30342]: Change background color to "#c0ffc0". Timestamp 2010-07-24 11:28:29. (artifact: 580c1f70 user: dan) | |
| 11:28 | Experimental code to measure memory consumed by database schemas and prepared statements. (check-in: 9aa30342 user: dan tags: experimental) | |
|
2010-07-23
| ||
| 22:26 | Fix a typo in the OS/2 vfs code. (check-in: a6bb2108 user: drh tags: trunk) | |
| 17:37 | Merge two leaves. (check-in: 54e5886d user: dan tags: trunk) | |
| 17:32 | Strenghten an assert() in malloc.c that helps to ensure that allocated memory is freed by the appropriate routine. (check-in: 80db61ac user: drh tags: trunk) | |
| 17:06 | Remove additional traces (mostly in comments) of the Table.dbMem field. (check-in: 5c58f44a user: drh tags: trunk) | |
| 15:55 | Add a test case to shared.test. No changes to production code. (check-in: bfb0dd3c user: dan tags: trunk) | |
| 15:41 | Remove the Table.dbMem variable, as it is no longer being used for its original purpose. (check-in: 6eddc6e6 user: dan tags: trunk) | |
| 08:31 | Remove the unused Schema.db variable. (check-in: 430be657 user: dan tags: trunk) | |
|
2010-07-22
| ||
| 22:40 | Add evidence marks to the where.c source file. Comment only - no code changes. (check-in: f3f9e8e5 user: drh tags: trunk) | |
| 17:55 | Make the sqlite3_db_status() routine threadsafe. (check-in: 241f7bd1 user: dan tags: trunk) | |
| 17:50 | • Fixed ticket [47110204]: should use collate nocase index even if column isn't declared nocase plus 1 other change (artifact: b9e1ea8e user: drh) | |
| 17:49 | Enhance the LIKE/GLOB query optimization so that it works as long as there is an index with the appropriate collating sequence and even if the default collating sequence of the column is different. Ticket [4711020446da7d93d99]. (check-in: 9f932655 user: drh tags: trunk) | |
| 15:44 | Add test/threadtest3.c, containing multi-thread tests implemented in C. (check-in: aad88cf5 user: dan tags: trunk) | |
| 12:34 | • Fixed ticket [e0901835]: like should use index when pattern has no wildcard plus 1 other change (artifact: e78b5e56 user: drh) | |
| 12:33 | Try to optimize LIKE and GLOB operators when RHS contains no wildcard. Ticket [e090183531fc274747] (check-in: 613a87d6 user: drh tags: trunk) | |
| 11:40 | Tweaks to the virtual table documentation contained in comments of sqlite3.h. No changes to actual code. (check-in: f5866d47 user: drh tags: trunk) | |
| 06:11 | • Ticket [5e0423b0] xAccess behaviour is not documented, changed in Windows 3.7.0 status still Open with 3 other changes (artifact: 6365a76d user: rogerb) ... 1 similar event omitted. | |
| 00:13 | • Edit [b36b105eab6fd319|b36b105e]: Change background color to "#ffc0d0". Edit check-in comment. Timestamp 2010-07-21 16:16:28. Add tag "release". (artifact: 144759b5 user: drh) | |
|
2010-07-21
| ||
| 22:22 | • Ticket [e0901835] like should use index when pattern has no wildcard status still Open with 3 other changes (artifact: 7a237684 user: drh) | |
| 22:21 | • Ticket [47110204] should use collate nocase index even if column isn't declared nocase status still Open with 3 other changes (artifact: 92b593d2 user: drh) ... 1 similar event omitted. | |
| 22:01 | • New ticket [e0901835] like should use index when pattern has no wildcard. (artifact: 5484eb7b user: aswift) | |
| 16:16 | Version 3.7.0 (check-in: b36b105e user: drh tags: trunk, release) | |
|
2010-07-20
| ||
| 20:23 | Added fix to os_win.c for race conditions from os_unix.c; added saving of errno in two places. (check-in: 13ed106c user: shaneh tags: trunk) | |