Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
1493 check-ins using file test/pcache.test version eebc4420
|
2010-11-19
| ||
| 04:22 | Lemon: Fixed compiler warning on Visual C++. (Closed-Leaf check-in: 892c5f37 user: icculus tags: lemon-update-2010) | |
| 04:19 |
Lemon: correcting previous win32 access() fix.
The previous fix caused link errors when compiled as C++. (check-in: bd03d333 user: icculus tags: lemon-update-2010) | |
| 03:55 | Lemon: Patched to compile with Visual Studio. (check-in: 0b945101 user: icculus tags: lemon-update-2010) | |
|
2010-10-02
| ||
| 01:01 | Backport the very lastest R-Tree in order to take advantage of its enhanced robustness to corrupt databases. (Leaf check-in: 29a681dd user: drh tags: apple-osx-3623) | |
| 01:00 | Fix the amalgamation builder so that it works with the rtree updates of the prior check-in. (Leaf check-in: 265b0b29 user: drh tags: branch-3.6.23) | |
|
2010-10-01
| ||
| 20:45 | Backport the RTree implementation from the trunk into the 3.6.23 branch. The code for the application-defined query boxes is still present but is disabled. The reason for this backport is to take advantage of recent enhancements to robustness to database corruption. (check-in: 854b23c1 user: drh tags: branch-3.6.23) | |
|
2010-08-24
| ||
| 12:05 | Pull the incremental_vacuum bug fix ([255f1eefa373153942c67b18b]) and the R-tree segfault bug fix ([7f2f71cc9e3c39093f09231f44]) into the 3.6.23 branch. Increase the version number to 3.6.23.3. (check-in: bcbdecd8 user: drh tags: branch-3.6.23) | |
| 02:10 | Cherrypick the R-tree invalid shadow-table big fix of [7f2f71cc9e3c39093f09231f44] into the apple-osx 3.6.23 branch. (check-in: 68103d91 user: drh tags: apple-osx-3623) | |
| 01:59 | Cherrypick the incremental_vacuum bug fix ([255f1eefa373153942c67b18b]) into the apple-osx 3.6.23 branch. (check-in: 60f16838 user: drh tags: apple-osx-3623) | |
|
2010-08-17
| ||
| 23:13 | Cherrypick the changes for enhancement requests [e090183531fc27474] (use indices on LIKE with no wildcards) and [4711020446da7d93d993] (use nocase index for LIKE even if the column is binary) into the 3.6.23.1 release of the Apple-OSX branch. (check-in: 220cca50 user: drh tags: apple-osx-3623) | |
|
2010-06-22
| ||
| 11:10 | Fix problem with handling of -vfs option in tcl interface. (check-in: 8f98e8b5 user: dan tags: trunk) | |
|
2010-06-21
| ||
| 21:45 | Minor tweaks to get the pager working for corner cases when the SQLITE_IOCAP_UNDELETABLE_WHEN_OPEN flag is set. (check-in: 82b38fca user: drh tags: trunk) | |
| 20:19 | Fix comment inside the pager_unlock() routine. Always close the master journal file before trying to delete it. (check-in: f820b815 user: drh tags: trunk) | |
| 18:29 | Add test cases to pager1.test and pager2.test. (check-in: cc9ddae6 user: dan tags: trunk) | |
| 12:47 | Merge the experimental UNDELETABLE_WHEN_OPEN optimization into the trunk. (check-in: ee0acef1 user: drh tags: trunk) | |
| 12:34 | Change things so that journal2.test works with ENABLE_ATOMIC_WRITE. (Closed-Leaf check-in: a64d96db user: dan tags: experimental) | |
| 07:45 | Add further pager tests. (check-in: 4104b175 user: dan tags: experimental) | |
| 06:00 | Merge latest trunk change. (check-in: f6d26e07 user: dan tags: experimental) | |
| 05:40 | Fix jrnlmode2.test so that it works on systems where UNDELETABLE_WHEN_OPEN is defined. (check-in: 59be370e user: dan tags: experimental) | |
|
2010-06-19
| ||
| 23:53 | Fix an uninitialized variable in os_unix.c. (check-in: 822a0283 user: drh tags: trunk) | |
| 19:06 | Fix an assert() failure that could occur if compiling with OMIT_SHARED_CACHE. (check-in: 3e76a9f2 user: dan tags: experimental) | |
| 18:12 | Change the name of IOCAP_SAFE_DELETE to IOCAP_UNDELETABLE_WHEN_OPEN. Have the xDeviceCharacteristics() method of the win32 VFS return this flag. (check-in: 5a5ff4e3 user: dan tags: experimental) | |
| 17:26 | Add tests to pager1.test and pagerfault.test. (check-in: f5df83fd user: dan tags: experimental) | |
| 15:10 | Change the unix VFS to always allocate shared-memory using a file in the same directory as the database. Otherwise, a chroot might cause different processes to use different shared memory files resulting in database corruption. (check-in: 2241788b user: drh tags: trunk) | |
| 11:30 | Add tests to pager1.test and pagerfault.test. (check-in: 58c0b5bf user: dan tags: experimental) | |
|
2010-06-18
| ||
| 18:59 | Add tests to pager1.test. (check-in: 582fca89 user: dan tags: experimental) | |
| 16:13 | Rearrange some code in OP_JournalMode to avoid incorrect returns from subsequent invocations of "PRAGMA journal_mode" following an IO error. (check-in: ce0a4a9b user: dan tags: experimental) | |
| 11:10 | Change the implementation of the unix implementation of xAccess() so that it returns 0 (does not exist) to an SQLITE_ACCESS_EXISTS query on a file that exists but is zero bytes in size. (check-in: 077b0e5b user: dan tags: experimental) | |
|
2010-06-17
| ||
| 17:05 | Apply [b9b11855e8] (the alternate fix to [fc62af4523]) to the trunk. (check-in: 9a949a3a user: dan tags: trunk) | |
| 16:44 | Merge trunk change [7c3a86b9c7]. (check-in: bd7bc4e0 user: dan tags: experimental) | |
| 16:44 | A different fix to [fc62af4523]. When changing from journal_mode=PERSIST or TRINCATE to some other rollback mode, delete the journal file only if a RESERVED lock can be obtained on the database file first. (check-in: b9b11855 user: dan tags: experimental) | |
| 16:08 | Fix the tkt-fc62af4523.test to work around non-randomness of the randomblob() function when in testing mode. (check-in: 7c3a86b9 user: drh tags: trunk) | |
| 11:36 | Fix bug in journal2.test. (check-in: c1e04f1d user: dan tags: experimental) | |
| 10:52 | Merge trunk updates with experimental branch. Also add tests to journal2.test. (check-in: 9f6ea1de user: dan tags: experimental) | |
| 10:42 | Do not delete the journal file in "PRAGMA journal_mode" commands. This fixes [fc62af4523]. (check-in: 1ec74591 user: dan tags: trunk) | |
| 10:24 | Add test case for [fc62af4523]. (check-in: cccd32c6 user: dan tags: trunk) | |
| 06:19 | Merge fix [f80c3f922a] with experimental changes. (check-in: 20133e9c user: dan tags: experimental) | |
| 02:13 | Bug fix: Only trust the database size number at offset 28 if the change counter at offset 24 matches the version number counter at offset 92. This prevents corruption in the case of two applications writing to the database where one is an older version of SQLite and the other is a newer version. (check-in: f80c3f92 user: drh tags: trunk) | |
|
2010-06-16
| ||
| 19:48 | Merge in changes up to and including the 3.6.23.1 release. (check-in: 21ca87f6 user: drh tags: apple-osx) | |
| 19:04 | Experimental change: On systems where it is not possible to unlink a file while one or more processes has it open (i.e. not unix), avoid closing the journal file each time the database is unlocked and reopening it at the start of each transaction. (check-in: bede8c8a user: dan tags: experimental) | |
| 12:30 | Add extra test cases to pager1.test. (check-in: ad320957 user: dan tags: trunk) | |
| 10:55 | Fix a memory leak that can occur in os_unix.c if an IO error occurs within the xUnlock method. (check-in: 6c5c04ee user: dan tags: trunk) | |
|
2010-06-15
| ||
| 19:07 | Rationalize a common pattern in tcl test cases into proc do_multiclient_test. (check-in: efe44564 user: dan tags: trunk) | |
| 18:00 | Fix a problem introduced into lock2.test by the previous commit. (check-in: c1c9f6fa user: dan tags: trunk) | |
| 17:44 | Add test file pager1.test, containing tests of inter-process locking in non-wal mode. (check-in: 6e43eed9 user: dan tags: trunk) | |
| 14:21 | Run extra iterations in wal3.test to ensure test coverage. (check-in: ea80cbe5 user: dan tags: trunk) | |
| 14:06 | Merge in [2c5e48a485]. Merge in [2c5e48a485]. Merge in [2c5e48a485]. (check-in: 03c0f4fd user: dan tags: trunk) | |
| 13:56 | Changes to the way tcl test scripts work. No changes to production code. (Closed-Leaf check-in: 2c5e48a4 user: dan tags: experimental) | |
| 12:09 | Minor tweaks to the interface documentation - adding hyperlinks. (check-in: 07b2fdd1 user: drh tags: trunk) | |
| 11:21 | Updates to comments and testcase() macros in wal.c. (check-in: 4d90cc0b user: drh tags: trunk) | |
|
2010-06-14
| ||
| 18:06 | Fix an error in the walfault.test script introduced by the previous check-in. (check-in: a2b9374f user: drh tags: trunk) | |
| 18:01 | Move the xShmMap method to in between xShmLock and xShmBarrier, since it seems to fit in there logically. (check-in: 58dfd83d user: drh tags: trunk) | |
| 17:09 | Merge the experimental shared-memory mmap-by-chunk changes into the trunk. (check-in: f295e7ed user: drh tags: trunk) | |
| 16:16 | Add the new xShmMap (formerly xShmPage) to os_win.c. (check-in: 13e7a824 user: dan tags: experimental) | |
| 14:07 | Remove xShmGet/Size/Release from the sqlite3_vfs structure. Change the name of xShmPage to xShmMap. Remove some code that is now unused from os_unix.c and some of the test VFS implementations. (check-in: fc0cabc1 user: dan tags: experimental) | |
| 11:49 | Update some comments in wal.c. No code changes. (check-in: 1ce9c92b user: dan tags: experimental) | |
| 11:18 | Change the interface to internal function walGetHash() to make it easier to follow. (check-in: 5e8e2e97 user: dan tags: experimental) | |
| 10:30 | Add the xShmPage method to the "crash" vfs in test6.c. (check-in: 1008f536 user: dan tags: experimental) | |
| 07:53 | Add some fault-injection tests to improve coverage. (check-in: 37b26d12 user: dan tags: experimental) | |
|
2010-06-12
| ||
| 12:02 | Fix some problems with handling IO errors on the experimental branch. (check-in: eade8bc2 user: dan tags: experimental) | |
|
2010-06-11
| ||
| 19:04 | Experimental change to the xShmXXX parts of the VFS interface. (check-in: ca68472d user: dan tags: experimental) | |
| 17:01 | Refactor and simplify the logic used to change journalmode. (check-in: 95cc3f6f user: drh tags: trunk) | |
|
2010-06-10
| ||
| 16:25 | Adding original art for the new SQLite logo. (check-in: af353bd8 user: drh tags: trunk) | |
| 14:07 | Add the SQLITE_TESTCTRL_PGHDRSZ verb for sqlite3_test_control() used to get the size of pcache headers for testing purposes. (check-in: 5d694f04 user: drh tags: trunk) | |
| 10:51 | Update the test VFS in test_journal.c so that it implements the xCurrentTimeInt64 method. (check-in: ec3b4a7e user: drh tags: trunk) | |
| 06:53 | Store the MemPage structure in memory following, instead of preceding, the page data for cached pages. This reduces the likelihood of a corrupt database page image causing SQLite to read past the end of a buffer. (check-in: 0ce42e76 user: dan tags: trunk) | |
| 01:07 | Remove a condition that must always be true. Replace it with an assert(). (check-in: 7abcd54e user: drh tags: trunk) | |
|
2010-06-09
| ||
| 19:01 | Fix test script problems causing errors in the "inmemory_journal" permutations. (check-in: ea05b7f9 user: dan tags: trunk) | |
| 16:58 | When restarting (wrapping) a log file, set all unused aReadMark[] slots to READMARK_NOT_USED instead of 0. Setting them to 0 does not cause a problem, but may cause SQLite to obtain and release a few more file locks than would otherwise be necessary. (check-in: 0797b10c user: dan tags: trunk) | |
| 15:47 | Fix for ticket [f973c7ac31]. (check-in: 6eb058dd user: dan tags: trunk) | |
| 14:45 | Shift the meaning of aReadMark[] back so that +1 offset from mxFrame is removed. Add the new READMARK_NOT_USED value (0xffffffff) instead of zero to signal an aReadMark[] that is not in use. (check-in: c576827d user: drh tags: trunk) | |
| 11:28 | Simpler fix for the race condition also fixed by [7c102c7b5f] (check-in: 3c2de820 user: dan tags: trunk) | |
| 11:02 | Fix for race condition in WAL locking code. (Closed-Leaf check-in: 7c102c7b user: dan tags: mistake) | |
|
2010-06-08
| ||
| 15:50 | Close database opened by tester.tcl when it is sourced in all.test. Because test scripts are now run in slave interpreters, this connection was not being closed by the first script run as it was previously. (check-in: b072e9f6 user: dan tags: trunk) | |
| 15:16 | Changes to wal3.test to support small default cache size settings. And builds that create autovacuum databases by default. (check-in: ac4cc257 user: dan tags: trunk) | |
|
2010-06-07
| ||
| 19:26 | In os_unix.c, call munmap() to unmap a shared-memory region prior to closing the associated file descriptor. (check-in: 95cc976f user: dan tags: trunk) | |
| 17:47 | Change all.test, quick.test and permutations.test so that they use a separate interpreter for each test file. (check-in: 7c33eb5a user: dan tags: trunk) | |
| 14:28 | Refactor some of the global variables and commands used by tester.tcl. (check-in: c2edf8e1 user: dan tags: trunk) | |
| 06:11 | Add a couple of extra coverage tests for wal.c. (check-in: cfe60254 user: dan tags: trunk) | |
|
2010-06-05
| ||
| 19:18 | Add further test cases for the logic in sqlite3WalBeginReadTransaction(). (check-in: a49713db user: dan tags: trunk) | |
| 18:34 | Clarify the purpose of a test for a race-condition in sqlite3WalBeginReadTransaction(). (check-in: c041c6a9 user: dan tags: trunk) | |
| 18:12 | Remove a condition from sqlite3WalRead() that is unreachable as of the changes to clear entries out of the wal-index hash tables on transaction or savepoint rollback. (check-in: 39420473 user: dan tags: trunk) | |
| 14:42 | Mark a condition in wal.c as ALWAYS(). (check-in: 3fe0cc78 user: dan tags: trunk) | |
| 11:53 | Add extra coverage test cases for wal.c. No changes to production code. (check-in: f9d4ae0e user: dan tags: trunk) | |
|
2010-06-04
| ||
| 18:38 | Clarify an assert in sqlite3WalExclusiveMode(). (check-in: 25585069 user: dan tags: trunk) | |
| 17:16 | Remove an unnecessary branch from wal.c. (check-in: 8e54786c user: dan tags: trunk) | |
| 15:59 | Add an English language error message to corresponding to SQLITE_PROTOCOL. "locking protocol". (check-in: ca327e32 user: dan tags: trunk) | |
| 12:22 | Add test for the code that detects an inconsistent pair of wal-index headers to wal2.test. (check-in: 157feba1 user: dan tags: trunk) | |
| 11:56 | If an attempt to sync the database file as part of a checkpoint fails, do not update the shared "nBackfill" variable. Otherwise, another process could wrap the log and overwrite content before it is synced into the database. (check-in: b813233d user: dan tags: trunk) | |
| 10:37 | Fix a problem where an SQLITE_BUSY in the checkpoint code was being treated as an IO error (abandoning, instead of just limiting, the checkpoint). (check-in: 02c4040c user: dan tags: trunk) | |
|
2010-06-03
| ||
| 19:10 | Fix another problem in test_vfs.c. (check-in: df7d5989 user: dan tags: trunk) | |
| 18:20 | Have test_vfs.c simulate IO errors in xShmLock. (check-in: fcbf7cf1 user: dan tags: trunk) | |
| 18:02 | Performance fix for winShmClose(). (check-in: ed7774de user: drh tags: trunk) | |
| 16:58 | Add extra tests for removing elements from wal-index hash tables as part of a rollback. (check-in: af3e598a user: dan tags: trunk) | |
| 12:35 | Delay opening the sub-journal until SQLite actually needs to write data to it. (check-in: c43deb33 user: dan tags: trunk) | |
| 12:09 | Remove global variables when compiled with SQLITE_OMIT_WSD (check-in: dd10a547 user: drh tags: trunk) | |
| 09:25 | Enhancements to test_vfs.c and walfault.test. (check-in: ac0de2f3 user: dan tags: trunk) | |
| 09:17 | If an error (OOM or SQLITE_FULL error) occurs while executing an SQL statement and a statement-transaction is automatically rolled back as a result, if a second error occurs during the statement rollback do a full transaction rollback instead. Otherwise the client can be left with an inconsistent cache. This can affect both WAL and rollback modes. (check-in: eb80ddc6 user: dan tags: trunk) | |
| 09:01 | If a malloc fails while allocating a savepoint object at the pager level, do not try to roll that savepoint back later on. (check-in: 91cb08ff user: dan tags: trunk) | |
|
2010-06-02
| ||
| 20:38 | When a lock fails due to an I/O error (not an SQLITE_BUSY) then the checkpoint should fail. (check-in: c6eb0a27 user: drh tags: trunk) | |
| 18:59 | Fix a problem with rolling back to a savepoint opened before the writer decided to wrap the log file. (check-in: 6b4aed6a user: dan tags: trunk) | |
| 17:15 | If an error occurs while writing frames to the log to commit a transaction, do not mark the pages as clean in the cache. Otherwise, the subsequent rollback does not roll them back (leaving the client to continue with a cache that makes it appear that the transaction was committed - inconsistent with the database on disk). (check-in: cb571c1b user: dan tags: trunk) | |
| 16:40 | Added implementation of vfslog_time() for Windows testing. (check-in: b3109f85 user: shaneh tags: trunk) | |
| 16:39 | Updated wal2.test for windows testing. (check-in: c3606f39 user: shaneh tags: trunk) | |
| 15:59 | Make sure the directory used by os_unix.c for shared-memory files is always the same, regardless of individual client permissions and environment variable settings. (check-in: f97bd520 user: drh tags: trunk) | |
| 15:43 | When walTryBeginRead() encounters an I/O error trying to set a lock (as opposed to SQLITE_BUSY) be sure to propagate that error back up the call stack. (check-in: aa2c2b67 user: drh tags: trunk) | |
| 14:45 | Changes to make WAL more robust against SHM locking failures and OOM errors. (check-in: ebf40413 user: drh tags: trunk) | |
| 14:43 | In the rowhash.test, make sure global variables are cleared prior to use. (check-in: 28efe0a4 user: drh tags: trunk) | |
| 12:58 | Undo one of the "const" markers from the previous check-in because the value needs to be changable, even if it is not actually ever changed in the current implementation. (check-in: c24ac356 user: drh tags: trunk) | |
| 05:53 | Add some 'const' markers to static data that is really constant. (check-in: e7073e23 user: dan tags: trunk) | |
|
2010-06-01
| ||
| 21:02 | The shared-memory used by WAL on linux now really is shared memory in /dev/shm. On other unix flavors, the file is in a temporary directory rather than in the same directory as the database. (check-in: fc18c4aa user: drh tags: trunk) | |
| 19:15 | Changes to the way faults are injected into xShmXXX VFS calls. (check-in: 716d99f3 user: dan tags: trunk) | |
| 17:46 | Change the OOM and IO error test cases in walfault.test so that each test case runs both types of error simulation. (check-in: b627e153 user: dan tags: trunk) | |
| 15:44 | Delay the decision to restart the log file until data is actually ready to be written to the log file (instead of at the start of a write transaction). (check-in: b1abfaaf user: dan tags: trunk) | |
| 15:24 | The incremental checkpoint feature is not perfect yet, but it is working well enough to merge it into the trunk. (check-in: 1d3e569e user: drh tags: trunk) | |
| 14:30 | Merge with [15abbc3416]. (Closed-Leaf check-in: f4b9003a user: dan tags: wal-incr-ckpt) | |
| 14:12 | If an attempt to get a read-lock on the WAL fails with SQLITE_BUSY_RECOVER, call the busy-handler at the btree level. (check-in: ce644965 user: dan tags: wal-incr-ckpt) | |
| 13:17 | Fix a bug introduced by the previous check-in but only seen if SQLITE_ENABLE_EXPENSIVE_ASSERTS is turned on. (check-in: 9c9ec899 user: drh tags: wal-incr-ckpt) | |
| 12:58 | Fix an off-by-one boundary-value issue in walCleanupHash(). (check-in: f039552e user: drh tags: wal-incr-ckpt) | |
| 11:08 | Do not hold the shared-memory mutex in os_unix.c if returning NULL. (check-in: 9622dd46 user: dan tags: wal-incr-ckpt) | |
| 10:44 | If the checkpoint fails to obtain an exclusive lock on one of the read-lock bytes, do not consider this an error. (check-in: 9e95e357 user: dan tags: wal-incr-ckpt) | |
| 07:51 | Fixes to the test cases in wal2.test. (check-in: cd5fbcbc user: dan tags: wal-incr-ckpt) | |
| 01:08 | Comment edits and cleanup in wal.c. No functional code changes. (check-in: e8e666ab user: drh tags: wal-incr-ckpt) | |
| 00:28 | Attempt to get the filectrl.test script running. (check-in: e46a8f2b user: drh tags: wal-incr-ckpt) | |
| 00:03 | Fix os_unix.c so that it will compile and build on a Mac. (check-in: bc707c83 user: drh tags: wal-incr-ckpt) | |
|
2010-05-31
| ||
| 20:28 | Fix issues with locking_mode=EXCLUSIVE in WAL. (check-in: 8deba0ce user: drh tags: wal-incr-ckpt) | |
| 18:24 | Add an "isInit" field in the wal-index header that must be non-zero for a valid header. Use this to detect an uninitialized wal-index. (check-in: a16fde19 user: drh tags: wal-incr-ckpt) | |
| 16:56 | Fix an inconsistent #ifdef in wal.c. Fix os_unix.c so that it does not allow moving an SHM lock directly exclusive to shared without going through unlocked. (check-in: 552658da user: drh tags: wal-incr-ckpt) | |
| 16:41 | Avoid dropping the checkpoint lock after a recovery run as a precursor to a checkpoint operation. (check-in: cc25cfa0 user: dan tags: wal-incr-ckpt) | |
| 16:17 | Zero the checkpoint header as the last step of successful WAL recovery. Avoid an unnecessary lock/unlock in WalBeginReadTransaction. (check-in: db3509c5 user: dan tags: wal-incr-ckpt) | |
| 16:10 | Get the new xShmLock interface design working on os_win.c. (check-in: 149a7082 user: drh tags: wal-incr-ckpt) | |
| 14:39 | Make sure WAL alway requests enough shared-memory space. (check-in: 138f1283 user: drh tags: wal-incr-ckpt) | |
| 14:28 | Remove the tests that enable tracing based on the existance of the "vdbe_*" files when in SQLITE_DEBUG mode. (check-in: cdbb4e7c user: drh tags: wal-incr-ckpt) | |
| 14:18 | Update another test case in wal.test. (check-in: 99fde69e user: dan tags: wal-incr-ckpt) | |
| 13:11 | Change WAL test cases to account for the improved concurrency in the new checkpoint logic. (check-in: 2d6f2485 user: drh tags: wal-incr-ckpt) | |
| 12:30 | Do not fail a checkpoint just because active readers prevent backfill. (check-in: 9aa4243e user: drh tags: wal-incr-ckpt) | |
| 12:15 | Add WALTRACE() macros to help with debugging and analysis. (check-in: 765c3318 user: drh tags: wal-incr-ckpt) | |
| 12:02 | Another minor fix to checkpoint on the WAL branch. (check-in: 853f1e3f user: dan tags: wal-incr-ckpt) | |
| 11:39 | Fix a bug in checkpoint introduced by [181ceb32ea]. (check-in: b499dbc8 user: dan tags: wal-incr-ckpt) | |
| 11:16 | Fix some asserts and other things in the new WAL branch. (check-in: 181ceb32 user: dan tags: wal-incr-ckpt) | |
| 06:38 | Changes to the way one of the WAL/OOM tests works. (check-in: 15abbc34 user: dan tags: trunk) | |
| 01:41 | WAL runs but quickly deadlocks. (check-in: ace58acb user: drh tags: wal-incr-ckpt) | |
|
2010-05-30
| ||
| 19:55 | Initial code for incremental checkpoint in WAL mode. This check-in compiles on unix and runs as long as you do not engage WAL mode. WAL mode crashes and burns. Consider this check-in a baseline implementation for getting the new capability up and running. (check-in: ef3ba7a1 user: drh tags: wal-incr-ckpt) | |
|
2010-05-29
| ||
| 08:40 | Add tests to fkey2.test to check that ON CONFLICT clauses do not affect SQLite's behaviour when an FK constraint is violated. (check-in: e9e5b100 user: dan tags: trunk) | |
| 06:18 | Minor changes to checksum related test cases. (check-in: 60c22bde user: dan tags: trunk) | |
|
2010-05-28
| ||
| 11:16 | Add a test to walcksum.test to stress the checksum calculation. (check-in: 778d0c17 user: dan tags: trunk) | |
|
2010-05-26
| ||
| 17:31 | Make sure the wal-index mapping is always large enough to cover the entire active area of the wal-index. (check-in: 42705bab user: drh tags: trunk) | |
| 15:06 | Change the semantics of xShmGet() such that it will never increase the size of shared memory. xShmSize() must be used to grow the size of shared memory. A shared memory segment size cannot be shrunk (except by dropping it). (check-in: 72de0073 user: drh tags: trunk) | |
|
2010-05-25
| ||
| 15:53 | Updated header comments in wal.c. No functional code changes. (check-in: 687632a6 user: drh tags: trunk) | |
| 15:23 | Add a busy-handler to a test case in walthread.test to prevent errors. (check-in: d3d348aa user: dan tags: trunk) | |
| 13:49 | Modify walcrash2.test so that it works with DEFAULT_AUTOVACUUM=1. (check-in: 77438882 user: dan tags: trunk) | |
| 13:40 | Update header comments in wal.c to correctly describe the WAL file format. Update the locking region offsets in os_unix.c and os_win.c and add assert() statement to verify that the locking region offsets are correct. (check-in: 40030c07 user: drh tags: trunk) | |
| 10:50 | If a writer exits unexpectedly in the middle of a transaction, have the following writer remove any wal-index hash-table entries left by the interrupted transaction. (check-in: ed77556a user: dan tags: trunk) | |
| 02:24 | Remove unreachable code associated with WAL from the pager. (check-in: 54c1718e user: drh tags: trunk) | |
|
2010-05-24
| ||
| 20:27 | OOM errors during an auto-checkpoint are benign. (check-in: 3d252ce5 user: drh tags: trunk) | |
| 20:24 | Remove the walmode.test case from the inmemory_journal test of permutations.test. (check-in: 27ba9c21 user: drh tags: trunk) | |
| 17:00 | Fix the wal2.test script so that it works in auto_vacuum mode. (check-in: 6a818afb user: drh tags: trunk) | |
| 13:57 | Change the checksum used in WAL files so that each frames checksum depends on the content of the WAL header and all frame headers and content up to and including the frame to which the checksum is attached. (check-in: 8a53f12c user: dan tags: trunk) | |
| 13:28 | Make sure a WAL frame of all zeros is detected as an invalid frame. (check-in: 02d99ad4 user: drh tags: trunk) | |
| 12:34 | Fix up test_osinst.c to work with SQLITE_OMIT_VIRTUALTABLE. (check-in: 51fd3815 user: drh tags: trunk) | |
| 10:39 | Change the WAL file format to support two kinds of checksums - one that is fast to calculate on little-endian architectures and another that is fast on big-endian architectures. A flag in the wal-header indicates which the file uses. (check-in: 65ba804d user: dan tags: trunk) | |
|
2010-05-22
| ||
| 08:22 | Add a couple of missing methods to test_osinst.c.. (check-in: 5c9e9c06 user: dan tags: trunk) | |
| 00:55 | Add several EXPENSIVE_ASSERT code blocks to validate the wal-index hash table. Fix the bugs that these code blocks fine. Rename walClearHash() to walCleanupHash() and simplify its interface. (check-in: 7aade899 user: drh tags: trunk) | |
|
2010-05-21
| ||
| 19:15 | Fix another bug in walClearHash(). (check-in: 40f80ffe user: dan tags: trunk) | |
| 18:24 | Comment out the SQLITE_FCNTL_SIZE_HINT from os_unix.c since it does not seem to provide any performance gain there. (check-in: 7d01309d user: drh tags: trunk) | |
| 16:23 | Correct an off-by-one bug in the previous commit. (check-in: 75a1130d user: dan tags: trunk) | |
| 15:31 | Remove entries from wal-index hash tables when a rollback or savepoint rollback occurs. (check-in: 36795c2b user: dan tags: trunk) | |
| 13:16 | Fix the wal-index header read routine so that it correctly detects a zero header as being malformed and in need of a wal-index rebuild. (check-in: 1a4eb3a3 user: drh tags: trunk) | |
|
2010-05-20
| ||
| 23:51 | Add a new xShmBarrier method to the VFS - a shared-memory fence operation. Implement the same in both unix and win32. Use it to make the WAL subsystem more robust. (check-in: 1bd011c9 user: drh tags: trunk) | |
| 21:21 | Make use of the extra information in the WAL header and frame header to enhance robustness. (check-in: 9580ecb7 user: drh tags: trunk) | |
| 16:45 | Convert the wal-header and frame-header to 24 bytes. Extra information in both headers is designed to enhance robustness after crashes, though the extra information is currently unused. This is a snapshot of a work in progress. (check-in: 66970643 user: drh tags: trunk) | |
|
2010-05-19
| ||
| 23:41 | Merge WIN32 WAL support into trunk. Still some issues with locking to resolve. (check-in: 43377663 user: shaneh tags: trunk) | |
| 22:09 | Cherry-pick the SQLITE_FCNTL_SIZE_HINT patch (check-in [2b7e3b4a30d6a7c4a8] and bump the version number to 3.6.23.2. (check-in: 776679af user: drh tags: branch-3.6.23) | |
| 20:30 | Disable the corruptE test which SQLITE_OVERSIZE_CELL_CHECK is enabled. This is a cherrypick merge from check-in [1dc4f9fe9b83467ba]. (check-in: a40d080c user: drh tags: branch-3.6.23) | |
| 19:39 | Add the SQLITE_FCNTL_SIZE_HINT to the os_win.c VFS. (check-in: 9894118d user: drh tags: trunk) | |
| 19:26 | Add the SQLITE_FCNTL_SIZE_HINT operator to sqlite3_file_control() and use it to give the VFS hints about the ultimate size of a database file when the file is growing. (check-in: 2b7e3b4a user: drh tags: trunk) | |
| 19:09 | Fix a bug in the new checkpoint computation. Also update the checkpoint algorithm in the test scripts to align with the new implementation. (check-in: 8b6056f2 user: drh tags: trunk) | |
| 18:08 | Revise the checksumming algorithm in wal.c. More variable refactoring. (check-in: 542b90eb user: drh tags: trunk) | |
| 17:49 | Refactoring some variable names in wal.c. (check-in: 1d201ff5 user: drh tags: trunk) | |
| 01:53 | Add a large comment to wal.c describing the WAL and wal-index file formats. (check-in: a71a22b5 user: drh tags: trunk) | |
|
2010-05-18
| ||
| 23:29 | Update the wal-index hash format so that hash-table space is reused following a rollback, thus preventing hash table overflows. Add assert()s to verify that hash tables do not overfill. Further refactoring of the wal-index code. (check-in: ada9a8c7 user: drh tags: trunk) | |
| 18:01 | Refactoring of the WalIterator implementation. (check-in: b5b60fdc user: drh tags: trunk) | |
| 13:27 | Mark the shared-memory in the WAL implementation as volatile. (check-in: 0a678790 user: drh tags: trunk) | |
| 12:56 | Comment clarifications in wal.c. (check-in: a029be10 user: drh tags: trunk) | |
|
2010-05-17
| ||
| 20:16 | Remove an unreachable test from wal.c. (check-in: 7162c456 user: drh tags: trunk) | |
|
2010-05-15
| ||
| 10:24 | Add a note to the documentation of sqlite3_wal_hook() to say not to return SQLITE_ROW or SQLITE_DONE. Or any other value that is not a valid SQLite error code. (check-in: 2f0b6cbf user: dan tags: trunk) | |
| 01:09 | Silently ignore errors trying to set the default journal_mode while doing an ATTACH. (check-in: b9af4ad2 user: drh tags: trunk) | |
|
2010-05-14
| ||
| 20:15 | Fix a typo that (by bad luck) was not a syntax error but which caused some important lines of code to be skipped when SQLITE_DEBUG was not used. (check-in: 9ef99d97 user: drh tags: trunk) | |
| 19:24 | Make sure the value of an INTEGER PRIMARY KEY column supplied to triggers and especially to FK constraints really contains the ROWID and not the NULL that is stored in the column itself. Ticket [dd08e5a988d00dec]. (check-in: 636f8609 user: drh tags: trunk) | |
| 16:34 | Pull in all the latest changes from the trunk. Update the win32 SHM methods to work with the new interface design. (Closed-Leaf check-in: 4b69f2cd user: drh tags: wal-win32) | |
| 14:52 | Simplifications to the SHM implementation in os_unix.c, taking advantage of the removal of the LinuxThreads mess. (check-in: d1debe5d user: drh tags: trunk) | |
| 12:43 | Simplify os_unix.c by removing support for LinuxThreads. Linux systems must either use NPTL or else not share database connections across threads. (check-in: e294b696 user: drh tags: trunk) | |
| 11:30 | Change the OSTRACEn() macro to OSTRACE(()). Also rename some internal fields and variables in os_unix.c to be more meaningful and searchable. (check-in: 5fe529f2 user: drh tags: trunk) | |
| 10:43 | Fix test_osinst.c so that it does not depend on gettimeofday() unless SQLITE_OS_UNIX is defined and NO_GETTOD is not defined. (check-in: 2a4014b7 user: dan tags: trunk) | |
| 08:39 | Add xShmXXX methods to test_osinst.c. (check-in: b8a9b37a user: dan tags: trunk) | |
|
2010-05-13
| ||
| 20:19 | Make debugging elements of the sqlite3_mutex object volatile and make them only appear when compiling with SQLITE_DEBUG. Ticket [51914f6acd2cb462]. (check-in: e823c60c user: drh tags: trunk) | |
| 15:44 | Allow sqlite3WalCheckpoint() to be called even if the WAL is in the SQLITE_SHM_READ state, as sometimes happens after an error. (check-in: 175b296f user: drh tags: trunk) | |
| 09:40 | Remerge mistaken test_osinst.c changes. (check-in: e8c10d3c user: dan tags: trunk) | |
| 09:32 | Fix mistake in test_osinst.c. (Closed-Leaf check-in: 4c3e1378 user: dan tags: mistake) | |
| 09:11 | Make sure xShmGet is never called while already holding the shared-memory segment. (check-in: 92ebdbc5 user: drh tags: trunk) | |
| 08:53 | The refactored of VFS SHM primitives are now working so merge the wal-refactor branch back into the trunk. (check-in: bce21c18 user: drh tags: trunk) | |
| 08:33 | Change the xShmSize() implementation in os_unix.c so that it will only increase and never decrease the size of a shared-memory segment. (Closed-Leaf check-in: 149d2ae4 user: drh tags: wal-refactor) | |
| 07:08 | Fix for a segfault that can follow a malloc failure. (check-in: 3cab9022 user: dan tags: wal-refactor) | |
| 06:19 | Modify the VFS in test_vfs.c to match the refactoring of the xShmXXX methods. (check-in: 25e72f81 user: dan tags: wal-refactor) | |
|
2010-05-12
| ||
| 19:02 | (Moved to the mistake branch due to compile errors in test_osinst.c.) Change the vfs instrumentation code in test_osinst.c to make it easier to deploy. Add a vtab implementation for reading the binary log file format. (check-in: ee13c884 user: dan tags: mistake) | |
| 18:30 | Fix for the test_devsym.c test module for the VFS-SHM refactoring. (check-in: 49d62933 user: drh tags: wal-refactor) | |
| 18:10 | A couple simple fixes to get wal.test mostly working. (check-in: c7445812 user: drh tags: wal-refactor) | |
| 18:01 | Refactoring the VFS-SHM methods used by WAL. This version compiles and runs non-WAL test cases but crashes and burns on wal.test. (check-in: 2b00152c user: drh tags: wal-refactor) | |
| 17:14 | Updates to WAL support in os_win.c: pulled in latest changes from src/os_unix.c; updated tracing; misc. improvements. (check-in: 76504726 user: shaneh tags: wal-win32) | |
| 13:50 | Remove the OP_Variable optimization of check-in [48b77b04935d894] since it can lead to malfunctions as described in ticket [26ff0c82d1e90]. (check-in: 7838163d user: drh tags: trunk) | |
| 06:54 | Remove a branch made redundant by the earlier exclusive-mode changes. (check-in: c501b2ed user: dan tags: trunk) | |
|
2010-05-11
| ||
| 16:29 | Have os_unix.c use the suffix "-wal-index" for wal-index files instead of "-wal-inde". (check-in: 885e854e user: dan tags: trunk) | |
| 14:00 | When an attempt to change journal_mode fails due to locks, be sure to leave internal structures in a consistent state. (check-in: cf3d1e0b user: drh tags: trunk) | |
| 12:19 | Changes so that WAL and exclusive-locking mode work together. (check-in: 71e7b1cf user: dan tags: trunk) | |
| 02:49 | Initial port of WAL VFS support from os_unix.c to os_win.c. (check-in: 111ad59f user: shaneh tags: wal-win32) | |
| 02:46 | Updates to WAL TCL test scripts to support running on Windows. (check-in: 6a563080 user: shaneh tags: trunk) | |
|
2010-05-10
| ||
| 19:51 | Fix an off-by-one error while constructing the name of the mmap file for the wal-index under os_unix.c. (check-in: 6e3735f7 user: drh tags: trunk) | |
| 18:10 | Merge [96d6eaf4d2] and [40b0a6357b]. (check-in: c67756c4 user: dan tags: trunk) | |
| 17:29 | Fix an uninitialized variable in OSX proxy locking. (check-in: 96d6eaf4 user: drh tags: trunk) | |
| 14:46 | Modify the format of the wal-index to use a hash table to index log file segments. (check-in: 40b0a635 user: dan tags: trunk) | |
| 14:10 | If an ATTACH command fails due to OP_JournalMode but still attaches the database, make sure VACUUM still detaches it when done. (check-in: 6ecdc7ba user: drh tags: trunk) | |
| 11:20 | Enhance the OP_JournalMode opcode with an ignore-errors option and use that option the ATTACH command. (check-in: 0bdea4cf user: drh tags: trunk) | |
|
2010-05-07
| ||
| 20:34 | When running a checkpoint while in locking_mode=EXCLUSIVE, be sure to move the wal-index lock to UNLOCK from READ prior to promoting to CHECKPOINT. (check-in: be114bdf user: drh tags: trunk) | |
| 20:02 | Change the xShmOpen() method of the VFS to take the database file name as its argument, not the WAL file name. Also fix some unrelated compiler warnings. (check-in: a65c2939 user: drh tags: trunk) | |
| 18:23 | Additional tweaks to API documentation. No functional code changes. (check-in: e1230782 user: drh tags: trunk) | |
| 13:57 | Updates to documentation on the C interfaces supporting WAL. No functional changes to code. (check-in: 681ba20c user: drh tags: trunk) | |
| 13:52 | Fix a recently introduced problem in wal.test. (check-in: 79b52d0f user: dan tags: trunk) | |
| 12:49 | Change wal.test and walhook.test so that they work with DEFAULT_AUTOVACUUM=1 builds. (check-in: 4cde9290 user: dan tags: trunk) | |
| 09:43 | Test another IO error case in wal.c. (check-in: 8b3b5ab8 user: dan tags: trunk) | |
| 06:59 | Fix a problem with checkpointing large log files created by an external process. (check-in: 8f94bde5 user: dan tags: trunk) | |
| 05:46 | Since walIndexTryHdr() can no longer encounter any error conditions, change tehe function signature so that it does not return an error code. This removes unreachable branches from other code. (check-in: 061a282c user: dan tags: trunk) | |
| 00:28 | Another attempt to fix permutations.test so that it excludes WAL tests from the journaltest permutation. (check-in: 58c40442 user: drh tags: trunk) | |
|
2010-05-06
| ||
| 22:21 | Remove savepoint6.test from the set of tests run by journaltest since savepoint6.test uses journal_mode=WAL. (check-in: 811b45a9 user: drh tags: trunk) | |
| 21:37 | The PRAGMA journal_mode=WAL; command now makes WAL the default journal mode for new databases added with ATTACH, so the behavior is consistent with the other journal modes. (check-in: c3520460 user: drh tags: trunk) | |
| 20:19 | Make sure to do a clean shutdown of the library upon existing the shell in order to remove WAL files. (check-in: c1762dda user: drh tags: trunk) | |
| 19:55 | Exclude all WAL tests from the journaltest permutation. (check-in: 46c97d7e user: drh tags: trunk) | |
| 19:20 | Changes to support SQLITE_OMIT_WAL. (check-in: 32a8501d user: shaneh tags: trunk) | |
| 19:04 | Updates to the VFS SHM locking documentation. (check-in: 9927ce42 user: drh tags: trunk) | |
| 18:48 | Instead of transitioning to RECOVER state from CHECKPOINT when a recovery is required, perform the recovery while holding the CHECKPOINT lock. (check-in: bb0b6021 user: dan tags: trunk) | |
| 18:27 | If recovery is run before a checkpoint, change back to a CHECKPOINT lock before performing the actual checkpoint. (check-in: dc98ee16 user: dan tags: trunk) | |
| 18:04 | Only pass -1 as the size parameter to xShmGet when the caller does not care what size mapping is returned. Only call xShmSize when the underlying allocation (not the mapping) should be resized. (check-in: 637f51dd user: dan tags: trunk) | |
| 17:28 | Fix a bug whereby an old snapshot could be checkpointed (and subsequent transactions discarded) if the last connection to disconnect from a WAL database happended to be holding an out-of-date wal-index header. (check-in: d0c0034b user: dan tags: trunk) | |
| 16:06 | After throwing an error to say that one cannot change into WAL mode within a tranactions, do not then go and change into WAL mode. (check-in: 56a17dae user: drh tags: trunk) | |
| 15:56 | Further coverage tests. (check-in: 98a9713e user: dan tags: trunk) | |
| 15:36 | Update configure scripts for WAL support. (check-in: 2edc5129 user: shaneh tags: trunk) | |
| 14:42 | Add tests to cover a couple of branches in wal.c. (check-in: 429f437f user: dan tags: trunk) | |
| 13:36 | Unset a tcl variable before reusing it in wal.test. (check-in: e83efb23 user: dan tags: trunk) | |
| 12:15 | Add tests for some lock-contention cases. (check-in: f0fcb9c9 user: dan tags: trunk) | |
| 11:56 | Remove the noop-mutex implementations of mutex_held() and mutex_notheld() since they are both unreachable. (check-in: 6767b62a user: drh tags: trunk) | |
| 11:55 | Add two text files containing pager design notes to the doc/ subfolder. (check-in: ed817fc8 user: drh tags: trunk) | |
| 11:32 | Add test cases to test the libraries handling of corrupt wal-index headers. (check-in: 9465b267 user: dan tags: trunk) | |
| 07:43 | Add tests for handling errors returned by xShm VFS methods. (check-in: fbbcacb1 user: dan tags: trunk) | |
|
2010-05-05
| ||
| 20:00 | Change the behavior of the sqlite3_wal_hook() callback. It should now return SQLITE_OK or an error code and the error code is propagated back up the stack. If a checkpoint is desired, the callback should invoke sqlite3_wal_callback() itself. (check-in: 1b14195e user: drh tags: trunk) | |
| 19:09 | Further simplifications to the SHM locking logic. Remove the SQLITE_SHM_QUERY option. Unify the WRITE to READ and RECOVER to READ transitions. (check-in: e9bfa6e4 user: drh tags: trunk) | |
| 19:04 | Test the handling of errors returned by the xShmXXX() APIs. (check-in: 72663123 user: dan tags: trunk) | |
| 18:46 | Do not compare page sizes on source and destination of backup until transactions are started and the page sizes are locked. This is a fix to check-in [7bd44794c4]. (check-in: ec715778 user: drh tags: trunk) | |
| 18:20 | Simplifications to the SHM locking implementation in os_unix.c. (check-in: 9de05bfb user: drh tags: trunk) | |
| 16:23 | Prohibit backup if the destination is using WAL and has a different page size from the source. (check-in: 7bd44794 user: drh tags: trunk) | |
| 15:33 | When closing a WAL database, if the exclusive lock on the database file is obtained and the database successfully checkpointed, delete the wal-index file from the file system. (check-in: 2ac5d96c user: dan tags: trunk) | |
| 04:11 | Make sure the page size of the main database is fixed following a failed VACUUM attempt. Otherwise, two consecutive failed VACUUM attempts with attempted page_size changes on encrypted or WAL databases could result in database corruption. (check-in: 208e7d5d user: drh tags: trunk) | |
| 03:39 | Do not attempt to modify the page_size when running VACUUM on a WAL database. (check-in: 58a1a59b user: drh tags: trunk) | |
| 00:22 | Get the previous mutex fix working with SQLITE_DEBUG and with the amalgamation. (check-in: df19928f user: drh tags: trunk) | |
| 00:05 | When the in single-threaded mode, the sqlite3_mutex_alloc() interface still returns a non-NULL value. The mutex doesn't do anything, but it tests non-NULL. This way, extensions (or VFSes) that use sqlite3_mutex_alloc() can tell the difference between an OOM error and mutexes being disabled. (check-in: 451fd175 user: drh tags: trunk) | |
|
2010-05-04
| ||
| 18:50 | When sqlite3PagerPagecount() is called without any locks, always return the physical file size, not the logical file size. (check-in: 4016b422 user: drh tags: trunk) | |
| 17:38 | Be sure to release all wal-index locks before closing the wal-index. (check-in: b15b67fa user: drh tags: trunk) | |
| 17:20 | Relax an over-zealous assert() in sqlite3WalUndo(). (check-in: 8f9d22d5 user: drh tags: trunk) | |
| 16:33 | Take care not to invoke the xShmClose method of the VFS with a NULL shared memory object. (check-in: 1465d384 user: drh tags: trunk) | |
| 15:20 | Add a test case to verify that log files containing pages that are not a power-of-two bytes in size are handled correctly. (check-in: c2bf693f user: dan tags: trunk) | |
| 14:47 | Fix problems with recovering wal files that use a page-size other than the default. (check-in: 1a391f3c user: dan tags: trunk) | |
| 11:06 | Fix a typo in walfault.test. (check-in: 232dbe8e user: dan tags: trunk) | |
| 10:36 | Test that the correct number of padding frames are appended to the log file after committing a transaction in synchronous=FULL mode. (check-in: a60104aa user: dan tags: trunk) | |
|
2010-05-03
| ||
| 19:20 | Unset some global TCL variables prior to use in test where prior tests can have those same variables set to an array value. (check-in: 49bef00e user: drh tags: trunk) | |
| 19:17 | Fix a bug in the computation of the current time for the alternative CURRENT_TIMESTAMP functions used when SQLITE_OMIT_DATETIME is defined. (check-in: e7ded46b user: drh tags: trunk) | |
| 18:51 | Documentation stubs for sqlite3_wal_autocheckpoint and sqlite3_wal_checkpoint. (check-in: 31376242 user: drh tags: trunk) | |
| 18:22 | Fix a bug in the xCurrentTime() method of the os_unix.c VFS. (check-in: 551df11a user: drh tags: trunk) | |
| 18:01 | Add a test to simulate an OOM during log recovery to walfault.test. (check-in: 7ed85e87 user: dan tags: trunk) | |
| 17:24 | Clear global variable "seconds" before use in walthread.test. (check-in: a35551a5 user: drh tags: trunk) | |
| 17:18 | Fix a couple of errors in WAL code that can occur in an OOM situation. (check-in: 9d3f3736 user: dan tags: trunk) | |
| 16:36 | Update the crash-test VFS in test6.c to pass-through the shared-memory methods to the real underlying VFS. This fixes the walcrash.test script. (check-in: ea09ff37 user: drh tags: trunk) | |
| 16:30 | Change the VFS definition so that all methods take a VFS object pointer as their first parameter. (check-in: 43b5b07f user: drh tags: trunk) | |
| 15:58 | Have sqlite3_wal_checkpoint() handle a zero-length string in the same way as a NULL pointer. Fix "PRAGMA wal_checkpoint" so that it checkpoints all attached databases. (check-in: 7fecd21f user: dan tags: trunk) | |
| 14:32 | Use VFS method xCurrentTimeInt64 instead of xCurrentTime when it is available. Provide an implementation of xCurrentTimeInt64 for os_unix.c. (check-in: ab77b3ae user: drh tags: trunk) | |
| 14:08 | Merge the write-ahead-logging changes into the trunk. (check-in: de9ae443 user: drh tags: trunk) | |
| 14:05 | If the sqlite3_wal_checkpoint() API is passed a NULL pointer in place of a database name, attempt to checkpoint all attached databases. (Closed-Leaf check-in: 27a5c09c user: dan tags: wal) | |
| 13:37 | Make sure the mutex is held while calling sqlite3ApiExit() in sqlite3_wal_checkpoint(). Other cleanup of WAL logic. (check-in: 11a85b82 user: drh tags: wal) | |
| 12:14 | Have sqlite3_wal_checkpoint() populate the database handle error message and error code (as returned by sqlite3_errmsg() and sqlite3_errcode()). (check-in: ff234cf5 user: dan tags: wal) | |
| 11:05 | Add the "PRAGMA wal_autocheckpoint" command. Rename "PRAGMA checkpoint" to "PRAGMA wal_checkpoint". (check-in: 714e5947 user: dan tags: wal) | |
| 08:19 | Merge two wal leaves. (check-in: 23c0e6c3 user: dan tags: wal) | |
| 08:04 | Add the sqlite3_wal_checkpoint() and sqlite3_wal_autocheckpoint() APIs. (check-in: 9803196d user: dan tags: wal) | |
|
2010-05-01
| ||
| 20:17 | Change the SHM interface so that it does not take the name of the shared object but rather the name of the WAL file and derives its own name from that. Remove the xShmDelete method from the VFS and replace it with a delete flag on xShmClose. (check-in: 94dea5f9 user: drh tags: wal) | |
| 18:23 | Remove the xShmPush and xShmPull methods from the VFS. (check-in: 69e07fdb user: drh tags: wal) | |
| 17:57 | Define an invariant to guarantee deadlock-free operation of SHM in os_unix.c and check that invariant with assert() statements. (check-in: 6af2dca7 user: drh tags: wal) | |
| 17:50 | Rework mutexes on the SHM implemention for os_unix to avoid a deadlock during WAL recovery. (check-in: 1a0f69be user: drh tags: wal) | |
| 16:40 | Support compile-time option SQLITE_OMIT_WAL, for building without WAL support. (check-in: 9b230c43 user: dan tags: wal) | |
| 11:33 | Fix a typo in the walmode.test test script introducted yesterday. (check-in: dcabc90c user: drh tags: wal) | |
| 11:19 | Fix the sqlite3_mutex_alloc() interface to return NULL (not segfault) when operating in threadsafe mode. (This is a general-purpose bug fix which really ought to be ported to trunk.) (check-in: 64840a3c user: drh tags: wal) | |
| 08:30 | Add test case demonstrating deadlock during recovery of very large log files. No fix yet. (check-in: 63ea318e user: dan tags: wal) | |
| 00:59 | Do not allow journal_mode=WAL if the underlying VFS does not support xShmOpen. (check-in: d1fcccec user: drh tags: wal) | |
|
2010-04-30
| ||
| 22:28 | Merge in changes from the trunk. (check-in: 76bf0eee user: drh tags: wal) | |
| 21:03 | Avoid assertion faults in queries using indices with redundant columns. Ticket [3dbdcdb14e7f41]. (check-in: 946e06cd user: drh tags: trunk) | |
| 17:47 | Change the SHM VFS logic in os_unix.c so that it does not hold an exclusive lock n the mapped memory when also holding a CHECKPOINT lock. This improves concurrency between readers and checkpointers. (check-in: 8660cda6 user: drh tags: wal) | |
| 17:28 | Remove some obsolete debugging parameters. (check-in: a012aed4 user: drh tags: wal) | |
| 17:05 | Fix a problem with resizing the wal-index mapping after the mapping has been extended by an external process. (check-in: 79d356fe user: dan tags: wal) | |
| 16:50 | Fix a problem with calculating the required size of the wal-index when appending frames. (check-in: 9526b111 user: dan tags: wal) | |
| 16:48 | In the debugging output for SHM-LOCK in os_unix.c, use symbolic names for the lock states rather than raw numbers. (check-in: 2afc33de user: drh tags: wal) | |
| 16:41 | Fix a problem in the previous commit. Both unixShm.lockState and unixShm.readLock should be set to READ when downgrading from a WRITE to a READ lock (check-in: 90119fec user: dan tags: wal) | |
| 16:38 | Fix a bug whereby upgrading from a READ_FULL to a WRITE lock, and then back to a read lock was dropping all WAL locks. (check-in: 55f5af5e user: dan tags: wal) | |
| 16:24 | Reapply commits [837d82a929] and [c05e7dca17] that were accidentally overwritten. (check-in: 598de527 user: dan tags: wal) | |
| 16:19 | Add xShmXXX() methods to the test VFS in test_devsym.test. (check-in: 1d203424 user: dan tags: wal) | |
| 16:12 | Correctly record the fact that the SHM lock reached PENDING if it did so but failed to reach CHECKPOINT. (check-in: d9250e84 user: drh tags: wal) | |
| 15:54 | Fix a couple uninitialized variables in the xShmLock method of the unix VFS. Improved debugging logic for xShmLock. (check-in: 69567c5f user: drh tags: wal) | |
| 15:49 | When closing a WAL connection, attempt an exclusive lock on the database file. If the lock is obtained, checkpoint the database and delete the wal and wal-index files. (check-in: c05e7dca user: dan tags: wal) | |
| 15:24 | If a reader attempts to upgrade to a writer, but is not reading the most recent database snapshot, return SQLITE_BUSY. (check-in: 837d82a9 user: dan tags: wal) | |
| 14:39 | Separate the concepts of underlying storage size and mapped size in the VFS shared-memory implementation. (check-in: 4cbe49f1 user: drh tags: wal) | |
| 11:43 | Add a missing walIndexUnmap() call to the checkpoint code. Change a couple of SQLITE_CANTOPEN constants to SQLITE_CANTOPEN_BKPT. (check-in: 1f9e8c5c user: dan tags: wal) | |
| 10:24 | Fix an assert() in sqlite3WalRead(). (check-in: 2e7a0050 user: dan tags: wal) | |
| 10:06 | Add missing mutexes to unixShmClose(). (check-in: a4741cb5 user: dan tags: wal) | |
| 09:52 | Fix a couple of assert() statements in os_unix.c and wal.c. Combine sqlite3WalIsDirty() with sqlite3WalUndo(). (check-in: a8f958be user: dan tags: wal) | |
| 09:32 | Add a missing walIndexUnmap() to sqlite3WalSnapshotOpen(). (check-in: 72b95fde user: dan tags: wal) | |
| 06:02 | Merge latest bugfix into wal branch. (check-in: 265e66e6 user: dan tags: wal) | |
| 05:57 | Zero the "immediate FK constraint counter" associated with a statement object when sqlite3_reset() is called. Fix for [c39ff61c43]. (check-in: f660be61 user: dan tags: trunk) | |
| 02:13 | The first 6 WAL tests now work. It's a start. (check-in: a92c1851 user: drh tags: wal) | |
|
2010-04-29
| ||
| 22:34 | Refactor wal.c to use the VFS. This check-in compiles and links and works ok as long as you leave WAL turned off, but WAL does not work. (check-in: 62db5fa3 user: drh tags: wal) | |
| 16:40 | Untested implementation of the shared-memory dead-man-switch. (check-in: 70661128 user: drh tags: wal) | |
| 15:17 | Progress towards a VFS that will support WAL. Locking code is in place but is untested. Still no support for the DMS. (check-in: 1bde41cf user: drh tags: wal) | |
| 14:58 | Close all open database connections at the end of wal.test. (check-in: 3cc55a75 user: dan tags: wal) | |
| 14:51 | Fix a bug in the WAL checkpoint code causing SQLite to use an inconsistent cache in a subsequent transaction. (check-in: d1cadeed user: dan tags: wal) | |
| 08:47 | Add tests to walthread.test. (check-in: 9e891e75 user: dan tags: wal) | |
|
2010-04-28
| ||
| 18:17 | Add a test to walthread.test for changing between WAL and rollback modes. (check-in: da229e44 user: dan tags: wal) | |
| 17:49 | Merge two "wal" leaves. (check-in: 13d2d5a6 user: dan tags: wal) | |
| 17:48 | Change walthread.test so that tests can be run with either multiple threads or multiple processes. (check-in: 25f85f68 user: dan tags: wal) | |
| 17:21 | Changes to the interface design for the xShmLock method of the VFS. (check-in: 348409de user: drh tags: wal) | |
| 14:42 | Rename the sqlite3_log_hook() to sqlite3_wal_hook(). Added comments to wal.h. (check-in: bbc38511 user: drh tags: wal) | |
|
2010-04-27
| ||
| 18:49 | Merge two "wal" leaves. (check-in: 8c2d43ba user: dan tags: wal) | |
| 18:43 | Add a test to check that a checkpointer releases its locks before returning, even if it has been prevented by a conflicting lock from checkpointing the database. (check-in: be44349d user: dan tags: wal) | |
| 14:37 | Fix a problem with identifying a corrupt database in WAL mode. (check-in: c8658da4 user: dan tags: wal) | |
| 11:49 | Add the xShmRelease() method to the VFS. The os_unix.c implementation of the shared memory logic is still non-functional. (check-in: ed715b47 user: drh tags: wal) | |
| 06:49 | Run trans.test and avtrans.test as part of the "wal" permutation. (check-in: c8893310 user: dan tags: wal) | |
| 05:42 | Fix a problem that occurs when one process causes the log-summary file to grow and then a second process attempts to read the database. (check-in: b51a5f8b user: dan tags: wal) | |
| 01:56 | Merge in recent changes from the trunk (check-in: 7a0ac682 user: drh tags: wal) | |
|
2010-04-26
| ||
| 19:17 | Optimization: Convert an ORDER BY clause into a no-op if the query also contains a GROUP BY clause that will force the same output order. (check-in: ca9d86ba user: drh tags: trunk) | |
| 17:42 | Do not attempt to set journal_mode=wal on :memory: or temp file databases. (check-in: 30d01344 user: dan tags: wal) | |
| 17:36 | Change the default_cache_size pragma to always store a positive value. (check-in: 36fb2cae user: drh tags: trunk) | |
| 17:30 | Identify the SQLite version meta-value entry in the db-header output of showdb. (check-in: 245d934b user: drh tags: trunk) | |
| 16:57 | Fixes for problems with small caches and SAVEPOINT rollback in WAL mode. (check-in: 6a944f02 user: dan tags: wal) | |
| 16:47 | Further enhancements to the showdb utility: A page number followed by "b" causes a btree decoding to occur on the page. (check-in: 2ff824e5 user: drh tags: trunk) | |
| 15:44 | Enhancements to the showdb.c utility program. Automatically detect the page size and adjust the display accordingly. Add the "dbheader" display option. (check-in: 23eb408b user: drh tags: trunk) | |
| 12:39 | Add the "wal" permutation to run existing test files savepoint.test and savepoint2.test in WAL mode. (check-in: 205e5d8a user: dan tags: wal) | |
| 10:40 | Add mutexes to fix a race condition in wal.c. This isn't a very good fix. (check-in: 3d159939 user: dan tags: wal) | |
| 00:19 | Change the names of the log.c and log.h source files to wal.c and wal.h. (check-in: 56fe5d76 user: drh tags: wal) | |
| 00:04 | Begin moving WAL-specific I/O into the VFS. This checkin contains VFS infrastructure but it is untested and is not yet hooked up to the WAL. The version number is bumped to 3.7.0 because of the VFS extension. (check-in: f5e615c2 user: drh tags: wal) | |
|
2010-04-25
| ||
| 20:58 | Remove an unreachable branch from the function initialization logic. (check-in: 064d3ddd user: drh tags: trunk) | |
|
2010-04-24
| ||
| 19:07 | Add comment explaining checksum mechanism. (check-in: 3e9ef515 user: dan tags: wal) | |
| 18:44 | Fix bugs in WAL mode rollback. (check-in: 31215969 user: dan tags: wal) | |
| 14:33 | Merge with [0291ed974d]. Merge with [0291ed974d]. Merge with [0291ed974d]. (check-in: a352f628 user: dan tags: wal) | |
| 14:02 | When commands such as ALTER TABLE and VACUUM use SQL internally, make sure they use only the built-in functions and not application-defined overrides for those functions. (check-in: 0291ed97 user: drh tags: trunk) | |
| 04:53 | Merge with [e79dac3c2f]. (check-in: 1e793d3a user: dan tags: wal) | |
| 04:49 | Reserve some space at the start of the log-summary file to apply locks to. (check-in: a031aa1f user: dan tags: wal) | |
|
2010-04-23
| ||
| 19:15 | Fixes and tests for backup of a WAL database. (check-in: 480d12db user: dan tags: wal) | |
| 14:29 | Merge of notify.c changes from experimental. Reset stored blocking connection when registered unlock_notify callback is cleared. (check-in: e79dac3c user: shaneh tags: trunk) | |
| 11:44 | Add very simple test cases for backup and VACUUM of WAL databases. More to come. (check-in: 1077d813 user: dan tags: wal) | |
|
2010-04-22
| ||
| 19:14 | Create a version of the log checksummer that works on big-endian platforms. Remove the 512KB size limit on the log-summary. (check-in: 5d6d4423 user: dan tags: wal) | |
| 06:27 | Further tests and changes related to switching between WAL and rollback modes. (check-in: 12363184 user: dan tags: wal) | |
|
2010-04-21
| ||
| 18:37 | Tests for (and changes to) the code to switch between WAL and rollback modes. (check-in: 9f4f933f user: dan tags: wal) | |
| 11:43 | If, after obtaining a SHARED lock, there exists a *-wal file in the file-system, use WAL mode. This is necessary to recover from a crash that damages the first page of the database file. (check-in: 33cabf27 user: dan tags: wal) | |
| 06:19 | Minor changes to test cases to account for the fact that databases with read/write versions of 2 are now understood. (check-in: 278ed41e user: dan tags: wal) | |
|
2010-04-20
| ||
| 18:53 | Use the read and write version fields of the database header to mark a database as operating in wal-mode. (check-in: 96bef18c user: dan tags: wal) | |
|
2010-04-19
| ||
| 18:03 | Add the sqlite3_log_hook() interface for scheduling checkpoints. (check-in: 9bda6014 user: dan tags: wal) | |
|
2010-04-17
| ||
| 18:50 | Add some comments regarding file-locks to log.c. (check-in: 9d51c3b7 user: dan tags: wal) | |
| 17:34 | Do not sync any files in wal mode if "PRAGMA synchronous=off" is set. If files are synced, pass either SQLITE_SYNC_FULL or SQLITE_SYNC_NORMAL to the xSync() callback as configured by "PRAGMA fullfsync". (check-in: 0ae91b00 user: dan tags: wal) | |
| 16:10 | It is possible for sqlite3.pUnlockConnection to be non-NULL when sqlite3.pBlockingConnection is NULL, as the notify1.test script demonstrates. (Closed-Leaf check-in: 95ff76ef user: drh tags: experimental) | |
| 15:45 | Merge with trunk commit [3e646e3f4c]. (check-in: 43463970 user: dan tags: wal) | |
| 15:42 | In synchronous=normal mode, do not sync the log after every transaction. In synchronous=full mode, sync the log and add any extra frames required to avoid blast-radius related problems after each transaction. (check-in: 9bc9b684 user: dan tags: wal) | |
| 12:53 | Change sqlite3_step() so that it automatically calls sqlite3_reset() instead of returning SQLITE_MISUSE when invoked on a prepared statement that previously returned any value other than SQLITE_ROW. (check-in: 3e646e3f user: drh tags: trunk) | |
| 12:31 | Enhancements to wal-mode locking scheme. (check-in: 8549c286 user: dan tags: wal) | |
|
2010-04-16
| ||
| 22:05 | Changes for branch coverage of notify.c. Fixed quirk of unlock_notify() where it would still think it was blocked after a callback was cleared (even after the transaction on the blocking connection was closed). (check-in: c54e8dad user: shaneh tags: experimental) | |
| 13:59 | Change the log file format to include a small (12 byte) header at the start of the file. (check-in: 9865d14d user: dan tags: wal) | |
| 11:30 | Fix bug in log recovery (last frame in log was being ignored). Also remove an incorrect assert statement. (check-in: 67d2a89e user: dan tags: wal) | |
|
2010-04-15
| ||
| 23:24 | Disable query flattening when the outer query is a compound SELECT and the inner query has a LIMIT clause. Ticket [02a8e81d44]. (check-in: f96782b3 user: drh tags: trunk) | |
| 16:45 | Allow writers to write dirty pages to the log mid-transaction in order to free memory. (check-in: ecd828f9 user: dan tags: wal) | |
| 13:33 | Merge two leaves on the WAL branch. (check-in: c9ed66cc user: dan tags: wal) | |
| 13:29 | The query planner fix of check-in [33b1f584ef] should have been on the trunk. (check-in: f538d759 user: drh tags: trunk) | |
| 12:36 | Fix a problem in the result set size estimation logic of the query planner - a problem introduced by the two previous changes. (check-in: 33b1f584 user: drh tags: wal) | |
| 10:58 | Change the way checksums are calculated. (check-in: 84955c2e user: dan tags: wal) | |
| 02:37 | Bring over the recent query planner enhancements from the trunk. (check-in: 82969f27 user: drh tags: wal) | |
| 01:04 | Further refinements to table order selection on join query planning. (check-in: defaf0d9 user: drh tags: trunk) | |
|
2010-04-14
| ||
| 19:01 | The query planner uses non-indexable WHERE clause terms to reduce the estimated number of output rows, then uses the estimated number of output rows as a tie-breaker when choosing table order. (check-in: b87cb0c2 user: drh tags: trunk) | |
| 18:50 | Add tests and fix bugs in WAL locking mechanism. (check-in: c18077f2 user: dan tags: wal) | |
| 18:06 | Add tests to check inter-process WAL locking. (check-in: 9435f313 user: dan tags: wal) | |
| 15:49 | Improve the logLockRegion() function in log.c. (check-in: 5e9dd3bd user: dan tags: wal) | |
| 11:23 | Fixes for locking issues in WAL mode. (check-in: a9617eff user: dan tags: wal) | |
|
2010-04-13
| ||
| 19:27 | Add experimental locking scheme. (check-in: 3f958e87 user: dan tags: wal) | |
| 15:30 | Fix an uninitialized variable in readDbPage of pager.c. (check-in: f4e1150f user: drh tags: wal) | |
| 11:56 | Remove old code to interpret an argument (no longer used) to "PRAGMA checkpoint". (check-in: 27dc5977 user: dan tags: wal) | |
| 11:45 | Fix other problems in the WAL test scripts. (check-in: dcb31181 user: dan tags: wal) | |
| 11:35 | Fix test script issues preventing walthread.test from running. (check-in: d90b1dde user: dan tags: wal) | |
| 11:18 | Fix a pager bug introduced by the wal code. (check-in: 4a0b29dc user: dan tags: wal) | |
| 06:20 | Sync wal branch with [012cf101bf]. (check-in: 9d690f24 user: dan tags: wal) | |
| 06:18 | Test that the rollback-hook is invoked if a commit-hook implementation returns non-zero (causing a rollback). Remove documentation comment that says otherwise from sqlite.h.in. (check-in: 012cf101 user: dan tags: trunk) | |
|
2010-04-12
| ||
| 20:54 | Add e_fts3.test to the set of tests excluded from the inmemory_journal exclusion list, since it does simulated OOM errors which trigger I/O errors and SQLite is unable to recover from I/O errors without a persistent journal. (check-in: e7e7127f user: drh tags: trunk) | |
| 19:44 | Recent changes cause the xSync method of the memory journal implementation to be used, so remove the assert(0). (check-in: 1f2cb940 user: drh tags: trunk) | |
| 19:43 | Clear a global variable in e_fts3.test before it is used. (check-in: ae89dfd2 user: drh tags: trunk) | |
| 19:32 | Remove duplicate secure_delete pragma implemention. (check-in: c15fda14 user: drh tags: wal) | |
| 19:05 | Add "log.h", which should have been part of the previous commit. (check-in: 8de9f010 user: dan tags: wal) | |
| 19:00 | Import experimental write-ahead-logging code. (check-in: 409d61ba user: dan tags: wal) | |
| 17:08 | Fix a problem where a process in exclusive mode could delete a hot-journal file without rolling it back from within sqlite3_close() or DETACH. This problem was introduced by commits from the past few days; it is not present in any release. (check-in: 51a61395 user: dan tags: trunk) | |
| 14:51 | Reset the simulated device in the test harness to its default configuration whenever it is restarted. (check-in: 562d20e6 user: drh tags: trunk) | |
|
2010-04-10
| ||
| 17:52 | Fix issues and clarify the operation of pager_playback_one_page(). A block comment in pager.c identifies 13 invariants on the pager subsystem. Ticket [9d68c883132c8]. (check-in: 09065976 user: drh tags: trunk) | |
|
2010-04-09
| ||
| 23:05 | Add a test case for the OOM-fault corruption issue. Ticket [9d68c883132c8]. (check-in: 0a64a937 user: drh tags: trunk) | |
| 15:34 | Candidate fix for a pager bug associated with sqlite3PagerMovepage(). (check-in: 8c046eb6 user: drh tags: trunk) | |
| 09:14 | Avoid unnecessary calls to sqlite3_step() with null prepared statements while parsing the schema. (check-in: ae04d2d1 user: drh tags: trunk) | |
|
2010-04-08
| ||
| 17:35 | Fix a typo in a comment. No changes to code. (check-in: 332cb551 user: drh tags: trunk) | |
| 17:28 | Fix code coverage problems in where.c. (check-in: b04a5282 user: drh tags: trunk) | |
| 16:30 | Additional automatic index tests. (check-in: 99d8e325 user: drh tags: trunk) | |
| 15:01 | New test cases for automatic indices. New testcase() macros associated with column-used bitmasks. (check-in: e1aa48ac user: drh tags: trunk) | |
| 15:01 | Fix the computation of the number of rows in a table during automatic index generation. (check-in: edeab06a user: drh tags: trunk) | |
| 14:38 | Fix an integer overflow bug in the automatic index logic. The bug was detected by the testcase() macros added in the previous check-in. Also add some more testcase() macros, since this issue points up their importance. (check-in: af97b488 user: drh tags: trunk) | |
| 14:15 | Add testcase() macros to the automatic index logic to insure that boundary cases are tested. (check-in: 72865478 user: drh tags: trunk) | |
| 11:35 | Make sure F_OK, R_OK, and W_OK are defined in test_demovfs.c. Define them manually if they are not. This is necessary to get the module to compile on some systems. (check-in: 7c8afc32 user: drh tags: trunk) | |
| 01:16 | Merge automatic index changes into the trunk. (check-in: bf7b8d86 user: drh tags: trunk) | |
| 00:40 | When constructing automatic indices do not include the same column more than once. (Closed-Leaf check-in: d067d9f7 user: drh tags: experimental) | |
|
2010-04-07
| ||
| 20:32 | Pull over the latest changes from the trunk. (check-in: e388fe8b user: drh tags: experimental) | |
| 20:29 | When rolling back a savepoint to the beginning of the transaction, make sure to initialize the database size in the btree layer correctly even if the database size field of the header is zeroed. (check-in: a3540c6a user: drh tags: trunk) | |
| 19:32 | Add an interface to the SQLITE_STMTSTATUS_AUTOINDEX status information to the TCL bindings. Add some simple automatic index test cases. (check-in: 1f404412 user: drh tags: experimental) | |
| 16:54 | Wrap all automatic index changes inside SQLITE_OMIT_AUTOMATIC_INDEX. Add the automatic_index PRAGMA to turn it on and off. (check-in: a811a47f user: drh tags: experimental) | |
| 14:59 | Make sure that all automatic indices are covering indices. Otherwise, the table and index might be used together in a query but the table could get out of sync with the automatic index through out-of-band changes. (check-in: 23643131 user: drh tags: experimental) | |
| 14:33 | Enhance comments on the SrcList object definition to better explain the operation of the SrcList.a[].colUsed field. No changes to code. (check-in: c0f67ea1 user: drh tags: experimental) | |
| 14:32 | Fix the trace output of the OP_Function opcode to show the correct register numbers. (check-in: 064c283e user: drh tags: experimental) | |
| 07:57 | Add file test_demovfs.c, containing a simple VFS implementation that demonstrates how writes to the journal file may be safely buffered by the VFS layer to improve performance on some embedded systems. (check-in: af3b72d9 user: dan tags: trunk) | |
|
2010-04-06
| ||
| 22:33 | Veryquick.test is now working. The SQLITE_STMTSTATUS_AUTOINDEX counter added. (check-in: abbf16e5 user: drh tags: experimental) | |
| 18:51 | Runs quicktest without hitting an assert now. Some tests get unexpected results still and there is a memory leak. (check-in: a8224448 user: drh tags: experimental) | |
| 18:28 | Progress toward getting automatic indices to work. Still failing in corner cases. (check-in: ac6d0fba user: drh tags: experimental) | |
| 15:57 | Automatically generate transient indices for tables in joins that would otherwise have to use a full table scan. (check-in: 1b2a0412 user: drh tags: experimental) | |
|
2010-04-05
| ||
| 15:11 | Minor comment changes to the OP_OpenEphemeral header. No changes to code. (check-in: 8e1d7ef4 user: drh tags: trunk) | |
|
2010-04-03
| ||
| 13:08 | Silently abandon a rollback if the journal has been truncated. (check-in: 7626f6c9 user: drh tags: trunk) | |
|
2010-04-02
| ||
| 13:25 | Check for a failure in the call to sqlite3PagerPagecount() from within sqlite3PagerWrite(). (Error discovered by valgrind.) (check-in: d64d181c user: drh tags: trunk) | |
| 12:55 | Take care to avoid a buffer over-read when adding a KEYINFO structure to the P4 parameter of an opcode. (check-in: 937b931a user: drh tags: trunk) | |
| 12:46 | If the size of the database according to the header is larger than the actual database file size, report that the database is corrupt. (check-in: 8eb1f244 user: drh tags: trunk) | |
| 03:39 | Pager changes for code coverage testing. (check-in: b04a45ea user: drh tags: trunk) | |
|
2010-04-01
| ||
| 18:33 | Disable the corruptE.test script when ENABLE_OVERSIZE_CELL_CHECK is on. (check-in: 1dc4f9fe user: drh tags: trunk) | |
| 16:15 | Zero the filesize-in-header word for bigfile.tcl. Remove bigfile.tcl from the journaltest permutation. (check-in: 63610c71 user: drh tags: trunk) | |
| 13:41 | Do not journal the content of freelist leaf pages. This restores an optimization that was accidently undone by check-in [12f7346c13] on 2009-01-20. (check-in: b1e7a1a6 user: drh tags: trunk) | |
| 02:47 | Remove a NEVER in sqlite3PagerPagecount() that can in fact be true. (check-in: 8f005337 user: drh tags: trunk) | |
| 02:22 | Restore the size information in the BtShared structure when a transaction rolls back. (check-in: 802aeb30 user: drh tags: trunk) | |
|
2010-03-31
| ||
| 23:20 | Make sure an OOM error on sqlite3PagerPagecount() is detected and reported out to the interface. (check-in: 8aecf302 user: drh tags: trunk) | |
| 20:29 | Report corruption if a database contains a reference that is out of range according to the filesize-in-header database size. (check-in: a91db0b1 user: drh tags: trunk) | |
| 17:47 | Clarification of the implementation of SQLITE_STMTSTATUS_FULLSCAN_STEP. No logical code changes - just commenting and layout changes to improve readability. (check-in: 06b9ca32 user: drh tags: trunk) | |
| 15:02 | Clear the column cache before populating aggregate accumulator registers. Fix fo r [883034dcb5]. (check-in: ffc23409 user: dan tags: trunk) | |
| 13:57 | Documentation enhancements for the sqlite3_log() interface. No functional changes to code. (check-in: ba13a111 user: drh tags: trunk) | |
| 11:56 | Include shell.c and sqlite3.def in the amalgamation ZIP archive. Ticket [e063139eb3f8] (check-in: b51fcac9 user: drh tags: branch-3.6.23) | |
| 11:52 | Include shell.c and sqlite3.def in the amalgamation ZIP archive. Ticket [e063139eb3f8] (check-in: 0077ed5c user: drh tags: trunk) | |
| 02:31 | Increase test coverage following the introduction of the new filesize-in-header logic. (check-in: dbc2c3c0 user: drh tags: trunk) | |
|
2010-03-30
| ||
| 22:58 | Continuing refinements to the filesize-in-header enhancement. (check-in: 79e22b95 user: drh tags: trunk) | |
| 15:51 | Add a NEVER macro around a test in btree.c that can not longer be true. (check-in: c117f992 user: drh tags: trunk) | |
| 12:40 | Do not crash when compiling an "INSERT INTO tbl DEFAULT VALUES" when tbl has either BEFORE or INSTEAD OF triggers attached to it. Ticket [f3162063fd]. (check-in: ca359a3e user: dan tags: trunk) | |
| 11:57 | Change an #if into an #ifdef. (check-in: 371bab5d user: drh tags: trunk) | |
|
2010-03-29
| ||
| 21:22 | Pull the experimental file-size-in-header changes into the trunk. (check-in: 0092b36c user: drh tags: trunk) | |
| 21:13 | The btree layer now tracks when a database is empty at the start of a transaction and rolls back to that state. (Closed-Leaf check-in: 01ef6c19 user: drh tags: experimental) | |
| 19:36 | Fix issues with the io.test script. There is still a problem with a ROLLBACK TO not working on an initially empty database files. And much more testing is needed. (check-in: 09786e2a user: drh tags: experimental) | |
|
2010-03-27
| ||
| 17:12 | Experimental changes that cause SQLite to use bytes 28..31 of the database header to determine the database size, rather than using the actual database size. This allows database space to be preallocated. (check-in: b844ac6f user: drh tags: experimental) | |
| 09:44 | Increase the estimated cost of using a virtual table as the outer loop of a join when there exists an ORDER BY clause that is not satisfied by the virtual table. Fix for [775b39dd3c]. (check-in: 9e075e70 user: dan tags: trunk) | |
|
2010-03-26
| ||
| 22:28 | Version 3.6.23.1 (check-in: b078b588 user: drh tags: release, branch-3.6.23) | |
| 21:53 | Pull in other fixes from the trunk: check-ins [bea9258643], [f186b6a619], and [bb591802ff]. (check-in: b1f342a6 user: drh tags: branch-3.6.23) | |
| 21:48 | Fix to the crash8.test test script. (check-in: f18a129a user: drh tags: branch-3.6.23) | |
| 17:42 | Enhance FTS3 to take advantage of the MAX() optimization. Cherrypick of [b7e42ae774]. (check-in: 4b65b480 user: drh tags: branch-3.6.23) | |
| 17:40 | Correctly handle strings with zero-length tokens in the FTS3 offsets() function. This is a cherrypick of [d37034f7fc]. (check-in: ff6ae4f0 user: drh tags: branch-3.6.23) | |
| 17:37 | After any rollback that modifies the database file, sync the database before deleting the rollback journal. This is a cherry-pick merge of check-ins [b21b911f23] and [f2326dad4e] (check-in: ca0bc2a2 user: drh tags: branch-3.6.23) | |
| 01:54 | Fix OOM error on ANALYZE with STAT2 enabled with zero-length data. Ticket [cbd054fa6b]. (check-in: c33b38b5 user: shaneh tags: trunk) | |
|
2010-03-24
| ||
| 15:57 | Fix for [9861b74ab9]. Correctly handle strings with zero tokens in the fts3 offsets() function. (check-in: d37034f7 user: dan tags: trunk) | |
| 11:55 | Attempt to suppress a warning from Borland Code Guard. (check-in: be278979 user: drh tags: trunk) | |
|
2010-03-23
| ||
| 18:24 | More commenting and documentation enhancements in FTS3. (check-in: 892e2867 user: drh tags: trunk) | |
| 15:46 | Fix up obsolete comments in FTS3 to conform to the latest nomenclature. Add new comments to better explain FTS3 operation. (check-in: 3e4a0082 user: drh tags: trunk) | |
| 15:29 | Close the auxiliary database db2 at the end of the crash8.test script. (check-in: 0fbdc431 user: drh tags: trunk) | |
|
2010-03-22
| ||
| 23:55 | Sync the database file after a rollback. This is a cherrypick merge of check-in [b21b911f23]. (Leaf check-in: 82dd61fc user: drh tags: branch-3.6.22) | |
| 17:38 | Add an extra test case for crash recovery. (check-in: f2326dad user: dan tags: trunk) | |
| 17:13 | After a rollback that writes or truncates the database file, sync the database file before invalidating the journal contents. Fix for [015d3820f2]. (check-in: b21b911f user: dan tags: trunk) | |
| 15:47 | Disable the "init.test" test script when compiled with SQLITE_THREADSAFE=0. (check-in: ad68fa65 user: drh tags: trunk) | |
| 04:32 | Change the formatting of the preprocessor macros surrounding the compileoption interface in sqlite.h.in. This is so the documentation produced by parsing this part of sqlite.h.in is formatted similarly to the docs for the deprecated interfaces (with the #ifndef visible on the webpage). (check-in: 161958d1 user: dan tags: trunk) | |
|
2010-03-19
| ||
| 19:23 | Put an ALWAYS() around a condition in pcache.c that must always be true in the current implementation. (check-in: 88810cd0 user: drh tags: trunk) | |
| 19:02 | Have the xUpdate() method of the rtree module virtual table implementation set the *pRowid output variable when inserting a row. Fix for [dd3ba6063a]. (check-in: f31d2248 user: dan tags: trunk) | |
| 16:55 | Store the SQLite version number of the last writer in bytes 96..99 of the header. (check-in: 146d4f5d user: drh tags: trunk) | |
| 16:52 | Make sure the in-journal boolean vector is cleared and released when ending a transaction. This fixes an obscure problem with the previous check-in. (check-in: 69d749d9 user: drh tags: trunk) | |
| 15:48 | Do not create, open, or initialize the rollback journal until something actually needs to be written into the journal. That way, expensive filesystem operations are avoided if the transaction ends up being a no-op. (check-in: b78e58ae user: drh tags: trunk) | |
| 13:59 | Rewrite a couple of queries used internally by FTS3 to take advantage of the "SELECT max(x) FROM ..." optimization. (check-in: b7e42ae7 user: dan tags: trunk) | |
|
2010-03-18
| ||
| 16:34 | Mark internal routines as "static" in FTS3. Ticket [b1552d0fa7]. (check-in: 39b34ac4 user: drh tags: trunk) | |
|
2010-03-15
| ||
| 20:53 | Improved error message when the right-hand operand of MATCH in FTS3 is not a valid search expression. Ticket [170872f1c3]. (check-in: bea92586 user: drh tags: trunk) | |
| 19:27 | Fix FTS3 so that it does not run illegal SQL and cause a spurious log message. Ticket [42d45a693e6]. (check-in: f186b6a6 user: drh tags: trunk) | |
|
2010-03-13
| ||
| 02:15 | Store the database size in pages in bytes 28..31 of the header. Currently this is for forensic use only, but it might be used in the future. (check-in: 59f75bba user: drh tags: trunk) | |
|
2010-03-12
| ||
| 16:32 | Add assert()s to mem2.c (activated by SQLITE_MEMDEBUG) which verify that memory alloctions that might have come from lookaside are always freed using a lookaside-aware free routine. (check-in: c2af2164 user: drh tags: trunk) | |
|
2010-03-10
| ||
| 23:13 | Move the ctime.c source module near the beginning of the list of files inserted into the amalgamation so that #defines that are overridden in other modules will not effect the values returned by sqlite3_compiletime_used(). (check-in: bb591802 user: drh tags: trunk) | |
| 21:42 | Add the SQLITE_DBSTATUS_CACHE_SIZE verb for sqlite3_db_status(). The highwater mark does not yet work on this verb. (check-in: 418c6d81 user: drh tags: trunk) | |
| 20:06 | Remove the "experimental" markings on most interfaces that have been around for multiple releases. (check-in: d670e8bd user: drh tags: trunk) | |
| 14:06 | Fix typo in a comment in the sqliteInt.h. No changes to code. (check-in: 1716821d user: drh tags: trunk) | |
|
2010-03-09
| ||
| 19:31 | Version 3.6.23 (check-in: 4ae453ea user: drh tags: trunk, release) | |
| 15:10 | Fix a test script issue affecting the TCL backup tests under Windows. (check-in: 20c400e7 user: shaneh tags: trunk) | |
| 14:41 | Version 3.6.23 release candidate 4 (check-in: 3445eda0 user: shaneh tags: trunk) | |
| 14:41 | Minor changes to silence compiler warnings on various MSVC builds. (check-in: 115c978b user: shaneh tags: trunk) | |
| 13:12 | Version 3.6.23 release candidate 3 (check-in: 747baea5 user: drh tags: trunk) | |
| 13:02 | Check-in [87a6a9970b] did not completely disable FTS4. This check-in aims to finish the job. (check-in: 0259b629 user: drh tags: trunk) | |
| 12:35 | Version 3.6.23 release candidate 2 (check-in: 73f4d292 user: drh tags: trunk) | |
| 12:35 | Disable FTS4 (which is incomplete and not ready for publication) to prevent accidental use. (check-in: 87a6a997 user: drh tags: trunk) | |
| 05:01 | Version 3.6.23 release candidate 1 (check-in: 30e9f966 user: drh tags: trunk) | |
|
2010-03-08
| ||
| 21:40 | Add NEVER and assert macros and explanatory comments for unreachable conditions. (check-in: ffb6596e user: drh tags: trunk) | |
| 15:17 | Correct an assert() in mem2.c (test code only). (check-in: 1a88e31b user: dan tags: trunk) | |
| 10:32 | Do not run the stmt.test script as part of the inmemory_journal permutation. (check-in: 049cadf9 user: dan tags: trunk) | |
|
2010-03-05
| ||
| 20:17 | Make sure the dbFileVers field in the Pager object is properly initialized even if there is an I/O error while reading its content off of disk. (check-in: 81ff698f user: drh tags: trunk) | |
| 18:46 | Change a condition in sqlite3VdbeMemShallowCopy() to avoid accessing an unitialized variable (doing so was not dangerous, but caused a valgrind error). (check-in: 4793c381 user: dan tags: trunk) | |
| 17:27 | Fix a one byte buffer overread in fts3. (check-in: 6e3e014a user: dan tags: trunk) | |
| 16:32 | Modify the vdbe so that the comparison operator opcodes do not modify the data type of operands. Fix for [aa92c76cd4]. (check-in: 8858042f user: dan tags: trunk) | |
| 13:53 | Adjustments to #ifdefs so that the #include of <sys/mount.h> is not included twice, because that confuses the amalgamation builder. (check-in: 27413fc8 user: drh tags: trunk) | |
| 13:41 | Adjust some #ifdefs so that the build works on Mac both with and without SQLITE_ENABLE_LOCKING_STYLE. (check-in: 0cc981f1 user: drh tags: trunk) | |
|
2010-03-04
| ||
| 22:36 | Second thoughts: Let Tiger users set SQLITE_ENABLE_LOCKING_STYLE=0. Was: Change SQLITE_ENABLE_LOCKING_STYLE to default off, even on Macs, since older Macs (Tiger) cannot compile the latest proxy locking sources. (Closed-Leaf check-in: 1a0fa8d1 user: drh tags: mistake) | |
| 17:58 | Fix for [9abd6aa831]. Do not overread a buffer used for an %q or %w conversion with a precision specifier. (check-in: 5e472896 user: dan tags: trunk) | |
| 17:11 | Fix an uninitialized variable in printf when compiling with SQLITE_OMIT_FLOATING_POINT. (check-in: 14ad62b9 user: drh tags: trunk) | |
| 16:12 | Suppress harmless compiler warnings. (check-in: fc2c63ac user: drh tags: trunk) | |
| 00:53 | Increase the maximum length of an sqlite3_log() result string. Provide more details on the statement abort log message. (check-in: 5322d1df user: drh tags: trunk) | |
|
2010-03-03
| ||
| 22:43 | Pull the latest Lemon updates from the lemon-update-2010 branch into the trunk. (check-in: 84d760bf user: drh tags: trunk) | |
| 22:40 | Modify the sqlite3_log() interface and implementation so that it never uses dynamic memory allocation - to avoid deadlocking when called while holding the memory allocator mutex. Cherry-pick merge of [28d1bc98d6]. (check-in: 6f368b54 user: drh tags: branch-3.6.22) | |
| 22:25 | Modify the sqlite3_log() interface and implementation so that it never uses dynamic memory allocation - to avoid deadlocking when called while holding the memory allocator mutex. (check-in: 28d1bc98 user: drh tags: trunk) | |
| 17:09 | Fixed compiler warning on Visual Studio. (check-in: 83495128 user: icculus tags: lemon-update-2010) | |
| 17:06 | Whoops, that shouldn't have been in the merge. Removed buggy code. (check-in: 64372800 user: icculus tags: lemon-update-2010) | |
| 17:00 | Use Strsafe() instead of strdup(). Fixes Visual Studio complaining. (check-in: 89d8f98e user: icculus tags: lemon-update-2010) | |
| 16:54 | Merged trunk to lemon-update-2010 branch. (check-in: 88b466ba user: icculus tags: lemon-update-2010) | |
| 16:02 | Fix some extra instances of the constants addressed by [83e47ca006]. (check-in: 0354ab27 user: dan tags: trunk) | |
| 15:18 | Factor out constant expressions in sqlite3GetVarint() to work around nuisance warning messages from the RVT compiler. (check-in: 83e47ca0 user: drh tags: trunk) | |
| 14:23 | Modify the command line tool to disable all mutexes. The command line tool is single-threaded. (check-in: 00e56790 user: dan tags: trunk) | |
| 08:18 | Fix another instance of the same warning-inducing expression in util.c. (Closed-Leaf check-in: 80c64e16 user: dan tags: mistake) | |
| 08:12 | Silence a compiler warning by using a constant value instead of a constant expression that some compilers mistakenly believe causes bitshift overflow. (check-in: 587109c8 user: dan tags: mistake) | |
| 00:02 | When TEMP files are in memory, also put the massive TEMP file used by the VACUUM command in memory. This is a cherry-pick merge of [9daf4e7d07] (check-in: e5342234 user: drh tags: branch-3.6.22) | |
|
2010-03-02
| ||
| 23:34 | When TEMP files are in memory, also put the massive TEMP file used by the VACUUM command in memory. (check-in: 9daf4e7d user: drh tags: trunk) | |
|
2010-02-26
| ||
| 22:05 | fix merge error and compiler warning (check-in: 5c0afe70 user: adam tags: apple-osx) | |
| 20:54 | More double free paranoia (check-in: 62dfb44b user: adam tags: apple-osx) | |
| 16:37 | Rename the sqlite_compile_option_*() SQL functions to sqlite_compileoption_*() for consistency with the C/C++ interface. (check-in: dd4962aa user: drh tags: trunk) | |
| 15:39 | Remove the compile_option pragma (retaining compile_options - with an "s"). Updates to documentation. (check-in: 733778df user: drh tags: trunk) | |
| 15:09 | Avoid a buffer overwrite that can occur with a corrupt database if secure-delete is enabled. (check-in: 7bdb1e05 user: dan tags: trunk) | |
| 13:07 | Avoid incorrect compiler warnings by doing a couple of needless variable initializations. (check-in: 8f29490d user: drh tags: trunk) | |
| 12:58 | Increase the version number to 3.6.23 in preparation for the next release. (check-in: b5835d3e user: drh tags: trunk) | |
| 01:46 | Minor changes to silence some compiler warnings with MSVC. (check-in: 4927c62d user: shaneh tags: trunk) | |
|
2010-02-25
| ||
| 23:44 | Fix to the database connection validity checker. Other test coverage enhancements. (check-in: 0f6291bb user: drh tags: trunk) | |
| 21:27 | Test coverage enhancements. Additional documentation detail on the new sqlite3_log() interface. (check-in: d986e928 user: drh tags: trunk) | |
| 19:09 | Add experimental fix for corruption detection problem. This may well be revised yet. (check-in: 1cc4be7e user: dan tags: trunk) | |
| 16:09 | Fixed numbering of a few tests; minor tweaks on others; added a couple new tests to stress previous simplifications to compile-time option reporting functions. (check-in: 9b18dfd1 user: shaneh tags: trunk) | |
| 15:28 | Minor simplifications to the compile-time option reporting functions and pragmas to facilitate coverage testing. (check-in: 633d8747 user: drh tags: trunk) | |
| 14:56 | Expire pragma statements when reset, even if they were not run to completion. (check-in: b8fbf427 user: drh tags: branch-3.6.22) | |
| 14:47 | Expire pragma statements when reset, even if they were not run to completion. (check-in: 78351d28 user: drh tags: trunk) | |
| 04:15 | Make sure the ON clause of a LEFT JOIN does not cause an index to be used to speed access to a table to the left of the join. Ticket [ebdbadade5] (check-in: f0ae251a user: drh tags: trunk) | |
| 02:32 | Merge in all of the logging enhancements. This is a cherrypick merge of the following check-ins: [103321e37a], [a8076aede3], [6d910245ad], [7c4cca6d1a], [edea3bb740], [1a6d4bb130], [a8c984c1d6], [69a493182f], and [1168763d2c]. (check-in: 46f406b2 user: drh tags: branch-3.6.22) | |
| 02:11 | Fix a segfault that is possible for an OOM fault in ATTACH when in shared-cache mode. This is a cherrypick merge of check-in [e37137376a]. (check-in: f73cc0e9 user: drh tags: branch-3.6.22) | |
| 02:04 | Enhance integrity_check to detect out-of-order rowids. This is a cherrypick merge of check-in [cae47c5b09]. (check-in: ecd22ef3 user: drh tags: branch-3.6.22) | |
| 01:54 | Merge in the fix for VACUUM on auto_vacuumed databases with large schemas. Check-in [86d50ce57f]. Ticket [da1151f97df]. (check-in: ce624e39 user: drh tags: branch-3.6.22) | |
| 01:47 | Start a new branch for release 3.6.22 plus limited enhancements and fixes. This first check-in includes the performance enhancements of check-ins [26cb1df735], [61a2c8d4d6], and [27dc5b1c52] (check-in: 95882598 user: drh tags: branch-3.6.22) | |
|
2010-02-24
| ||
| 21:44 | Add testcase() macros beside each sqlite3_log() call to make sure it is tested with both logging enable and disabled. (check-in: 1168763d user: drh tags: trunk) | |
| 19:36 | Changes to compile time option diags to report values for some defines. Added test cases to TCL test suite (ctime.test). (check-in: dd480f62 user: shaneh tags: trunk) | |
| 19:23 | Add a sqlite3_log() call on anonymous constraint failures. Fix the output of test cases having to do with improved reprepare reporting. Fix the VACUUM command to report more helpful error messages when things go wrong. (check-in: 69a49318 user: drh tags: trunk) | |
| 18:40 | Fix an incorrect ALWAYS() macro in vdbeapi.c. Fix the output of a few test cases that changed due to better error propagation out of reprepare. (check-in: a8c984c1 user: drh tags: trunk) | |
| 18:25 | Remove an unused variable from rtree. (check-in: 8e60d399 user: drh tags: trunk) | |
| 17:15 | Enhancements to the way errors are reported up when an automatic statement reprepare fails. (check-in: 1a6d4bb1 user: drh tags: trunk) | |
| 15:10 | Fix an assertion fault that occurs when two different virtual tables are used in a single UPDATE statement. Ticket [d2f02d37f52b]. (check-in: ff61e0fd user: drh tags: trunk) | |
|
2010-02-23
| ||
| 21:08 | Fix the stmt.test test script so that it works with SQLITE_TEMP_STORE=3. (check-in: 8bf710ce user: drh tags: trunk) | |
| 20:32 | Merge the compile-time option introspection interfaces into the trunk. (check-in: 9f429434 user: drh tags: trunk) | |
| 20:11 | Continuing improvements to error reporting and the sqlite3_log() routine. (check-in: edea3bb7 user: drh tags: trunk) | |
| 20:08 | Refactored compile time option diagnostic support to use strings identifiers instead of bitmask. (Closed-Leaf check-in: 76a0294a user: shaneh tags: compile_opts) | |
| 18:05 | Fix a segfault that can occur if a malloc fails in ATTACH in shared-cache mode. (check-in: e3713737 user: dan tags: trunk) | |
| 17:36 | Remove the obsolete sqlite3SafetyOn() mechanism. Add additional logging output for CORRUPT, and CANTOPEN errors. (check-in: 7c4cca6d user: drh tags: trunk) | |
| 10:56 | Fix a segfault that can occur if a malloc fails in ATTACH in shared-cache mode. (check-in: 875f8fa3 user: dan tags: compile_opts) | |
| 05:17 | Removed checking of some compile options from the sqlite3_compileopts() API. (check-in: 833da702 user: shaneh tags: compile_opts) | |
| 04:19 | Test new api to report which options (defines) were used to compile SQLite. (check-in: 84c97569 user: shaneh tags: compile_opts) | |
| 01:47 | Add the ".log" command to the shell. This shows that some errors are logged multiple times and other errors are never logged at all. (check-in: 6d910245 user: drh tags: trunk) | |
| 01:01 | Remove the ".genfkey" command from the shell. We've had foreign-key support in the core since 3.6.19. Anyone who still needs the simulated foreign-key triggers can always generate them with an older version of the shell. (check-in: c4401fc9 user: drh tags: trunk) | |
|
2010-02-22
| ||
| 23:17 | Fix an assertion-fault/segfault problem that comes up when trying to VACUUM an auto-vacuumed database with a large schema. Ticket [da1151f97df244]. (check-in: 86d50ce5 user: drh tags: trunk) | |
| 19:37 | Merge in the massive clean-up and ANSI-fication of Lemon carried out by Ryan Gordon. There are no functional changes to SQLite itself - Lemon still generates exactly the same parsing automaton from exactly the same grammar. (check-in: 1e8b8420 user: drh tags: trunk) | |
| 19:32 | Log all error messages if logging is enabled. (check-in: a8076aed user: drh tags: trunk) | |
|
2010-02-19
| ||
| 04:28 | Changes to pragma integrity_check to check rowid order. Tests of same in corruptE.test. (check-in: cae47c5b user: shaneh tags: trunk) | |
|
2010-02-18
| ||
| 18:45 | Add a new, experimental logging interface designed to aid in debugging of deeply embedded projects that use SQLite. (check-in: 103321e3 user: drh tags: trunk) | |
| 08:19 | Allow statements like "REPLACE INTO tbl(rowid) VALUES(...)" to run without a statement journal as long as there are no triggers, foreign keys or indexes. (check-in: 0e422580 user: dan tags: trunk) | |
|
2010-02-17
| ||
| 20:31 | Report error if the grammar has multiple %type lines for the same nonterminal. (check-in: 721f33e7 user: icculus tags: lemon-update-2010) | |
| 20:22 | Fixed % formatting in some printf-style strings. (check-in: f96add89 user: icculus tags: lemon-update-2010) | |
| 20:19 | Corrected error message (cut-and-paste bug). (check-in: b3839f2a user: icculus tags: lemon-update-2010) | |
| 19:34 | Merge latest changes from the trunk into the apple-osx branch. (check-in: 4b72e39b user: drh tags: apple-osx) | |
| 17:48 | Consistent use of #ifdef for SQLITE_HAS_CODEC to avoid confusion. (check-in: 34a3413a user: shaneh tags: trunk) | |
| 05:40 | Removed unused variables. (check-in: ca570a02 user: icculus tags: lemon-update-2010) | |
| 04:19 | Avoid using the internal printf routine for round(x,y) in the common case where y==0. (check-in: d76ad8b3 user: shaneh tags: trunk) | |
| 03:57 | More rounding tests. (check-in: 3863638b user: shaneh tags: trunk) | |
|
2010-02-16
| ||
| 22:06 | Remove duplicated assignment in realloc. (check-in: c419955d user: shaneh tags: trunk) | |
| 22:00 | Added additional rounding tests. (check-in: b8ee76bc user: shaneh tags: trunk) | |
| 16:09 | Removed %expect directive, on drh's advice. (check-in: b43ac330 user: icculus tags: lemon-update-2010) | |
| 16:07 | Patched to compile. Accidentally removed va_list declaration. (check-in: 673d470c user: icculus tags: lemon-update-2010) | |
| 12:18 | Change a C++ style comment in os_unix.c to use normal C style commenting. (check-in: 7a193309 user: dan tags: trunk) | |
| 10:59 | Fix for [e9a9fde1f4]. When opening an existing rtree, determine the node size by inspecting the root node of the r-tree structure (instead of assuming it is a function of the page-size). (check-in: ebc9433f user: dan tags: trunk) | |
|
2010-02-15
| ||
| 18:03 | Fix the ALTER TABLE RENAME command so that it converts FOREIGN KEY constraints in ATTACH-ed and in TEMP tables as well as in the main database. Ticket [13336e9c3c8c3f]. (check-in: ab197d0a user: drh tags: trunk) | |
| 16:54 | When creating a trigger on a main database table when there is a TEMP table with the same name, make sure the trigger is bound to the main table. Ticket [985771e11612]. (check-in: ec914af3 user: drh tags: trunk) | |
| 15:47 | Fix a compiler warning in shell.c. Updates to comments in trigger.c. (check-in: c727601e user: drh tags: trunk) | |
| 15:17 | Fix a typo in the documentation for the OP_OpenPseudo opcode. (check-in: 9acfb069 user: drh tags: trunk) | |
| 00:01 | Removed the 80-char line formatting for ErrorMsg(), on dhr's suggestion. (check-in: 132ea4e5 user: icculus tags: lemon-update-2010) | |
|
2010-02-14
| ||
| 17:14 | Make Lemon able to compile as C++ code. (check-in: 75a1060b user: icculus tags: lemon-update-2010) | |
| 05:42 |
Added option to not clip error output.
This is useful for IDEs and other tools that benefit from full path information, so they can jump directly to the error line in the source code. (check-in: 90602030 user: icculus tags: lemon-update-2010) | |
| 05:34 |
Delete output files if this is a failed run.
Otherwise, the fail will stop a Makefile from progressing, but if you immediately run the build again, Make will think the output files are up to date, since they are newer (albeit incomplete/incorrect). (check-in: e38c08d9 user: icculus tags: lemon-update-2010) | |
| 05:19 |
Added %expect directive, to consider a certain number of conflicts "correct."
This has the side effect of changing the process exit code to never overflow. (check-in: d8bab8cf user: icculus tags: lemon-update-2010) | |
| 00:48 |
Added -T option, to specify a template filename on the command line.
The default is still "lempar.c", though. (check-in: e6cbe1e5 user: icculus tags: lemon-update-2010) | |
|
2010-02-13
| ||
| 02:31 | Merged tracing and initialization changes from mutex_unix.c into mutex_w32.c. (check-in: 942aa1f6 user: shaneh tags: trunk) | |
|
2010-02-12
| ||
| 22:43 | Added experimental mutex implementation using interlocked ops and spin locks. (Closed-Leaf check-in: f3676b24 user: shaneh tags: experimental) | |
| 19:46 | Enhancements to the secure_delete pragma to make it easier to use. (check-in: 2bb38bb9 user: drh tags: trunk) | |
| 18:18 | Allow the secure-delete setting to be changed at run-time using a pragma. The SQLITE_SECURE_DELETE compile-time option determines the default setting. (check-in: f72f8a87 user: drh tags: trunk) | |
|
2010-02-10
| ||
| 05:33 | Fix a segfault in fts3 that may occur if the snippet, offsets or matchinfo functions are used in a query that does not contain a MATCH clause. (check-in: 14ab536a user: dan tags: trunk) | |
|
2010-02-05
| ||
| 18:00 | Make sure file descriptors are closed before unlinking in VxWorks. (check-in: f64b81f1 user: drh tags: trunk) | |
| 16:28 | Avoid truncating non-in-memory sub-journals when releasing a savepoint for a small performance improvement. (check-in: 27dc5b1c user: shaneh tags: trunk) | |
| 14:12 | Remove the use of 64-bit math in the offset computations of the OP_Column opcode for a small performance improvement. (check-in: 61a2c8d4 user: drh tags: trunk) | |
|
2010-02-04
| ||
| 17:38 | Fix a performance glitch that appears for large transactions. (check-in: 26cb1df7 user: drh tags: trunk) | |
|
2010-02-03
| ||
| 19:55 | Add a new full-text search variant that tracks the total number of documents and document sizes, to make ranking search results easier. Currently called FTS4. (check-in: 1b6e6094 user: drh tags: trunk) | |
|
2010-02-01
| ||
| 15:47 | Add log.c. (Closed-Leaf check-in: 7b31373f user: dan tags: wal) | |
| 15:47 | Create new branch named "wal" (check-in: 784b718e user: dan tags: wal) | |
|
2010-01-31
| ||
| 15:14 | Fix a C++-ism that snuck into sqlite3_finalize(). (check-in: fb96f2a2 user: drh tags: trunk) | |
| 14:18 | If an OOM occurs while setting the page size for the TEMP database, be sure that error gets reported back out to the interface layer. (check-in: 6487e70a user: drh tags: trunk) | |
|
2010-01-28
| ||
| 19:56 | Tweaks for consistency to the SEE and CEROD API declarations in sqlite3.h. (check-in: 299f7456 user: shaneh tags: trunk) | |
|
2010-01-26
| ||
| 01:25 | Make the TEMP file tables use the page size set for the main database. Ticket [b80eeab588c4]. Also copy over the changes from apple-osx check-in [7c3bede3f2]. (check-in: 5dcfb0c9 user: drh tags: trunk) | |
| 01:14 | Updated open mask to include SQLITE_OPEN_AUTOPROXY (check-in: 7c3bede3 user: adam tags: apple-osx) | |
|
2010-01-22
| ||
| 15:48 | Fix two similar problems in fts3 that meant that an OOM error could cause a memory leak. (check-in: 701ef64b user: dan tags: trunk) | |
|
2010-01-21
| ||
| 23:11 | Fix a segfault that can occur when the LHS of a LIKE operator has an undefined collating sequence. Ticket [1258875e07553]. (check-in: a82e6b45 user: drh tags: trunk) | |
| 01:53 | Redesign the string to numeric value caster so that it is more likely to work on unusual floating point hardware. (check-in: 8bb1104c user: drh tags: trunk) | |
|
2010-01-20
| ||
| 14:25 | Fix a problem with handling OOM errors in fts3. (check-in: f9c54e95 user: dan tags: trunk) | |
| 13:20 | Align the os_unix.c source file with the version found on trunk. (check-in: fa0f6c14 user: drh tags: apple-osx) | |
| 13:07 | Move the Apple OS-X VFS changes into the trunk. (check-in: 571594bf user: drh tags: trunk) | |
| 01:26 | Update the Apple OS-X branch to include all of the latest changes in trunk. (check-in: 96499b1d user: drh tags: apple-osx) | |
| 01:20 | Manually copy over the rebustness fixes from the apple-osx branch. (check-in: 095c74ea user: drh tags: trunk) | |
|
2010-01-19
| ||
| 23:50 | robustness fixes for preventing a finalized statement from being reused (check-in: a7a0c8d6 user: adam tags: apple-osx) | |
|
2010-01-18
| ||
| 18:17 | Fix a problem with the CLI where not all SQL commands were being echoed. Added tests of same to tools/shell2.test. Ticket [eb620916be]. (check-in: 7080ae3b user: shaneh tags: trunk) | |
|
2010-01-15
| ||
| 17:33 | Fix test numbering in fts3snippet.test. (check-in: 34f68546 user: dan tags: trunk) | |
| 17:25 | Add coverage tests for fts3_snippet.c. Also fixes related to the same. (check-in: 5e9d8cca user: dan tags: trunk) | |
|
2010-01-14
| ||
| 11:45 | Add a test to e_fts3.test for the matchinfo example in fts3.html. (check-in: e5336edc user: dan tags: trunk) | |
| 11:17 | Fix some matchinfo related test problems in fts3rnd.test. (check-in: da7abe05 user: dan tags: trunk) | |
| 00:39 | Comment cleanup in sqliteInt.h. (check-in: 67c3aea5 user: drh tags: trunk) | |
|
2010-01-13
| ||
| 17:39 | Make the sqlite3IsNaN() function a macro that evaluates to FALSE when compiled with SQLITE_OMIT_FLOATING_POINT. (check-in: 56f232d5 user: drh tags: trunk) | |
| 16:43 | When SQLITE_OMIT_FLOATING_POINT is defined, the floating-point formats in the sqlite3_*printf() functions should pull an int64 off of the parameter list and ignore it. (check-in: 3fd6f9ad user: drh tags: trunk) | |
| 16:25 | When SQLITE_OMIT_FLOATING_POINT is defined, make sure the result of a mathematical operation is always tagged as an integer. (check-in: e12da0d3 user: drh tags: trunk) | |
| 15:15 | Make the doubleToInt64() routine a pass-through when using OMIT_FLOATING_POINT. (check-in: 41716718 user: drh tags: trunk) | |
| 14:08 | Add tests to backup.test to verify that SQLite behaves as expected when the source database is modified mid-backup. (check-in: 985d3bec user: dan tags: trunk) | |
| 04:22 | Comment out unused code when SQLITE_OMIT_FLOATING_POINT is selected. Added testcase() macros to boundary conditions in util.c routines. (check-in: dfc6595d user: drh tags: trunk) | |
| 00:04 | Make the sqlite3_result_double() interface work the same as sqlite3_result_int64() when SQLITE_OMIT_FLOATING_POINT is defined. (check-in: 61df5986 user: drh tags: trunk) | |
|
2010-01-12
| ||
| 23:54 | Make sure internal queries are well ordered, even when the reverse_unordered_selects pragma is in effect. Ticket [eb942c64a157]. (check-in: 0209f171 user: drh tags: trunk) | |
| 19:28 | Fix the implementation of CURRENT_TIMESTAME and CURRENT_DATE when SQLITE_OMIT_DATETIME_FUNCS is defined. (check-in: eb98265b user: drh tags: trunk) | |
| 17:57 | Update comments in fts3_snippet.c. (check-in: dd9689b7 user: dan tags: trunk) | |
| 17:04 | Use #ifdefs to disable unused code when SQLITE_OMIT_FLOATING_POINT is defined. (check-in: 66bab856 user: drh tags: trunk) | |
|
2010-01-11
| ||
| 18:26 | Add a few documentation evidence comments to the built-in function implementations. (check-in: 8bd0f814 user: drh tags: trunk) | |
| 12:00 | Modify snippets code to run more efficiently. And to avoid a bug relating to snippets based on full-text queries that contain duplicate terms. (check-in: a2b1183d user: dan tags: trunk) | |
|
2010-01-09
| ||
| 07:33 | Fix handling of an OOM error in the fts3 offsets() function. Fix a couple of snippet related test cases in e_fts3.test. (check-in: 14dc46a7 user: dan tags: trunk) | |
|
2010-01-08
| ||
| 23:01 | Update comments in fts3.c to more accurately describe the doclist format. (check-in: e424a030 user: drh tags: trunk) | |
| 04:50 | Added option to restore_jrnl.tcl utility to hex dump journal pages. (check-in: 08c545f0 user: shaneh tags: trunk) | |
|
2010-01-07
| ||
| 22:02 | Minor tweaks to restore_jrnl.tcl utility script. (check-in: b97aca12 user: shaneh tags: trunk) | |
| 21:49 | Small tool that attempts to repair a journal header. (check-in: 05b18b5f user: shaneh tags: trunk) | |
| 15:17 | Fix the expression comparison logic to take the COLLATE operator into account. Ticket [360c6073e197] (check-in: 44bb1bfe user: drh tags: trunk) | |
| 11:27 | Changes to test code so that testfixture compiles when OMIT_SHARED_CACHE and OMIT_UTF16 are defined. (check-in: d6ee5ff6 user: dan tags: trunk) | |
| 10:54 | Fixes to problems in FTS3 snippet() function found by th3 tests. (check-in: 3b5ccd26 user: dan tags: trunk) | |
| 03:53 | Another attempt at fixing the table generator in lemon. Again, this does not effect the SQLite grammar. (check-in: e22c090f user: drh tags: trunk) | |
|
2010-01-06
| ||
| 18:36 | Fix a segfault that can occur following an OOM in the FTS3 snippet() function (check-in: c7e5966e user: dan tags: trunk) | |
| 17:19 | Change the fts3 snippet function to return (hopefully) more relevant snippets in less time. (check-in: 8a208223 user: dan tags: trunk) | |
| 13:12 | Update the OS-X branch to include all trunk changes through version 3.6.22. (check-in: 541e2b48 user: drh tags: apple-osx) | |
| 13:07 | Fix an issue with lemon generating incorrect grammars. This issue does not effect SQLite. (check-in: 077a6bee user: drh tags: trunk) | |
| 00:00 | Version 3.6.22 (check-in: 28d0d771 user: drh tags: trunk, release) | |
|
2010-01-05
| ||
| 14:18 | Removed compiler warnings in MSVC. (check-in: 4ecb4725 user: shaneh tags: trunk) | |
| 13:40 | Make sure new pages are zeroed even when loading a freelist page using the noContent option. This prevents a harmless valgrind warning. (check-in: e47e2133 user: drh tags: trunk) | |
| 04:59 | Fix a problem in the shell tool. In some cases sqlite3_errmsg() was being called before sqlite3_finalize(), causing error messages to be more generic than they should be. (check-in: e5d07045 user: dan tags: trunk) | |
| 03:30 | In the debugging memory allocator, initialize new memory allocations to pseudo-randomness in an effort to find problems with memcmp() of structures that have uninitialized pad bytes. (check-in: 6462817b user: drh tags: trunk) | |
| 00:14 | Fix a case in os_unix.c where two structures that might have uninitialized padding bytes are compared using memcmp(). (check-in: e02f2556 user: drh tags: trunk) | |
|
2010-01-04
| ||
| 13:30 | Version 3.6.22 Release Candidate 1 (check-in: 9d8ab0f1 user: drh tags: trunk) | |
|
2010-01-02
| ||
| 19:02 | Add experimental implementation of FTS3 functions matchinfo() and snippet() (not enabled by default). (check-in: 51f7ee84 user: dan tags: trunk) | |
| 03:46 | Fix a bug in the new sqlite3_test_control case of the previous check-in. (check-in: 3b77701b user: drh tags: trunk) | |
| 03:21 | Add a new sqlite3_test_control() verb that facilitates testing that all keywords are shown in the documentation. (Two keywords were found to be missing while testing the change.) (check-in: d3cdc4b1 user: drh tags: trunk) | |
|
2010-01-01
| ||
| 18:57 | Add evidence marks associated with autoincrement. (check-in: 0e918c54 user: drh tags: trunk) | |
|
2009-12-31
| ||
| 20:35 | Add comments and an assert() to help clarify the operation of the sqlite3VdbeList() routine used to implement EXPLAIN. (check-in: e1ccdb93 user: drh tags: trunk) | |
| 19:48 | Minor changes to test scripts to support various SQLITE_OMIT options. (check-in: 97f8a886 user: shaneh tags: trunk) | |
| 19:06 | Changes to remove warnings in MSVC build. (check-in: 6cf76c2a user: shaneh tags: trunk) | |
| 15:51 | Change evidence marks to track requirement number changes in the foreign-key documentation. (check-in: 6b4fcd81 user: drh tags: trunk) | |
|
2009-12-30
| ||
| 14:19 | Change the version number to 3.6.22. (check-in: 96919a46 user: drh tags: trunk) | |
| 14:12 | Adjustments to column cache handling in order to restore 100% branch test coverage. (check-in: cc6b959b user: drh tags: trunk) | |
| 01:13 | Remove some code in the column cache that is no longer used. Replace it with an assert(). (check-in: 1f890efb user: drh tags: trunk) | |
| 00:12 | Remove obsolete comment text from expr.c. No functional code changes. (check-in: 84058be0 user: drh tags: trunk) | |
|
2009-12-29
| ||
| 23:39 | Within the special new.* and old.* tables of a trigger, recognize all the original table names even if those names overload the "rowid", "oid", or "_rowid_" special names. Ticket [34d2ae1c6d0]. (check-in: 1a0e5fa9 user: drh tags: trunk) | |
| 10:32 | Fix a bug triggered by optimizing an FTS3 table when there are no segments on disk but pending terms in the hash table. (check-in: a3b50e4f user: dan tags: trunk) | |
|
2009-12-24
| ||
| 16:00 | Immediately purge entries from the column cache when the associated register undergoes an affinity change. Ticket [eb5548a849]. Enhance the SQLITE_TESTCTRL_OPTIMIZATIONS setting of sqlite3_test_control so that it can disable the column cache for testing purposes, in an effort to prevent future problems of a similar nature to this one. (check-in: ea4e57e1 user: drh tags: trunk) | |
|
2009-12-23
| ||
| 18:06 | Fix the main.mk makefile so that the sqlrr extension is built into the amalgamation correctly. (check-in: a3204d8a user: drh tags: apple-osx) | |
|
2009-12-22
| ||
| 23:52 | Move the query flattener turn-off from a pragma to an sqlite3_test_control() call. Make provisions (not yet implemented) to turn off other optimizers using the same call. (check-in: 4a97c623 user: drh tags: trunk) | |
| 18:56 | Add the experimental FTS3 matchinfo() function. Provides details of the match that may be used for result ranking and other purposes. (check-in: 37a1de02 user: dan tags: trunk) | |
| 00:29 | Add a pragma to disable the query flattener - for use during testing. (check-in: 1d8550e5 user: drh tags: trunk) | |
|
2009-12-21
| ||
| 08:53 | Change many comments in e_fkey.test to include the full text of the corresponding statement in foreignkeys.html. (check-in: 4b489ecb user: dan tags: trunk) | |
|
2009-12-20
| ||
| 15:00 | Fix a problem in FTS3 phrase queries. Add tests to check that it really is fixed. (check-in: 956de051 user: dan tags: trunk) | |
|
2009-12-17
| ||
| 22:17 | Modified statement machine in sqlite3_complete() to return 0 on empty string. Added/updated tests for same. Ticket [356c885b0b]. (check-in: 76eca729 user: shaneh tags: trunk) | |
| 22:12 | Fixed some TCL test cases to work if SQLITE_OMIT_TRIGGER is defined. (check-in: 85e3c734 user: shaneh tags: trunk) | |
| 21:07 | Updated and added new test cases for CLI. (check-in: 29e3c8da user: shaneh tags: trunk) | |
| 21:07 | In shell.c (CLI), modified local_getline() to remove '\r' as well as '\n' from end of lines. This provides consistency between MSVC and CYGWIN builds. (check-in: b1b48be1 user: shaneh tags: trunk) | |
| 21:05 | Fix typo in comment of prepare.c. (check-in: 333c3ffe user: shaneh tags: trunk) | |
| 03:49 | Disable trace when recursively running commands while vacuuming. (check-in: 69a1348a user: drh tags: trunk) | |
| 02:13 | Change the REGEXP function in ICU to require exactly 2 arguments. (check-in: c34cf23e user: drh tags: trunk) | |
|
2009-12-16
| ||
| 23:46 | Merge the latest changes on trunk (and especially the fix for the (xANDy)OR(z) bug) into apple-osx. (check-in: 5754a3a5 user: drh tags: apple-osx) | |
| 23:43 | Remove unreachable code that was added by the fix to the "(xANDy)OR(z)" bug in the previous check-in. (check-in: 04915562 user: drh tags: trunk) | |
| 23:28 | Fix for the "(x AND y) OR z" bug backported to version 3.6.21. (Leaf check-in: 86734732 user: drh tags: branch-3.6.21) | |
| 22:10 | Ensure WHERE clause terms involving tables on the right end of a join are not prematurely evaluated when tables on the left end of the join make use of the OR-clause optimization. Fix for ticket [31338dca7e]. (check-in: 2c2de252 user: drh tags: trunk) | |
| 14:49 | Modify some comments in e_fkey.test. No changes to code or tests. (check-in: 1ba4f412 user: dan tags: trunk) | |
|
2009-12-14
| ||
| 17:42 | Add a few simple evidence comments to the tokenizer. No functional changes. (check-in: 0daec309 user: drh tags: trunk) | |
| 15:17 | Handle an unhandled OOM error code that occurs in fts3 with SQLITE_TEST defined. (check-in: 82b06999 user: dan tags: trunk) | |
| 14:49 | Fix a broken assert() in fts3_write.c. Also fix a couple of compiler warnings in fts3 code. (check-in: a730a05b user: dan tags: trunk) | |
|
2009-12-13
| ||
| 23:03 | Additional corrections to documentation comments. No functional code changes. (check-in: b3740b0a user: drh tags: trunk) | |
| 22:20 | Minor documentation updates. No functional changes. (check-in: 6ae7e40b user: drh tags: trunk) | |
|
2009-12-12
| ||
| 23:57 | Fix some documentation comments in sqlite.h.in. No functional code changes. (check-in: c16b9bec user: drh tags: trunk) | |
| 19:15 | Tests to cover a few extra branches in fts3.c. (check-in: 06b72b00 user: dan tags: trunk) | |
| 16:04 | Further fts3 coverage tests. (check-in: d2a8c0f6 user: dan tags: trunk) | |
| 13:58 | Rename tkt-d82e3f3721.txt to use the (correct) .test suffix. (check-in: 68cccd62 user: drh tags: trunk) | |
| 13:16 | Extra tests for coverage of fts3 code. (check-in: eee921a9 user: dan tags: trunk) | |
| 09:51 | Add coverage test cases for fts3. (check-in: 8fcb0478 user: dan tags: trunk) | |
|
2009-12-11
| ||
| 23:11 | Additional changes to C-language interface documentation. (check-in: 1342916f user: drh tags: trunk) | |
| 16:03 | Change the fts3 test interface used to configure the advisory node size parameter. (check-in: 87fc0ce1 user: dan tags: trunk) | |
| 12:29 | Rationalize some code in fts3 used by optimize operations, queries of the pending-terms hash table and segment merges. Add the "INSERT INTO tbl(tbl) VALUES('optimize')" syntax. (check-in: 29476da3 user: dan tags: trunk) | |
| 07:07 | Add comment to fts3rnd.test to explain how the test works. (check-in: 6b740c7c user: dan tags: trunk) | |
| 03:44 | Extensive edits to the comments in the sqlite.h.in source file to identify testable statements of truth about the C-language interface. (check-in: ea884e1e user: drh tags: trunk) | |
|
2009-12-10
| ||
| 21:11 | Avoid declaring and freeing test variables if SQLITE_TEST not defined. (check-in: c2d22960 user: shaneh tags: trunk) | |
| 18:38 | Remove some unused macros and types from fts3.c. (check-in: 83a80efe user: dan tags: trunk) | |
| 18:29 | Fix handling of "WHERE col MATCH ? AND docid = ?" clauses in fts3. (check-in: 6cbbae84 user: dan tags: trunk) | |
| 18:20 | Fix an OOM related problem in the snippet() and offsets() functions of fts3. (check-in: 61efff41 user: dan tags: trunk) | |
| 16:04 | Changes to fts3 to avoid flushing data to disk within a SELECT statement. (check-in: 48c0db0e user: dan tags: trunk) | |
| 01:17 | Only declare the sqlite3_mutex_held() and sqlite3_mutex_notheld() interfaces in the header file if NDEBUG is not defined. (check-in: ee9b1c05 user: drh tags: trunk) | |
|
2009-12-09
| ||
| 21:43 | Restore the incorrect legacy ON INSERT syntax to the parser as a no-op. That way, older databases that actually use this meaningless syntax will still be readable. (check-in: 54b955c3 user: drh tags: trunk) | |
| 18:22 | Added test cases for the multi-way USING and NATURAL JOIN fix. Ticket [f74beaabde]. (check-in: 0b34ab25 user: drh tags: trunk) | |
| 17:36 | The USING clause and NATURAL JOIN look at all tables to the left when searching for a match, not just the one table to the immediate left. Tables further to the left are preferred. Fix for ticket [f74beaabde]. Still need to add test cases to complete the ticket. (check-in: b558e96f user: drh tags: trunk) | |
| 14:39 | Improve test coverage of fts3.c. (check-in: 56b6432f user: dan tags: trunk) | |
| 05:30 | Mark fts3ReallocOrFree and fts3InitVtab as static. Ticket [ff44d82f3b]. (check-in: a9038306 user: dan tags: trunk) | |
|
2009-12-08
| ||
| 22:16 | Change to order of two branches in vtab.c in order to a restore 100% test coverage. (check-in: fdfdc777 user: drh tags: trunk) | |
| 19:58 | Add comments to better explain the two-pass memory allocation approach for prepared statements. (check-in: 0e5e18ea user: drh tags: trunk) | |
| 19:05 | Add tests to improve coverage of fts3. Associated bugfixes. (check-in: f0eac417 user: dan tags: trunk) | |
| 15:35 | Avoid pointer aliasing in the allocSpace() routine in vdbeaux.c. (check-in: d6ae2751 user: drh tags: trunk) | |
| 15:16 | Begin updating API documentation to use the new MD5-hash based automatic requirements numbering. Comment changes only. Many similar changes will follow. (check-in: 973c5c86 user: drh tags: trunk) | |
| 14:34 | Allow zero-length names for tables, columns, and indices. Fix for ticket [78e04e52eaf]. (check-in: 7b1cfd65 user: drh tags: trunk) | |
| 13:44 | Add test cases for ticket [abe728bbc311]. (check-in: 4eb9bf0b user: drh tags: trunk) | |
| 13:36 | Avoid dereferencing a null pointer when doing a multi-database comment where one or more of the databases is an ATTACH-ed TEMP database. Ticket [abe728bbc311d]. (check-in: f43e37e0 user: drh tags: trunk) | |
| 02:06 | Add evidence marks for the abs() and soundex() SQL functions. (check-in: 003f3ed1 user: drh tags: trunk) | |
|
2009-12-07
| ||
| 23:53 | Merge all changes associated with the version 3.6.21 release into the OS-X branch. (check-in: ad08794d user: drh tags: apple-osx) | |
| 16:39 | Version 3.6.21 (check-in: 1ed88e9d user: drh tags: trunk, release) | |
| 16:26 | Remove a redundant line from fts3. (check-in: cd50acf3 user: dan tags: trunk) | |
| 16:23 | Add extra tests for parsing of whitespace in tokenizer declarations,. (check-in: 5030ba2e user: dan tags: trunk) | |
| 16:18 | Move some SQLITE_TEST code down to avoid mixing code and variable declarations. Fix a test for high-order bit handling in sqlite3Fts3InitTokenizer(). (check-in: fad43d29 user: shaneh tags: trunk) | |
| 14:48 | Version 3.6.21 release candidate 2. (check-in: 78f6baff user: drh tags: trunk) | |
| 12:34 | Add some tests for OR, AND and NOT operations to fts3rnd.test. Add tests to check that errors are returned when bad arguments are passed to fts3 functions snippet, offsets and optimize. Minor fix for the same (check-in: 5811df3f user: dan tags: trunk) | |
|
2009-12-06
| ||
| 03:35 | Enhanced detection of database corruption in btree.c:allocateSpace(). (check-in: 5a511f98 user: drh tags: trunk) | |
|
2009-12-05
| ||
| 18:34 | Fix to check-in [f221f31eff]: Make sure local variable rc is declared even when compiled without SQLITE_DEBUG. (check-in: 7a9a3532 user: drh tags: trunk) | |
| 18:16 | Add test case for the obscure corruption detection added by the previous commit. (check-in: f221f31e user: dan tags: trunk) | |
| 18:03 | Catch an obscure case of database corruption. (check-in: 0f9b17a6 user: dan tags: trunk) | |
| 14:29 | Fix another bug in 3-way NEAR queries. (check-in: 3bb13a06 user: dan tags: trunk) | |
| 11:37 | Fix some problems with FTS3 and 3-way NEAR queries. (check-in: 23dc6fb5 user: dan tags: trunk) | |
|
2009-12-04
| ||
| 23:10 | Add the SQLITE_4_BYTE_ALIGNED_MALLOC compile-time option which tells some assert() statements that the underlying system only requires 4-byte alignment of 8-byte data objects like double or int64 and that system malloc() only guarantees 4-byte alignment of returned pointers. (check-in: 08faee68 user: drh tags: trunk) | |
| 22:51 | Remove an assert() in btree.c:releasePage() that is not necessarily true when the btree is recovering from an OOM that occurs in the middle of a balance(). (check-in: 04fc9c76 user: drh tags: trunk) | |
| 19:07 | Fix a problem involving a 3-way NEAR query. (check-in: 507890a9 user: dan tags: trunk) | |
| 14:25 | In the OSTRACE debugging macros in os_unix.c, identify the VFS currently in use. (check-in: 082b8da0 user: drh tags: trunk) | |
| 14:11 | Modify [2ad1461f25] to avoid leaving a prepared statement in "active" state following an OOM error in FTS3. (check-in: 69c21ee4 user: dan tags: trunk) | |
| 13:43 | Fix an FTS3 problem where an OOM error was not being propagated back out to the top-level interface. (check-in: 2ad1461f user: drh tags: trunk) | |
| 08:41 | Reset the state of the asynchronous IO module at the end of the tkt-94c04eaadb.test script. (check-in: e8ae7e98 user: dan tags: trunk) | |
| 05:41 | Do not run fts3rnd.test as part of quick.test. (check-in: 3bb64842 user: dan tags: trunk) | |
| 01:44 | Clear a conflicting TCL variable in the fts3rnd.test test script. (check-in: f9047b2e user: drh tags: trunk) | |
|
2009-12-03
| ||
| 21:26 | Fix an assert() in btree.c that might have fired on a corrupt database file. (check-in: 5ab2b3e3 user: drh tags: trunk) | |
| 19:40 | Remove a NEVER() from btree.c that could occur in a very obscure tested evaluation with an I/O error on fstat(). (check-in: d5861d9f user: drh tags: trunk) | |
| 17:36 | Fix an incorrect assert() in fts3.c. Add further fts3 tests. (check-in: 75863c2d user: dan tags: trunk) | |
| 06:26 | Updates to FTS3 to correct compiler warnings under MSVC. (check-in: 37495b55 user: shaneh tags: trunk) | |
| 04:40 | Fix a couple of compiler warnings under MSVC. (check-in: e3aa0870 user: shaneh tags: trunk) | |
| 04:32 | Fix cut-and-paste error affecting a couple of attach tests on Windows. (check-in: 59e2e2c6 user: shaneh tags: trunk) | |
| 01:01 | Add links from C/C++ interfaces to their equivalent SQL functions. Comment changes only - no changes to code. (check-in: acce10f7 user: drh tags: trunk) | |
|
2009-12-02
| ||
| 20:25 | Change an error message in FTS3 to avoid using an uninitialized variable. (check-in: 620a8a2b user: drh tags: trunk) | |
| 18:16 | Fix an uninitialized value read in sqlite3async.c. (check-in: 1cf2136c user: dan tags: trunk) | |
| 18:03 | Fix compiler warnings in test_hexio.c. (check-in: 6a2e0725 user: drh tags: trunk) | |
| 16:09 | Modify a test in fts3b.test to reflect the fact that the docid field may now be updated. (check-in: ef5b7450 user: dan tags: trunk) | |
| 14:44 | Clear the Pager.dbModified flag when unlocking the database. Assert that it is clear when locking the database. (check-in: d17ec16b user: dan tags: trunk) | |
| 02:49 | Make sure a variable is cleared before use in the where8.test script. (check-in: b3436529 user: drh tags: trunk) | |
|
2009-12-01
| ||
| 22:09 | Add testcase() macros to make sure boundary cases in BEFORE UPDATE triggers are well tested. (check-in: 27175caa user: drh tags: trunk) | |
| 18:46 | Fix a faulty function prototype in test_intarray.c. (check-in: 110956a6 user: drh tags: trunk) | |
| 17:08 | Use #include "sqlite3.h" instead of <sqlite3.h> in fts3Int.h. (check-in: 7737db49 user: dan tags: trunk) | |
| 17:05 | Add typedefs for i16 and u8 to fts3Int.h when not building via the amalgamation method. (check-in: fa56c1c4 user: dan tags: trunk) | |
| 15:04 | Add tests for the remaining examples in fts3.in. (check-in: e381cd52 user: dan tags: trunk) | |
| 14:31 | Reorder function declarations in mutex_os2.c. This is a blind change - we have no capability of testing on OS/2. Ticket [97214a34d814] (check-in: c40e4ef0 user: drh tags: trunk) | |
| 13:57 | Changes to the TCL interface header to allow it to be compiled independently from the amalgamation. (check-in: 58113932 user: drh tags: trunk) | |
| 13:48 | Open a savepoint within the FTS3 optimize() function. (check-in: 4924fbb2 user: dan tags: trunk) | |
| 12:00 | Fix a segfault that can occur when querying an empty FTS3 table. Also restore the rowid/docid conflict handling to work as it did in version 3.6.20. (check-in: c022f66b user: dan tags: trunk) | |
|
2009-11-30
| ||
| 19:48 | Test coverage improvements in the FTS3 porter stemmer. (check-in: 6d112bfd user: drh tags: trunk) | |
| 08:55 | Add test cases for examples recently added to documentation file fts3.html. (check-in: 498922cc user: dan tags: trunk) | |
|
2009-11-28
| ||
| 21:33 | Updates to snippet() and offsets() functions of FTS3 so that they work sanely following an OOM fault. (check-in: b939a37a user: drh tags: trunk) | |
| 17:23 | Change FTS3 to detect when the RHS of the MATCH opertor encounters an OOM during string format conversion and report back an SQLITE_NOMEM error. (check-in: 31eed4f8 user: drh tags: trunk) | |
| 17:07 | Remove all benign OOM failure opportunities from the FTS3 hash table implementation. All OOM faults cause SQLITE_NOMEM to be returned. (check-in: 80754d38 user: drh tags: trunk) | |
| 15:35 | Add a test case for creating an FTS3 table with no module arguments or opening/closing brackets in the CREATE VIRTUAL TABLE statement. (check-in: a9cba7ea user: dan tags: trunk) | |
| 13:46 | Initialize a variable (unnecessarily) to avoid a compiler warning. (check-in: db65fd59 user: drh tags: trunk) | |
| 12:40 | Fix a bug introduced by the fts3 refactoring (segfault when creating a table with zero module args). Also a fix to handle an OOM error. (check-in: eada284b user: dan tags: trunk) | |
|
2009-11-27
| ||
| 18:31 | Change the test numbers in e_fkey.test so that they are in order. (check-in: ca73be44 user: dan tags: trunk) | |
| 12:14 | Remove a C++ism accidentally added to fts3.c. (check-in: 97d33241 user: dan tags: trunk) | |
| 12:12 | Move [7d30880114] to the trunk. Add optimizations to reduce the number of opcodes used for BEFORE UPDATE triggers. (check-in: 1b7c5250 user: dan tags: trunk) | |
|
2009-11-26
| ||
| 14:01 | Simplifications to the sqlite3_trace() bound parameter substitution logic. (check-in: cb4b9286 user: drh tags: trunk) | |
|
2009-11-25
| ||
| 22:42 | Make sure the new fts3Int.h header file is added to the amalgamation. (check-in: f25558f3 user: drh tags: trunk) | |
| 21:05 | Support zeroblob in trace output. (check-in: 5086bf8e user: drh tags: trunk) | |
| 19:35 | Get trace with parameter insertion working for UTF16 databases. (check-in: 01d5451a user: drh tags: trunk) | |
| 18:03 | Initial check-in of code that inserts tokenizations of the values for bound parameters into the output of sqlite3_trace(). (check-in: 545cfb3b user: drh tags: trunk) | |
| 16:53 | Enhance the %q, %Q, and %w printf conversions so that the precisions specifies the length of the input. (check-in: 3ba77313 user: drh tags: trunk) | |
|
2009-11-24
| ||
| 19:02 | Recomputed unchanged column values in an UPDATE statement after running BEFORE triggers, in case the triggers have modified any of those values. (Closed-Leaf check-in: 7d308801 user: drh tags: experimental) | |
| 16:26 | Fix a tracing macro in the VDBE in the VColumn opcode. (check-in: cb74b81b user: drh tags: trunk) | |
| 02:37 | Fix SQLITE_SECURE_DELETE so that it overwrites content on the root pages of tables and indices that are dropped using the truncate optimization. (check-in: 02ef972a user: drh tags: trunk) | |
|
2009-11-23
| ||
| 21:23 | Make sure registers computed for the VFilter opcode are marked invalid after the VFilter opcode finishes. Ticket [16fbf14cb2]. (check-in: 83dc7d38 user: drh tags: trunk) | |
| 15:59 | Rearrange the order of two branches in order to restore 100% branch coverage after recent changes to in-memory database processing. (check-in: 6ce6ff44 user: drh tags: trunk) | |
| 14:39 | Change so that deleting an Expr structure requires only one frame per level in the expression tree, not two. (check-in: a4380ab3 user: dan tags: trunk) | |
| 13:17 | Bump the version number to 3.6.21. (check-in: 39214aee user: drh tags: trunk) | |
|
2009-11-21
| ||
| 03:03 | Add some missing comments to fts3 files. No source code changes. (check-in: b6402b20 user: dan tags: trunk) | |
| 01:33 | Fix a bug in transaction rollback during backup when the CODEC is enabled that can result in a corrupt backup file. This is only a problem when SQLITE_HAS_CODEC is defined. (check-in: f3398fc4 user: drh tags: trunk) | |
|
2009-11-20
| ||
| 18:48 | When moving pages as part of autovacuum on an in-memory database, make sure that the source location is journalled so that a ROLLBACK can occur. Part of the fix for ticket [564d412f15a00] (check-in: 2f42f91f user: drh tags: trunk) | |
| 17:23 | Merge leaf accidentally created by [1c4984c62f]. (check-in: cae949ce user: dan tags: trunk) | |
| 17:18 | Add a test to make sure a database can be attached to a single handle twice if not in shared-cache mode. (check-in: 1c4984c6 user: dan tags: trunk) | |
| 16:13 | Fix a bug in LIMIT 0 for compound SELECT statement. The problem was introduced by recent enhancements and has not appeared in any release. (check-in: c6ed7e2a user: drh tags: trunk) | |
| 15:02 | If the root pages numbers of the internal schema are adjusted due to autovacuum on a DROP TABLE statement and that statement later aborts (for example, due to an OOM error) then reset the internal schema at the conclusion of the statement. Partial fix for ticket [564d412f15a]. (check-in: e493b093 user: drh tags: trunk) | |
| 13:18 | Avoid unnecessary page cache allocations when move a page while autovacuuming an in-memory database, since the allocation might fail making it impossible to rollback the transaction. (check-in: 9a429349 user: drh tags: trunk) | |
| 05:05 | Improve comments and other things in fts3_write.c. (check-in: 1cf0e3cc user: dan tags: trunk) | |
| 02:24 | Minor optimizations to fts3 code. (check-in: b456eacb user: dan tags: trunk) | |
|
2009-11-19
| ||
| 18:30 | Merge leaves [7cd178a72a] and [598727e6da]. (check-in: 4115c0c2 user: dan tags: trunk) | |
| 18:28 | Fix a performance regression introduced while reworking the fts3 code. (check-in: 7cd178a7 user: dan tags: trunk) | |
| 15:44 | Remove an unreachable branch from the OP_Affinity suppression logic. (check-in: 598727e6 user: drh tags: trunk) | |
| 15:26 | Merge fts3-refactor into the main trunk. (check-in: 551dbe30 user: dan tags: trunk) | |
| 15:25 | Fix some fts3 related issues with the autoconf and amalgamation build systems. (Closed-Leaf check-in: 3b179247 user: dan) | |
| 14:57 | Merge with [4bd4330709]. (check-in: 7a46d1eb user: dan tags: fts3-refactor) | |
| 14:52 | Merge the fts3-refactor branch with the trunk. (check-in: c8d2bd37 user: dan tags: fts3-refactor) | |
| 14:48 | Fix a bug introduced with recent optimizations: The unary minus operator is TK_UMINUS, not TK_MINUS. (check-in: 4bd43307 user: drh tags: trunk) | |
| 00:15 | Fix problems introduced into fts3 as part of the refactoring. (check-in: fa0998e1 user: dan tags: fts3-refactor) | |
|
2009-11-18
| ||
| 23:01 | Performance improvement by avoiding unnecessary calls to memset(). (check-in: 85940468 user: drh tags: trunk) | |
| 15:35 | Add some missing comments and fix some other issues in fts3 code. (check-in: 2fe579e7 user: dan tags: fts3-refactor) | |
| 01:25 | Suppress more instances of unnecessary OP_IsNull and OP_Affinity opcodes. (check-in: bf6c0bd1 user: drh tags: trunk) | |
|
2009-11-17
| ||
| 23:59 | Enhancements to the VDBE opcode loop to reduce the number of machine-code instructions evaluated by about 10%. (check-in: 9744ffb3 user: drh tags: trunk) | |
| 18:31 | Code generator tries to avoid pointless OP_IsNull and OP_Affinity opcodes. (check-in: ebb0c8a3 user: drh tags: trunk) | |
| 12:52 | Improvements to the way fts3 reads the full-text index. (check-in: 45c051e7 user: dan tags: fts3-refactor) | |
|
2009-11-16
| ||
| 23:36 | Make sure the output of sqlite3OsFullPathname is initialized even if the fails due to I/O or OOM errors. (check-in: e4943adb user: drh tags: trunk) | |
| 22:54 | Fix a reference to freed memory that can occur following an OOM error in where.c. (check-in: 929b6047 user: drh tags: trunk) | |
| 21:28 | Back out part of the change in [23ea2b700fd6d28d] since TH3 reveals some problems in OOM situations. (check-in: 15d215d6 user: drh tags: trunk) | |
| 16:36 | Add a few extra coverage test cases for fts3. (check-in: f29c8fca user: dan tags: fts3-refactor) | |
| 15:11 | Use the sqlite3CtypeMap to improve the speed and reduce the size of the logic in the tokenizer that recognizes identifiers. (check-in: 1c9243b0 user: drh tags: trunk) | |
| 03:13 | Additional optimizations in sqlite3VdbeRecordCompare(). (check-in: 23ea2b70 user: drh tags: trunk) | |
| 02:14 | Optimizations to the sqlite3VdbeRecordCompare() routine help it to run 12.5% faster for some traces. (check-in: 74c76f0b user: drh tags: trunk) | |
|
2009-11-15
| ||
| 06:50 | Fixes to fts3 integrity check code. (check-in: d3cae986 user: dan tags: fts3-refactor) | |
|
2009-11-14
| ||
| 23:22 | Optimizations to the main loop inside sqlite3VdbeExec() to help VDBE byte code run a few percent faster. (check-in: d622ac6a user: drh tags: trunk) | |
| 18:41 | Further OOM testing for fts3 code. Add Tcl code implementing an integrity-check for fts3. (check-in: c27d46b3 user: dan tags: fts3-refactor) | |
| 18:04 | Make the sqlite3VdbeExec() function about 2% faster by storing the opcode array in a local variable. (check-in: 8bd3cc82 user: drh tags: trunk) | |
|
2009-11-13
| ||
| 20:52 | Performance improvements and size reductions on the OP_Seek* family of VDBE opcodes. (check-in: ed820f45 user: drh tags: trunk) | |
| 19:43 | Simplify the OPFLG processing logic in the VDBE for a speed boost in the VDBE processing loop and a reduction in code size. (check-in: 3352b3eb user: drh tags: trunk) | |
| 17:36 | Start reworking fts3 code to match the rest of SQLite (code conventions, malloc-failure handling etc.). (check-in: 30a92f11 user: dan tags: fts3-refactor) | |
| 17:05 | Force 8-byte memory alignment on memory allocated for VDBE cursors. (check-in: bdc45ba7 user: drh tags: trunk) | |
| 10:04 | Changes to test_intarray.c so that it build with SQLITE_OMIT_VIRTUALTABLE. Added testcase() macros to expr.c. (check-in: f0599d28 user: drh tags: trunk) | |
|
2009-11-12
| ||
| 20:39 | Shift more OP_Found opcodes over to using the unpacked format, for improved performance. (check-in: 6705ab1a user: drh tags: trunk) | |
| 19:59 | Enhance the OP_Found and OP_NotFound opcodes so that they can accept an array of registers as an unpacked record in addition to a record built using OP_MakeRecord. Use this to avoid OP_MakeRecord calls during IN expression processing. (check-in: b9eab885 user: drh tags: trunk) | |
| 17:52 | Factor out the IN operator code generation into a subroutine. Use this subroutine to implement both logic and branching versions of the IN operator. (check-in: fcff5b7e user: drh tags: trunk) | |
| 13:32 | Factor out the code generator for BETWEEN into a subroutine. (check-in: 5735f60b user: drh tags: trunk) | |
| 05:04 | Update comment for substrFunc(). Added additional SUBSTR() test cases. (check-in: d7b3801d user: shaneh tags: trunk) | |
| 04:26 | Suppress unnecessary OP_Noop instructions on when the right table of a LEFT JOIN uses the index-only optimization. (check-in: e8aec08b user: drh tags: trunk) | |
| 03:46 | Suppress excess OP_Null opcodes caused by binary IS or IS NOT operators that are converted into unary ISNULL or NOTNULL operators. (check-in: cff1b36a user: drh tags: trunk) | |
| 03:13 | Adjustments to the implementation of LIMIT so that it uses fewer opcodes. (check-in: 39d5b292 user: drh tags: trunk) | |
|
2009-11-11
| ||
| 23:58 | If the sector size is unspecified (initially 0) then set it to 512, not to 32. (check-in: 8861b5c1 user: drh tags: trunk) | |
| 23:48 | Remove an obsolete (and incorrect) comment from sqlite3MemCompare(). No changes to code. (check-in: c15b70b3 user: drh tags: trunk) | |
| 20:53 | Tweaks to the SUBSTR() function to make it dramatically faster in the common case where the input string is large but the 2nd and 3rd arguments are small positive integers. (check-in: 5a474a86 user: drh tags: trunk) | |
| 13:17 | Allow media sector sizes as small as 32. The former minimum size was 512. (check-in: 5a32bfc1 user: drh tags: trunk) | |
| 04:17 | Modified CLI to raise an error when extra command line options are passed. Added tests to verify correct handling, as well as other basic handling of command line options. Ticket [f5cb008a65]. (check-in: 09b4f19f user: shaneh tags: trunk) | |
| 01:14 | Additional test cases for the coalesce() and ifnull() functions. (check-in: d0591258 user: drh tags: trunk) | |
| 00:24 | Generate VDBE code for the built-in COALESCE() and IFNULL() functions. This allows unused arguments to never be evaluated, which is a performance win when the unused argument is a subquery. (check-in: 30055b25 user: drh tags: trunk) | |
|
2009-11-10
| ||
| 17:55 | Additional intarray test cases and tweaks to comments. (check-in: 0c8db5d8 user: drh tags: trunk) | |
| 17:24 | Initial implementation and test cases for the "intarray" virtual table. (check-in: 0e7d2925 user: drh tags: trunk) | |
| 17:07 | Fix for datbase name being ignored with the -batch option in the CLI. Ticket [aeff892c57]. (check-in: 8097c64a user: shaneh tags: trunk) | |
| 01:30 | Remove the obsolete "$Id:$" RCS identifier strings from the source code. (check-in: f6c045f6 user: drh tags: trunk) | |
| 01:13 | Remove from the TCL interface some old code associated with sqlite3_analyzer. (check-in: 01f2271e user: drh tags: trunk) | |
| 01:12 | Omit some code that is not used when SQLITE_SECURE_DELETE is defined. (check-in: 5a9e7463 user: drh tags: trunk) | |
|
2009-11-09
| ||
| 19:30 | Fix for lock structure sharing with AFP-style locking (check-in: 62f15c0a user: adam tags: apple-osx) | |
|
2009-11-06
| ||
| 17:20 | Improved shell "dot" command argument handling. Ticket [f12a9eeedc]. (check-in: 9fb69919 user: shaneh tags: trunk) | |
| 04:13 | Fix the backup API so that a backup from an empty database to a non-empty database works. Ticket [0bf974bdf9]. The only changes are in assert() statements. (check-in: ddb71cd9 user: drh tags: trunk) | |
| 03:33 | Merged accidental fork. (check-in: d410cd4e user: shaneh tags: trunk) | |
| 03:22 | Add some very basic tests for the commands available from the CLI. (check-in: 6bf43338 user: shaneh tags: trunk) | |
|
2009-11-05
| ||
| 18:46 | Recover an ophaned branch that resulted from some ill-advised "shunning". This check-in should be identical to [bd443f39a9] except that it should be on the trunk. (check-in: 38eaf1ab user: drh tags: trunk) | |
| 18:31 | Fix an undefined variable on non-Mac builds. (check-in: 9552de8f user: drh tags: apple-osx) | |
| 17:27 | Do not limit the trigger recursion depth when SQLITE_SMALL_STACK is defined since trigger recursion uses the heap, not the stack. (Closed-Leaf check-in: bd443f39 user: drh tags: shunning-error) | |
| 16:29 | Add a comment related to code coverage. No changes to code. (check-in: 6bdc2c99 user: drh tags: shunning-error) | |
| 04:14 | Update maxRow size after shrinking the row buffer (CSV). (Closed-Leaf check-in: d474195a user: shaneh tags: csv_ext) | |
| 04:01 | Removed line limit on rows. (check-in: ec08b15f user: shaneh tags: csv_ext) | |
| 02:34 | Initial implementation of a virtual table for CSV files. (check-in: 90e63b7d user: shaneh tags: csv_ext) | |
|
2009-11-04
| ||
| 23:02 | Fix a typo on the ynVar typedef that unfortunately caused 3.6.20 to be released with a "bug" that causes excess memory usage for Expr objects. This bug should be benign in the sense that the correct answer is still always obtained - it only uses more more memory than is really necessary. (check-in: f1c09aca user: drh tags: shunning-error) | |
| 21:49 | Do not accept a codec no an in-memory database. (check-in: 6f3081a8 user: drh tags: shunning-error) | |
| 18:51 | Fix documentation typos. (check-in: 590d4ac1 user: drh tags: shunning-error) | |
| 13:30 | Version 3.6.20 (check-in: eb7a544f user: drh tags: trunk, release) | |
| 13:17 | Create a new datatype "ynVar" to hold "nVar" (number of variable) values. This is normally 16-bits can can be 32-bits if SQLITE_MAX_VARIABLE_NUMBER is large enough. (check-in: 24a4d520 user: drh tags: trunk) | |
| 01:56 | Report errors back up to the ATTACH command if the codec fails to initialize. (check-in: a6c69610 user: drh tags: trunk) | |
|
2009-11-03
| ||
| 22:34 | Customizations for OSX contributed by Apple. (check-in: 941a01eb user: adam tags: apple-osx) | |
| 19:42 | Fix compiler warnings on MSVC build. (check-in: 01c4b5b8 user: shane tags: trunk) | |
| 19:18 | Enhancements to lemon to generate more compact action tables and to avoid making array bounds tests that can never fail on action table calculations. (check-in: 27d8e684 user: drh tags: trunk) | |
| 13:08 | Force all qsort() calls in mkkeywordhash.c to be stable so that we get predictable results on different platforms. (check-in: 3b02df27 user: drh tags: trunk) | |
| 13:02 | Adjust the lemon implementation so that it always computes the same PDA regardless of qsort() implementation on the host platform. In other words, make all sorts in lemon stable. (check-in: d66a0f31 user: drh tags: trunk) | |
| 01:22 | All SQLITE_MAX_VARIABLE_NUMBER to exceed 32767. The sizes of some structures increase when the compile-time parameter is configured this way. (check-in: 16a24b44 user: drh tags: trunk) | |
|
2009-11-02
| ||
| 18:44 | Bug fix in the recent changes to mkopcodeh.awk. (check-in: 6610cac4 user: drh tags: trunk) | |
| 18:14 | Update the generator AWK script for opcodes.h so that it always generates opcode numbers in the same order and is not dependent on the hash order of opcode names within AWK. (check-in: e730195f user: drh tags: trunk) | |
| 18:01 | Do not insert the date and time of generation into the amalgamation. (check-in: 9f6cf13d user: drh tags: trunk) | |
| 17:40 | Update the configure script for version 3.6.20. (check-in: f19cb105 user: drh tags: trunk) | |
| 15:51 | Comment changes in support of better documentation. No changes to code. (check-in: 79ab4714 user: drh tags: trunk) | |
|
2009-11-01
| ||
| 19:26 | Fix compiler warnings. Better comments on some variables associated with LIKE/GLOB processing in the query planner. (check-in: cfa2db0f user: drh tags: trunk) | |
|
2009-10-30
| ||
| 14:27 | Increase the version number to 3.6.20. (check-in: 61295274 user: drh tags: trunk) | |
| 13:26 | Version 3.6.16.1 (Leaf check-in: 2a832b19 user: drh tags: release, branch_3_6_16) | |
| 13:25 | Avoid storing the result register for EXISTS and SELECT in any field of the Expr object - simply return the register number as the return value of the function that codes those expressions. (check-in: 7253f8fa user: drh tags: trunk) | |
|
2009-10-29
| ||
| 18:38 | Fix a 16-bit integer overflow that might occur in statements that use both an EXISTS clause and IN operator with a RHS holding in excess of 32K entries. (check-in: 65a1f133 user: drh tags: branch_3_6_16) | |
| 13:48 | Reduce the memory required by prepared statements that use the IN operator with a list on the right and an INTEGER PRIMARY KEY on the left. (check-in: e7984ad2 user: drh tags: trunk) | |
|
2009-10-27
| ||
| 18:06 | Minor updates and corrections to comments in btreeInt.h. No changes to code. (check-in: ad949fa1 user: drh tags: trunk) | |
|
2009-10-26
| ||
| 22:32 | Remove additional obsolete code from the sqlite3_release_memory() logic. (check-in: 2eeb6ed8 user: drh tags: trunk) | |
| 22:08 | Remove some obsolete code within #if 0 that was causing developer concern. (check-in: 1d64e945 user: drh tags: trunk) | |
|
2009-10-24
| ||
| 15:51 | Fix a compiler warning that occurs when building the TCL interface. (check-in: fc36034b user: drh tags: trunk) | |
| 03:04 | Another attempt at getting the pTail pointer computations correct for sqlite3_prepare16(). Ticket [3fe897352e]. (check-in: a96b4e8c user: drh tags: trunk) | |
| 02:06 | In shell, modified to not print informational messages during startup while in "batch mode". Ticket [2cb66577f6]. (check-in: e3898e25 user: shane tags: trunk) | |
| 02:00 | In shell, changed ".tables" command to not add '%' to specified pattern. This is consistent with other commands that take a LIKE pattern (.dump, .schema, .indices). Updated internal help to better describe LIKE pattern usage. Fixed a few more inconsistencies in error messages. Ticket [6da68f691b]. (check-in: 7201244e user: shane tags: trunk) | |
| 01:55 | Check-in [19064d7cea] broke the pTail return on sqlite3_prepare16() when the SQL contained surrogates. This check-in restores correct function. Part of ticket [3fe897352e]. (check-in: 65b1e3a4 user: drh tags: trunk) | |
|
2009-10-23
| ||
| 18:15 | Make sure that UTF16 to UTF8 conversions to not read past the end of the UTF16 input buffer if the last two bytes of the UTF16 happen to be the first half of a surrogate pair. Ticket [3fe897352e] (check-in: 19064d7c user: drh tags: trunk) | |
| 01:27 | In shell, ensure that do_meta_command() returns consistent error values. Adjusted the text of some error message to be more consistent. Ticket [beb2dd69ad]. (check-in: 1ebac9ed user: shane tags: trunk) | |
| 00:37 | In shell, modified "import" handling to ensure error code returned correctly on exit. Ticket [bd770b2c52]. (check-in: 009efad0 user: shane tags: trunk) | |
|
2009-10-22
| ||
| 21:23 | In shell, reworked .header and .echo handling. Updated shell_exec() to (really) handle multiple statements. Tickets [72adc99de9], [7b61b6c6ce], and [eb620916be]. (check-in: 790402c1 user: shane tags: trunk) | |
| 20:52 | Merge the MD5 checksum logic into the TCL interface. This facilitates building a tclsh that contains both SQLite and MD5. The plan is to use this augmented tclsh to help build the documentation. (check-in: a024c0a8 user: drh tags: trunk) | |
| 20:50 | Add some sample requirement implementation comments to where.c. (check-in: 9854ad00 user: drh tags: trunk) | |
| 18:12 | In shell, in shell_exec() logic, use type info if available when outputting in "insert" mode for other types in addition to blobs. Changed shell_exec() to use sqlite_prepare_v2(). Ticket [72adc99de9] and [7b61b6c6ce]. (check-in: ab99faca user: shane tags: trunk) | |
| 17:30 | Updated shell to output blobs in X'1234' form when in "insert" mode. Ticket [72adc99de9]. (check-in: a2ad9e63 user: shane tags: trunk) | |
| 00:20 | Add markings on test evidence for R-30323-21917. (check-in: e51af74c user: drh tags: trunk) | |
| 00:14 | Mention the sqlite3_column_count() function on the same documentation page with all of the other sqlite3_column... interfaces. (check-in: 5354ace5 user: drh tags: trunk) | |
|
2009-10-21
| ||
| 14:33 | Merge accidental fork back to trunk. (check-in: f63397ad user: shane tags: trunk) | |
| 14:11 | Modified shell to call output_html_string() for header values in .mode html. Updated output_html_string() to support the 5 basic HTML entities (<,>,&,',"). Ticket [dc3a6e5b31]. (check-in: c8815fcd user: shane tags: trunk) | |
| 13:53 | Made detectSchemaProblem function and aSchemaTable array static. Ticket [6f8b0c5855]. (check-in: c6f22f3e user: shane tags: trunk) | |
| 13:48 | Add evidence marks to parse.y. Fix a broken evidence link in e_fkey.tcl. (check-in: 0441d28e user: drh tags: trunk) | |
| 03:56 | For the shell, changed the output of the errors on lines 2910 and 2914 to goto stderr. I left the timer values going to stdout as all of the other interactive shell commands (startup banner, help, etc.) go to stdout and I felt this was more consistent. Ticket [43db771bb2]. (check-in: 8a8eeb0c user: shane tags: trunk) | |
| 03:42 | For Windows version of shell, add support for .timer command using the GetProcessTimes() API if available (in the same way getrusage() is used on UNIX.) Ticket [89668ca167]. (check-in: 83216fbe user: shane tags: trunk) | |
| 02:00 | For Windows VFS, modified xGetLastError() to call FormatMessage() with the FORMAT_MESSAGE_IGNORE_INSERTS option. Additionally updated to ensure strings are returned as UTF8. Ticket [39c85e8a4e]. (check-in: 761396f8 user: shane tags: trunk) | |
|
2009-10-20
| ||
| 15:27 | Clarify the use of sqlite3_shutdown(). Ticket [f9af981dd2a2]. Comment changes only - no changes to code. (check-in: b2aa48b5 user: drh tags: trunk) | |
| 15:01 | Fix the VACUUM command so that when the MAIN and TEMP databases have identically named tables, content from the TEMP database does not accidently overwrite MAIN database content. Ticket [d82e3f3721296e905d7e8c9dee718f71a826b0eb]. Note that this bug could potentially lead to database corruption. (check-in: 0bbd8d72 user: drh tags: trunk) | |
| 14:23 | Remove all mention of sqlite3_next_stmt() from the documentation of sqlite3_close(). (check-in: f494ed38 user: drh tags: trunk) | |
| 14:08 | Update documentation on the sqlite3_prepare_v2() interface. No changes to code. (check-in: 3b00e769 user: drh tags: trunk) | |
| 07:01 | Fix cases where modifying bound variables using the clear_bindings() or transfer_bindings() APIs can invalidate a query plan. (check-in: e6808c98 user: dan tags: trunk) | |
|
2009-10-19
| ||
| 22:41 | Remove an unreachable condition in the WHERE clause processing of bound parameters and replace it with an assert(), for test coverage. (check-in: 3bcd78a1 user: drh tags: trunk) | |
| 20:15 | Fix an incorrect assert() in vdbeUnbind(). (check-in: 651c1efb user: drh tags: trunk) | |
| 18:30 | Fix an assert() that may fail if sqlite3_step() is called on a statement after a previous call has already returned SQLITE_SCHEMA. (check-in: 63bf7345 user: dan tags: trunk) | |
| 18:11 | Remove the sqlite3_reoptimize() API. The same functionality is now provided automatically to queries prepared using prepare_v2(). (check-in: 2c50b3d5 user: dan tags: trunk) | |
| 15:52 | When generating WHERE clause terms internally for NATURAL and USING joins, identify the table by its position in the FROM list, not by its name or alias. Fix for [b73fb0bd64]. (check-in: 6fe63711 user: dan tags: trunk) | |
| 07:50 | Use 64-bit arithmetic in the xRead() method of asyncRead. Fix for [94c04eaadb]. (check-in: ca3e41b0 user: dan tags: trunk) | |
|
2009-10-17
| ||
| 14:19 | Fix a problem in the icu.test script. (check-in: 8a21fdaf user: dan tags: trunk) | |
| 13:13 | Make sure the target register is allocated before adding the bogus OP_Variable opcode in isLikeOrGlob(). Bugfix for the new sqlite3_reoptimize() logic. (check-in: e74f8dc4 user: drh tags: trunk) | |
| 08:26 | Add documentation comment for new experimental API sqlite3_reoptimize(). (check-in: c48f8ffe user: dan tags: trunk) | |
|
2009-10-16
| ||
| 16:21 | Merge the sqlite3_reoptimize() changes into the trunk. (check-in: 50136840 user: drh tags: trunk) | |
| 15:59 | Fix a problem in the analyze3.test script. (Closed-Leaf check-in: 61174aea user: dan tags: experimental) | |
| 15:19 | Remove a mutex reference leak from sqlite3_reoptimize(). (check-in: 9f093706 user: dan tags: experimental) | |
| 15:05 | Comment cleanup in btree.c. (check-in: 32966ba4 user: drh tags: trunk) | |
| 14:55 | Experimental fix for [f777251dc7]. This may be changed yet. (check-in: 174477bc user: dan tags: trunk) | |
| 13:23 | Fix comment typos in btreeInt.h. (check-in: 550566a7 user: drh tags: trunk) | |
|
2009-10-15
| ||
| 23:03 | When a forced rollback occurs, cause all active statements on the same database connection to halt immediately with an SQLITE_ABORT error code. This is a partial fix to ticket [f777251dc7]. (Closed-Leaf check-in: f256bc07 user: drh tags: trunk, experimental) | |
| 19:45 | On any rollback, immediately interrupt all running statements on the same database connection. This is a partial fix for ticket [f777251dc7]. (Closed-Leaf check-in: c304b44c user: drh tags: experimental) | |
| 18:35 | Add the experimental sqlite3_reoptimize() API. (check-in: 9bd6f3d8 user: dan tags: experimental) | |
|
2009-10-14
| ||
| 11:33 | Version 3.6.19 (check-in: c1d499af user: drh tags: trunk, release) | |
|
2009-10-13
| ||
| 22:47 | Update the configure script for version 3.6.19. (check-in: ac19dbc6 user: drh tags: trunk) | |
| 19:19 | Fixed warnings with MSVC compile. (check-in: e0aa50c3 user: shane tags: trunk) | |
| 18:49 | Fix a typo in the test script added for ticket [5ee23731f15]. (check-in: d0f55b5c user: drh tags: trunk) | |
| 18:38 | Update the TCL wrapper to provide a non-NULL objProc pointer to the Tcl_NRCreateCommand() interface. The TCL gurus say this is needed to support legacy TCL extensions. (check-in: 1b3cfa01 user: drh tags: trunk) | |
| 15:42 | Add a test case to verify that ticket [5ee23731f15] has been fixed. (check-in: f894ebf8 user: drh tags: trunk) | |
| 15:36 | When a write transaction fails to being because the database is read-only, fail right away. Do not give the error code an opportunity to be overwritten with SQLITE_OK. Ticket [5ee23731f15]. (check-in: bc110117 user: drh tags: trunk) | |
| 13:08 | Suppress some harmless compiler warnings. (check-in: 5d381daa user: drh tags: trunk) | |
| 12:48 | Make sure mem3.c releases its mutex on an sqlite3_shutdown() call. (check-in: 16254ad5 user: drh tags: trunk) | |
| 10:39 | Add further test cases to e_fkey.test. (check-in: c85e5874 user: dan tags: trunk) | |
|
2009-10-12
| ||
| 22:30 | Add asserts as evidence that all FK constraints are either immediate or deferred. (check-in: 634ef4fc user: drh tags: trunk) | |
| 20:01 | When processing FK constraints on changes to parent tables, make sure the names of child tables appear correctly in the output of EXPLAIN QUERY PLAN. (check-in: a4ab5215 user: drh tags: trunk) | |
| 18:57 | Add more test cases to e_fkey.test. (check-in: 5633cb2b user: dan tags: trunk) | |
| 16:02 | Fix some mappings from e_fkey.test. (check-in: 171c6713 user: dan tags: trunk) | |
| 15:25 | Update e_fkey.test to match the latest version of foreignkeys.html. (check-in: a2ca9f1a user: dan tags: trunk) | |
| 11:27 | Extra test cases mapped to statements in foreignkeys.html. (check-in: ffa6207d user: dan tags: trunk) | |
| 08:41 | Add further test cases to e_fkey.test. (check-in: d61cc0e1 user: dan tags: trunk) | |
|
2009-10-10
| ||
| 15:49 | Add tests to e_fkey.test. (check-in: 5ec07fee user: dan tags: trunk) | |
| 14:29 | Update requirement text for sqlite3_open_blob(). Update evidence comment in e_fkey.test to conform to the latest requirement phrasing. (check-in: 89c548ac user: drh tags: trunk) | |
|
2009-10-09
| ||
| 18:16 | Add test cases to e_fkey.test. (check-in: e391dfe7 user: dan tags: trunk) | |
|
2009-10-08
| ||
| 19:02 | Test a couple of the examples in foreignkeys.html. (check-in: a2fb7902 user: dan tags: trunk) | |
| 17:42 | Add test cases to test file "e_fkey.test". (check-in: 23e0f61a user: dan tags: trunk) | |
| 13:42 | Fix a problem in the legacy genfkey code to do with column names that require quoting. (check-in: 273ccbf1 user: dan tags: trunk) | |
| 11:57 | Add tests to test/e_fkey.test. (check-in: 533703d7 user: dan tags: trunk) | |
| 06:24 | Update triggerC.test to match the change to the way the trigger-recursion limit is handled. (check-in: 58069252 user: dan tags: trunk) | |
| 01:43 | Fix a segfault that can occur following an OOM fault while renaming a table holding foreign key constraints. (check-in: 33973814 user: drh tags: trunk) | |
|
2009-10-07
| ||
| 23:42 | Use memcpy() rather than structure assignment so that memcmp() can later be used for comparison. Ticket [8550ecca70] (check-in: 56f609da user: drh tags: trunk) | |
| 18:41 | Add some tests of statements in foreignkeys.html. (check-in: 83828679 user: dan tags: trunk) | |
| 16:04 | Add a missing OP_Close opcode to VDBE programs that check for FK constraint violations. (check-in: 5caa4a2b user: dan tags: trunk) | |
| 11:29 | Update the header comment for sqlite3_blob_open to note that it is not possible to write to a column that is indexed, part of a primary key or, if foreign key support is enabled, part of a foreign key. (check-in: 2fd487fd user: dan tags: trunk) | |
| 10:43 | Update header file comments describing the sqlite3_changes and sqlite3_total_changes functions to describe how changes made by foreign key actions are handled. (check-in: 4c342c02 user: dan tags: trunk) | |
|
2009-10-06
| ||
| 14:59 | Change tclsqlite.c to use the Tcl_NRxxx() APIs in Tcl versions 8.6 and later. (check-in: e9f72f1d user: dan tags: trunk) | |
|
2009-10-05
| ||
| 05:40 | Fix some errors in the guttman versions (disabled by default) of the algorithms in rtree.c. (check-in: 64bad00b user: dan tags: trunk) | |
|
2009-10-03
| ||
| 07:04 | Handle an SQLITE_IGNORE returned when requesting authorization to read parent key columns by pretending the parent key columns contain NULL values. (check-in: 3c24df38 user: dan tags: trunk) | |
|
2009-10-02
| ||
| 15:29 | Add one more authentication test to fkey2.test to cover an untested branch. (check-in: e4fa8be7 user: dan tags: trunk) | |
| 14:23 | When inserting a row into a child table, invoke the authorization callback to request permission to read the parent key columns. (check-in: 9842f2d5 user: dan tags: trunk) | |
| 06:35 | Add a test to check that the incrblob API cannot be used to write to an IPK column. Also a comment to explain why the incrblob code does not need to check if a column is part of a parent key before writing to it. (check-in: dca2a7f6 user: dan tags: trunk) | |
|
2009-10-01
| ||
| 17:13 | Add test cases to cover the branches added by the previous commit. (check-in: aaa005b6 user: dan tags: trunk) | |
| 16:09 | If an update does not modify any child or parent key columns, omit foreign key processing for the statement. (check-in: edff3500 user: dan tags: trunk) | |
| 04:35 | Add tests to check that FK support interacts with count-changes correctly. (check-in: 5b836615 user: dan tags: trunk) | |
|
2009-09-30
| ||
| 08:11 | Remove unreachable branches from fkey.c. Add a few tests to cover the remaining branches. (check-in: 0a0a17d8 user: dan tags: trunk) | |
| 04:28 | Change a couple of comments to use "SQLITE_MUTEX_OMIT" instead of OMIT_MUTEX. (check-in: b733e939 user: dan tags: trunk) | |
|
2009-09-29
| ||
| 16:38 | Use the affinity and collation sequence associated with the parent key when finding child table rows to apply a foreign key action to. (check-in: 9a4b7ec2 user: dan tags: trunk) | |
| 15:41 | Check that a unique index uses the default collation sequences for each column before using it as part of a foreign key constraint operation. (check-in: 64154174 user: dan tags: trunk) | |
| 11:28 | Remove an unused variable from fkey.c. (check-in: 582bd768 user: dan tags: trunk) | |
| 06:33 | Various fixes so that "make test" works with OMIT_FOREIGN_KEY, OMIT_TRIGGER and OMIT_ALTER_TABLE. (check-in: e4eb227b user: dan tags: trunk) | |
|
2009-09-28
| ||
| 18:52 | Ignore foreign key mismatch errors while compiling DROP TABLE commands. (check-in: 5b4d4637 user: dan tags: trunk) | |
| 14:49 | Fix the DROP TABLE command so that it cannot be used to bypass foreign key constraints (if foreign key support is enabled). (check-in: 8353808c user: dan tags: trunk) | |
| 11:54 | Fix some foreign key constraint related problems that occur when a row refers to itself. (check-in: 9e503e2d user: dan tags: trunk) | |
|
2009-09-26
| ||
| 17:51 | When ALTER TABLE RENAME TO is used to change the name of a table that is the parent table of a foreign key constraint, modify that foreign key constraint to use the new table name. (check-in: b4a10c39 user: dan tags: trunk) | |
|
2009-09-25
| ||
| 17:03 | Avoid checking if an insert or delete has "fixed" an outstanding FK constraint violation if the constraint counter indicates that the database contains no such violations. (check-in: 519144ac user: dan tags: trunk) | |
| 12:00 | Fix comments in fkey2.c to reflect the immediate-constraint-counter approach. (check-in: 9fd54b0a user: dan tags: trunk) | |
| 11:26 | Prevent ALTER TABLE from being used to add a column with a REFERENCES clause and a non-NULL default value while foreign key support is enabled. (check-in: 353b1b18 user: dan tags: trunk) | |
|
2009-09-24
| ||
| 18:19 | Fixes and test cases to make sure the affinity and collation sequence associated with the parent key is used when comparing it with a child key value. (check-in: 76e8e74b user: dan tags: trunk) | |
| 16:52 | Fix a bug in the code for REPLACE conflict handling on IPK columns when there are no indexes on the table. Triggers and foreign key processing were being bypassed. (check-in: beb2094f user: dan tags: trunk) | |
| 15:16 | Fix a problem in fkey.c when a child key is the IPK of its table. (check-in: 8a000a8a user: dan tags: trunk) | |
| 14:27 | Remove the ALWAYS() added by the previous check-in. It is not needed. (check-in: 36568963 user: drh tags: trunk) | |
| 13:48 | Comment enhancements. Add an ALWAYS macro in fkey.c to indicate a branch that is always taken. (check-in: d5b714cf user: drh tags: trunk) | |
| 11:31 | Ensure foreign key related processing takes place when rows are deleted from the database by REPLACE conflict handling. (check-in: 3f40c142 user: dan tags: trunk) | |
| 10:42 | Use sqlite3FkOldmask() in delete.c instead of assuming that foreign key constraints always require all columns of the deleted row to be stored in registers. (check-in: ce554a39 user: dan tags: trunk) | |
| 09:05 | Remove unused parameter from sqlite3CodeRowTrigger(). Fix header comments for this function and CodeRowTriggerDirect(). (check-in: 0443f7c9 user: dan tags: trunk) | |
| 00:09 | Make sure a transaction is available for rollback whenever a REDUCE conflict resolution occurs and there is the possibility to ABORT. Ticket [4a03edc4c8c] (check-in: f0c56fa9 user: drh tags: trunk) | |
|
2009-09-23
| ||
| 18:49 | More fkey tests. (check-in: 2d544bd5 user: shane tags: trunk) | |
| 18:07 | Fix a problem with FK constraints that implicitly map to a composite primary key. (check-in: e0a48d53 user: dan tags: trunk) | |
| 17:31 | Fix a problem in the fkey_malloc.test script. (check-in: 0ce1efa4 user: dan tags: trunk) | |
| 17:30 | Do not check immediate foreign key constraints until the end of the statement. This matches the postgres behaviour. (check-in: 1a32149c user: dan tags: trunk) | |
| 15:51 | Modify the ".dump" command on the CLI so that it always issues a PRAGMA foreign_keys=OFF at the top of the output. (check-in: 0755b9b6 user: drh tags: trunk) | |
| 14:45 | Change the version number to 3.6.19. Fix a couple of incorrect testcase() macros associated with the new IS and IS NOT operators in expr.c. (check-in: e3b73394 user: drh tags: trunk) | |
| 13:59 | Disallow using the implicit rowid column as the parent key of an FK constraint. (check-in: a5f7e71f user: dan tags: trunk) | |
| 13:39 | Add test cases for the IS and IS NOT operator. (check-in: 101ed58c user: drh tags: trunk) | |
| 12:06 | Add extra tests for foreign key support. (check-in: 7d086afe user: dan tags: trunk) | |
| 08:43 | Add missing comments to fkey.c. Also, change the terminology used for comments and names in fkey.c from "referenced/referencing" to "parent/child". This is arguably less correct, but is easier to follow. (check-in: 540c2d18 user: dan tags: trunk) | |
| 03:01 | Do not run ON UPDATE actions of a foreign key constraint unless at least one column value really does change. (check-in: 71ac8e28 user: drh tags: trunk) | |
| 02:29 | Generalize the IS and IS NOT operators so that their right-hand side can be an arbitrary expression and not simple the constant NULL. They work like = and <> except that NULL values compare equal to one another an unequal to everything else. (check-in: 98853f61 user: drh tags: trunk) | |
|
2009-09-22
| ||
| 20:08 | Factor the UMINUS and UPLUS tokens out of the parser so that the parser tables can go back to using 8-bit values instead of 16-bit values. (check-in: 3fc938c9 user: drh tags: trunk) | |
| 19:53 | The NO ACTION action really should be no-action and not RESTRICT. (check-in: 55b263fa user: drh tags: trunk) | |
| 16:55 | Fix another OOM related problem in fkey.c. (check-in: e2bc51bc user: dan tags: trunk) | |
| 16:08 | Fix a problem with ON DELETE SET DEFAULT actions. (check-in: 94069950 user: dan tags: trunk) | |
| 15:53 | Fix an OOM related crash in fkey.c. (check-in: 635d6a77 user: dan tags: trunk) | |
| 13:25 | Allow specific exclusion of localtime_s() usage on Windows. (check-in: 216bcda7 user: shane tags: trunk) | |
| 07:13 | Fix compilation errors when OMIT_FOREIGN_KEY is defined. (check-in: c447cb37 user: dan tags: trunk) | |
|
2009-09-21
| ||
| 18:56 | Fix a problem with foreign key constraints that map from and IPK column. (check-in: 84129052 user: dan tags: trunk) | |
| 16:34 | Fix compilation with both OMIT_UTF16 and ENABLE_STAT2 defined. Ticket [56928bd084]. (check-in: cd850d49 user: dan tags: trunk) | |
| 16:06 | Fix compilation with OMIT_TRIGGER defined. Ticket [1ff6d29030]. (check-in: fb6ceed3 user: dan tags: trunk) | |
|
2009-09-19
| ||
| 17:59 | Add fkey.c to the autoconf and amalgamation build systems. (check-in: aab7a4b3 user: dan tags: trunk) | |
| 17:00 | Check in implementation of foreign key constraints. (check-in: d5d39981 user: dan tags: trunk) | |
|
2009-09-17
| ||
| 00:41 | When coding a trigger, assume that the "oldmask" requires all columns until we know otherwise. That pessimistic assumption assures that all necessary parameters are available on a cascading delete trigger. Ticket [e25d9ea771f] (check-in: 03e464be user: drh tags: trunk) | |
|
2009-09-14
| ||
| 23:47 | Make sure sufficient space is allocated for at least one VDBE cursor when autoincrement is used. Ticket [a696379c1f088]. (check-in: 2b2a1ef7 user: drh tags: trunk) | |
|
2009-09-11
| ||
| 18:28 | Set the execute permission bit on "configure". (check-in: f8c02b2b user: drh tags: trunk) | |
| 17:59 | Make sure the (unsupported) configure script in the source tree has execute permission prior to constructing a tarball. Ticket [65bbb65a47ce]. (check-in: 5e0e941f user: drh tags: trunk) | |
| 14:05 | Version 3.6.18 (check-in: b084828a user: drh tags: trunk, release) | |
|
2009-09-10
| ||
| 22:30 | Omit the mutexIsInit test variable when mutexes are disabled. (check-in: 3ea10434 user: drh tags: trunk) | |
| 20:23 | Fixed some warnings and an assert in the WINCE code when compiling under MSVC. (check-in: f42ec993 user: shane tags: trunk) | |
| 19:20 | Make sure the threadsOverrideEachOthersLocks variable is only referenced by linux systems on a threadsafe compile. (check-in: e9d064bd user: drh tags: trunk) | |
| 19:09 | Cleanup of warnings on MSVC build. (check-in: 5588cc96 user: shane tags: trunk) | |
| 18:26 | Update an r-tree test to account for changes in the query planner. (check-in: e5ce66d4 user: dan tags: trunk) | |
| 18:17 | Define SQLITE_TEST when compiling mem5.c for testfixture. (check-in: 6b16f3c9 user: dan tags: trunk) | |
| 18:04 | Fix a problem in test script corrupt.test. (check-in: dad2b74a user: dan tags: trunk) | |
| 17:45 | Add assert() statement to verify that new mutexes are not allocated when the mutex subsystem is uninitialized. (check-in: 1183c533 user: drh tags: trunk) | |
| 16:14 | Fix a problem with the sqlite3VdbeMayAbort() assert failing following an OOM. (check-in: b3027863 user: dan tags: trunk) | |
| 10:15 | Remove an ALWAYS macro around an expression that is sometimes false. (check-in: f2a9ee72 user: dan tags: trunk) | |
| 02:54 | Reduce default SQLITE_MAX_TRIGGER_DEPTH when SQLITE_SMALL_STACK is defined. (check-in: 913fb70e user: shane tags: trunk) | |
|
2009-09-09
| ||
| 18:46 | Trying to open a transaction in one thread and close it in another is a misuse with LinuxThreads. Doing so may cause memory and file-descriptors to be leaked. Update an assert() and some test cases to account for this. (check-in: ef99eb57 user: dan tags: trunk) | |
| 16:10 | Suppress some harmless compiler warnings. (check-in: f0c72a53 user: drh tags: trunk) | |
| 15:29 | Bug fix in the SQLITE_OPEN_PRIVATECACHE option added a few minutes ago. (check-in: f3a0f23b user: drh tags: trunk) | |
| 15:06 | Add some tests that involve BEFORE UPDATE or BEFORE DELETE triggers messing with the row being updated or deleted. SQLite behaviour in this scenario is undefined, so the tests just check that the behaviour is relatively sane and there are no crashes. (check-in: 9a4c59a2 user: dan tags: trunk) | |
| 14:48 | Update the README file to state that TCL is required in order to run the makefiles for SQLite. Ticket [7d96113ff]. (check-in: ba20091a user: drh tags: trunk) | |
| 14:17 | Added SQLITE_OPEN_SHAREDCACHE and SQLITE_OPEN_PRIVATECACHE flags as possibilities to sqlite3_open_v2(), to override the global shared cache mode setting. Ticket [9fd0bc36639c15] (check-in: f509eb8b user: drh tags: trunk) | |
| 12:41 | Increase the size of bitvec objects to 1024 bytes on 64-bit systems. (check-in: f1272b90 user: drh tags: trunk) | |
| 11:43 | Add a test case to show that 29ab7be99f is fixed. (check-in: 135d656a user: dan tags: trunk) | |
| 11:37 | Add a complicated assert() to check that statement transactions are opened when required. (check-in: 28aa1f4e user: dan tags: trunk) | |
|
2009-09-08
| ||
| 19:15 | Combine the OP_Statement and OP_Transaction opcodes. (check-in: aec9dbd8 user: dan tags: trunk) | |
| 15:55 | If recursive-triggers are enabled, fire DELETE triggers if database rows are removed as a result of OR REPLACE conflict resolution. (check-in: 85cb0c94 user: dan tags: trunk) | |
| 13:40 | Additional simplifications in support of structural testing. (check-in: 4ab8c841 user: drh tags: trunk) | |
| 02:27 | Further simplifications to the VDBE - removing functionality that is no longer used. (check-in: b271e166 user: drh tags: trunk) | |
| 01:14 | Code simplifications, especially to the pseudo-table logic, and comment improvements. (check-in: 52449a95 user: drh tags: trunk) | |
|
2009-09-07
| ||
| 18:14 | Add the SQLITE_LIMIT_TRIGGER_DEPTH option to sqlite3_limit(). (check-in: d7dc8b43 user: drh tags: trunk) | |
| 17:41 | Merge 1958db4493 and 26cd015c0e. (check-in: ecbe0832 user: dan tags: trunk) | |
| 17:32 | Fix a bug where a condition was incorrectly being compiled out when SQLITE_OMIT_VIRTUALTABLE was defined. (check-in: 1958db44 user: dan tags: trunk) | |
| 04:38 | Warning cleanup from the MSVC compile. (check-in: 26cd015c user: shane tags: trunk) | |
| 03:48 | Attempt to handle numbers at the limits of the IEEE 754 range better (nan.test now passes). (check-in: 2a74ae19 user: shane tags: trunk) | |
| 02:46 | Changes to internal AtoF to improve "accuracy" when measured against built-in GCC atof(). (check-in: 438e1577 user: shane tags: trunk) | |
|
2009-09-03
| ||
| 16:23 | Add the "unix-wfl" VFS that does whole-file locking in order to help NFS do better cache coherency. (check-in: 2aeab80e user: drh tags: trunk) | |
| 01:18 | Do not reuse function parameters in subsequent expressions since the function call might have triggered a text encoding change. Fix for ticket [2ea2425d34be]. (check-in: f22e3887 user: drh tags: trunk) | |
|
2009-09-02
| ||
| 19:04 | Documentation updates for the version info #defines and C interfaces. (check-in: 69055e9b user: drh tags: trunk) | |
|
2009-09-01
| ||
| 17:28 | Fix a problem in main.c. SQLITE_NoRecTriggers -> SQLITE_RecTriggers. (check-in: 5ac69ed6 user: dan tags: trunk) | |
| 17:11 | Change "PRAGMA disable_recursive_triggers" to "PRAGMA recursive_triggers". Also a fix for compiling with OMIT_TRIGGER defined. (check-in: e016cca3 user: dan tags: trunk) | |
| 16:39 | Merge in changes to support recursive invocation of triggers (disabled by default). (check-in: c1b388c3 user: dan tags: trunk) | |
| 16:19 | Add new test script triggerC.test. (check-in: a2a062a4 user: dan tags: trunk) | |
| 12:16 | More fixes and comment updates. (check-in: 38a9327b user: dan tags: trunk) | |
|
2009-08-31
| ||
| 16:09 | More thorough initialization of a buffer when writing the rollback journal header, in order to prevent a harmless valgrind warning. CVSTrac ticket #4039 (check-in: 5e2f5496 user: drh tags: trunk) | |
| 15:27 | More fixes for test cases. (check-in: 85d9f23b user: dan tags: trunk) | |
| 08:22 | Fix some authorization callback problems. (check-in: 8a746fbf user: dan tags: trunk) | |
| 05:39 | Merge with main branch. (check-in: 6e09e287 user: dan tags: trunk) | |
| 05:23 | Fix another test problem and some instances where an OOM may cause a segfault. (check-in: 31199db0 user: dan tags: trunk) | |
|
2009-08-30
| ||
| 11:42 | Fixes for new triggers scheme. (check-in: 9eb91efd user: dan tags: trunk) | |
|
2009-08-28
| ||
| 18:53 | Changes to support recursive triggers. (check-in: 9b9c1921 user: dan tags: trunk) | |
| 00:49 | Add a new source code logo gif. (check-in: 6abcba10 user: drh tags: trunk) | |
|
2009-08-27
| ||
| 17:56 | Patch to the named semaphore locking mechanism used by VxWorks. (check-in: 609c5341 user: drh tags: trunk) | |
|
2009-08-26
| ||
| 00:26 | Updates to the pluggable page cache documentation. (check-in: 3085ad76 user: drh tags: trunk) | |
|
2009-08-25
| ||
| 16:28 | Remove an unreachable branch from where.c in order to restore 100% branch test coverage. Add assert() and testcase() macros to verify that the branch is unreachable. (check-in: 58db7e71 user: drh tags: trunk) | |
| 15:56 | Initialize variables differently in the range processing logic of where.c in order to make sure variables are always initialized even following an OOM error. (check-in: 3fb3686a user: drh tags: trunk) | |
| 14:59 | Attempt to clarify the meaning of a "parameter" in the sqlite3_bind() API documentation. (check-in: 9389e6a7 user: drh tags: trunk) | |
| 14:24 | Make sure that the output of EXPLAIN is right when the P4 argument of an opcode is of type P4_MEM with the MEM having type MEM_Blob. (check-in: 6a623e6c user: drh tags: trunk) | |
| 13:53 | Avoid calling sqite3VdbeRecordCompare() with uninitialized memory following an OOM. Doing so is harmless, but it worries valgrind. (check-in: d33b9eb2 user: drh tags: trunk) | |
| 13:10 | Fix an uninitialized variable in sqlite3VdbeIdxRowid(). (check-in: 6e787e18 user: drh tags: trunk) | |
| 12:11 | Merge together the os_unix.c fix of [aa6acfa8ca] and the trigger fix of [dee1b8eb40]. (check-in: 1e2c6e13 user: drh tags: trunk) | |
| 05:57 | Fix a bad interaction between "proxy-locking" and aa6acfa8ca. (check-in: 2a5c9e1d user: dan tags: trunk) | |
|
2009-08-24
| ||
| 19:37 | Update the version number to 3.6.18, even though the actual release of that version is still a few weeks away. (check-in: c0ea4e9a user: drh tags: trunk) | |
| 18:57 | Fix some errors in aa6acfa8ca. (check-in: 82d1934a user: dan tags: trunk) | |
| 01:35 | Run the column cache in a new context when generating code for trigger programs. Fix for ticket [efc02f9779]. (check-in: dee1b8eb user: drh tags: trunk) | |
|
2009-08-22
| ||
| 19:17 | Remove an obsolete documentation file left over from SQLite version 1.0. (check-in: f7eb1efc user: drh tags: trunk) | |
| 11:39 | Fix a problem in os_unix.c where a malloc failure could lead to a leaked file descriptor. (check-in: aa6acfa8 user: dan tags: trunk) | |
|
2009-08-21
| ||
| 17:18 | When a database file is opened, try to find an unused file descriptor to reuse. This change affects unix (and other systems that use os_unix.c) only. Fix for cvstrac ticket #4018. (check-in: 9b4d9ab6 user: dan tags: trunk) | |
| 13:22 | Change the expression code generator to account for the fact that the new sqlite3AtoF() never returns NaN. Also, clarification of a comment in where.c. (check-in: 75f596a0 user: drh tags: trunk) | |
| 08:29 | Add assert() statements to os_unix.c to check that the mutex is held when it should be. (check-in: 11a669b6 user: dan tags: trunk) | |
| 02:13 | Updated sqlite3AtoF() that performs slightly better with GCC, and significantly better with MSVC. (check-in: f084f5a8 user: shane tags: trunk) | |
| 02:07 | Updates for test_init.c for configure and other consistency changes. (check-in: 3ba316e9 user: shane tags: trunk) | |
|
2009-08-20
| ||
| 23:05 | Remove a redundant OOM test from the sqlite_stat2 analysis loader. (check-in: d30ceeb8 user: drh tags: trunk) | |
| 20:05 | Simplifications to the range-scan logic in support of structural testing. (check-in: bbbee810 user: drh tags: trunk) | |
| 18:14 | Continuing refinements of the range-scan optimizations in where.c. The range scores are changed from an integer 1..9 to 0..100. (check-in: f0c24b5f user: drh tags: trunk) | |
| 16:11 | Change the code that collects samples for sqlite_stat2 so that the first sample taken is the (nRow/(2*SQLITE_INDEX_SAMPLES))th entry in the index, where nRow is the total number of index entries. (check-in: cbfe6e9d user: dan tags: trunk) | |
| 13:45 | Incremental code and comment cleanup in where.c. There is more to be done. (check-in: 4a5d9550 user: drh tags: trunk) | |
| 02:49 | Set the "type" correctly of built-in BINARY collating sequences for UTF16. (check-in: 167644f3 user: drh tags: trunk) | |
| 02:34 | All the sqlite3GetCollSeq() function to specify an arbitrary text encoding. (check-in: 4ee44322 user: drh tags: trunk) | |
|
2009-08-19
| ||
| 22:14 | Make sure that sqlite3ValueFromExpr() sets the "type" field of the sqlite3_value object correctly when the value is a floating point number. (check-in: 2467eb68 user: drh tags: trunk) | |
| 17:17 | Change the way an Expr that has been converted to TK_REGISTER stores its original opcode. The previous method, overloading Expr.iColumn, did not work in all cases. (check-in: 71ea9d0e user: dan tags: trunk) | |
| 16:34 | Fix an assert() failure that may follow an OOM error. (check-in: 14a715c5 user: dan tags: trunk) | |
| 16:21 | Merge 29cafcfdcc and a6f39181a7. (check-in: 740a93e8 user: dan tags: trunk) | |
| 15:57 | Documentation improvements in sqlite.h.in. No changes to code. (check-in: a6f39181 user: drh tags: trunk) | |
| 15:34 | Add some tests that use the sqlite_stat2 table in shared-cache mode. (check-in: 29cafcfd user: dan tags: trunk) | |
| 14:42 | Merge with 709e16145e. (check-in: 6c8a0e2b user: dan tags: trunk) | |
| 14:34 | Clean up sqlite_stat2 related code. Add test cases. (check-in: aa728e06 user: dan tags: trunk) | |
| 09:09 | Fix a problem where a buffer allocated from a lookaside pool was being released using the system free(). (check-in: 67207a15 user: dan tags: trunk) | |
| 08:18 | Add the SQLITE_ENABLE_STAT2 macro. If this is not defined at build-time, the stat2 table is not created, populated, or used. (check-in: 362665e8 user: dan tags: trunk) | |
| 04:24 | Adjust sync count for SQLITE_DISABLE_DIRSYNC compiler option. (check-in: 709e1614 user: shane tags: trunk) | |
|
2009-08-18
| ||
| 16:24 | Change the sqlite_stat2 schema to be more flexible. (check-in: ded9dec6 user: dan tags: trunk) | |
| 16:05 | Unknown functions in the DEFAULT clause of a table cause an error when the DEFAULT value is needed. Ticket [2d401a94287b5]. (check-in: 093917d7 user: drh tags: trunk) | |
| 15:33 | Move the allocation of the memsys5 mutex into the initializer. (check-in: 4e377a09 user: drh tags: trunk) | |
| 14:48 | Fix obscure issues with the memsys5 memory allocator. Arrange that the xRealloc() interface to memory allocators is only called with a value that has been through xRoundup(). (check-in: 577bd6f1 user: drh tags: trunk) | |
| 12:16 | When shutting down the memsys5 memory allocator, be sure to clear the mutex pointer in case the next startup does not use a mutex because it is configured differently. (check-in: d4e7e2d8 user: drh tags: trunk) | |
| 01:54 | Fix a bug in mem5.c which would cause an infinite loop on an attempt to allocate more than 1073741824 bytes of contiguous memory. Also, some cleanup of mem5.c. More work to do on this. (check-in: 783b751a user: drh tags: trunk) | |
|
2009-08-17
| ||
| 17:06 | First version of sqlite_stat2 (schema forces exactly 10 samples). (check-in: dd96bda2 user: dan tags: trunk) | |
| 16:01 | Always call sqlite3_malloc() in sqlite3OsInit(), even when not compiled with SQLITE_TEST. (check-in: b98a8706 user: drh tags: trunk) | |
| 15:52 | Move error simulation code from the sqlite3_os_init() functions into a wrapper. (check-in: 67ad21ab user: dan tags: trunk) | |
| 15:31 | More documentation and comment updates for sqlite3_initialize/shutdown interface changes to handle failures. (check-in: 32509bc7 user: shane tags: trunk) | |
| 15:16 | Add tests to check that sqlite recovers from an error in sqlite3_initialize() correctly. (check-in: 904a371c user: dan tags: trunk) | |
| 13:42 | Enhanced documentation and minor code tweaks in preparation for hardening the sqlite3_initialize/shutdown interfaces against initialization failures. (check-in: 98c49e61 user: drh tags: trunk) | |
| 09:29 | Add a test to verify that 3.6.17 releases read-locks obtained to read the database schema even if the connection is not currently in auto-commit mode. (check-in: 78dfe832 user: dan tags: trunk) | |
|
2009-08-14
| ||
| 18:18 | Update the amalgamation builder so that it avoids putting redundant SQLITE_API macros on declarations. (check-in: 0d5b0587 user: drh tags: trunk) | |
| 17:53 | Work around an over-zealous optimization in GCC 4.3.3. See CVSTrac ticket #4027. (check-in: 9cbe3654 user: drh tags: trunk) | |
| 17:01 | Fix a case where SQLite may write past the end of a buffer as a result of a corrupted database file. (check-in: 43321a55 user: dan tags: trunk) | |
| 16:15 | Fix to the "publish.sh" script that was broken by the previous check-in. (check-in: 34c21210 user: drh tags: trunk) | |
| 16:01 | Incorporate fossil-scm version information into the build. Add the SQLITE_SOURCE_ID macro to the header. Add the sqlite3_sourceid() interface. Add the sqlite_source_id() SQL function. (check-in: 302dabe9 user: drh tags: trunk) | |
|
2009-08-13
| ||
| 20:15 | Fix typos in comments used to generate documentation. CVSTrac tickets #4028 and #4029. (check-in: 1fc73b91 user: drh tags: trunk) | |
| 19:54 | Tweak to the new whereB.test file to make it more consistent. (check-in: 06098505 user: drh tags: trunk) | |
| 19:21 | If a binary operator in a WHERE clause that should be performed with no affinity conversions applied to its operands (see http://www.sqlite.org/datatype3.html) is optimized by index lookup, do not apply any conversions to the key value before looking it up in the index. Fix for [93fb9f89d6]. (check-in: e72186f2 user: dan tags: trunk) | |
| 18:14 | Enhancements to the whereB.test to check more affinity corner cases. (check-in: 10484598 user: drh tags: trunk) | |
| 17:14 | Add a test case for the affinity problem reported by ticket [93fb9f89d6]. (check-in: 149ec24e user: drh tags: trunk) | |
| 15:42 | Merge the accidental fork. (check-in: 86a06dd0 user: drh tags: trunk) | |
| 15:13 | Fix a typo on a comment in sqlite3VdbeIntegerAffinity(). (check-in: b5a709d3 user: drh tags: trunk) | |
| 07:09 | Changes to the query planner that improve the order in which tables/indexes are scanned in join queries. (check-in: 19f799b3 user: dan tags: trunk) | |
|
2009-08-12
| ||
| 15:34 | Fixed some compiler warnings in WINCE only sections when using the MSVC compiler. (check-in: 1f0a93e1 user: shane tags: trunk) | |
| 11:45 | Make sure the large-file support macros occur first in sqliteInt.h. Fix for CVSTrac ticket #4022. (check-in: b0848925 user: drh tags: trunk, cvs-to-fossil-cutover) | |
|
2009-08-11
| ||
| 05:50 | Instead of adding SQLITE_API macros sqlite3.h while creating sqlite3.c, add them as part of generating the sqlite3.h target, while copying from sqlite.h.in to sqlite3.h. This fixes a build problem that was causing multiple SQLITE_API macros to be added to a single line of sqlite3.h, and the problem cited by #4016. (CVS 6971) (check-in: 7f481074 user: danielk1977 tags: trunk) | |
|
2009-08-10
| ||
| 17:48 | Reduce the maximum supported sector size to 64KiB. (CVS 6970) (check-in: 8e1e0ba2 user: drh tags: trunk) | |
| 15:16 | Version 3.6.17 (CVS 6969) (check-in: 36650102 user: drh tags: trunk, release) | |
| 10:18 | Update a couple of tests in corrupt7.test to account for a change in the error message text emitted by SQLite. (CVS 6968) (check-in: 3172a296 user: danielk1977 tags: trunk) | |
| 04:37 | Exclude a static function in test1.c from the build when OMIT_DEPRECATED is defined. This avoids a compiler warning that occurs with some compilers. (CVS 6967) (check-in: ace49950 user: danielk1977 tags: trunk) | |
| 04:26 | Change mallocI.test to close a database connection that was being left open. (CVS 6966) (check-in: 4f0096d7 user: danielk1977 tags: trunk) | |
| 03:57 | Removed some warnings from the MSVC compile. (CVS 6965) (check-in: ee034cfc user: shane tags: trunk) | |
| 03:23 | Fixed "Interlocked" logic error that prevented the "static" mutexes from being released when sqlite3_shutdown() is called. Ticket #4011. (CVS 6964) (check-in: 04cafb7e user: shane tags: trunk) | |
|
2009-08-08
| ||
| 18:01 | Suppress various compiler warnings. (CVS 6963) (check-in: 257e9b57 user: drh tags: trunk) | |
|
2009-08-07
| ||
| 16:56 | Enhancement to (6844). If sqlite3_exec() is to return SQLITE_NOMEM because it fails to allocate space for an error message, set the database error-code to SQLITE_NOMEM too. (CVS 6962) (check-in: 8293fa19 user: danielk1977 tags: trunk) | |
|
2009-08-06
| ||
| 18:36 | Mark the rtreeUpdate function as static. (CVS 6961) (check-in: b6bdfdc6 user: danielk1977 tags: trunk) | |
| 17:43 | Alternative fix for ticket #3810. This is a replacement for check-in (6956). (CVS 6960) (check-in: ba1afc04 user: drh tags: trunk) | |
| 17:40 | Change a hyperlink label on shared cache mode documentation. No changes to code. (CVS 6959) (check-in: 3d08ca0e user: drh tags: trunk) | |
|
2009-08-05
| ||
| 04:08 | Simplified conditional for read-only lock in winceLockFile() function. Ticket #3991. (CVS 6958) (check-in: a1c184cd user: shane tags: trunk) | |
|
2009-08-01
| ||
| 18:22 | Make sure all database connections are closed at the end of the test case for ticket #3810. (CVS 6957) (check-in: 9bca5a0e user: drh tags: trunk) | |
| 16:27 | Return a better error message when problems are encountered parsing a TEMP trigger or TEMP view that references objects in other databases that have been modified or dropped. Ticket #3810. (CVS 6956) (check-in: 102785b9 user: drh tags: trunk) | |
| 15:54 | Add a testcase for ticket #3810. (CVS 6955) (check-in: 29972f74 user: drh tags: trunk) | |
| 15:09 | Fix a segfault following OOM that was introduced by check-in (6949) which was a fix for ticket #3997. (CVS 6954) (check-in: 359d78e1 user: drh tags: trunk) | |
| 15:04 | Increment the version number in preparation for the next release. (CVS 6953) (check-in: 12d9b873 user: drh tags: trunk) | |
|
2009-07-31
| ||
| 06:14 | Fix a bug in where.c that could cause SQLite to write to freed memory while compiling a query with many terms in the WHERE clause. (CVS 6952) (check-in: 8161af84 user: danielk1977 tags: trunk) | |
|
2009-07-29
| ||
| 06:04 | Add a test to mallocI.test to show that a shared lock is correctly released if an OOM condition occurs while parsing a database schema. (CVS 6951) (check-in: 5a82620e user: danielk1977 tags: trunk) | |
|
2009-07-28
| ||
| 16:44 | Add the experimental API sqlite3_strnicmp(). Modify fts3 so that in terms like 'column_name:token' the column_name is interpreted in a case-insenstive fashion. Ticket #3996. (CVS 6950) (check-in: 4571aa9e user: danielk1977 tags: trunk) | |
| 13:30 | When flattening a SELECT query, do not discard collation sequences explicitly attached to expressions in the parent query. Fix for #3997. (CVS 6949) (check-in: b2784cf5 user: danielk1977 tags: trunk) | |
| 08:43 | Fixed reversed labels in WHERETRACE() statement. Debugging code only. (CVS 6948) (check-in: 614a8d83 user: shane tags: trunk) | |
|
2009-07-27
| ||
| 14:15 | If there is a hot-journal in the file-system, a connection with journal_mode=memory set may have to open it to effect rollback. Account for this in pager_end_transaction(). This prevents an assert from failing in the in-memory journal permutation test. (CVS 6947) (check-in: 224bc69a user: danielk1977 tags: trunk) | |
| 11:41 | Do not simulate OOM conditions in the sqlite3OsXXX() calls if the underlying file is an in-memory journal file. (CVS 6946) (check-in: d4868117 user: danielk1977 tags: trunk) | |
| 10:05 | When extracting values from a record to use in an UPDATEd version of that record, apply OP_RealAffinity if required. Fix for #3992. (CVS 6945) (check-in: 3616766a user: danielk1977 tags: trunk) | |
|
2009-07-25
| ||
| 22:13 | The noReadlock property on a pager object implies the readOnly property. Use this fact to simplify the logic in sqlite3PagerSharedLock(). (CVS 6944) (check-in: 886e665f user: drh tags: trunk) | |
| 19:31 | Remove obsolete assert and testcase macros related to ticket #2565 (CVS 6943) (check-in: f0024b31 user: drh tags: trunk) | |
| 17:42 | Adjust memory size computations to avoid a conditional that is always false on 64-bit systems. (CVS 6942) (check-in: 04211e6a user: drh tags: trunk) | |
| 17:39 | Change an unreachable condition in PagerSharedLock to a NEVER(...). Add an assert to pager_error() to show that it is never called to put an in-memory pager to the error-state. (CVS 6941) (check-in: 1d931f77 user: danielk1977 tags: trunk) | |
| 17:33 | Change a condition to a '%' operator in bitvec.c. (CVS 6940) (check-in: 10250fe5 user: drh tags: trunk) | |
| 17:08 | More simplifications to pager.c. (CVS 6939) (check-in: 533d6cc6 user: drh tags: trunk) | |
| 15:24 | Simplifications and comment improvements to pager.c in support of coverage testing. (CVS 6938) (check-in: 5b70b5c1 user: drh tags: trunk) | |
| 14:18 | Add a NEVER() on an always-false conditional in pager.c. Make sure the text of the master journal file is zero-terminated before trying to process it - to prevent a buffer overrun in strlen(). (CVS 6937) (check-in: cc9430e3 user: drh tags: trunk) | |
| 13:42 | Remove the "proc zeroblob" implementation from incrvacuum2.test. It is no longer required as of (6906) and, as #3988 points out, unreliable. (CVS 6936) (check-in: 03153831 user: danielk1977 tags: trunk) | |
| 11:46 | Add files that should have gone in with the previous checkin. (CVS 6935) (check-in: 64c7afc7 user: danielk1977 tags: trunk) | |
| 11:40 | Add a complex assert() that checks a pager invariant. (CVS 6934) (check-in: 2e08ad7b user: danielk1977 tags: trunk) | |
| 04:12 | Add assert() statements to more tightly constrain the state of pager.c. Remove the three pager*.test scripts since they violate the constraints asserted above by modifying the state of the system in ways that it cannot be modified in a live system. (CVS 6933) (check-in: 3b6d370e user: drh tags: trunk) | |
| 00:13 | Remove two more overzealous assert() statements. (CVS 6932) (check-in: 788aa9dd user: drh tags: trunk) | |
| 00:10 | Remove an overzealous assert() from pager.c. (CVS 6931) (check-in: 4a924b47 user: drh tags: trunk) | |
|
2009-07-24
| ||
| 21:23 | Fix an issue with the previous check-in, (6929). (CVS 6930) (check-in: e5165c80 user: drh tags: trunk) | |
| 19:01 | Further simplifications to pager.c in support of structural coverage testing. (CVS 6929) (check-in: 866216c0 user: drh tags: trunk) | |
| 17:58 | Allow virtual tables to be used in shared-cache mode. (CVS 6928) (check-in: 5d9e767a user: danielk1977 tags: trunk) | |
| 16:32 | Simplifications to pager.c in support of structural coverage testing. (CVS 6927) (check-in: 7222ad26 user: drh tags: trunk) | |
| 12:35 | Simplifications and comment improvements in pager.c. (CVS 6926) (check-in: 2d2f42ca user: drh tags: trunk) | |
|
2009-07-23
| ||
| 01:44 | Simplifications and comment enhancements on btree.c. (CVS 6925) (check-in: 5ba880dd user: drh tags: trunk) | |
|
2009-07-22
| ||
| 18:24 | Modify mksqlite3c.tcl so that it inserts SQLITE_API macros into sqlite3.h. Ticket #3983. (CVS 6924) (check-in: 803ec79f user: danielk1977 tags: trunk) | |
| 18:07 | Simplifications to btree.c in support of coverage testing. (CVS 6923) (check-in: 375fd6f9 user: drh tags: trunk) | |
| 16:41 | Avoid leaving a suspect page in the page-cache if an error occurs during sqlite3PagerAcquire(). (CVS 6922) (check-in: 2594f01c user: danielk1977 tags: trunk) | |
| 14:08 | Fix a double-free that can occur after database corruption is detected. (CVS 6921) (check-in: 5260e8b1 user: danielk1977 tags: trunk) | |
| 13:19 | Add a NEVER() macro around a pager.c conditional that must always be false in the current implementation. (CVS 6920) (check-in: 1e510d7f user: drh tags: trunk) | |
| 07:27 | Changes to test code (only) to allow veryquick.test to run with OMIT_UTF16 builsd. Ticket #3985. (CVS 6919) (check-in: 7f3fc7f2 user: danielk1977 tags: trunk) | |
| 02:02 | Remove an ALWAYS() from a condition that can in fact be false under very obscure circumstances. (CVS 6918) (check-in: 38083228 user: drh tags: trunk) | |
| 00:35 | Further simplifications to btree.c, especially the sqlite3BtreeKey() and sqlite3BtreeData() functions. New assert() statements added to verify that these routines are called correctly. (CVS 6917) (check-in: 96cfd079 user: drh tags: trunk) | |
|
2009-07-21
| ||
| 19:25 | Change getAndInitPage() (btree.c) to use only PagerAcquire(), not PagerLookup() and PagerAcquire(). (CVS 6916) (check-in: 0b41dfc0 user: danielk1977 tags: trunk) | |
| 19:02 | Additional simplifications to btree.c in support of coverage testing. (CVS 6915) (check-in: 716fccea user: drh tags: trunk) | |
| 15:33 | Remove an assert() in btree.c which is no longer true due to changes in the error reporting behavior of ptrmapPut(). (CVS 6914) (check-in: 110998f1 user: drh tags: trunk) | |
| 11:52 | Changes to btree.c in support of coverage testing. (CVS 6913) (check-in: 4cf23e9e user: drh tags: trunk) | |
|
2009-07-20
| ||
| 19:30 | Reverse the order of two conditionals in a test in order to achieve coverage of them both. Also: clarifications to comments in btree.c. (CVS 6912) (check-in: a159e9d2 user: drh tags: trunk) | |
| 17:11 | Modify various routines inside btree.c to take a pointer to the return-code as a parameter and to no-op if the return-code storage location already contains a non-zero code. (CVS 6911) (check-in: 7dcf2a78 user: drh tags: trunk) | |
| 12:33 | Use an ALWAYS() on a conditional in btmutex.c that is always true. (CVS 6910) (check-in: 609022ca user: drh tags: trunk) | |
| 12:25 | Corrections to the comment that is inserted at the top of the amalgamation. Ticket #3981. (CVS 6909) (check-in: 7572991b user: drh tags: trunk) | |
| 11:32 | Always provide a name for the TEMP database even if SQLite is compiled using SQLITE_OMIT_TEMPDB. (CVS 6908) (check-in: 0ba17c04 user: drh tags: trunk) | |
|
2009-07-18
| ||
| 20:01 | Code tweaks in support of full-coverage testing. (CVS 6907) (check-in: 22c6dbfd user: drh tags: trunk) | |
| 14:36 | Add some code to sqlite3VdbeMemSetZeroBlob() so that it works (less efficiently) even when OMIT_INCRBLOB is defined. Ticket #3977. (CVS 6906) (check-in: f5f7259d user: danielk1977 tags: trunk) | |
| 11:52 | When the asynchronous IO backend opens a file with the EXCLUSIVE flag set, make sure only a single file-descriptor is opened (not one for reading and one for writing). This change fixes #3978. (CVS 6905) (check-in: 630e669b user: danielk1977 tags: trunk) | |
| 08:30 | Changes to test scripts so that (make test) works when OMIT_INCRBLOB is defined. (CVS 6904) (check-in: 1dd834a3 user: danielk1977 tags: trunk) | |
|
2009-07-17
| ||
| 17:25 | Fix a subtle problem that can occur when a transaction is automatically rolled back, and afterwards a VM that had opened a statement transaction within the abandoned transaction is finalized. Fixes an assert() that was failing in rtree3.test. (CVS 6903) (check-in: 910df46e user: danielk1977 tags: trunk) | |
| 16:54 | Return a meaningful error message if a keyword is used as an rtree table column name. Ticket #3970. (CVS 6902) (check-in: 046efe46 user: danielk1977 tags: trunk) | |
| 14:37 | Skip all tests in tableapi.test if SQLITE_OMIT_GET_TABLE defined. Ticket #3975. (CVS 6901) (check-in: 0219a543 user: shane tags: trunk) | |
| 11:44 | Code simplifications in support of structural testing. (CVS 6900) (check-in: fb1b955d user: drh tags: trunk) | |
|
2009-07-16
| ||
| 18:21 | Code simplifications and comment improvements in support of structural coverage testing. (CVS 6899) (check-in: 94525179 user: drh tags: trunk) | |
| 12:41 | Remove an unused variable from sqlite3ExprListDup(). Ticket #3973. (CVS 6898) (check-in: 17c05aef user: drh tags: trunk) | |
|
2009-07-15
| ||
| 18:15 | Fix a potential database corruption following DROP TABLE when the pending byte page corresponds to a ptrmap page. This situation cannot happen in a real deployment - but it still needs to be fixed. (CVS 6897) (check-in: 6242db39 user: drh tags: trunk) | |
| 17:25 | Simplifications to btree.c in support of structural testing. Renamed the "skip" field of the BtCursor object to "skipNext" to make it easier to search for places where it is used. (CVS 6896) (check-in: d3897235 user: drh tags: trunk) | |
| 16:30 | Remove an assert() from vdbeaux.c that might not be true if the database file is corrupt. (CVS 6895) (check-in: a42dc51e user: drh tags: trunk) | |
| 11:26 | Added the SQLITE_TESTCTRL_RESERVE option to sqlite3_test_control(). Simplifications to btree.c to facilitate structural coverage testing. (CVS 6894) (check-in: cfbeda1b user: drh tags: trunk) | |
|
2009-07-14
| ||
| 18:35 | Add comments and assert() statements to show that the return value from sqlite3BtreeKeySize() and sqlite3BtreeData() usually do not matter. Ticket #3968. Also remove a NEVER() macro that can sometimes be true - discovered while testing the previous change. (CVS 6893) (check-in: 0c710c1b user: drh tags: trunk) | |
| 17:48 | Simplifications to the PRAGMA integrity_check logic. Remove unreachable code. (CVS 6892) (check-in: 17749fa5 user: drh tags: trunk) | |
| 14:15 | Simplifications and tweaks to vdbeaux.c resulting from structural testing. (CVS 6891) (check-in: fa49666f user: drh tags: trunk) | |
| 02:33 | Make sure the IN operator works with zeroblobs. Ticket #3965. Other simplifications associated with structural testing. (CVS 6890) (check-in: 25dd3422 user: drh tags: trunk) | |
|
2009-07-13
| ||
| 15:52 | Remove the SQLITE_CommitBusy flag. This was an attempt to block recursion on the sqlite3_commit_hook() interface. But such recursion is explicitly disallowed, so the flag is pointless. (CVS 6889) (check-in: 1c2bfc43 user: drh tags: trunk) | |
| 13:18 | Change an unreachable condition in btree.c to a NEVER(). (CVS 6888) (check-in: 9f800e11 user: danielk1977 tags: trunk) | |
| 11:22 | In sqlite3PagerWrite(), do not set the PGHDR_NEED_SYNC flag on a page if an IO error occured while attempting to journal it. (CVS 6887) (check-in: b9be365d user: danielk1977 tags: trunk) | |
| 09:41 | Remove an assert ("assert( subpage>0 )") from btree.c that may not be true for a corrupt database. Also add comments and other assert() statements to btree.c function moveToRoot(). (CVS 6886) (check-in: 3151dab9 user: danielk1977 tags: trunk) | |
| 07:30 | Remove a case from BtreeMovetoUnpacked() that is unreachable as of (6881). (CVS 6885) (check-in: 39ce2097 user: danielk1977 tags: trunk) | |
|
2009-07-12
| ||
| 02:32 | Simplifications to btree.c to help facilitate coverage testing. (CVS 6884) (check-in: dfb146ef user: drh tags: trunk) | |
| 02:31 | When an OOM fault occurs while moving a page for autovacuum in an in-memory database, be sure to undo the page move prior to returning the error, to avoid corrupting the in-memory database file. (CVS 6883) (check-in: d0964b93 user: drh tags: trunk) | |
|
2009-07-11
| ||
| 18:26 | Remove an unused parameter from the accessPayload() function in btree.c. (CVS 6882) (check-in: 7deb6568 user: drh tags: trunk) | |
| 17:39 | Fix a case where deleting a row from a corrupt database could cause an assert to fail. (CVS 6881) (check-in: 6994b41a user: danielk1977 tags: trunk) | |
| 17:04 | Bug fix in the computation of the number of pages to autovacuum when nReserve is greater than zero. (CVS 6880) (check-in: 618a83d6 user: drh tags: trunk) | |
| 13:13 | Unwind some complex conditions in sqlite3BtreeDelete() into separate "if" statements. (CVS 6879) (check-in: d99bde9c user: drh tags: trunk) | |
| 11:45 | Remove another unreachable branch from btree.c. (CVS 6878) (check-in: b0853100 user: danielk1977 tags: trunk) | |
| 06:55 | Modify the integrity-check code so that each b-tree page inspected is reinitialized while doing so, even if an initialized copy exists in the cache. This prevents an assert from failing when running integrity_check on a corrupt database. (CVS 6877) (check-in: 709576c6 user: danielk1977 tags: trunk) | |
| 05:06 | Remove unreachable condition from btree.c. (CVS 6876) (check-in: 47b40fef user: danielk1977 tags: trunk) | |
|
2009-07-10
| ||
| 16:51 | Fix a pager reference count leak in btree that occurs on an OOM following a database page size change. (CVS 6875) (check-in: c6dfc8bd user: drh tags: trunk) | |
| 09:24 | Fix a double-free that can occur when using the fts3 legacy syntax '-' operator. Add tests for the same operator. Ticket #3960. (CVS 6874) (check-in: c19d419e user: danielk1977 tags: trunk) | |
| 02:52 | Simplifications to the PRAGMA integrity_check logic in btree.c. (CVS 6873) (check-in: 440c573c user: drh tags: trunk) | |
|
2009-07-09
| ||
| 13:25 | Fix two faulty assert() statements btree. (It was possible for the assertions to be false given a corrupt database file as input.) (CVS 6872) (check-in: 9403f04d user: drh tags: trunk) | |
| 11:36 | Remove an unused branch from btree.c. (CVS 6871) (check-in: 45efb78a user: danielk1977 tags: trunk) | |
| 08:27 | In btree.c, have setSharedCacheLock() assume that the b-tree is sharable instead of checking to see if it is. The caller has already performed this check. (CVS 6870) (check-in: 3190eefd user: danielk1977 tags: trunk) | |
| 05:07 | Restore (btree_set_cache_size) to test3.c. This is used on unix for the tcl "crash tests". Make some functions no longer used outside of btree.c static. (CVS 6869) (check-in: 1d96ce8c user: danielk1977 tags: trunk) | |
| 03:20 | removed the test function sqlite3BtreeFlags() (test code only); (CVS 6868) (check-in: 579ba6c8 user: shane tags: trunk) | |
| 02:48 | removed unused functions from test3.c (test code only); (CVS 6867) (check-in: 0eb69e8d user: shane tags: trunk) | |
| 02:24 | Simplifications and additional testcase() macros for btree.c. (CVS 6866) (check-in: 1b8c3a82 user: drh tags: trunk) | |
|
2009-07-08
| ||
| 18:50 | Fix a typo in (6861): "ifndef SQLITE_TEST" -> "ifdef SQLITE_TEST". (CVS 6865) (check-in: 890dbab2 user: danielk1977 tags: trunk) | |
| 18:45 | Revert part of (6860) that was accidentally checked in. (CVS 6864) (check-in: 11e295cc user: danielk1977 tags: trunk) | |
| 18:12 | Add another boundary-condition testcase() macro to btree.c. Correct the comment on modifyPagePointer(). (CVS 6863) (check-in: fed66e86 user: drh tags: trunk) | |
| 16:54 | Improvements to comments in btree.c:freeSpace(). Added a testcase() macro to verify a boundary case is tested in sqlite3BtreeInitPage(). (CVS 6862) (check-in: 5ac991b3 user: drh tags: trunk) | |
| 15:14 | Comment out sqlite3BtreeFlags() when not in debug mode. Tweaks to database corruption detection logic in freeSpace(). (CVS 6861) (check-in: 196cbf82 user: drh tags: trunk) | |
| 13:55 | Fix error handling in sqlite3BtreePutData(). (CVS 6860) (check-in: 86eba4f1 user: danielk1977 tags: trunk) | |
| 08:05 | Remove an ALWAYS clause that could be false if an IO error within a statement executed by a virtual table caused an emergency rollback. (CVS 6859) (check-in: 0cc705c2 user: danielk1977 tags: trunk) | |
| 01:49 | Simplifications to sqlite3BtreeInsert() and allocateSpace(). Added many testcase() macros to verify boundary conditions in btree.c. (CVS 6858) (check-in: aab82a22 user: drh tags: trunk) | |
|
2009-07-07
| ||
| 17:38 | Improvements to corrupt database detection in defragmentPage(). (CVS 6857) (check-in: 87bbc8d6 user: drh tags: trunk) | |
| 15:47 | Do not load the root-page of a b-tree table/index when opening a cursor. Instead, allow it to be loaded when the cursor is first used (in function moveToRoot()). Also move the root-page flags sanity checks that were a part of the OP_OpenRead/OpenWrite opcodes into the moveToRoot() function. (CVS 6856) (check-in: 06dcfe72 user: danielk1977 tags: trunk) | |
| 13:56 | If an error occurs in PagerBegin(), call pager_end_transaction() to reset the internal state of the pager object. (CVS 6855) (check-in: ea7ed166 user: danielk1977 tags: trunk) | |
| 11:39 | Fix a reference count problem following database corruption detection in the ptrmapPut() routine of btree.c. (CVS 6854) (check-in: ec1e27e3 user: drh tags: trunk) | |
| 02:44 | Simplifications to btree.c and vdbeaux.c in support of structural coverage testing. (CVS 6853) (check-in: fc2a40a9 user: drh tags: trunk) | |
|
2009-07-06
| ||
| 21:54 | Fix the pager so that it remembers the name of the journal for an in-memory database as an empty string. This is important when committing a multi-database transaction that includes one or more attached memory databases. (CVS 6852) (check-in: 31a5e819 user: drh tags: trunk) | |
| 19:03 | Fix the backup logic so that it detects a failure to sqlite3BtreeUpdateMeta() and reports that failure back up to the top level. (CVS 6851) (check-in: 471b92f2 user: drh tags: trunk) | |
| 18:56 | Make the sqlite3BtreeMoveto function static, since it is only used from within btree.c. Remove unused function lockBtreeWithRetry from btree.c. (CVS 6850) (check-in: 30d5ec62 user: danielk1977 tags: trunk) | |
| 00:44 | Simplifications and comment cleanup in vdbeaux.c. (CVS 6849) (check-in: 1636e783 user: drh tags: trunk) | |
|
2009-07-04
| ||
| 17:16 | Remove unreachable code from function btreeCursor() in btree.c. (CVS 6848) (check-in: c76a366e user: danielk1977 tags: trunk) | |
| 15:41 | Remove a redundant branch from btree.c. (CVS 6847) (check-in: 133357d2 user: danielk1977 tags: trunk) | |
|
2009-07-03
| ||
| 22:54 | Remove two unused fields from the Parse structure. The TCL interface now removes leading whitespace from SQL statements before parsing them. (CVS 6846) (check-in: 983cb692 user: drh tags: trunk) | |
| 19:19 | Remove unused code (that was commented out) from prepare.c. Remove an incorrect NEVER() macro. Fix the comment on schemaIsValid() to correctly describe its return behavior. (CVS 6845) (check-in: 90deb658 user: drh tags: trunk) | |
| 19:18 | Cause sqlite3_exec() to return SQLITE_NOMEM if it cannot allocate memory to hold the error message for some other error. (CVS 6844) (check-in: e9849292 user: drh tags: trunk) | |
| 17:23 | Remove an unused variable from the test code in test3.c. (CVS 6843) (check-in: 9e35bec9 user: drh tags: trunk) | |
| 17:09 | Add a version of the Lemon parser template "lempar.c" to the src/ subfolder. This new parser template contains a couple of NEVER() macros that disable tests that are needed for general grammars but not for the specific grammar used by SQLite. SQLite builds with the modified lempar.c. (CVS 6842) (check-in: d426cc64 user: drh tags: trunk) | |
| 16:25 | Simplify the way the read-uncommitted flag is handled to facilitate test coverage. (CVS 6841) (check-in: e2112d61 user: danielk1977 tags: trunk) | |
| 15:37 | Make explicit the restrictions on UPDATE, DELETE, and INSERT statement syntax within triggers. Ticket #3947. (CVS 6840) (check-in: c8bf40df user: drh tags: trunk) | |
| 12:57 | Test the result of pthread_create() and do not call pthread_join() if the thread creation failed. Ticket #3933. (CVS 6839) (check-in: 304c5110 user: drh tags: trunk) | |
|
2009-07-02
| ||
| 18:40 |
Fix to sqlite3AuthRead to accommodate "new" or "old" references that are used in a context where a column reference may also be used (i.e. "SELECT new. | |
| 17:21 | When a b-tree transaction is committed when there are open cursors, downgrade shared-cache write-locks to read-locks instead of relinquishing all locks. Fix for #3942. (CVS 6837) (check-in: 611e704f user: danielk1977 tags: trunk) | |
| 07:47 | Cause opening a transaction on a sharable b-tree module automatically obtain a read-lock on page 1. This means there is no way for sqlite3BtreeGetMeta() to fail. (CVS 6836) (check-in: e3c055f1 user: danielk1977 tags: trunk) | |
| 05:23 | Fix for compilation with OMIT_AUTOVACUUM defined. Ticket #3940. (CVS 6835) (check-in: 8f0591ae user: danielk1977 tags: trunk) | |
|
2009-07-01
| ||
| 18:09 | Change to test code and scripts allowing memsys3 to be tested using a build that has both memsys3 and memsys5 enabled. (CVS 6834) (check-in: a00aeec2 user: danielk1977 tags: trunk) | |
| 18:04 | Fix a 1-byte buffer overwrite that can occur when a virtual table overloads an SQL function. (CVS 6833) (check-in: bfe336a8 user: danielk1977 tags: trunk) | |
| 16:12 | Report an error if a USING or ON clause is specified following a from-list element that is not to the right of a join operator. Fix for #3846. (CVS 6832) (check-in: 29b48972 user: danielk1977 tags: trunk) | |
| 14:56 | Do not ignore alias "a" in a query of the form "SELECT ... FROM (...) AS a" Fix for #3935. Also expand upon (6751) to fix some similar obscure memory leaks. (CVS 6831) (check-in: 42f9d1e5 user: danielk1977 tags: trunk) | |
|
2009-06-29
| ||
| 06:00 | Cause incremental-blob read/write operations lock shared-cache tables in the same way as normal SQL read/writes. Add complex assert statements to make sure tehe correct shared-cache locks are held when accessing the database. Eliminate some redundant checks from btree.c. (CVS 6830) (check-in: f17ef378 user: danielk1977 tags: trunk) | |
|
2009-06-27
| ||
| 14:10 | Version 3.6.16 (CVS 6829) (check-in: ff691a6b user: drh tags: trunk, release) | |
| 11:17 | Fix an instance where sqlite3JumpHere() might be called with a negative address following an OOM fault. (CVS 6828) (check-in: 49f22e55 user: drh tags: trunk) | |
| 00:48 | Fix a bug in sqlite3_realloc() - if called with a size of more than 2147483392 it returns 0 but it also releases the prior allocation. (CVS 6827) (check-in: 653df0af user: drh tags: trunk) | |
|
2009-06-26
| ||
| 18:35 | Remove incorrect NEVER() macros from malloc.c. The allocations can be exceeded using sqlite3_malloc() and sqlite3_realloc(). (CVS 6826) (check-in: 0d345e59 user: drh tags: trunk) | |
| 16:32 | Removed a few compiler warnings under MSVC. (CVS 6825) (check-in: f01a9fc3 user: shane tags: trunk) | |
| 15:14 | Remove the priorNewRowid field from the sqlite3 structure. Use the last_insert_rowid as the initial value when searching for a new random rowid in the OP_NewRowid opcode. (CVS 6824) (check-in: 96729b2d user: drh tags: trunk) | |
| 14:17 | Skip a couple tests requiring working 64bit int support on platforms without that support. Test code change only. (CVS 6823) (check-in: d8c8145c user: shane tags: trunk) | |
| 14:04 | When sqlite3VdbeAddOp() fails due to an OOM fault, return an address of 1 rather than zero so that there appears to always be a prior instruction. This prevents a negative jump-to address computation in where.c following an OOM fault. (CVS 6822) (check-in: fe02339c user: drh tags: trunk) | |
| 12:30 | Fix a test script problem in exclusive.test. (CVS 6821) (check-in: a52ef5a4 user: danielk1977 tags: trunk) | |
| 12:15 | Fix the new pager.c changes so that they compile with IOTRACE defined. Fix an out-of-order variable definition in vdbeaux.c. (CVS 6820) (check-in: ac145028 user: drh tags: trunk) | |
| 10:39 | Another change to test_journal.c to account for (6817). Again, only test code has changed. (CVS 6819) (check-in: 58884b6c user: danielk1977 tags: trunk) | |
| 09:01 | Update test_journal.c to account for (6817). Changes to test code only. (CVS 6818) (check-in: 542ee8cc user: danielk1977 tags: trunk) | |
| 07:12 | Avoid writing the 8-byte journal-header magic until the journal-header is synced. In persistent journal-mode, this prevents any old content that follows an unsynced journal-header from being interpreted as part of the rollback journal. (CVS 6817) (check-in: a5ecffcf user: danielk1977 tags: trunk) | |
|
2009-06-25
| ||
| 16:11 | Fix a problem with a return code being ignored in insertCell(). (CVS 6816) (check-in: bb5f1c01 user: danielk1977 tags: trunk) | |
| 11:50 | Get the notify2.test running again by removing an ALWAYS() that was mistakenly inserted and by taking care not to report corruption following an SQLITE_LOCKED failure during schema parsing. (CVS 6815) (check-in: 6e9140a7 user: drh tags: trunk) | |
| 09:40 | Replace an "if" condition that is always true in btree.c with an assert(). (CVS 6814) (check-in: daadbd9a user: danielk1977 tags: trunk) | |
| 01:47 | Improved handling of oversize string and blob errors. Other simplifications in support of full coverage testing. (CVS 6813) (check-in: 8b340766 user: drh tags: trunk) | |
|
2009-06-24
| ||
| 13:16 | Remove an ALWAYS() macro from vdbe.c that sometimes fails on thread tests. (CVS 6812) (check-in: 52eac078 user: drh tags: trunk) | |
| 13:13 | Disable the new malloc-37 test when locking_mode==exclusive. The test basically works, but sometimes reports "database is locked" instead of "out of memory" and that messes up the error reporting. (CVS 6811) (check-in: 3b8f9dec user: drh tags: trunk) | |
| 11:08 | Remove a declaration of the sqlite3Assert() function that was deleted by the previous check-in. (CVS 6810) (check-in: ee20b6a2 user: drh tags: trunk) | |
| 10:26 | Remove the sqlite3Assert() function. The ALWAYS() and NEVER() macros call assert() directly when compiled with SQLITE_DEBUG. (CVS 6809) (check-in: d8fc373f user: drh tags: trunk) | |
| 05:40 | Simplify things by rolling the functionality of balance_shallower() into balance_nonroot(). (CVS 6808) (check-in: 11750c6a user: danielk1977 tags: trunk) | |
|
2009-06-23
| ||
| 20:28 | Enhance autoincrement so that it works with triggers that also do autoincrement inserts, even multiple inserts into the same table. Ticket #3928 (CVS 6807) (check-in: 1330993d user: drh tags: trunk) | |
| 16:40 | Remove a condition from balance_nonroot() that is always true. (CVS 6806) (check-in: c5dc80e6 user: danielk1977 tags: trunk) | |
| 15:43 | Call balance_shallower() from balance_nonroot() instead of from balance(). This simplifies coverage testing a bit. (CVS 6805) (check-in: da9893e2 user: danielk1977 tags: trunk) | |
| 14:39 | Update the configure script for version 3.6.16 (CVS 6804) (check-in: b614e554 user: drh tags: trunk) | |
| 14:18 | Update the version number in preparation for the next release. (CVS 6803) (check-in: 207c4a02 user: drh tags: trunk) | |
| 14:15 | Simplifications to vdbe.c to promote better test coverage. (CVS 6802) (check-in: 3ffc93d7 user: drh tags: trunk) | |
| 11:53 | Add a test to verify that #3929 is fixed. (CVS 6801) (check-in: 16c1ae9b user: danielk1977 tags: trunk) | |
| 11:22 | Fix a failing assert in btree.c. The same bug was causing a spurious SQLITE_CORRUPT return when compiled without SQLITE_DEBUG. (CVS 6800) (check-in: 47ec8749 user: danielk1977 tags: trunk) | |
|
2009-06-22
| ||
| 19:05 | Simplifications to vdbe.c and it service routines in support of coverage testing. (CVS 6799) (check-in: 308f2e61 user: drh tags: trunk) | |
| 18:03 | Remove a condition that is always true from btree.c. (CVS 6798) (check-in: 3ec8b37a user: danielk1977 tags: trunk) | |
| 12:05 | Make sure sqlite3BtreeCreateTable() returns an error code if the largest rootpage field is invalid in auto_vacuum mode. Update file-format comments to describe all of the "meta" values. (CVS 6797) (check-in: bda38f5e user: drh tags: trunk) | |
| 11:10 | More simplifications to vdbe.c. Remove a NEVER() from vdbe.c that is possible after all. (CVS 6796) (check-in: c8f009bd user: drh tags: trunk) | |
| 05:43 | If an OOM error occurs just after obtaining a shared lock on the database file, release the lock before returning. (CVS 6795) (check-in: 971a9650 user: danielk1977 tags: trunk) | |
| 00:55 | Simplifications to vdbe.c in support of coverage testing. (CVS 6794) (check-in: 16680f05 user: drh tags: trunk) | |
|
2009-06-20
| ||
| 18:52 | When recovering from the error-state in pagerSharedLock(), do not search for a hot-journal in the file-system if the pager is configured to use an anonymous temp file as the journal. (CVS 6793) (check-in: 4b46805c user: danielk1977 tags: trunk) | |
| 11:54 | If after obtaining an exclusive lock to rollback a hot-journal file it is found that the journal file has been deleted, do not return an SQLITE_BUSY error. Just downgrade the lock and continue with the current operation. This eliminates a spurious SQLITE_BUSY error caused by a race condition. (CVS 6792) (check-in: 9a066600 user: danielk1977 tags: trunk) | |
|
2009-06-19
| ||
| 22:50 | Improved documentation on the VFS methods. Ticket #3925. (CVS 6791) (check-in: f66fc771 user: drh tags: trunk) | |
| 22:23 | Do now allow a COMMIT to occur while even a single write operating is in progress. (CVS 6790) (check-in: 605520ec user: drh tags: trunk) | |
| 18:32 | Fix the comment on the sqlite3PExpr() subroutine. Ticket #3924. (CVS 6789) (check-in: 4bb96db6 user: drh tags: trunk) | |
| 17:50 | If a call is made to sqlite3PagerAcquire when there are no outstanding references to any pages and the pager is in the error-state, try to exit the error-state at this point. Previously this was only attempted if the pager was configured to use exclusive mode. (CVS 6788) (check-in: 62db08bc user: danielk1977 tags: trunk) | |
| 16:44 | Make sure group_concat() ignores initial NULL values. (CVS 6787) (check-in: 90c3b23c user: drh tags: trunk) | |
| 14:06 | Reorganize and cleanup the prepared statement object. Remove code that has been commented out for ages and is no longer relevant to anything. (CVS 6786) (check-in: 37ae5f5e user: drh tags: trunk) | |
| 02:25 | Fix a possible segfault when backing up an encrypted database. (CVS 6785) (check-in: 7d8f178c user: drh tags: trunk) | |
| 00:33 | Make sure the sqlite3VdbeGetOp() subroutine works even when SQLite is compiled with SQLITE_OMIT_TRACE. (CVS 6784) (check-in: 1c8ec2a6 user: drh tags: trunk) | |
|
2009-06-18
| ||
| 20:52 | Changes to the backup functionality so that it works better with codecs. (CVS 6783) (check-in: 87d6796d user: drh tags: trunk) | |
| 17:22 | Move codec management from database connections into the pager so that it will work together with shared cache. (CVS 6782) (check-in: ed08b53c user: drh tags: trunk) | |
| 11:34 | Update the filefmt.test test script to conform to the previous check-in. (CVS 6781) (check-in: 40a40e81 user: drh tags: trunk) | |
| 11:29 | Reduce the minimum usable page space to 480 byte from 500 bytes. (CVS 6780) (check-in: db4fb835 user: drh tags: trunk) | |
| 00:41 | Remove the P3 operand from OP_IsNull since it was not being used. (CVS 6779) (check-in: 767ef1e4 user: drh tags: trunk) | |
|
2009-06-17
| ||
| 22:50 | Avoid an assertion fault if an out-of-memory error occurs while trying to run the string-concatentation operator on a zero-blob. (This is an absurd thing to do, but even so, we still should not fault.) (CVS 6778) (check-in: 0def0b76 user: drh tags: trunk) | |
| 21:42 | Fix a problem in vdbe.c that could cause a double-free of memory if the SQLITE_LIMIT_LENGTH is changed after a statement is prepared but before it is run. Also remove debugging statements from tkt3841.test. (CVS 6777) (check-in: 3d7327fd user: drh tags: trunk) | |
| 16:20 | Use caution to avoid integer overflow when doing real to integer affinity operations. Ticket #3922. (CVS 6776) (check-in: 39255946 user: drh tags: trunk) | |
| 13:57 | Fix a potential segfault following database corruption. Problem made visible by check-in (6772). (CVS 6775) (check-in: 69eb0ff8 user: drh tags: trunk) | |
| 13:09 | Suppress some -Wextra compiler warnings from GCC. (CVS 6774) (check-in: 59ec937c user: drh tags: trunk) | |
| 11:49 | Fix a bug affecting secure-delete mode introduced by (6768). (CVS 6773) (check-in: a433ca82 user: danielk1977 tags: trunk) | |
| 11:13 | Fix for #3918. Also, fix the TRACE macros in balance_nonroot(). (CVS 6772) (check-in: 368e44ec user: danielk1977 tags: trunk) | |
| 01:17 | A minor simplification to the tokenizer. (CVS 6771) (check-in: 18f2076a user: drh tags: trunk) | |
| 00:35 | Changes to prepare.c and callback.c to promote better test coverage. (CVS 6770) (check-in: 08cca219 user: drh tags: trunk) | |
|
2009-06-16
| ||
| 17:49 | Changes sqlite3_prepare_v2() (and related routines) so that if it fails due to a missing table and the schema is out of date, it retries once before returning SQLITE_SCHEMA. Other changes to prepare.c to facilitate coverage testing. (CVS 6769) (check-in: 256ec3c6 user: drh tags: trunk) | |
| 16:50 | Changes to balance_nonroot() and BtreeDelete() to simplify delete operations and reduce stack/heap usage while balancing b-tree structures. (CVS 6768) (check-in: 092b276e user: danielk1977 tags: trunk) | |
| 14:15 | Fix a link error and warning that can occur in where.c when compiling under MSVC with SQLITE_OMIT_VIRTUALTABLE defined. Ticket #3914. (CVS 6767) (check-in: 793c93be user: shane tags: trunk) | |
| 04:35 | Leave out a redundant line in build.c. Ticket #3916. (CVS 6766) (check-in: d9f997ae user: danielk1977 tags: trunk) | |
| 04:31 | Remove a duplicate assert from btree.c. Ticket #3915. (CVS 6765) (check-in: d89542cd user: danielk1977 tags: trunk) | |
|
2009-06-15
| ||
| 23:15 | Additional updates to the symbol resolver and expression tree walker to facilitate test coverage. (CVS 6764) (check-in: a49c2d4b user: drh tags: trunk) | |
| 20:45 | Changes to allow a clean compile with no floating-point instructions generated when SQLITE_OMIT_FLOATING_POINT is used. (CVS 6763) (check-in: 1591b832 user: drh tags: trunk) | |
| 18:32 | Updates to resolve.c to facilitate full coverage testing. (CVS 6762) (check-in: c50f0f28 user: drh tags: trunk) | |
| 16:27 | Fix compiler warnings with MSVC build. (CVS 6761) (check-in: e1e64307 user: shane tags: trunk) | |
| 03:28 | Version 3.6.15 (CVS 6760) (check-in: aff34826 user: drh tags: trunk, release, version-3.6.15) | |
| 00:04 | Update the configuration script for version 3.6.15. (CVS 6759) (check-in: 14df65dc user: drh tags: trunk) | |
|
2009-06-14
| ||
| 12:47 | Detect database corruption and abort balance_nonroot() before the corruption causes problems. (CVS 6758) (check-in: 3448b39d user: drh tags: trunk) | |
|
2009-06-12
| ||
| 17:46 | Change lemon so that it does not generate yytestcase() macros on destructor cases since destructors are commonly unreachable in a reasonable grammar. For the reduce-rule switch, gather all no-ops into the "default:" case. (CVS 6757) (check-in: caebfe82 user: drh tags: trunk) | |
| 15:47 | The previous check-in ((6755)) put in more yytestcase() macros than are called for. This check-in addresses that problem. (CVS 6756) (check-in: a5b182f9 user: drh tags: trunk) | |
| 13:53 | The lemon parser generator now inserts yytestcase() macros on reduce action and on each destructor, to verify that all have been executed. yytestcase() is a no-op by default but can be set to something more useful inside of %include. (CVS 6755) (check-in: fe9c9177 user: drh tags: trunk) | |
| 12:50 | In lemon: omit unused entries from the end of the yyFallback array. (CVS 6754) (check-in: 9cfbe2ba user: drh tags: trunk) | |
| 12:04 | Enable the SQLITE_LIMIT_FUNCTION_ARG limiter. (CVS 6753) (check-in: a17e3317 user: drh tags: trunk) | |
| 11:42 | Remove a C++-ism from the code. Ticket #3912. (CVS 6752) (check-in: bc729bc3 user: drh tags: trunk) | |
| 03:47 | An additional memory leak in the parser fixed. Ticket #3911. (CVS 6751) (check-in: 6197b492 user: drh tags: trunk) | |
| 03:27 | Make sure nested queries with USING clauses do not leak memory. Preliminary fix for ticket #3911. (CVS 6750) (check-in: bd341a10 user: drh tags: trunk) | |
| 02:27 | Modifications to the parser to eliminate unreachable code. (CVS 6749) (check-in: 457e0b24 user: drh tags: trunk) | |
|
2009-06-11
| ||
| 17:32 | Honor the SQLITE_OPEN_READONLY flag even on a :memory: or temporary database, even though such a database is pointless. Ticket #3908. (CVS 6748) (check-in: 73ba841a user: drh tags: trunk) | |
| 17:04 | Add an ALWAYS() to a conditional in where.c. (CVS 6747) (check-in: 1fdb3e1e user: drh tags: trunk) | |
| 00:47 | Changes to reenable codec operation and to handle memory allocation failures within a codec. (CVS 6746) (check-in: 43a6ca98 user: drh tags: trunk) | |
|
2009-06-10
| ||
| 19:33 | Do not let the reverse_unordered_selects pragma force the use of an index that would not otherwise be used. Ticket #3904. Also: remove an test which is always true. (CVS 6745) (check-in: 78a391dc user: drh tags: trunk) | |
| 11:07 | Remove a NEVER() that can sometimes occur on an OOM error. (CVS 6744) (check-in: c27f23bb user: drh tags: trunk) | |
| 09:11 | Minor tweak to balance_quick() to save a few instructions. (CVS 6743) (check-in: 9ace9959 user: danielk1977 tags: trunk) | |
|
2009-06-09
| ||
| 19:53 | Remove the actual checks that verify that memory pools (such as set up using SQLITE_CONFIG_HEAP) are 8-byte aligned. But document that 8-byte alignment is required. (CVS 6742) (check-in: e9b55ccc user: drh tags: trunk) | |
| 18:58 | Fix compiler warnings with MSVC build. (CVS 6741) (check-in: 0bd84e73 user: shane tags: trunk) | |
| 18:14 | Avoid a false failure report in the test scripts caused by the vdbe stack "compression" script. (CVS 6740) (check-in: 9d332989 user: shane tags: trunk) | |
| 18:02 | Require that the buffer specified by SQLITE_CONFIG_HEAP be 8-byte aligned. (CVS 6739) (check-in: 18b78068 user: drh tags: trunk) | |
| 18:01 | Changes to tokenize.c to facilitate full coverage testing. (CVS 6738) (check-in: 5e8c48cf user: drh tags: trunk) | |
| 15:25 | Change savepoint4.test to limit database growth. Otherwise the test can take a very long time to run, depending on the prng. (CVS 6737) (check-in: ed1d4f47 user: danielk1977 tags: trunk) | |
| 14:27 | Modify mkkeywordhash.c so that it puts the original text of each token in a comment beside the "testcase()" macros. (CVS 6736) (check-in: a7d0817c user: drh tags: trunk) | |
| 13:42 | Enable cell overflow checking in sqlite3BtreeInitPage() using the compile-time option SQLITE_ENABLE_OVERSIZE_CELL_CHECK. Change the tests so that they recognize different error messages depending on the setting of this macro. (CVS 6735) (check-in: 56bff6eb user: drh tags: trunk) | |
| 11:34 | Avoid calling fillInCell() to create an internal node intkey cell from balance_nonroot(). A single putVarint() does the same thing more quickly. (CVS 6734) (check-in: 2e5d42ae user: danielk1977 tags: trunk) | |
| 10:37 | Only do the cell overread checks in sqlite3BtreeInitPage if SQLITE_OVERREAD_CHECK is defined at compile-time. (CVS 6733) (check-in: 49f544eb user: drh tags: trunk) | |
| 09:41 | Do not clear the MemPage.nFree variable when insertCell() adds an overflow cell to a page. Not doing this means balance_quick() can avoid a call to sqlite3BtreeInitPage(). (CVS 6732) (check-in: 8f1c1f61 user: danielk1977 tags: trunk) | |
|
2009-06-08
| ||
| 19:44 | Additional comments to clarify the operation of the LIKE optimizer in where.c. (CVS 6731) (check-in: cc9c1217 user: drh tags: trunk) | |
| 17:11 | Clarification of the operation of the OR-term optimizer in where.c. (CVS 6730) (check-in: 6b42dc3d user: drh tags: trunk) | |
| 14:49 | Change the btree balance code so that it does not call balance_nonroot() recursively. (CVS 6729) (check-in: 7863db90 user: danielk1977 tags: trunk) | |
| 12:52 | Increase the version number to 3.6.15 in preparation for the next release. (CVS 6728) (check-in: 456ea541 user: drh tags: trunk) | |
|
2009-06-07
| ||
| 23:45 | Disable the LIKE optimization if the affinity of the LHS column is not TEXT. Ticket #3901. (CVS 6727) (check-in: a255c645 user: drh tags: trunk) | |
|
2009-06-06
| ||
| 19:21 | Update the error message on one of the corruption tests to account for the fact that we are finding the corruption sooner. (CVS 6726) (check-in: ba9848e7 user: drh tags: trunk) | |
| 16:08 | Fix an error in the malloc.test test script causing a small buffer overread. (CVS 6725) (check-in: 4d129bee user: danielk1977 tags: trunk) | |
| 15:17 | Make sure the result of sqlite3VdbeGetOp() is not used as an array following an OOM error, since after (6691) it might be a single-entry dummy opcode. (CVS 6724) (check-in: 127b1398 user: drh tags: trunk) | |
| 14:13 | When building in debug mode with gcc, force the nullMem variable in function columnMem() to be aligned to an 8-byte boundary. Otherwise an assert() statement may fail. (CVS 6723) (check-in: 3fd6c72d user: danielk1977 tags: trunk) | |
|
2009-06-05
| ||
| 18:44 | Detect when database corruption causes an attemp to set the pointer map value for a pointer map page and report the corruption. (CVS 6722) (check-in: 3ae4880b user: drh tags: trunk) | |
| 17:09 | Remove leftover "breakpoint" commands from test scripts. Also remove blank lines at the end of scripts. (CVS 6721) (check-in: 1fef16ec user: drh tags: trunk) | |
| 16:46 | Do not cast an unsigned int into an int for comparison. (CVS 6720) (check-in: 302ab855 user: drh tags: trunk) | |
| 14:17 | Take care that a corrupt variable-length integer does not cause 32-bit integer overflow when parsing a record format, nor cause excessively large memory allocations. (CVS 6719) (check-in: 38b20327 user: drh tags: trunk) | |
|
2009-06-04
| ||
| 19:06 | Earlier detection of freelist corruption in the page allocation routines. (CVS 6718) (check-in: e557c8e5 user: drh tags: trunk) | |
| 17:02 | Fix an assertion fault in balance_quick() that occurs when an interior btree node has zero cells due to database corruption. Also update the corrupt7.test result vectors for a couple of cases where the error report on database corruption changed due to earlier detection. (CVS 6717) (check-in: 1335e444 user: drh tags: trunk) | |
| 16:14 | If the root page of a btree is empty and is also not a leaf page and the page is not page 1, then report database corruption. (CVS 6716) (check-in: 52b02ca5 user: drh tags: trunk) | |
| 14:46 | Reduce the stack usage of balance_quick(). (CVS 6715) (check-in: e7ad68e9 user: danielk1977 tags: trunk) | |
| 02:47 | Minor updates to comments in test scripts. (CVS 6714) (check-in: 453ff88f user: shane tags: trunk) | |
| 02:46 | Minor tweak to test script to ensure every byte in file touched at least once. (CVS 6713) (check-in: bc6c2d89 user: shane tags: trunk) | |
| 00:11 | Fix two instances where a malformed header in a database file (in other words a corrupt database file) can lead to an assertion fault. (CVS 6712) (check-in: 4e92c7f2 user: drh tags: trunk) | |
|
2009-06-03
| ||
| 21:04 | Change the pcache1.c implementation so that the "header" occurs at the end of page buffer, not at the beginning. This insures that the 20 bytes immediately following the page buffer are mapped if a read of the page buffer overruns due to a malformed cell. (CVS 6711) (check-in: c54de1f5 user: drh tags: trunk) | |
| 17:26 | Add corruptD.test, a container for testing the "cell overflow" problem. Also shuffle a small amount of code in BtreeInitPage() to check that the page header pointer to the start of the cell offset array is set to a sane value. (CVS 6710) (check-in: 7fa5d3cb user: danielk1977 tags: trunk) | |
| 11:25 | Define a set of constants to use as the "index" argument to sqlite3BtreeGetMeta and UpdateMeta. This makes some parts of the code easier to follow. (CVS 6709) (check-in: 6dbf4eca user: danielk1977 tags: trunk) | |
| 01:24 | Additional changes to reduce stack usage. The SQLITE_SMALL_STACK compile-time option is now available. (CVS 6708) (check-in: baea79fd user: drh tags: trunk) | |
|
2009-06-02
| ||
| 21:31 | Further reductions in the amount of stack space required. (CVS 6707) (check-in: 04bad9eb user: drh tags: trunk) | |
| 16:06 | Tweaks to vdbe.c to further reduce stack space requirements. (CVS 6706) (check-in: 6f2aab3f user: drh tags: trunk) | |
| 15:47 | Add a test case for ticket #3893 and ticket #3894. (CVS 6705) (check-in: 2472f6db user: drh tags: trunk) | |
| 15:21 | Add the vdbe-compress.tcl script which automatically refactors the sqlite3VdbeExec() routine to use less stack space. Use this script when constructing the amalgamation. (CVS 6704) (check-in: 7f433918 user: drh tags: trunk) | |
|
2009-06-01
| ||
| 19:53 | Avoid allocating large objects on the stack in the incremental BLOB I/O interface. (CVS 6703) (check-in: ea7dfde7 user: drh tags: trunk) | |
| 18:18 | Malloc for space to hold the Parse object in sqlite3_prepare() and friends. Or, if compiled with SQLITE_USE_ALLOCA, obtain space for the object from alloca(). (CVS 6702) (check-in: c7c0c58e user: drh tags: trunk) | |
| 17:10 | Added assert to verify winMutex system is inited before "static" mutexes are used. (CVS 6701) (check-in: b126013b user: shane tags: trunk) | |
| 17:06 | Ensure that "static" mutexes are deleted on shutdown so that they are not leaked across restarts. (CVS 6700) (check-in: 4e75897e user: shane tags: trunk) | |
| 16:53 | Fix compiler warnings with MSVC build. (CVS 6699) (check-in: 07915885 user: shane tags: trunk) | |
| 16:42 | Fix a couple of Windows only issues with test scripts. (CVS 6698) (check-in: 0f6bc5e1 user: shane tags: trunk) | |
|
2009-05-31
| ||
| 21:21 | Code simplifications to facilitate coverage testing following the recent changes to the Expr object. (CVS 6697) (check-in: ee9a144d user: drh tags: trunk) | |
| 17:16 | Fix the ".genfkey" command of the CLI so that it does not leak memory if sqlite3_realloc() fails. Ticket #3891. (CVS 6696) (check-in: a028d69c user: drh tags: trunk) | |
|
2009-05-30
| ||
| 23:35 | Fix typo reported by ticket #3888. Other minor edits to facilitate coverage testing. (CVS 6695) (check-in: 164adf26 user: drh tags: trunk) | |
| 20:49 | Simplifications to the expr.c source module resulting from structural testing. (CVS 6694) (check-in: f760bb7a user: drh tags: trunk) | |
| 14:16 | Fix for ticket #3887. Other unrelated bits of code cleanup in the expr.c source module. (CVS 6693) (check-in: 00e7250f user: drh tags: trunk) | |
| 10:46 | In vacuum.c use "PagerIsMemdb(X)" instead of "PagerFile(X)->pMethods==0". (CVS 6692) (check-in: 874295ce user: danielk1977 tags: trunk) | |
|
2009-05-29
| ||
| 19:00 | Enhance the sqlite3VdbeGetOp() routine so that it always returns a pointer to VdbeOp, even following an OOM fault. This simplifies error handling in callers. Cleanup to the column cache logic and the expr.c source module. (CVS 6691) (check-in: ede06dac user: drh tags: trunk) | |
| 14:39 | Refinements in expression handling logic that have resulted from recent structural testing. (CVS 6690) (check-in: bd89f2c2 user: drh tags: trunk) | |
| 11:57 | Still more refinements to the hasHotJournal() fix of ticket #3883. (CVS 6689) (check-in: 726b425e user: drh tags: trunk) | |
| 10:55 | The hasHotJournal() fix of check-in (6687) causes some minor problems in various alternative operating modes, such as locking_mode=EXCLUSIVE. This additional patch attempts to fix those concerns. Ticket #3883. (CVS 6688) (check-in: a2ba61d9 user: drh tags: trunk) | |
| 00:30 | Modify the hasHotJournal() routine to return a false-positive if it is unable to open the journal file to check its header due to a race condition. Processing downstream of hasHotJournal() already knows how to deal with false-positives. Ticket #3883. (CVS 6687) (check-in: d6b5d8e1 user: drh tags: trunk) | |
|
2009-05-28
| ||
| 21:04 | Remove references to deleted function sqlite3ExprRegister(). Changes to the expr.c source module to promote better testing. (CVS 6686) (check-in: 6ae4ad6e user: drh tags: trunk) | |
| 14:34 | Fix a NULL pointer dereference following an OOM error in the column name resolver. (CVS 6685) (check-in: 3b461425 user: drh tags: trunk) | |
| 12:49 | Make sure the column name resolver does not try to duplicate Expr.u.zToken when the Expr.u.iValue union member is the element being used. (CVS 6684) (check-in: 376ecf0d user: drh tags: trunk) | |
| 11:05 | Workaround the need for internal API sqlite3BtreeCursorDb(). It was only being used for an assert() statement, which can be done differently. (CVS 6683) (check-in: 5904af6e user: danielk1977 tags: trunk) | |
| 01:00 | Additional refinements to Expr handling. Restore compression of trigger expressions. Change Expr.zToken to Expr.u.zToken and added Expr.u.iValue. Remove an unnecessary ExprDup from CHECK constraint processing. And so forth. (CVS 6682) (check-in: 4ac2bdfb user: drh tags: trunk) | |
|
2009-05-27
| ||
| 10:31 | Simplifications to the Expr object: Remove Expr.span completely and convert Expr.token into a char* Expr.zToken. Also simplify the Token object by removing the Token.dyn and Token.quoted fields. (CVS 6681) (check-in: 7cb1c3ba user: drh tags: trunk) | |
|
2009-05-25
| ||
| 14:17 | Fix the rtree test module so that it works even if the ext/ subfolder is omitted from the tree. (CVS 6679) (check-in: 086206e1 user: drh tags: trunk) | |
| 11:46 | When pinning a temp register after it is reused by the column cache, make sure all instances of that register in the cache are pinned so that the register is never reused for a different purpose. Ticket #3879. (CVS 6676) (check-in: 5f358e63 user: drh tags: trunk) | |
| 11:46 | Add a test case for ticket #3879. (CVS 6675) (check-in: 5b9b66f4 user: danielk1977 tags: trunk) | |
|
2009-05-24
| ||
| 21:59 | Additional clarification of the 3rd parameter to sqlite3_create_function(). Documentation change only; no changes to code. Ticket #3875. (CVS 6674) (check-in: dd75e376 user: drh tags: trunk) | |
| 11:08 | In the previous, the upper bound on the number of function arguments is 127, not 255. Ticket #3876 (CVS 6673) (check-in: 4fe4c5fa user: drh tags: trunk) | |
| 11:07 | Correct and clarify the documentation on the third parameter to sqlite3_create_function(). (CVS 6672) (check-in: 6b7929ed user: drh tags: trunk) | |
|
2009-05-22
| ||
| 15:43 | Do not transform a WHERE clause of the form "a = ? OR a = ?" to "a IN (?, ?)" if "a" is a column of a virtual table. Ticket #3871. (CVS 6671) (check-in: fad88e71 user: danielk1977 tags: trunk) | |
| 11:12 | Add an assert() to pcache1.c to double-check that page cache buffer memory is never allocated if pcache1 is not enabled. Ticket #3872 (CVS 6670) (check-in: 93369d91 user: drh tags: trunk) | |
| 11:10 | Additional changes to pcache1 to prevent the page buffer memory from being configured if pcache1 is not enabled. Ticket #3872. (CVS 6669) (check-in: 5153ad19 user: drh tags: trunk) | |
| 10:53 | Make sure sqlite3_shutdown() completely disables the default pager cache mechanism in pcache1.c. Ticket #3872. Also fix some comments associated with configuring the page cache. (CVS 6668) (check-in: 6240992c user: drh tags: trunk) | |
| 01:02 | Add omitted word in the documentation for total_change_count(). ticket #3870. (CVS 6667) (check-in: 3f6fd16f user: drh tags: trunk) | |
| 01:00 | Fix an 8-byte alignment problem on HP/UX. Ticket #3869 (CVS 6666) (check-in: fc64f850 user: drh tags: trunk) | |
|
2009-05-21
| ||
| 20:41 | Simplifications to expr.c in support of full coverage testing. (CVS 6665) (check-in: 309adb5e user: drh tags: trunk) | |
| 15:15 | In the CLI in the ".dump" command, do not attempt to clear the sqlite_sequence table until the first row of content of that table is seen. Ticket #3867 (CVS 6664) (check-in: bedd5ad1 user: drh tags: trunk) | |
| 14:51 | In the CLI, always "dump" the sqlite_sequence table last. Ticket #3867 (CVS 6663) (check-in: a0028d48 user: drh tags: trunk) | |
| 04:42 | Add conditional 'extern "C"' block to sqlite3async.h. Ticket #3866. (CVS 6662) (check-in: e4d1b117 user: danielk1977 tags: trunk) | |
|
2009-05-20
| ||
| 20:10 | Fix a memory leak that occurs when an out-of-memory error occurs while preparing a statement that has multiple virtual table updates within triggers. Other virtual table changes to support full-coverage testing. (CVS 6661) (check-in: 02b77a58 user: drh tags: trunk) | |
| 16:22 | Remove unused, undocumented, and untested error reporting logic from the xFindFunction interface in virtual tables. (CVS 6660) (check-in: 55d6ced2 user: drh tags: trunk) | |
| 02:40 | Updates to the extension loading logic to support full coverage testing. (CVS 6659) (check-in: 929cfbc6 user: drh tags: trunk) | |
|
2009-05-19
| ||
| 19:04 | Changes to select.c to facilitate full coverage testing. (CVS 6658) (check-in: f678ed34 user: drh tags: trunk) | |
| 14:21 | Improvements to the SQLITE_INT_TO_PTR macro to reduce the number of warnings. For some platforms it might be necessary to compile with the -DHAVE_STDINT_H flag. Ticket #3860. (CVS 6657) (check-in: 1b0ee9d1 user: drh tags: trunk) | |
| 03:21 | Update the showdb.c utility so that it works with databases larger than 4GiB. (CVS 6656) (check-in: 9675b622 user: drh tags: trunk) | |
|
2009-05-18
| ||
| 16:04 | Fix memdb.test so that it works when OMIT_AUTOVACUUM is defined. (CVS 6649) (check-in: 8c3c0c24 user: danielk1977 tags: trunk) | |
| 15:46 | Changes to select.c in support of full coverage testing. (CVS 6647) (check-in: e225f365 user: drh tags: trunk) | |
| 13:34 | Fix a typo on a comment associated with check-in (6641) and ticket #3860. (CVS 6646) (check-in: 43051e03 user: drh tags: trunk) | |
|
2009-05-17
| ||
| 15:29 | Remove debugging code accidently left in check-in (6642). Ticket #3862. (CVS 6645) (check-in: 0d974613 user: drh tags: trunk) | |
| 15:26 | Take care to free the memory for the ORDER BY clause on SELECT statements containing errors and contained within triggers. Ticket #3863. (CVS 6644) (check-in: dd665eac user: drh tags: trunk) | |
| 12:07 | Fix a typo in the sqlite3_blob_open() documentation. (CVS 6643) (check-in: 92866482 user: drh tags: trunk) | |
| 02:06 | Tweaks and simplifications to select.c to facility full coverage testing. (CVS 6642) (check-in: e3ccbc69 user: drh tags: trunk) | |
|
2009-05-16
| ||
| 17:38 | Attempt to provide a version of the SQLITE_INT_TO_PTR macro that works on both llvm-gcc-4.2 and MSVC. Ticket #3860. (CVS 6641) (check-in: ddee7ff2 user: drh tags: trunk) | |
|
2009-05-15
| ||
| 14:41 | Re-enable file locking in async4.test. (CVS 6640) (check-in: f7098187 user: danielk1977 tags: trunk) | |
|
2009-05-14
| ||
| 22:37 | Documentation enhancements for sqlite3_update_hook(), sqlite3_commit_hook(), and the various incremental blob I/O routines. No changes to code. (CVS 6639) (check-in: e8ca932d user: drh tags: trunk) | |
| 19:26 | Change the sqlite3_backup_finish() interface so that calling it with a NULL pointer is a harmless no-op. One other change to backup.c to facilitate full coverage testing. (CVS 6638) (check-in: 06bc8917 user: drh tags: trunk) | |
| 03:21 | Documentation updates to better explain the SQLITE_OPEN_EXCLUSIVE flag. Ticket #3855. (CVS 6637) (check-in: 6bd44f36 user: shane tags: trunk) | |
|
2009-05-13
| ||
| 22:58 | Additional simplifications to build.c in support full coverage testing. (CVS 6636) (check-in: 913d1114 user: drh tags: trunk) | |
| 17:35 | Remove incorrect NEVER() macro in the previous check-in (6634). (CVS 6635) (check-in: c6849579 user: drh tags: trunk) | |
| 17:21 | Changes to build.c to help facilitate full coverage testing. (CVS 6634) (check-in: 264ace50 user: drh tags: trunk) | |
| 14:46 | Do not run async4.test as part of the "journaltest" permutation. It specifically tests that the asynchronous IO backend can be configured to violate the assumptions made by journaltest. (CVS 6633) (check-in: b19bf39b user: danielk1977 tags: trunk) | |
| 07:52 | When locking a database file, check if any backup objects need to be restarted even if the page cache is empty. Ticket #3858. (CVS 6632) (check-in: 90309008 user: danielk1977 tags: trunk) | |
|
2009-05-12
| ||
| 18:00 | Remove debugging comments left in the prior check-in by mistake. (CVS 6631) (check-in: 82070560 user: drh tags: trunk) | |
| 17:46 | Move the sqlite3MinimumFileFormat() routine from build.c over to alter.c since alter.c is the only code that uses it. Additional changes and cleanup to build.c to facility coverage testing. (CVS 6630) (check-in: 6fe3750a user: drh tags: trunk) | |
| 13:35 | Make sure the SQLITE_OPEN_EXCLUSIVE flag is ignored by sqlite3_open_v2(). That flag is to be used by the VFS interface only. Ticket #3855. (CVS 6629) (check-in: c37f0279 user: drh tags: trunk) | |
| 00:40 | Tweaks to build.c to facilitate coverage testing. (CVS 6628) (check-in: 6572b438 user: drh tags: trunk) | |
|
2009-05-11
| ||
| 23:38 | If a virtual table constructor returns SQLITE_NOMEM, set the mallocFailed flag in the corresponding database connection to insure that the error is propagated back to the top layer. (CVS 6627) (check-in: 5635666f user: drh tags: trunk) | |
| 20:53 | Rework the logic that generates a schema for tables created using "CREATE TABLE ... AS SELECT ...". Instead of trying to copy the raw datatype string from the right-hand side, just make the type one of TEXT, INT, REAL, NUM, or nothing. This is much simpler than trying to parse and quote datatype strings. Other minor implifications to build.c are bundled with this change. (CVS 6626) (check-in: 33cf8359 user: drh tags: trunk) | |
| 18:22 | Enhance the parser to allow nested parentheses in the module argument of a CREATE VIRTUAL TABLE statement. (CVS 6625) (check-in: 93772bd7 user: drh tags: trunk) | |
|
2009-05-09
| ||
| 23:29 | Convert an always-true condition in hash.c into an assert(). (CVS 6624) (check-in: 42372999 user: drh tags: trunk) | |
| 18:59 | Add the SQLITE_TESTCTRL_ASSERT and SQLITE_TESTCTRL_ALWAYS codes for the sqlite3_test_control() interface. (CVS 6623) (check-in: 38df91c2 user: drh tags: trunk) | |
| 15:17 | Change sqlite_blob_open() so that it zeros the output pBlob pointer when it fails. The other sqlite3_blob interfaces accept a NULL pointer as input. (CVS 6622) (check-in: 999d507b user: drh tags: trunk) | |
| 00:18 | Changes to the trigger.c module that facilitate full coverage testing. (CVS 6621) (check-in: 567ccc68 user: drh tags: trunk) | |
|
2009-05-08
| ||
| 11:34 | Fix a warning in the osx-specific part of os_unix.c. Ticket #3847. (CVS 6620) (check-in: 254ca327 user: danielk1977 tags: trunk) | |
| 06:52 | Correctly update the cache page-count when discarding pages in the xTruncate() method of the default pcache implementation. Otherwise the effective configured size of the cache shrinks each time xTruncate() is called. Fix for #3844. (CVS 6619) (check-in: 88211cee user: danielk1977 tags: trunk) | |