SQLite

Timeline
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

100 ancestors of d03d480544e1f782

2007-03-08
12:23
Change the documentation to be more specific about the restriction on the use of keywords. Ticket #2264. (CVS 3676) (check-in: d03d4805 user: drh tags: trunk)
2007-03-06
16:03
Catch an out-of-memory condition in vacuum code. (Bug in (3373)). (CVS 3675) (check-in: 302ec768 user: danielk1977 tags: trunk)
15:53
Fixes to the overflow-chain optization of (3672). (CVS 3674) (check-in: d6986d1e user: drh tags: trunk)
13:46
Use heap instead of stack for large buffers in the pager. Fix for #2262. (CVS 3673) (check-in: dfe1dffa user: danielk1977 tags: trunk)
11:42
Do not read the last page of a overflow chain when deleting that chain. Just add the page to the freelist. This reduces I/O. (CVS 3672) (check-in: 6db945f7 user: drh tags: trunk)
2007-03-04
13:15
Changes to the btree and pager that reduce the amount of I/O when dealing with the freelist. (1) Avoid journaling pages of a table that is being deleted. (2) Do not read the original content of pages being pulled off of the freelist. (CVS 3671) (check-in: 2ba5be31 user: drh tags: trunk)
2007-03-02
08:12
Handle the case where the estimated cost of a virtual table scan is larger than SQLITE_BIG_DBL. Ticket #2253. (CVS 3670) (check-in: 52885ed8 user: danielk1977 tags: trunk)
07:27
Disable the "SELECT max(rowid) ..." optimization for virtual tables. Ticket #2250. (CVS 3669) (check-in: ddb4d0af user: danielk1977 tags: trunk)
06:24
Minor fixes so that testfixture builds without IO tracing enabled. (CVS 3668) (check-in: 8d3829cd user: danielk1977 tags: trunk)
2007-03-01
00:29
Additional I/O Tracing support. (CVS 3667) (check-in: ed915f57 user: drh tags: trunk)
2007-02-28
06:14
Work around incompatibilities in the windows printf() routine within the new I/O tracing logic. (CVS 3666) (check-in: ceb3a07f user: drh tags: trunk)
04:47
Add the undocumented and experimental I/O tracing interface. This interface is likely to change and may be completely abandoned in the near future. (CVS 3665) (check-in: 007ca283 user: drh tags: trunk)
2007-02-27
02:01
Improvements to OS layer tracing on the unix backend. (CVS 3664) (check-in: 3ad96dbe user: drh tags: trunk)
2007-02-24
15:29
Add comments to sqlite3ExprCompare() to clarify its operation. Ticket #2216. (CVS 3663) (check-in: fba0a1e5 user: drh tags: trunk)
15:18
Additional test cases added. Improvements to the INSERT transfer optimization. (CVS 3662) (check-in: 2bf5475b user: drh tags: trunk)
13:53
Additional tests and some improvements to the INSERT transfer optimization. More testing is needed. (CVS 3661) (check-in: 83098581 user: drh tags: trunk)
13:23
Make sure the INSERT xfer optimization does not trigger if the CHECK constraints on the two tables are not identical. Ticket #2252. (CVS 3660) (check-in: 6fc18275 user: drh tags: trunk)
11:52
Make sure that integer values are converted to real when pulled from a REAL table column by GROUP BY. Ticket #2251. Also make sure default values are correctly expanded. There may be other places in the code where this issue comes up - we need to look further. (CVS 3659) (check-in: e11bbf17 user: drh tags: trunk)
2007-02-23
23:13
Disable the OR optimization if it would conflict with column affinity coercions. Ticket #2249. Additional cleanup and testing of the OR optimization. (CVS 3658) (check-in: 908daaa9 user: drh tags: trunk)
14:20
Clarifications to the datatype3.html document. (CVS 3657) (check-in: 4692a85e user: drh tags: trunk)
03:00
Clarify the documentation on how comparisons occur in an IN operator. Fix the comparison operators when both sides of an IN operator are expressions (ticket #2248). Changes to main.mk for adding FTS2 into the standard build also got mixed in with this check-in by mistake. (CVS 3656) (check-in: da81725c user: drh tags: trunk)
00:14
Fix typos in test naming (was using 'e' instead of 'f'). (CVS 3655) (check-in: 16cb00ad user: shess tags: trunk)
2007-02-22
23:06
Make the depth of the pushdown automaton stack generated by lemon changable at compile-time using -DYYSTACKDEPTH=<number>. (CVS 3654) (check-in: d8845ac1 user: drh tags: trunk)
2007-02-21
17:04
Modify the fix for #2244 to avoid using a vdbe label. (CVS 3653) (check-in: f7ce74b2 user: danielk1977 tags: trunk)
16:52
Fix for an UPDATE on a virtual table when the WHERE clause matches zero rows. Ticket #2244. (CVS 3652) (check-in: 43bf7978 user: danielk1977 tags: trunk)
16:44
Add comments to the top of keywordhash.h. (CVS 3651) (check-in: 0aa9ed5b user: drh tags: trunk)
2007-02-20
15:21
Enhance the documentation to clarify that SQLite is not reentrant through the authorization callback function. Ticket #2242. (CVS 3650) (check-in: b18a758a user: drh tags: trunk)
2007-02-17
14:59
merge the second portion of the cross-compile cleanup (CVS 3649) (check-in: 309f2de6 user: vapier tags: trunk)
14:46
regenerate (CVS 3648) (check-in: 20ac0f36 user: vapier tags: trunk)
14:31
punt custom handling of cross-compile logic and move towards the standard autotool way of things (CVS 3647) (check-in: 9817b433 user: vapier tags: trunk)
14:28
make the readline checks more flexible and handle cross-compiling logic (CVS 3646) (check-in: 4626c84b user: vapier tags: trunk)
2007-02-14
12:32
In the corrupt2.test script, change uses of (open ... a) to (open ... RDWR) to work around inconsistencies in behavior across various platforms. (CVS 3645) (check-in: 50e86b03 user: drh tags: trunk)
09:19
Use OP_VColumn instead of OP_Column when querying virtual tables for values to save in aggregate context records. #2230. (CVS 3644) (check-in: cb78f7cb user: danielk1977 tags: trunk)
2007-02-13
15:01
Add a new algorithm for handling INSERT which reduces fragmentation on a VACUUM. Ticket #2075. More testing needed. (CVS 3643) (check-in: 9f56a878 user: drh tags: trunk)
14:11
Changes to the script that generates download.html so that it recognizes FTS2 modules. (CVS 3642) (check-in: 06c22de2 user: drh tags: trunk)
13:04
Version 3.3.13 (CVS 3641) (check-in: 286c4eb3 user: drh tags: trunk)
12:49
Expand the cursor index in SrcList_item to 32-bits to accomodate really huge VDBE programs resulting from deeply nested triggers. This is an interim fix until we rework triggers to use subroutines. (CVS 3640) (check-in: 22769e2d user: drh tags: trunk)
02:12
Changed my mind. Convert "GUID" back to "UUID". (CVS 3639) (check-in: c2588405 user: drh tags: trunk)
02:07
Change "UUID" to "GUID" in the documentation. (CVS 3638) (check-in: dddd31f4 user: drh tags: trunk)
02:03
Changes in preparation for version 3.3.13. (CVS 3637) (check-in: 7799a0b6 user: drh tags: trunk)
01:41
Explanation of the "fragmentation" line is added to the help message at the end of the output of sqlite3_analyzer. (CVS 3636) (check-in: 31e5073c user: drh tags: trunk)
01:38
Additional fixes to the new fragmentation feature of sqlite3_analyzer. (CVS 3635) (check-in: 82aed271 user: drh tags: trunk)
2007-02-10
19:22
Changes to support fragmentation analysis in sqlite3_analyzer. (CVS 3634) (check-in: bd6bc3b8 user: drh tags: trunk)
2007-02-07
13:09
Explicit collations always override implicit collations. This is backwards compatible since SQLite has not previously supported explicit collations. Need to add tests of this new behavior. (CVS 3633) (check-in: 3638823a user: drh tags: trunk)
01:06
Change the coding of PRAGMA count_changes so that it uses memory cells of the VM rather than the stack, to avoid problems with leftovers on the stack interfering with other operations. Ticket #2217. (CVS 3632) (check-in: 2bd4b62a user: drh tags: trunk)
01:01
http://www.sqlite.org/cvstrac/tktview?tn=2219

When creating fts tables in an attached database, the backing tables are created in database 'main'. This change propagates the appropriate database name to the routines which build sql statements.

Note that I propagate the database name and table name separately. I briefly considered just making the table name be "db.table", but it didn't fit so well in the model used to store the table name and other information, and having the db name passed separately seemed a bit more transparent. (CVS 3631) (check-in: 283385d2 user: shess tags: trunk)

2007-02-06
23:41
Additional tests to give full coverage testing to ticket #2211. (CVS 3630) (check-in: ecb1f2fd user: drh tags: trunk)
13:26
When optimizing out an ORDER BY clause due to uniqueness constraints, make sure unused terms to the right in the ORDER BY clause to not reference other tables in a join. Ticket #2211. Additional test cases needed before closing this ticket. (CVS 3629) (check-in: 912faf18 user: drh tags: trunk)
11:11
Check the return value of lseek() in os_unix.c to make sure it really worked. (CVS 3628) (check-in: e4408dd1 user: drh tags: trunk)
2007-02-05
14:21
Set the MEM_Term flag when an internal string has a nul-terminator appended to it. Fix for #2213. (CVS 3627) (check-in: fc969ad9 user: danielk1977 tags: trunk)
2007-02-02
12:44
Minor tweaks to collating sequences. We'll hold of making major changes until 3.4.0, since we'll likely end up with some minor technical imcompatibilities. (CVS 3626) (check-in: 9740aa95 user: drh tags: trunk)
12:33
Fix typos in the pragma documentation. Ticket #2210. (CVS 3625) (check-in: f0c66ad8 user: drh tags: trunk)
2007-02-01
23:02
First cut at adding the COLLATE operator. Regression tests pass (or at least the quick set does) and a few new tests have been added. But many more tests are needed. Rules for combining collations need to be worked out. (CVS 3624) (check-in: 85cca7cd user: drh tags: trunk)
01:53
Fix a bug in the copy method of the TCL interface. Ticket #2201. (CVS 3623) (check-in: 93626396 user: drh tags: trunk)
01:40
Allow up to 64 tables in a join (the number of bits in a long long int). The old limit was 32 tables. (CVS 3622) (check-in: 505dce8f user: drh tags: trunk)
2007-01-31
23:37
Fixed incorrect typecast for flock structure ptr in fcntl() call in sqlite3TestLockingStyle() Restored previous fullfsync behavior, try fsync() if fcntl(fd, F_FULLFSYNC, 0) returns an error. (CVS 3621) (check-in: f044c5f4 user: aswift tags: trunk)
2007-01-29
17:58
Replace the randomHex() function with separate functions randomBlob() and hex(). (CVS 3620) (check-in: f5ad74a9 user: drh tags: trunk)
15:50
Add the randomhex() function as a built-in. (CVS 3619) (check-in: a6001589 user: drh tags: trunk)
2007-01-28
21:42
Implement the platform specific part of the shared library interface on OS/2 (CVS 3618) (check-in: 027251a6 user: pweilbacher tags: trunk)
21:12
Adapt returns of the os2Read() function to those of other platforms using checkin (3549) to prevent possible corruption (CVS 3617) (check-in: ba76107c user: pweilbacher tags: trunk)
2007-01-27
14:30
Version 3.3.12 (CVS 3616) (check-in: fc660703 user: drh tags: trunk)
14:26
Adjustments to the pragma test for OS-X. (CVS 3615) (check-in: 4d0d2063 user: drh tags: trunk)
14:11
More adjustments to the pragma test for windows. (CVS 3614) (check-in: 79a65436 user: drh tags: trunk)
14:04
Adjustment to tests for PRAGMA integrity_check so that they run under windows. (CVS 3613) (check-in: a79914cc user: drh tags: trunk)
13:38
Increment the version number in preparation for the 3.3.12 release. (CVS 3612) (check-in: 24e465af user: drh tags: trunk)
13:37
Fix a documentation comment in vdbe.c. Ticket #2188. (CVS 3611) (check-in: 3eec703b user: drh tags: trunk)
02:38
Parser changes to allow parenthesized numerical arguments to pragmas, in support of the integrity_check enhancement of check-in (3609) and ticket #2176. (CVS 3610) (check-in: ab6322bf user: drh tags: trunk)
02:24
Limit the number of errors returned by PRAGMA integrity_check to 100 by default. Specify an alternative limit using an argument to the pragma. Ticket #2176. (CVS 3609) (check-in: d564a039 user: drh tags: trunk)
2007-01-26
21:08
Make sure the vdbeInt.h file is not #included multiple times. Ticket #2194. (CVS 3608) (check-in: 93edd3b0 user: drh tags: trunk)
19:31
Fix prototypes for the parser. Syntactic change only - no effect on object code. Ticket #2193. (CVS 3607) (check-in: 02990fab user: drh tags: trunk)
19:23
Add code to select.c for printing the contents of parse-tree structures. The code is normally omitted. You must compile with -DSQLITE_TEST or -DSQLITE_DEBUG to enable it. (CVS 3606) (check-in: 1b26d687 user: drh tags: trunk)
19:04
Fix a (harmless) assertion fault on nested views where the inner views are compound selects. Ticket #2192. (CVS 3605) (check-in: 942e7193 user: drh tags: trunk)
13:08
Make sure the loadext.c module works even if all SQLITE_OMIT macros are defined. Tickets #2184 and #2069. (CVS 3604) (check-in: e1fae43c user: drh tags: trunk)
00:51
Work around issues with certain C++ compilers by adding a typedef to sqlite3.h. Ticket #2191. (CVS 3603) (check-in: 31e8cf7f user: drh tags: trunk)
2007-01-25
16:56
Fix an additional problem with the IS NULL optimization on LEFT JOINs. Ticket #2189. See also ticket #2177. (CVS 3602) (check-in: 358dd82d user: drh tags: trunk)
2007-01-24
03:46
Do not run the fts2i.test unless the FTS2 module is available. (CVS 3601) (check-in: 310f6858 user: drh tags: trunk)
03:43
Do not run test fts1i.test if FTS1 is not installed. (CVS 3600) (check-in: 810c0176 user: drh tags: trunk)
2007-01-22
13:07
Version 3.3.11 (CVS 3599) (check-in: 66cbbe04 user: drh tags: trunk)
13:06
Prepare for the release of version 3.3.11 (CVS 3598) (check-in: daf28830 user: drh tags: trunk)
13:02
Fix a pragma test so that it works in directories that have spaces in their names. (CVS 3597) (check-in: 071c957a user: drh tags: trunk)
2007-01-19
22:59
http://www.sqlite.org/cvstrac/tktview?tn=2166,35

Calling UPDATE against an fts table in a UTF-16 database inserts corrupted data into the database. The UTF-8 data is being inserted directly. This appears to happen because sqlite3_ value_text() destructively coerces a value to UTF-8, and it's never converted back when updating the table. This works around the problem by rearranging things so that the update happens before the coercion. (CVS 3596) (check-in: 4f2ab4b6 user: shess tags: trunk)

01:06
Make sure the IS NULL optimization introduced by check-in (3494) correctly handles a LEFT JOIN where the a term from the right table of the join uses an IS NULL constraint. Ticket #2177. This check-in also adds the new test cases that were suppose to have been added with (3494) but which were mistakenly omitted. (CVS 3595) (check-in: 335863e4 user: drh tags: trunk)
2007-01-16
18:19
Additional memory initialization in lemon - bug reported from wireshark. Ticket #2172. Note this problem only comes up with certain grammars, and does not impact SQLite. On the other hand, it might cause SQLite to run slower. So we might want to revisit this change at some point. (CVS 3594) (check-in: d537aa5e user: drh tags: trunk)
03:09
Make sure the parser generated by lemon always calls destructors on unused symbols, even if there is no action following the rule. Ticket #2171. (CVS 3593) (check-in: 0c5c1b93 user: drh tags: trunk)
2007-01-12
23:43
Fix a bug in sqlite3_prepare_v2() introduced by checkin (3578) and reported on the mailing list. (CVS 3592) (check-in: 87248470 user: drh tags: trunk)
23:09
Fix a bug in lemon. The code in which the error appears might well be unreachable. The changes does not effect the code that is generated for the SQLite parser. Ticket #2164. (CVS 3591) (check-in: f4ee5d83 user: drh tags: trunk)
2007-01-10
18:56
Make the AreFileApisANSI() call a macro with a value of 1 on wince. Ticket #2163. (CVS 3590) (check-in: a72c950c user: drh tags: trunk)
13:34
More homepage typo fixes. (CVS 3589) (check-in: 85b5d446 user: drh tags: trunk)
13:32
Fix a typo on the homepage. (CVS 3588) (check-in: a0b42991 user: drh tags: trunk)
12:58
Version 3.3.10 (CVS 3587) (check-in: 204a212a user: drh tags: trunk)
12:57
Corrections to the documentation on sqlite3_exec(). Ticket #2161. (CVS 3586) (check-in: afd33f62 user: drh tags: trunk)
12:54
Corrections to the documentation for sqlite3_busy_handler(). Ticket #2160. (CVS 3585) (check-in: 9614c0f9 user: drh tags: trunk)
2007-01-09
23:13
Documentation updates prior to version 3.3.10. Among the changes is a fix for ticket #2148 (CVS 3584) (check-in: 686beffa user: drh tags: trunk)
17:18
Fix the windows OS layer so that it returns detailed IOERR error codes. (CVS 3583) (check-in: 4b36de46 user: drh tags: trunk)
15:32
Work around win2k problems so that single-character filenames can be used. Ticket #2151. (CVS 3582) (check-in: c4eb2100 user: drh tags: trunk)
15:06
Documentation and tests to show that the cause of a parsing error is available on sqlite3_errmsg after sqlite3_step return SQLITE_SCHEMA. (CVS 3581) (check-in: 31a661d4 user: drh tags: trunk)
15:02
Updates tests and documentation to justify the behavior observed in ticket #2158. No changes to code. (CVS 3580) (check-in: e1f7cf06 user: drh tags: trunk)
14:37
Do not use the symbol "interrupt" since that is a reserved word in OpenWatcom. Ticket #2159. (CVS 3579) (check-in: 9960ba57 user: drh tags: trunk)
14:01
When an automatic re-prepare occurs, take care not to reset the internal schema symbol table. Ticket #2156. This change also includes some debugging enhancements. (CVS 3578) (check-in: 43fe7fc1 user: drh tags: trunk)
2007-01-08
22:40
Additional tests of sqlite3_prepare_v2. (CVS 3577) (check-in: b0650aa6 user: drh tags: trunk)
21:07
Fix a critical bug in sqlite3_prepare_v2 that can lead to segfaults when the schema changes and the statement is recompiled automatically. Ticket #2154. (CVS 3576) (check-in: 3401388d user: drh tags: trunk)