SQLite

Timeline
Login

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

100 check-ins occurring around ba058ce90a2ba.

2025-03-19
14:57
Correct part of [7126a51e] which changed the implicit return value of a test function. (check-in: 505d9e49f7 user: stephan tags: cygwin-fixes)
14:52
Add some explicit db close calls to work around a process-reaping timing problem on cygwin builds. (check-in: 7126a51ed8 user: stephan tags: cygwin-fixes)
13:49
Test script patches from Jan Nijtmans: always use -DUSE_TCL_STUBS in buildtclext.tcl and improvements in how TRG is populated in testrunner.tcl. (check-in: aa192203aa user: stephan tags: cygwin-fixes)
13:13
Merge trunk into the cygwin-fixes branch for the shell test improvements on Windows. An MSC build's make test now has zero failures. (check-in: 2de4efe10c user: stephan tags: cygwin-fixes)
11:53
Avoid running test cases involving ANSI control characters or Unicode on Windows in a slave interpreter, as that combination does not work. (check-in: c7fd71c77f user: stephan tags: trunk)
10:14
Merge trunk into the cygwin-fixes branch. (check-in: 34eadd374b user: stephan tags: cygwin-fixes)
2025-03-18
20:28
Fix two obscure logic problems that cause incorrect answers, found by a third-party fuzzer. (check-in: 1a8f763c31 user: drh tags: branch-3.49)
20:15
Change the generate_series() table-valued function so that its rowid is just an alias for its value. This allows it to be used as the RHS operand of a RIGHT JOIN. This fixes the issue raised by forum post 1e17219c88. (check-in: 77db4d85e7 user: drh tags: trunk)
19:21
Fix a problem that could occur when the RHS of an IN operator was a compound SELECT featuring an ORDER BY on a subquery that was flattened into one of the component SELECTs introduced by [baa83b460c677c21]. Forum post /forumpost/1e17219c88. (check-in: 7101ccd533 user: dan tags: trunk)
13:52
Internal doc touchups in ext/wasm/mkwasmbuilds.c. No functional changes. (check-in: 47d34260e7 user: stephan tags: trunk)
12:42
Further slight simplification of the run-fuzzcheck rules. (Edit: moving to mistake branch because it having to build the binaries before detecting a missing FUZZDB arg is annoying.) (Leaf check-in: aac59804cc user: stephan tags: mistake)
12:31
Very slight simplification of the run-fuzzcheck rules. (check-in: c858a39fad user: stephan tags: trunk)
11:36
General updates to autosetup/README.md. (check-in: be8ad5cf57 user: stephan tags: trunk)
10:28
Update the docs in tool/mkccode.tcl to reflect that it's more generic than it was when the docs were written. Change the shebang line to use /bin/env tclsh instead of a hard-coded tclsh path. (check-in: 9300f7f42d user: stephan tags: trunk)
2025-03-17
15:13
Prevent integer overflow when parsing NEAR queries in FTS5. (check-in: 1a5283d7da user: drh tags: trunk)
14:59
Add support for the --with-wasi-sdk configure flag to the autoconf build. (check-in: 44880fa3f0 user: stephan tags: trunk)
2025-03-16
14:05
Fix a long-standing filename digest computation bug in the OPFS SAHPool VFS which caused all VFS-stored filenames to have a digest value of 0. See /forumpost/042d53c928382021 for full details. (check-in: 493cbe7450 user: stephan tags: trunk)
13:09
Add --asan-fsanitize=... configure flag to the canonical build to optionally set -fsantize flags for the fuzzcheck-asan tool. Teach proj-check-fsanitiz to fail for flags which the compiler emits any warning for, for reasons described in its comments. (check-in: 013730e9b9 user: stephan tags: trunk)
12:27
Configure-internal doc cleanups. No functional changes. (check-in: be3a2e6311 user: stephan tags: trunk)
11:24
Consolidate some much-duplicated run-fuzzcheck recipe code in main.mk. (check-in: c0d9b9fad3 user: stephan tags: trunk)
00:13
Rework the run-fuzzcheck makefile target so that it better exploit parallelism. Test case "make -j16 run-fuzzcheck FUZZDB=20250222.db" went from 596 seconds down to 107 seconds. (check-in: 18bda13e19 user: drh tags: trunk)
2025-03-15
23:42
Enhance the fuzzcheck testing tool with new command-line options: --brief, and --slice M N. (check-in: e64132723d user: drh tags: trunk)
20:35
Merge the latest trunk enhancements into the reuse-schema branch. (check-in: 271e0373a8 user: drh tags: reuse-schema)
20:32
Merge the latest trunk enhancements into the bedrock branch via the wal2 intermediary. (check-in: b1860b6372 user: drh tags: bedrock)
20:28
Merge the latest trunk enhancements into the wal2 branch. (check-in: c8d8f613ed user: drh tags: wal2)
20:25
Merge the latest trunk enhancements into the begin-concurrent branch. (check-in: 624225499c user: drh tags: begin-concurrent)
19:55
Make use of the C99 flexible array feature, when available, so that the -fsanitize=bounds-strict option can be used, when available. Forum thread 311dbf9a1c. (check-in: d4307a0d43 user: drh tags: trunk)
19:00
Work around compilers that do not understand flexible arrays, in the recovery extension and in the fuzzcheck test module. (Closed-Leaf check-in: f101c46cf8 user: drh tags: flex-array)
18:26
Fix alignment problems on Linux with -m32 and on Mac PPC. (check-in: 8a91aeca60 user: drh tags: flex-array)
16:58
Speed up parsing of very long fts3 query expressions. (check-in: 2dd5b6895a user: dan tags: trunk)
15:19
Configure-internal build cleanups (no functional changes). Add EXTRA_SRC to the deps of sqlite3.c. (check-in: 8afb8bbce8 user: stephan tags: trunk)
13:50
-fsanitize is a CFLAG, not LDFLAG, so rename some vars accordingly and simplify the feature check to not run the linker. (check-in: 44f2c64ec1 user: stephan tags: flex-array)
13:36
For fuzzcheck-asan, dynamically determine the list of -fsanitize flags to use based on configure-time feature tests. (check-in: b70f9cc815 user: stephan tags: flex-array)
13:11
Omit the -fsanitize=bounds-strict for now, as that is still not widely implemented. In particular, it does not work on Macs. (check-in: 3e1c2ac781 user: drh tags: flex-array)
13:04
Use flexible arrays in the recovery extension and in the fuzzcheck test program. Adjust the unix makefile to use -fsanitize=bounds-strict when building fuzzcheck-asan. (check-in: 6ea6a6b211 user: drh tags: flex-array)
12:22
Use flexible arrays whereever appropriate in FTS5. (check-in: 16dfc415b6 user: drh tags: flex-array)
00:11
Convert the Fts5Sorter.aIdx field to a flexible array. (check-in: 28ac776a23 user: drh tags: flex-array)
2025-03-14
23:57
Turn Fts5Colset.aiCol into a flexible array. (check-in: 0c4d9c7474 user: drh tags: flex-array)
23:20
In FTS3, rename the MatchinfoBuffer.aMatchinfo field to aMI, to avoid confusing it with MatchInfo.aMatchinfo. Make aMI a flexiable array. (check-in: bb00b97398 user: drh tags: flex-array)
21:15
Fix one of two flexible arrays in FTS3. (check-in: ddfa87c179 user: drh tags: flex-array)
20:19
Use flexible arrays for RTREE. (check-in: 2b41776179 user: drh tags: flex-array)
19:07
KeyInfo is now an indeterminate size, so we cannot declare a variable of that type, only a pointer to an instance of that type. (check-in: 37b687dc2d user: drh tags: flex-array)
18:10
Make use of the flexible-array feature of C99, when available, to try to pacify -fsanitize=strict-bounds. This check-in fixes the core. There is more yet to do in FTS3, RTREE, and in FTS5. (check-in: 6fd6b32d06 user: drh tags: flex-array)
12:37
Fix an internal doc typo reported in forum post e25e581f917. (check-in: fa6f6ccdff user: stephan tags: trunk)
11:14
Minor doc corrections for the sahpool-digest fix and merge in current trunk. (Closed-Leaf check-in: 500f2e6ec7 user: stephan tags: sahpool-digest)
09:34
Cherrypick the [2b582c0097e33] doc addition, which was initially committed to the wrong branch. (check-in: f786de8d18 user: stephan tags: trunk)
2025-03-13
18:51
Fix the generate_series extension for the case where the termination value is not an even multiple of the step from the start value and there is also a value=NNN constraint in the WHERE clause. Forum post bf2dc8e9 (check-in: 75e72e3b0d user: drh tags: trunk)
2025-03-12
15:17
The --echo flag on the CLI also echos dot-commands provided on the command-line. (check-in: 6ec0c03b95 user: drh tags: trunk)
11:41
Document that sqlite_update_hook() can unset the current hook by passing a NULL callback, to address forum post 652aef4747. (check-in: 2b582c0097 user: stephan tags: cygwin-fixes)
2025-03-11
15:46
Avoid running test cases involving ANSI control characters or Unicode on Windows in a slave interpreter, as that combination does not work. (Closed-Leaf check-in: f6745a7355 user: drh tags: windows-ansi)
12:19
The substitute "puts" command used by the Windows implementation of sqlite3_analyzer must invoke fflush() after each line of output. Otherwise the output can be truncated when redirected into a file. (check-in: ba058ce90a user: drh tags: trunk)
11:24
Teach testrunner.tcl to distinguish Cygwin from Windows. This gets it running but then all downstream tests fail, at least in part because of the build target name discrepancies between the platform which requires .exe and those which do not. (check-in: 2861788e47 user: stephan tags: cygwin-fixes)
2025-03-10
22:31
Ensure that the TEMP database has been initialized at the beginning of a call to sqlite3_open_blob() for the TEMP database. Fix for the issue reported by forum post 0a556d619b. (check-in: 2cfccdbe08 user: drh tags: trunk)
17:35
Merge trunk into the cygwin-fixes branch. (check-in: f3d14433f2 user: stephan tags: cygwin-fixes)
17:28
Add an explicit db close to test/walsetlk.test to work around an unjustified test failure on Windows when the walsetlk tests are run in the same invocation of testfixture.exe in Windows. (check-in: f418de1093 user: stephan tags: trunk)
15:35
Patch an unnecessary size_t-to-int conversion which warns on some compilers. (check-in: 8681dfdb59 user: stephan tags: cygwin-fixes)
15:27
Ensure that the 'clean' target for Makefile.msc cleans up the testdir dir, analog to how the canonical makefile works, as stale files there have been leading to test failures. (check-in: a591c01855 user: stephan tags: cygwin-fixes)
15:15
Reimplement pathname conversion on Cygwin. MSC test passes/failures are on par with trunk and 'make test' running on cygwin is down to a single failure. (check-in: 57a4de0914 user: stephan tags: cygwin-fixes)
14:10
Tweak [6f24da264cf8d] so that it has no side-effects on autoconf/Makefile.msc. (check-in: d3d322f122 user: stephan tags: cygwin-fixes)
10:39
Disable the [d1ba200234f40b84|count-of-view optimization] if any subquery is DISTINCT, as the optimization does not work in that case. (check-in: bae270b988 user: drh tags: branch-3.49)
10:36
Fix the build by rerunning test/mkpragmatab.tcl. (check-in: 167d6314c5 user: drh tags: branch-3.49)
10:32
Disable the [d1ba200234f40b84|count-of-view optimization] if any subquery is DISTINCT, as the optimization does not work in that case. Bug reported by forum post a860f5fb2e. (check-in: d7013b6393 user: drh tags: trunk)
09:54
Add tf.bat target to Makefile.msc to facilitate running of individual test scripts via testfixture.exe. (check-in: 6f24da264c user: stephan tags: cygwin-fixes)
2025-03-08
06:53
Test app/script patches from Jan Nijtmans for cygwin. Add/replace many more sentinel arguments for TCL variadic functions which specifically need a NULL trailing argument. With this, 'make test' on cygwin runs to completion for me, with 134 of 329227 tests failing. MSC build fails the same 13 tests which fail for me in trunk. (check-in: 84d8fcfca1 user: stephan tags: cygwin-fixes)
2025-03-07
06:54
os_win.c patches for Cygwin (described in the cygwin-fixes branch wiki page). Makefile.msc's test target on trunk fails the same 13 tests as this branch. (check-in: 655c60d192 user: stephan tags: cygwin-fixes)
2025-03-06
14:40
Merge trunk into the cygwin-fixes branch (which was inadvertently branched too early). (check-in: a7863b2e2f user: stephan tags: cygwin-fixes)
13:49
Cygwin-centric test script tweaks from Jan Nijtmans. (check-in: 9868615b7d user: stephan tags: cygwin-fixes)
13:38
Cygwin-centric fixes from Jan Nijtmans. (check-in: 205979813c user: stephan tags: cygwin-fixes)
09:29
Back out the most significant part of [5c28a17253e2f], as Cygwin is a hybrid. With SQLITE_OS_UNIX it will use POSIX locking, which will misinteract with apps using Windows-style locking. (check-in: 44adf8f387 user: stephan tags: trunk)
09:08
Various typo fixes reported by Daniel Dumitriu. No functional changes. (check-in: 37e6ec7774 user: stephan tags: trunk)
07:48
Additional 'array index is signed char' warning cleanups for the shell and its embedded extensions, analog to [44bd44532d]. (check-in: f31042595b user: stephan tags: trunk)
07:09
Fix a tcl typo in the previous checkin which triggers an error on one machine but not another. (check-in: 646c2821ad user: stephan tags: trunk)
06:25
Have the configure script report cygwin as SQLITE_OS_UNIX instead of SQLITE_OS_WIN, per off-list discussion with Jan Nijtmans. (check-in: 5c28a17253 user: stephan tags: trunk)
2025-03-05
19:39
Have the recover module add "PRAGMA foreign_keys = off" to SQL output. Have the shell tool add ".dbconfig defensive off". (check-in: dcfe3d3292 user: dan tags: trunk)
18:18
Always ignore comments in the schema of a database, even if SQLITE_DBCONFIG_ENABLE_COMMENTS is turned off. (check-in: 373ae3f4de user: drh tags: trunk)
17:26
Update the recovery extension so that it works with encrypted databases. (check-in: b0b66f2115 user: dan tags: trunk)
17:12
Use AtomicRead() and AtomicWrite() to access the pcache1_g.bUnderPressure global, to forestall unnecessary angst from thread analyzers. (check-in: 41ec85637a user: drh tags: trunk)
16:35
Improvement output for ".schema --indent" in the CLI when the schema contains partial indexes with long and complicated WHERE clauses. (check-in: defd7187ff user: drh tags: trunk)
16:30
Improvement output for ".schema --indent" in the CLI when the schema contains partial indexes with long and complicated WHERE clauses. (Leaf check-in: 94c6e8d971 user: drh tags: wrong-branch)
13:39
Update Lemon so that it accepts filename arguments to directives like "%include". Ex: "%include <./subdir/file-to-include.txt>". This was capability needed by pikchr and backported here for safe keeping. SQLite does not need this capability, though it doesn't hurt to have it in tree. (check-in: 52b97a7ad5 user: drh tags: reuse-schema)
2025-03-04
21:37
Minor configure script cleanups which started out as cygwin-specific fixes but ended up just being minor platform-agnostic cleanups. (check-in: 2cda90410a user: stephan tags: trunk)
21:31
Roll back part of [6d87a8efe]: the check for tclsh90.exe (as opposed to tclsh9.0.exe) is incorrect on these platforms (it's an msvc build). Also remove an unused function added in that checkin. (Closed-Leaf check-in: dc84976a7c user: stephan tags: cygwin-fixes)
21:25
tclsqlite3 patch from Christian Werner: replace FILE handles with TCL channels for the db copy command. (check-in: ea1f7f8de4 user: stephan tags: trunk)
19:38
Teach autosetup how to find tclsh v9.0 on cygwin. (check-in: 6d87a8efe5 user: stephan tags: cygwin-fixes)
17:06
Fix pointless comparison of an unsigned integer to less than zero in Lemon. (check-in: 559560da45 user: drh tags: reuse-schema)
07:29
Makefile doc updates. Remove a couple extraneous targets. No functional changes. (check-in: 5a3e22999f user: stephan tags: trunk)
06:29
Minor doc typo fixes via forum post 65bd941da8. (check-in: 6f6a03e93c user: stephan tags: trunk)
05:16
Document the EXTRA_SRC makefile var and add --amalgamation-extra-src=list to the canonical-build configure script as the formal way to pass that at configure-time. (check-in: 44de0ec29a user: stephan tags: trunk)
2025-03-03
16:53
Merge the latest trunk enhancements into the reuse-schema branch. (check-in: b51690e180 user: drh tags: reuse-schema)
16:01
Merge the latest trunk enhancements into the bedrock branch through the wal2 intermediary. (check-in: 2033763007 user: drh tags: bedrock)
15:57
Merge the latest trunk enhancements into the wal2 branch. (check-in: 6baea938f0 user: drh tags: wal2)
15:53
Merge all the latest trunk enhancements into the begin-concurrent branch. (check-in: 051a6b159e user: drh tags: begin-concurrent)
14:15
Internal configure script cleanups. Resolve an as-yet-hypothetical corner case involving the --dev flag mixed with custom CFLAGS containing SQLITE_ENABLE... or SQLITE_OMIT... flags. Fix an unrelated API doc typo reported in forum post 606ea661df. (check-in: 0554c00f32 user: stephan tags: trunk)
11:48
Minor configure script doc tweaks. This is also a note that the basic elements of the build run as-is on Haiku OS Beta 5 but the tcl bits do not (for lack of tclConfig.sh). (check-in: acf9babf0e user: stephan tags: trunk)
2025-03-02
20:29
Ensure that detection of control characters by comparison against 0x1f uses unsigned characters. Forum post 4c344ca61f. (check-in: b7c5ce8421 user: drh tags: trunk)
2025-03-01
23:44
Use SQLITE_EXTRA_INIT_MUTEXED instead of SQLITE_EXTRA_INIT for the SQLITE_WASM_EXTRA_INIT feature, as suggested in forum post 14183b98fc0b1dea. This doesn't make a functional difference now - this is in the name of future-proofing against eventual threading support in wasm. (check-in: 46479c2e30 user: stephan tags: trunk)
11:47
Add an assert() in vdbesort.c to help both humans and static analyzer AIs understand why a particular array cannot overflow. (check-in: a14d4afd50 user: drh tags: trunk)
2025-02-28
15:54
In the CLI, ensure that all calls to ctype.h macros are cast to unsigned char. Apparently cygwin needs this. (check-in: 44bd44532d user: drh tags: trunk)
05:01
Bring the autosetup-related bits up to par with trunk. (check-in: 47fe1c5978 user: stephan tags: branch-3.49)
04:58
Latest upstream autosetup. Fixes two minor corner-case issues which we haven't yet seen in practice. (check-in: 5c2438a719 user: stephan tags: trunk)