Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
514 check-ins using file src/whereexpr.c version 36b47f7261
2019-04-03
| ||
17:54 | Add a defense-in-depth NEVER() test to the WAL cleanup code. (Leaf check-in: 1109942a50 user: drh tags: branch-3.27) | |
2019-03-13
| ||
17:20 | Avoid allocating excessive registers for the PARTITION BY expressions when processing window functions. (check-in: 180be26623 user: dan tags: window-functions) | |
15:29 | Remove rows from the ephemeral table used by window functions once they are no longer required. (check-in: 6ad5531920 user: dan tags: window-functions) | |
08:28 | Minor optimization in sqlite3WindowCodeStep(). (check-in: b1322ffb6e user: dan tags: window-functions) | |
2019-03-12
| ||
18:28 | Allow real values to be used in PRECEDING and FOLLOWING expressions for RANGE window frames. (check-in: 25ff7091cb user: dan tags: window-functions) | |
15:21 | Expand on header comment for sqlite3WindowCodeStep(). Further simplify the implementation of the same. (check-in: 5129bcc996 user: dan tags: window-functions) | |
2019-03-11
| ||
19:50 | Remove "cache mode" from the window frame code generator. Handle the same cases by editing the window frame specification itself. (check-in: 0812635383 user: dan tags: window-functions) | |
18:17 | Simplify the windows frame code some. Add a comment explaining some of the VM code generated by sqlite3WindowCodeStep(). (check-in: 6bd1a07949 user: dan tags: window-functions) | |
11:12 | Fix problems with "RANGE ... ORDER BY <expr> DESC" window frames. (check-in: e7bced731a user: dan tags: window-functions) | |
2019-03-09
| ||
20:49 | Add support for RANGE window frames. Some cases still do not work. (check-in: ffc32b246d user: dan tags: window-functions) | |
07:38 | Merge latest trunk changes into this branch. (check-in: 53ea550ce7 user: dan tags: window-functions) | |
2019-03-08
| ||
20:57 | Add simple tests for GROUPS window frames. (check-in: 2872702dac user: dan tags: window-functions) | |
20:26 | Fix a harmless bug in the Makefile. (check-in: 62ac9cb07f user: drh tags: trunk) | |
20:02 | Finish consolidation of window frame code. Add untested support for GROUPS frames. (check-in: 954bf36993 user: dan tags: window-functions) | |
16:06 | Merge the latest enhancements and fixes from trunk. (check-in: 543ecb30d6 user: drh tags: begin-concurrent-pnu-wal2) | |
15:57 | Merge the latest enhancements and fixes from trunk. (check-in: bf20d4ffcb user: drh tags: wal2) | |
15:39 | Add the bgckpt.c extension to Makefile.in and Makefile.msc. (check-in: 3712d625d6 user: drh tags: wal2) | |
15:30 | Merge all the latest enhancements from trunk. (check-in: 667cce3dce user: drh tags: begin-concurrent-pnu) | |
14:07 | Add new test cases in test/fuzzdata8.db (check-in: 97bdc090f3 user: drh tags: trunk) | |
14:02 | Patch makes balance_nonroot() slightly smaller and faster and more robust against corrupt databases. (check-in: 174da3c32c user: drh tags: trunk) | |
13:27 | Fix another segfault in fts3 that could occur with corrupt records. (check-in: fcc24edde4 user: dan tags: trunk) | |
01:52 | Detect an attempt to drop a btree that is not within the bounds of the database file and abort early with an SQLITE_CORRUPT error, to avoid problems later on in the process. (check-in: 235a9698f5 user: drh tags: trunk) | |
2019-03-07
| ||
20:47 | Fix other "ROWS BETWEEN" cases on this branch. (check-in: a5f68f6647 user: dan tags: window-functions) | |
19:26 | Modify new window functions function so that cursors are stepped immediately after each operation, instead of immediately before. (check-in: 093d2b25f1 user: dan tags: window-functions) | |
2019-03-06
| ||
21:04 | Simplify the window function code generator some more. (check-in: 45cbd3b449 user: dan tags: window-functions) | |
17:12 | Improvements to the way built-in window functions are handled. (check-in: e8eee566df user: dan tags: window-functions) | |
14:53 | Add the sqlite3_stmt_isexplain() interface. (check-in: ee642d3e27 user: drh tags: trunk) | |
14:08 | Add an "|| CORRUPT_DB" term to an assert() that might be false if the database is corrupt. Also add a branch to have sqlite3PagerMovepage() return SQLITE_CORRUPT in that case. (check-in: b0d5cf40bb user: drh tags: trunk) | |
2019-03-05
| ||
23:49 | Fix a potential 32-bit integer overflow in the "showdb" utility program when it is trying to interpret a corrupt database file. (check-in: 3803e75038 user: drh tags: trunk) | |
19:29 | Extend windowCodeStep() to handle any ROWS PRECEDING/FOLLOWING frame specification. (check-in: af0ea13635 user: dan tags: window-functions) | |
16:53 | Add a script to tool/ that will extract the sqlite3.h header file from an sqlite3.c amalgamation. (check-in: 38d2e510cd user: drh tags: trunk) | |
14:47 | New dbfuzz2 test cases added to test/fuzzdata7.db (check-in: 25975e1fb2 user: drh tags: trunk) | |
14:39 | If the initial byte of a record is a large value, then always used the generalized record comparison routine sqlite3VdbeRecordCompare() rather than one of the specialized versions that are optimized for shorter records. (check-in: c1ac00706b user: drh tags: trunk) | |
2019-03-04
| ||
21:08 | Merge trunk changes into this branch. (check-in: 9b4d561f68 user: dan tags: window-functions) | |
21:07 | Support some "ROWS BETWEEN N PRECEDING AND M FOLLOWING" window functions without caching entire partitions. (check-in: e7a91f1228 user: dan tags: window-functions) | |
07:25 | Merge latest trunk changes into this branch. (check-in: 97a9604d85 user: dan tags: reuse-schema) | |
07:15 | Fix a problem preventing compilation with SQLITE_OMIT_UTF16. (check-in: 906d1fd865 user: dan tags: trunk) | |
2019-03-02
| ||
20:09 | Improved comments on the fossildelta.c extension. (check-in: d7d23f0450 user: drh tags: trunk) | |
15:25 | Fix the SQLITE_DIRECT_OVERFLOW_READ compile-time option so that it works with SQLITE_HAS_CODEC. (check-in: fd085e9260 user: drh tags: trunk) | |
12:17 | Remove a redundant parameter extraction call from the percentile.c extension. (check-in: d49c32e6e7 user: drh tags: trunk) | |
2019-03-01
| ||
21:37 | Merge recent changes from trunk. (check-in: f270245a7b user: drh tags: apple-osx) | |
21:33 | The fts3_tokenizer() function returns NULL if the SQLITE_DBCONFIG_ENABLE_FTS_TOKENIZER setting is disabled, which is is by default. (check-in: f5732f4caf user: drh tags: trunk) | |
21:12 | Fix a minor comment typo. No code changes. (check-in: 9a0a93c89d user: drh tags: trunk) | |
18:27 | Merge the latest enhancements from trunk. (check-in: e64ded7b04 user: drh tags: apple-osx) | |
18:07 | In a query that uses a partial index, the expression that is the WHERE clause of the partial index must always be true. Use this fact to avoid evaluating identical terms in the WHERE clause of the query. (check-in: 9b2879629c user: drh tags: trunk) | |
2019-02-28
| ||
20:10 | Add the ".parameter" command to the CLI. (check-in: 1f9fa58541 user: drh tags: trunk) | |
17:29 | Add the new "bind_fallback" method to the "sqlite3" object in the TCL interface. (check-in: c7f70b6d96 user: drh tags: trunk) | |
14:09 | New test case loaded into test/fuzzdata8.db. (check-in: 00ae0c6c48 user: drh tags: trunk) | |
13:41 | Avoid an assert() failure and an undefined left-shift operation in the fts3 snippet() function that could occur if the database was corrupt. (check-in: d5250db632 user: dan tags: trunk) | |
13:21 | Add an "|| CORRUPT_DB" term to an assert() statement in the btree defragmenter. (check-in: 6bfd8655f0 user: drh tags: trunk) | |
2019-02-27
| ||
23:05 | Enable the LIKE optimization when the ESCAPE keyword is present and the PRAGMA case_sensitive_like pragma is ON. (check-in: 6ae4b8c525 user: drh tags: trunk) | |
19:59 | Fix the readfile() UDF so that it returns an empty BLOB, not an OOM error, when reading an empty file. (check-in: 0edad5339e user: drh tags: trunk) | |
16:38 | Add the "-returntype" option to the "db function" Tcl method. (check-in: 789a492b68 user: dan tags: trunk) | |
15:26 | Verify that fts5 auxiliary functions cannot be used in aggregate queries. (check-in: 122330dba3 user: dan tags: trunk) | |
2019-02-26
| ||
19:16 | Reformat the white-space in a few test cases so that the tests work with Tcl 8.6.9. (check-in: 29d02bf2fa user: dan tags: trunk) | |
18:21 | Minor change to dbfuzz2 that allows it to be compiled against older versions of SQLite. (check-in: 6d39d6a68b user: drh tags: trunk) | |
17:52 | New test cases added to test/fuzzdata8.db. (check-in: 61fdfc57b9 user: drh tags: trunk) | |
17:49 | Allocate a few extra bytes for the pager temp page as an overrun buffer while processing corrupt database files. (check-in: e7aca0714b user: drh tags: trunk) | |
16:36 | Update doc/shared_schema.md to describe the shell tool ".shared-schema" dot-command. (check-in: 0ce2092a59 user: dan tags: reuse-schema) | |
16:17 | Use unsigned integers to count the number of pages in a freelist during an integrity_check, to avoid any possibility of a signed integer overflow. (check-in: 05b87e0755 user: drh tags: trunk) | |
16:13 | Add new test file reuse5.test. For testing the shell tool ".shared-schema" command. (check-in: fa6008d547 user: dan tags: reuse-schema) | |
16:11 | Fix a harmless compiler warning that only comes up during debug builds. (check-in: 848869ced9 user: drh tags: trunk) | |
15:43 | Add the ".shared-schema check|fix DB1 DB2..." command to the shell tool. For checking if a database is eligible to share an in-memory with the main database, and for fixing small problems that prevent it from being so. (check-in: 7d8e8a9572 user: dan tags: reuse-schema) | |
12:29 | Update the configure-generated makekfile to generate a portable pkgIndex.tcl. (check-in: 35cbee7fe4 user: drh tags: trunk) | |
2019-02-25
| ||
19:23 | Fix a comment in build.c. (check-in: d6a9bff6f5 user: dan tags: reuse-schema) | |
18:43 | Enhance the command-line shell to accept bound parameters, using the TEMP table named "$Parameters" to look up the values for bound parameters. (check-in: 7c941ce5bc user: drh tags: trunk) | |
18:07 | Merge latest trunk changes into this branch. (check-in: 5c1cf30859 user: dan tags: reuse-schema) | |
17:54 | Where possible, avoid loading all schemas into memory for PRAGMA statements. (check-in: 64f97530ad user: dan tags: reuse-schema) | |
16:06 | Version 3.27.2 (check-in: bd49a8271d user: drh tags: release, branch-3.27, version-3.27.2) | |
15:55 | Internally, remove all references to a Window object that belongs to an expression in an ORDER BY clause if that expression is converted to an alias of a result-set expression. (check-in: a21ffcd817 user: drh tags: branch-3.27) | |
15:43 | This was originally intended to be the 3.27.2 release, but then I realized that I left out an important patch. (Closed-Leaf check-in: 6c5e5e2ee0 user: drh tags: mistake) | |
14:52 | Update comments on the fossildelta extension. No changes to code. (check-in: 9da4fb59b2 user: drh tags: trunk) | |
14:25 | Fix a typo in the documentation for sqlite3_total_changes(). (check-in: 4febdfb37b user: drh tags: branch-3.27) | |
14:16 | Fix a typo in the documentation for sqlite3_total_changes(). (check-in: 8474c1560e user: drh tags: trunk) | |
2019-02-23
| ||
20:48 | Fix a couple of assert() statements in fts3 that can be false if the database is corrupt. (check-in: 84162af521 user: dan tags: trunk) | |
00:56 | Import from trunk the new fix to ticket [df46dfb631f75694] in which all ephemeral tables used as the RHS of an IN operator be index btrees and never table btrees so that they can always be reused. (check-in: 0e64ac122a user: drh tags: branch-3.27) | |
00:21 | Check-in [fa792714ae62fa98] is incorrect. Add a test case to refute it and also a fix to make it right. Then add an alternative fix to ticket [df46dfb631f75694] in which all ephemeral tables used as the RHS of an IN operator be index btrees and never table btrees so that they can always be reused. (check-in: d3915230e3 user: drh tags: trunk) | |
00:08 | Remove a testcase() macro which is now unreachable due to the contraction of the use of IN_INDEX_ROWID. (Closed-Leaf check-in: 90c5a17cd5 user: drh tags: tkt-df46dfb631) | |
2019-02-22
| ||
23:29 | Make all ephemeral tables built to hold the RHS of an IN operator be index-btrees, never table-btrees, regardless of whether or not they are used as IN_INDEX_LOOP or IN_INDEX_MEMBERSHIP. That way, the same ephmerial table can be reused for both cases. (check-in: c2d50df8fd user: drh tags: tkt-df46dfb631) | |
21:33 | Check-in [fa792714ae62fa980] is not a valid fix for ticket [df46dfb631f75694], as the new test case in this check-in demonstrates. The fix here causes test cases for the [df46dfb631f75694] bug to fail again, so this check-in is on a branch. A new fix is needed for [df46dfb631f75694]. (check-in: 0d456456da user: drh tags: tkt-df46dfb631) | |
19:24 | Internally, remove all references to a Window object that belongs to an expression in an ORDER BY clause if that expression is converted to an alias of a result-set expression. Fix for [4feb3159c6]. (check-in: 579b66eaa0 user: dan tags: trunk) | |
17:44 | Merge latest trunk changes into this branch. (check-in: 001771afd0 user: dan tags: reuse-schema) | |
16:18 | In sqlite3NestedParse() be sure to detect all SQLITE_NOMEM and SQLITE_TOOBIG errors and to distinguish between them. (check-in: 73056b314b user: drh tags: trunk) | |
15:42 | Modify sqlite3_str_finish() and sqlite3VMPrintf() so that they always return NULL on any OOM or SQLITE_LIMIT_LENGTH error. (check-in: e7144ffd21 user: drh tags: trunk) | |
2019-02-21
| ||
18:11 | Fix OOM handling in sqlite3NestedParse(). (Leaf check-in: e6e9dd5c17 user: dan tags: nested-parse-oom) | |
16:41 | Detect oversized strings in the OP_String opcode even if the P4 argument is originally UTF8 and has to be converted to UTF16 to match the database file and that conversion causes the string to become shorter and cross below SQLITE_LIMIT_LENGTH threshold. This might fix an OSSFuzz problem that we have been so far unable to reproduce. (check-in: c13d563925 user: drh tags: trunk) | |
2019-02-20
| ||
19:20 | Changes to the unix VFS that allegedly enable it to build of Fuchsia. We have no way of testing this. (check-in: be21a6416d user: drh tags: trunk) | |
19:06 | Disable unix-specific features of dbfuzz2 when compiling on windows. (check-in: af84774d1e user: drh tags: trunk) | |
18:44 | Further test cases and fixes for SQLITE_OPEN_SHARED_SCHEMA. (check-in: ba0ab042f4 user: dan tags: reuse-schema) | |
18:13 | Avoid the use of function pointers in columnName(), as function pointers appear to be a source of consternation to LLVM. (check-in: c48f6f39c5 user: drh tags: trunk) | |
17:36 | Add test and fixes for SQLITE_OPEN_SHARED_SCHEMA mode. (check-in: 9a78d89c84 user: dan tags: reuse-schema) | |
13:55 | Progress handler improvements: (1) Invoke the callback after OP_Program opcodes (2) Invoke the callback multiple times in a row to catch up after a long run of no progress checks. (check-in: 0c5db18d79 user: drh tags: trunk) | |
13:23 | Merge the latest performance enhancements and fixes from trunk. (check-in: be71d1e24c user: drh tags: apple-osx) | |
13:14 | Increment the version number to 3.27.2. (check-in: a70d67d8af user: drh tags: branch-3.27) | |
13:12 | When an IN operator drives a query loop, mark it as "CODED" so that it will not be used afterwards for a (pointless) membership test. (check-in: 55e38d53ad user: drh tags: branch-3.27) | |
12:52 | When an IN operator drives a query loop, mark it as "CODED" so that it will not be used afterwards for a (pointless) membership test. This is a better fix for ticket [df46dfb631f75694] than the previous fix that is now on a branch as it preserves the full optimization of check-in [e130319317e76119]. (check-in: fa792714ae user: drh tags: trunk) | |
03:38 | Back off the optimization of check-in [e130319317e76119] slightly so that it only applies to IN operators that are used for membership tests. Proposed fix for ticket [df46dfb631f75694]. (Closed-Leaf check-in: b5f90bfe62 user: drh tags: tkt-df46dfb631) | |
2019-02-19
| ||
20:29 | Enhance the xBestIndex method on delta_parse() to return SQLITE_CONSTRAINT if no delta argument is supplied. (check-in: f16d127c3b user: drh tags: trunk) | |
20:19 | Add the delta_parse(DELTA) table-valued function to the fossildelta extension. (check-in: d91fcc267b user: drh tags: trunk) | |
18:39 | Add the fossildelta.c extension in ext/misc with implementations of the Fossil delta functions. (check-in: b80cafa6f8 user: drh tags: trunk) | |
18:00 | Improve error messages caused by corrupt database schemas in OPEN_SHARED_SCHEMA mode. (check-in: 8ac75b8a88 user: dan tags: reuse-schema) | |
17:45 | Fix a potential memory leak in RBU if the rbu_fossil_delta() SQL function is misused. Misuse never happens in a working RBU system, so this is not a particularly important fix. (check-in: 12517d1b15 user: drh tags: trunk) | |
16:42 | Fix an assert() that might not be true if the database file is corrupt. (check-in: f2d400db4d user: drh tags: trunk) | |
13:51 | New assert() to verify the TF_HasprimaryKey flag. (check-in: 0abace8a7a user: drh tags: trunk) | |
13:08 | Always use the symbolic name SQLITE_IDXTYPE_PRIMARYKEY rather than a hard-coded magic number 2. (check-in: 55d43adc89 user: drh tags: trunk) | |
2019-02-18
| ||
18:16 | Ensure that creating temp schema items does not cause an OPEN_SHARABLE_SCHEMA connection to load all schemas into memory. (check-in: 88cbf54eee user: dan tags: reuse-schema) | |
12:16 | Fix an assert() that might not be true if the %_segdir table of FTS3 contains corrupt entries. (check-in: a6509ca3dc user: drh tags: trunk) | |
2019-02-16
| ||
22:45 | Fix an assert() that might be false for a corrupt database. (check-in: 0cfb888723 user: drh tags: trunk) | |
17:27 | Add support for chaining of WINDOW definitions. (check-in: c155125fd5 user: dan tags: window-functions) | |
2019-02-15
| ||
19:36 | Enhance the virtual table in test_schemapool.c so that it can be used to check that SHARED_SCHEMA connections are not allocating and freeing schemas when they should not be. (check-in: cb236cb985 user: dan tags: reuse-schema) | |
19:00 | Fix a problem with eponymous virtual tables and SHARED_SCHEMA databases. Also, after preparing statements that require all database schemas (REINDEX, ANALYZE, CREATE, DROP and some PRAGMA statements), do not allow the database connection to return more than one schema to each schema-pool. (check-in: ecf6251ec0 user: dan tags: reuse-schema) | |
11:54 | Revert the rearrangement of VDBE code in [219b39e14] so that vdbe.c matches trunk. Since the new call to sqlite3Init() in OP_ParseSchema was removed, the rearrangement no longer provides any performance advantage. (check-in: 03c4f00317 user: dan tags: reuse-schema) | |
2019-02-14
| ||
21:04 | Fix SQLITE_DBSTATUS_SCHEMA_USED so that it works with SQLITE_OPEN_SHARED_SCHEMA connections. (check-in: d43b3c056c user: dan tags: reuse-schema) | |
18:38 | Change the name of the SQLITE_OPEN_REUSE_SCHEMA flag to SQLITE_OPEN_SHARED_SCHEMA. (check-in: 7257fcc8c9 user: dan tags: reuse-schema) | |
17:59 | Fix a typo in shared_schema.md. (check-in: e47a5aea76 user: dan tags: reuse-schema) | |
17:51 | Add documentation file doc/shared_schema.md to describe the change on this branch. (check-in: a5f8067dde user: dan tags: reuse-schema) | |
15:56 | Merge latest trunk into this branch. (check-in: 577d163836 user: dan tags: reuse-schema) | |
15:47 | Add missing comments and fix other code issues in the new functions in callback.c. (check-in: 441cabb62f user: dan tags: reuse-schema) | |
15:27 | Improved oversized cell detection when updating ptrmap pages in balance_nonroot(). (check-in: aa61435a4c user: drh tags: trunk) | |
2019-02-13
| ||
19:17 | Fix for sqlite3_table_column_metadata() on REUSE_SCHEMA databases. (check-in: 53220ad780 user: dan tags: reuse-schema) | |
18:29 | Avoid crashing after parsing a corrupt schema with a REUSE_SCHEMA connection. (check-in: b102148e71 user: dan tags: reuse-schema) | |
15:51 | Fix a problem with the incrblob API and reusable schemas. (check-in: 34f0f96f47 user: dan tags: reuse-schema) | |
14:06 | Rearrange the code in the VDBE to help out the C-compiler optimizer. And fix a harmless compiler warning. (check-in: 219b39e149 user: drh tags: reuse-schema) | |
13:48 | Performance optimization in the VDBE, and a fix for a harmless compiler warning. (Closed-Leaf check-in: e002666ac7 user: drh tags: reuse-schema-vdbe-opt) | |
08:40 | Fix a problem with OPEN_REUSE_SCHEMA connections reloading the temp schema. (check-in: 7c2ec2d4cf user: dan tags: reuse-schema) | |
2019-02-12
| ||
22:58 | Ensure that the nProgressLimit variable is always initialized in sqlite3VdbeExec(), even if the routine jumps to its exit processing early. (check-in: 167b91df77 user: drh tags: trunk) | |
21:04 | Enhancement the progress callback mechanism so that the progress callback is always invoked at least once at the end of a prepared statement if the opcode count has been exceeded. This makes the progress callback more effective at limiting run times. This check-in also includes and unrelated performance enhancement to OP_Column. (check-in: 68cce272e7 user: drh tags: trunk) | |
20:58 | Add tests for creating temp schema objects with SQLITE_OPEN_REUSE_SCHEMA connections. (check-in: 8c07b609fc user: dan tags: reuse-schema) | |
19:20 | Share schemas between databases attached to the same database handle. (check-in: ea611d7cba user: dan tags: reuse-schema) | |
16:58 | Further performance improvements to btreeInitPage(). (check-in: 93ae382e97 user: drh tags: trunk) | |
15:51 | Increase the version number to 3.28.0 for the next release cycle. (check-in: 6eb38c59a8 user: drh tags: trunk) | |
12:25 | Register the "vfsstat" virtual table immediately when loading the vfsstat.c extension. (check-in: 9bd92afd0c user: drh tags: trunk) | |
01:28 | New test cases in test/fuzzdata8.db. (check-in: ab2356f580 user: drh tags: trunk) | |
01:04 | Defer computing the number of bytes of free space on a btree page until that value is actually needed. (check-in: 177f5f40ea user: drh tags: trunk) | |
00:58 | Change an assert() into a NEVER(), since the condition is difficult to prove with certainty. Improved comment on the MemPage.nFree field. (Closed-Leaf check-in: fec071b89d user: drh tags: deferred-free-space) | |
2019-02-11
| ||
22:50 | Do not invoke btreeComputeFreeSpace() when not necessary. (check-in: f11b0ed474 user: drh tags: deferred-free-space) | |
20:13 | Merge latest trunk changes into this branch. (check-in: dbedd81bf2 user: dan tags: reuse-schema) | |
19:34 | Add eponymous virtual table "schemapool". For inspecting the current contents of the schema-pool. (check-in: 2ebeb74783 user: dan tags: reuse-schema) | |
19:27 | Improvements to comments. No changes to code. (check-in: 16eb0fc8f8 user: drh tags: deferred-free-space) | |
17:22 | Fix test cases. Avoid unnecessary tests for MemPage.nFree less than zero. (check-in: 480a5090be user: drh tags: deferred-free-space) | |
16:12 | Fix another segfault that could occur in fts5 with a corrupted database. (check-in: 09e337386b user: dan tags: trunk) | |
13:25 | Fix typos in comments in fts5.h. No changes to code. (check-in: c19e97bc92 user: dan tags: trunk) | |
12:51 | Updates to a comment to keep it in alignment to the changes on this branch. (check-in: 7499f74351 user: drh tags: deferred-free-space) | |
11:34 | Build and test fixes for SQLITE_OMIT_WINDOWFUNC builds. (check-in: 8824115a97 user: dan tags: trunk) | |
11:04 | Avoid using a "return void_function();" statement in void function sqlite3WalSnapshotUnlock(). It provokes a warning from some compilers. (check-in: 4a4952d38e user: dan tags: trunk) | |
01:58 | Add an assert() in an attempt to repro an ASAN warning from OSSFuzz. (check-in: 7b41222428 user: drh tags: trunk) | |
2019-02-09
| ||
22:33 | Fix a page-cache reference leak in the btree balancer when there is a corrupt database. (check-in: 9285899120 user: drh tags: deferred-free-space) | |
21:06 | Defer computing the MemPage.nFree value of an in-memory btree page until it is actually needed, since for many pages it is never needed. This checkin works sufficiently to prove the concept, but still has issues with exception handling. (check-in: 1d43ee4000 user: drh tags: deferred-free-space) | |
19:23 | Change a few assert() statements in fts3 that might fail if the database is corrupt. (check-in: db74a56af7 user: dan tags: trunk) | |
17:47 | Fix virtual table support for SQLITE_OPEN_REUSABLE_SCHEMA connections. (check-in: 3ca8856a7b user: dan tags: reuse-schema) | |
2019-02-08
| ||
22:34 | Small performance improvement and size reduction for pageFindSlot() - the routine in btree.c that locates a free slot for a cell on a btree page. (check-in: 1969372ac7 user: drh tags: trunk) | |
20:55 | Use a fast compiler-provided memory barrier exclusively, if such a memory barrier is available. (Leaf check-in: df66fec9bc user: drh tags: faster-memory-barrier) | |
19:30 | Add test cases and fix problems on this branch. (check-in: 2b2e9f81cd user: dan tags: reuse-schema) | |
17:28 | Further simplifications to sqlite3VdbeMemSetStr(). (check-in: 1d21295707 user: drh tags: trunk) | |
15:59 | Change the sqlite3VdbeMemGrow() routine so that it no longer guarantees a minimum size of 32 bytes. That minimum is no longer required, and without the extra check for the minimum size, the routine runs faster. (check-in: 5c499da8a4 user: drh tags: trunk) | |
14:59 | Merge the fix for ticket [4e8e4857d32d401f], so that this branch now contains release 3.27.1 plus the extra patch to preserve ROWID values on VACUUM. (check-in: 0cdae60ed7 user: drh tags: apple-osx) | |
14:55 | Give the sqlite3 object a pointer to the current Parse so that if an OOM occurs, it can automatically set the Parse.rc value to SQLITE_NOMEM. This avoids a frequent extra test of db.mallocFailed in the innermost parser loop. (check-in: 5c6638040b user: drh tags: trunk) | |
13:17 | Version 3.27.1 (check-in: 0eca3dd3d3 user: drh tags: release, branch-3.27, version-3.27.1) | |
12:46 | Cherrypick the fix for ticket [4e8e4857d32d401f] from trunk. (check-in: d5d944d794 user: drh tags: branch-3.27) | |
12:44 | Increase the version number to 3.27.1. (check-in: b94d5428bb user: drh tags: branch-3.27) | |
04:15 | Do not do the optimization that attempts to pull expression values from an index on that expression when processing a multi-index OR (see check-in [a47efb7c8520a0111]) because the expression transformations that are applied become invalid when the processing moves off of the current index and on to the next index. Fix for ticket [4e8e4857d32d401f]. (check-in: 440a7cda00 user: drh tags: trunk) | |
2019-02-07
| ||
19:28 | Pull in the change that causes VACUUM to preserve rowid values. (check-in: c9af2f71bf user: drh tags: apple-osx) | |
19:07 | Change VACUUM so that it preserves ROWID values. (check-in: 49e5d11d46 user: drh tags: trunk) | |
18:49 | Merge the offical 3.27.0 release. (check-in: 628291641f user: drh tags: apple-osx) | |
17:02 | Version 3.27.0 (check-in: 97744701c3 user: drh tags: trunk, release, version-3.27.0) | |
15:09 | Experimental change to VACUUM so that it preserves existing ROWID values. (Closed-Leaf check-in: f264015f84 user: drh tags: vacuum-preserves-rowid) | |
00:17 | In the fuzzcheck test program, make sure the isspace() and related routines are not given values outside the range of -1 to 255. (check-in: f298328771 user: drh tags: trunk) | |
2019-02-06
| ||
20:49 | Ensure all bytes of the PgHdr1 structure are initialized. This avoids a valgrind error when running corruptI.test. (check-in: a505e34d4e user: dan tags: trunk) | |
20:12 | Update an assert() in test_journal.c to take the new "VACUUM INTO" functionality into account. (check-in: 0e6249cb95 user: dan tags: trunk) | |
19:42 | Avoid attempting to delete a file while it is still open in walfault2.test. (check-in: 9d54a63d6e user: dan tags: trunk) | |
18:08 | Update test file dbfuzz001.test so that it works with SQLITE_ENABLE_OVERSIZE_CELL_CHECK builds. (check-in: 4371a0c46e user: dan tags: trunk) | |
15:23 | Add -DSQLITE_ENABLE_DESERIALIZE to the default testing configuration in test/releasetest.tcl. (check-in: 433d6ef637 user: dan tags: trunk) | |
13:48 | Do not attempt to test the shell ".trace" command in SQLITE_OMIT_TRACE builds. (check-in: 3839e90981 user: dan tags: trunk) | |
13:41 | Fix a problem preventing testfixture from being built with SQLITE_OMIT_VIRTUALTABLE. (check-in: d2cca36adc user: dan tags: trunk) | |
01:18 | When extending the size of SrcList objects, do not attempt to use extra space on the end of the allocation, as that provides scarcely any performance benefit but does greatly complicate cross-platform testing. (check-in: 3087a0c31e user: drh tags: trunk) | |
00:55 | If the query flattener detects an error, cause the SELECT code generator to abort immediately. (check-in: 3d3b142f10 user: drh tags: trunk) | |
00:11 | Back out check-in [008112bcef561a8], reenabling deprecated PRAGMAs when the SQLITE_OMIT_DEPRECATED compile-time option is used, because way too many TH3 tests depend on those deprecated PRAGMAs. (check-in: 9dbf512d1c user: drh tags: trunk) | |
2019-02-05
| ||
20:51 | Omit the ".progress" command from the shell if it is compiled with the SQLITE_OMIT_PROGRESS_CALLBACK option. (check-in: 4d0a949fd9 user: drh tags: trunk) | |
19:52 | Fix a harmless compiler warning in the memtrace.c extension. (check-in: 2f468da4e9 user: drh tags: trunk) | |
19:51 | Merge latest trunk into this branch. (check-in: c089cc4fbe user: dan tags: reuse-schema) | |
19:48 | Fix compiler warnings on Windows. (check-in: 4978ee8b54 user: drh tags: trunk) | |
16:53 | Update the tester.tcl --malloctrace option so that it uses eu-addr2line instead of addr2line. (check-in: 79c073878d user: dan tags: trunk) | |
14:36 | The IS NOT NULL operator does not imply that the operand is never NULL. Fix for ticket [5948e09b8c415bc45da5cf] (check-in: d840e9bb02 user: drh tags: trunk) | |
12:16 | Keep the temp_store_directory and data_store_directory pragmas even when compiling with SQLITE_OMIT_DEPRECATED. (check-in: 93386a7c97 user: drh tags: trunk) | |
08:55 | Update test file "resetdb.test" so that it works with the "prepare" permutation. (check-in: 95d338124b user: dan tags: trunk) | |
2019-02-04
| ||
21:10 | Early detection of implausibly sized records to avoid unnecessary large memory allocations. (check-in: 2c8769c69f user: drh tags: trunk) | |
19:52 | Mention the new -memtrace command-line option in the -help output of the CLI. (check-in: ada91aefe3 user: drh tags: trunk) | |
19:50 | Add the -memtrace option to dbfuzz2. (check-in: 67fecbc79d user: drh tags: trunk) | |
19:45 | Enhance dbfuzz2 so that with the -v option it shows the return code and error message for any failing SQL statements. (check-in: 3a127ef9f7 user: drh tags: trunk) | |
19:12 | Ensure that the sqlite3_exec() callback gets the correct number of columns for a query, even if the schema changes out from under it. (check-in: a16ffb5a4b user: drh tags: trunk) | |
16:42 | Do not run shmlock.test as part of the inmemory_journal permutation. (check-in: d3fea69cbb user: dan tags: trunk) | |
16:36 | Update the test/fuzzdata8.db database with new interesting cases from dbsqlfuzz. (check-in: e467e4c8fa user: drh tags: trunk) | |
16:25 | Fix problems with SAVEPOINT and related commands within transactions on a corrupt database that include fts5 or some other virtual tables. (check-in: b0da6e0dd1 user: dan tags: trunk) | |
14:53 | Add the ".progress" command to the CLI. (check-in: ad84ca24ff user: drh tags: trunk) | |
14:00 | Fix another potential buffer overread in fts5. (check-in: 14fba4cc5d user: dan tags: trunk) | |
11:54 | Fix another corruption case in fts5. (check-in: e608085d3f user: dan tags: trunk) | |
2019-02-03
| ||
07:46 | Fix an assertion failure in fts3 triggered by a corrupt database. (check-in: 560ebd4269 user: dan tags: trunk) | |
2019-02-02
| ||
18:24 | New testcase added to test/fuzzdata8.db. (check-in: 2f5c1925b9 user: drh tags: trunk) | |
15:59 | Honor key query parameters for SEE on the URI filename for ATTACH and VACUUM INTO. (check-in: 2e01096b89 user: drh tags: trunk) | |
15:05 | Give a link to the URI Filename document in the documentation to the various sqlite3_uri_parameter() interfaces. (check-in: 69b0b2ee53 user: drh tags: trunk) | |
13:47 | Fix a buffer overread in fts5 debugging scalar function fts5_decode(). (check-in: 54f2399fb2 user: dan tags: trunk) | |
01:27 | Fix harmless compiler warning. (check-in: dddda685f3 user: mistachkin tags: trunk) | |
2019-02-01
| ||
21:08 | Slight adjustment to the printf formatter large memory allocation detector so that it does not overestimate the amount of space needed for oversize %d conversions. (check-in: 1aee70d6de user: drh tags: trunk) | |
20:29 | Prevent the printf formatter from doing large memory allocations - larger than either the size of the static buffer for interfaces like sqlite3_snprintf(), or larger than SQLITE_LIMIT_LENGTH for interfaces that are associated with a database connection. This helps to prevent DOS attacks on products that let hostile sources inject arbitrary SQL. It also helps fuzzers run faster and more effectively. (check-in: 179e5d4605 user: drh tags: trunk) | |
18:46 | Performance improvement in the parsing of options to %-formats in the printf implementation. (check-in: 40d8f8ae87 user: drh tags: trunk) | |
15:06 | Ensure that the Walker.pParse structure is initialized when walking the source tree to gather the WINDOW clauses on a query with window functions. (check-in: 4ca9d5d53d user: drh tags: trunk) | |
14:54 | New test cases added to test/fuzzdata8.db. (check-in: e5924939c9 user: drh tags: trunk) | |
14:50 | Improve the strict enforcement of cell sizes in balancing from check-in [12713f320b2c1def] so that it also works with table-btrees in addition to index-btrees. (check-in: ef27e7a087 user: drh tags: trunk) | |
14:40 | Fix an assert() in fts5 that could fail if the database is corrupt. (check-in: 55f06aa3f8 user: dan tags: trunk) | |
13:34 | Fix another corrupt-database-handling problem in fts5. (check-in: 9876ae1b80 user: dan tags: trunk) | |
2019-01-31
| ||
16:09 | Remove an incorrect assert() from the deserialize in-memory database VFS. (check-in: 80151d7e3b user: drh tags: trunk) | |
15:38 | Add the SQLITE_CONFIG_MEMDB_MAXSIZE configuration option for configuring the default maximum size of an in-memory database created using sqlite3_deserialize(). This is necessary to make the interface reasonably testable. (check-in: cb72ee0478 user: drh tags: trunk) | |
14:37 | Fix another buffer overrun that could occur when quering a corrupt database using an fts5vocab table. (check-in: b80119bd60 user: dan tags: trunk) | |
02:42 | If the amalgamation file "sqlite3.c" is generated using the SQLITE_ENABLE_UPDATE_DELETE_LIMIT option, but then subsequently (mis-)compiled without that option, and then a particular syntax error could cause a memory leak, but for this fix. (check-in: 1bb17c863a user: drh tags: trunk) | |
01:39 | Improvements to "PRAGMA parser_trace=ON": Make it a flag pragma, and then use the flag to show the complete SQL text at the beginning of the parse. (check-in: 507c43537f user: drh tags: trunk) | |
2019-01-30
| ||
19:50 | Change the error message slightly for when two or more indexes have the same rootpage. (check-in: 916c52dad4 user: drh tags: trunk) | |
19:12 | When parsing the schema, check for duplicate index root page numbers before adding the index to the schema hash table. (check-in: 6b360bc0b9 user: dan tags: trunk) | |
18:47 | Add test case for the previous commit. (check-in: 197edb235b user: dan tags: trunk) | |
18:33 | Detect if two indexes of the same table share a common rootpage while parsing the schema, and throw an error immediately. (check-in: f21c6f1343 user: drh tags: trunk) | |
16:58 | Fix an off-by-one error when parsing the names of indexes that do not have arguments in the index_usage utility. (check-in: dc794d8f51 user: drh tags: trunk) | |
15:47 | Add the --progress, --using, and -q options to the index_usage utility program. (check-in: a5e6be7cbc user: drh tags: trunk) | |
14:01 | Enhancements to the index_usage utility program. (check-in: 19c739b4a8 user: drh tags: trunk) | |
12:15 | Fix another buffer overread in fts5 that may occur when accessing a corrupt database. (check-in: 760d14374d user: dan tags: trunk) | |
2019-01-29
| ||
19:17 | Merge enhancements from trunk. (check-in: 31efcc3775 user: drh tags: apple-osx) | |
16:54 | Fix harmless compiler warnings. (check-in: 9a93c68a76 user: drh tags: trunk) | |
16:47 | Load all the latest dbsqlfuzz finds into test/fuzzdata8.db. (check-in: e744d2dd93 user: drh tags: trunk) | |
16:41 | More robust handling of corrupt database file in the rebalance operation of the btree logic. (check-in: 97704cb7d2 user: drh tags: trunk) | |
16:34 | Avoid an integer overflow in the fts5 snippet() function triggered by a corrupt database record. (check-in: 7c862c4655 user: dan tags: trunk) | |
15:30 | Avoid a crash when the fts5 snippet() function (or similar) is used with a special query pattern, like '*id' or '*reads'. (check-in: 9d58a15737 user: dan tags: trunk) | |
11:42 | Fix a buffer overrun triggered by a merge operation on a corrupt fts5 database. (check-in: 7ee3cd5b2c user: dan tags: trunk) | |
02:37 | Set a low limit on the length of strings and blobs in the OSSFuzz module, in an effort to avoid timeouts when the fuzzer does things like "randomblob(1e12)". (check-in: 119d1609fa user: drh tags: trunk) | |
2019-01-28
| ||
19:06 | Fix the query planner so that it is able to use an index on a CAST expression. (check-in: 2c886f3d49 user: drh tags: trunk) | |
18:58 | Fix a performance regression caused by the previous commit. (Closed-Leaf check-in: c4db0ad12d user: drh tags: index-on-cast) | |
18:08 | Make indexes on CAST(...) expressions work. (check-in: 3ef711d98f user: dan tags: index-on-cast) | |
16:50 | Fix a buffer overread in fts3 that could occur when accessing a corrupt database. (check-in: a9faf90339 user: dan tags: trunk) | |
13:27 | Do not assume that text parameters passed to fts4aux queries do not contain embedded nul characters. (check-in: df04859a99 user: dan tags: trunk) | |
11:54 | Add a new test case to fuzzdata8.db. (check-in: ce8e279a7d user: drh tags: trunk) | |
00:42 | Change the definition of the BtCursor.skipNext field so that it is undefined for states CURSOR_VALID and CURSOR_INVALID. (check-in: 6089d5dba7 user: drh tags: trunk) | |
2019-01-27
| ||
19:50 | Improvements to the ".eqp trace" command in the CLI so that it ensures that the schema has been read prior to activing the tracing feature, to avoid cluttering the output with a trace of the schema parse. (check-in: 42687d45aa user: drh tags: trunk) | |
02:45 | New test cases form dbsqlfuzz (check-in: 9cf8ebd141 user: drh tags: trunk) | |
02:41 | Window functions that can abort should indicate this, so that if they are used in DML statement, a statement journal will be used. (check-in: 0ea05a0eb9 user: drh tags: trunk) | |
01:11 | Update fuzzcheck so that with the --load-dbsql options it screens its inputs to ensure that they are valid dbsqlfuzz cases. Add new dbsqlfuzz finds to the test/fuzzdata8.db database. (check-in: 004f7d9bbe user: drh tags: trunk) | |
2019-01-26
| ||
23:34 | Ensure that the btree cursor is correctly set up prior to backing it up as part of a delete operation on a divider cell in an index. (check-in: a3ea1a822d user: drh tags: trunk) | |
21:05 | Fix a broken assert() in fts3. Also some test script issues causing failures with builds that do not support fts3. (check-in: d59567dda2 user: dan tags: trunk) | |
20:41 | Fix another typo in fts3_common.tcl causing test failures in builds that do not support fts3. (Closed-Leaf check-in: 773b598608 user: dan tags: rollback-abort) | |
20:35 | Fix a typo in fts3corrupt4.test causing it to fail if the build does not support fts3. (check-in: 5016303421 user: dan tags: rollback-abort) | |
19:09 | Fix a broken assert() in fts3. (check-in: b8dd2d67b5 user: dan tags: rollback-abort) | |
18:10 | If a write statement fails with OE_Abort, but there is no statement journal, roll the entire transaction back instead. (check-in: 75a8ed7a42 user: dan tags: rollback-abort) | |
17:47 | Fix "PRAGMA journal_mode" so that if it fails because there is a transaction open, it does not roll that transaction back. (check-in: 9f39cb5b81 user: dan tags: rollback-abort) | |
16:34 | Roll back the transaction if a write statement fails with OE_Abort but there is no open statement transaction. (check-in: d536be698d user: dan tags: rollback-abort) | |
15:40 | Add the ".eqp trace" command to the CLI when using SQLITE_DEBUG, as a convenient shorthand for "PRAGMA vdbe_debug=ON" but with automatic indentation feature for program listings provided by the CLI. (check-in: 626502faa1 user: drh tags: trunk) | |
2019-01-25
| ||
20:09 | Do not accidently truncate zeroblob values when doing an arithmetic operation. Fix for ticket [bb4bdb9f7f654b0bb9f34cfba]. (check-in: 13f6942eb0 user: drh tags: trunk) | |
19:29 | Reinitialize debugging information on registers of a trigger prior to each invocation of the trigger. This prevents false-positives from the sqlite3VdbeMemAboutToChange() test logic. (check-in: 98b3b97573 user: drh tags: trunk) | |
18:17 | Fix harmless compiler warnings. (check-in: 3ff220ad37 user: drh tags: trunk) | |
18:17 | Fix the assert_fts3_nc() macro so that it works for test builds that omit SQLITE_DEBUG. (check-in: b10df2b569 user: drh tags: trunk) | |
17:51 | Fix harmless compiler warnings in fuzzcheck (check-in: a825396657 user: drh tags: trunk) | |
17:26 | Fix a segfault that could follow an OOM when querying a table that has one or more columns with default values "true" or "false". (check-in: 202f9919c2 user: dan tags: trunk) | |
16:54 | Fix asan warnings in fts5 triggered by corrupt databases - passing NULL to memcmp, out-of-range left-shift values and signed integer overflow. (check-in: 93f8ec146d user: dan tags: trunk) | |
14:48 | Add an assert_fts3_nc() macro to fts3 - for assert() conditions that are only true when it is guaranteed that the fts3 database is not corrupt. (check-in: 3498908cd7 user: dan tags: trunk) | |
14:23 | Extend fuzzcheck so that it can process dbsqlfuzz cases. Add a collection of interesting dbsqlfuzz cases to the standard test suite. (check-in: e2991a7ecf user: drh tags: trunk) | |
14:16 | Fix the xFetch method of the "memdb" VFS (used by deserialize) so that it is robust against corrupt database file. (Closed-Leaf check-in: 2c1ef40e78 user: drh tags: dbsqlfuzz-in-fuzzcheck) | |
13:42 | Fix a couple of assert() statments in btree.c that could fail with corrupt databases. (check-in: 5eb5e8289f user: dan tags: trunk) | |
13:03 | In fuzzcheck, activate vdbe_debug for dbsqlfuzz cases when using the -vvvvv verbosity level or above. (check-in: 2e6f7c2ace user: drh tags: dbsqlfuzz-in-fuzzcheck) | |
04:43 | Improved automatic detection of dbsqlfuzz cases in fuzzcheck. (check-in: 1ef24e89c9 user: drh tags: dbsqlfuzz-in-fuzzcheck) | |
04:00 | Add the ability to process dbsqlfuzz cases in fuzzcheck and add an initial set of interesting dbsqlfuzz cases. (check-in: fb9074ff45 user: drh tags: dbsqlfuzz-in-fuzzcheck) | |
2019-01-24
| ||
17:41 | Fix a buffer overread in fts3 that could occur in a prefix query on a corrupted database. (check-in: d0d5689371 user: dan tags: trunk) | |
16:51 | While PRAGMA full_column_names is off (the default) do not make changes to the short_column_names pragma when computing the column names for subqueries. (Leaf check-in: 718ead555b user: drh tags: literal-column-names) | |
16:27 | Fix a problem with running ALTER TABLE on a schema that contains expressions of the type "col IN ()" (empty set on RHS of IN operator). (check-in: 2d9cd06715 user: dan tags: trunk) | |
16:07 | Omit deprecated PRAGMAs when compiling with the -DSQLITE_OMIT_DEPRECATED option. (check-in: 008112bcef user: drh tags: trunk) | |
15:51 | Make sure the column name flags are restored correctly after an error inside of sqlite3ResultSetOfSelect(). (check-in: b1601db7ad user: drh tags: trunk) | |
15:16 | Fix a potential problem with "INSERT INTO ... SELECT * FROM" (or VACUUM) statements on a corrupted database. (check-in: db4b4c2c1e user: dan tags: trunk) | |
14:16 | Change a integer variable in sqlite3VdbeRecordUnpack() to unsigned in order to avoid any possibility of an integer overflow. (check-in: 1b536f6fd8 user: drh tags: trunk) | |
13:36 | Use memmove() instead of memcpy() in a place where buffers might overlap if the database file is badly corrupted, to prevent warnings from ASAN and valgrind. (check-in: 65ad6c55f1 user: drh tags: trunk) | |
04:44 | Remove an unreachable branch. (check-in: 4c976f1713 user: drh tags: trunk) | |
2019-01-23
| ||
20:31 | Fix an assert() in vdbemem.c that could fire if the database was corrupt. (check-in: a70958cd7d user: dan tags: trunk) | |
19:50 | Fix a problem with renaming a table within a schema that contains a composite query that uses a column alias as an ORDER BY term. (check-in: 2ca6b8f84e user: dan tags: trunk) | |
19:25 | Stricter enforcement of cell sizes when doing balancing operations on the btree, in order to catch file corruption sooner. (check-in: 12713f320b user: drh tags: trunk) | |
19:17 | Fix another fts5 crash that can occur if the database is corrupted. (check-in: 44ce8baa47 user: dan tags: trunk) | |
16:59 | Fix problems with sub-selects in WINDOW definitions. Also rename-column operations when the column being renamed appears in a WINDOW definition that is part of a VIEW or TRIGGER. (check-in: 0387cb3add user: dan tags: trunk) | |
12:19 | Fix a buffer overwrite triggered by a prefix query on a corrupt fts5 table. (check-in: 1d8172a94b user: dan tags: trunk) | |
2019-01-22
| ||
21:17 | Fix a buffer overrun that could occur in fts5 if a prefix query is made on a corrupt database. (check-in: 1abc441564 user: dan tags: trunk) | |
20:18 | Do not run shmlock.test as part of the journaltest permutation. (check-in: 5b7d0c784e user: dan tags: trunk) | |
16:44 | Use _strdup() instead of strdup() on Windows builds of the CLI, to avoid a compiler warning reported on the mailing list. (check-in: a7126a4f4f user: drh tags: trunk) | |
16:43 | Fix the SQLITE_DESERIALIZE_READONLY feature so that it does not cause an assertion fault in the pager. (check-in: b9eccef782 user: drh tags: trunk) | |
16:11 | Update dbfuzz2 to set a maximum database size of 100MiB by default, but with the new --max-db-size N option to change that limit. (check-in: 21d6bb78ef user: drh tags: trunk) | |
16:06 | Enhancements to deserialize: (1) Add the SQLITE_FCNTL_SIZE_LIMIT file control to set a maximum size for an in-memory database, defaulting to SQLITE_MEMDB_DEFAULT_MAXSIZE or 1GiB. (2) Honor the SQLITE_DESERIALIZE_READONLY flag. (3) Enhance the TCL interface to support -maxsize N and -readonly BOOLEAN. (4) Add the --maxsize option to the ".open" command and on the command-line for the CLI. (check-in: 30f08d5888 user: drh tags: trunk) | |
13:45 | Make sure cursors are opened on all indexes for an UPDATE OR REPLACE regardless of whether or not the indexes are partial or contain columns that might need to be updated. (check-in: e148cdad35 user: drh tags: trunk) | |
12:21 | Fix another segfault caused by a corrupt fts3 database. (check-in: ba3b841272 user: dan tags: trunk) | |
02:34 | Ensure that a key comparison does not read a collating sequence past the end of the KeyInfo, even if the key field of an index is corrupted by having a string in the last column instead of the ROWID. (check-in: 058a8006dc user: drh tags: trunk) | |
2019-01-21
| ||
23:18 | Enhance the btree search routine so that it does early detection of impossibly large keys and thereby avoids a large malloc() call. (check-in: 3ecaaee69f user: drh tags: trunk) | |
17:57 | Avoid deferencing a freed pointer following an OOM or SQLITE_CORRUPT error in the fts3 xDestroy method. (check-in: 505ed9a478 user: dan tags: trunk) | |
16:12 | Remove a faulty assert() from fts3. (check-in: 6c33a303eb user: dan tags: trunk) | |
16:01 | Make sure the extra bytes allocated for the saved cursor position by check-in [160b1e31c0f27257] are initialized to zero. (check-in: 2737564929 user: drh tags: trunk) | |
14:55 | Add the --memtrace option to the CLI. (check-in: a1e12fa2a8 user: drh tags: trunk) | |
14:49 | Minor fix the fallocate.test module change from [7cd56cad5efead5] (check-in: 94fb7a4700 user: drh tags: trunk) | |
13:47 | Add the --max-data and --max-as options to dbfuzz2. Also cause dbfuzz2 to show its maximum RSS size upon exit in standalone mode with the -v option. (check-in: 7ce93e824a user: drh tags: trunk) | |
2019-01-20
| ||
00:03 | Add the --max-stack option to dbfuzz2. (check-in: c11ae4fed8 user: drh tags: trunk) | |
2019-01-19
| ||
15:55 | Avoid a division-by-zero error in fts5 caused by a corrupt database. (check-in: 928e622178 user: dan tags: trunk) | |
15:27 | Remove a broken assert() triggered by a "PRAGMA max_page_count = N" invocation, where N is larger than the number of pages in the database image, but smaller than the number of pages in the database file. (check-in: 7cd56cad5e user: dan tags: trunk) | |
14:07 | Fix a problem with using ALTER TABLE to rename a table or column when the database schema contains a trigger or view that itself contains an expression "<column> AND 0". (check-in: 908ff7fffa user: dan tags: trunk) | |
2019-01-18
| ||
21:17 | Fix a memory leak introduced by [55c5d72a]. (check-in: fbd681dce2 user: dan tags: trunk) | |
21:12 | Fix a crash in the fts5vocab module caused by including a "term < NULL" term in a WHERE clause. (check-in: 9e717c4377 user: dan tags: trunk) | |
21:03 | Fix an infinite loop caused by a corrupt database in fts3. Also an undefined left-shift in fts5. (check-in: 55c5d72af9 user: dan tags: trunk) | |
20:15 | Avoid passing a NULL pointer to memcpy in fts5, even if the database is corrupt. (check-in: acccc9808f user: dan tags: trunk) | |
19:33 | Typecasts added to the GeoPoly extension to avoid harmless UBSAN warnings. (check-in: a1f6a093ac user: drh tags: trunk) | |
19:26 | Fix problems causing undefined left-shift operations in the fts3 snippet() function. (check-in: b90dbaed30 user: dan tags: trunk) | |
18:52 | Avoid integer overflow when computing the array of a bounding box with the rtree_i32 virtual table. (check-in: b352f1590d user: drh tags: trunk) | |
17:53 | Handle oversize floating point values carefully when converting to integers for the '%' binary operator. (check-in: 048add13fc user: drh tags: trunk) | |
16:06 | Fix a fairly obscure problem allowing an "ALTER TABLE RENAME col TO ..." statement to modify the schema in such a way as to break a reference within a trigger program. (check-in: 64bec9e621 user: dan tags: trunk) | |
14:53 | Use the full 64-bit integer value in the argument to randomblob(). (check-in: 05df5f7aea user: drh tags: trunk) | |
2019-01-17
| ||
20:19 | Disable the xfer-optimization if the two tables have the same root page due to schema corruption. (check-in: f31b3bd2a6 user: drh tags: trunk) | |
20:06 | Fix another corruption related crash in fts5. (check-in: 4538d9afe4 user: dan tags: trunk) | |
19:33 | Add the SQLITE_ENABLE_EARLY_CURSOR_CLOSE compile-time option which causes read cursors to be closed after their usefulness ends during a two-pass UPDATE. (check-in: 7def6c8edd user: drh tags: trunk) | |
19:11 | Fix a buffer overwrite that could occur when running an fts5 prefix query against a corrupt database. (check-in: 3910b5639d user: dan tags: trunk) | |
17:39 | Fix problems with joining two or more fts5_vocab tables that access the same underlying fts5 table. (check-in: 49956395e1 user: dan tags: trunk) | |
15:47 | Sync up with the latest enhancements on trunk. (check-in: 475a179a27 user: drh tags: apple-osx) | |
15:40 | Revamp the SrcList allocator routines to be methods of Parse instead of being methods of the "sqlite3" object, so that they can leave better error messages when the SrcList object grows too large. (check-in: df08d472b0 user: drh tags: trunk) | |
14:34 | Limit the size of SrcList objects to 200 entries (compile-time configurable using -DSQLITE_MAX_SRCLIST=n). The maximum number of tables in a join has always been 64, so this is not a real constraint on capability. Limiting the size of a SrcList prevents DOS attacks (discovered by OSSFuzz) using crazy nexted CTE joins. (check-in: 7cac614d5d user: drh tags: trunk) | |
04:40 | Fix a corner-case for the logic that cause an insert of a NULL into an INTEGER PRIMARY KEY column to be converted into a valid integer key, when the NULL results from a CASE expression that lacks an ELSE clause. (check-in: 9a425051e7 user: drh tags: trunk) | |
03:43 | Ensure that the variable-length integer decoder in FTS3 never tries to left-shift a negative number. (check-in: 10ffc1fe00 user: drh tags: trunk) | |
01:06 | Fix a failure in sqlite3ExprCompare() (check-in: 835e2cc55f user: drh tags: trunk) | |
2019-01-16
| ||
20:48 | Fix a problem with fix [b4b57413]. (check-in: ca7b7aaed0 user: dan tags: trunk) | |
19:44 | Fix a problem in the fts3 matchinfo() function with corrupt database handling. (check-in: 24ed5fb6ae user: dan tags: trunk) | |
19:26 | Fix a problem in the code generator for sorting results with SRT_EphemTab and a LIMIT clause. (check-in: 49fcde2f1f user: drh tags: trunk) | |
14:58 | Fix a problem with renaming a column that is used as part of an ORDER BY on a compound SELECT within a database view or trigger. (check-in: b4b5741366 user: dan tags: trunk) | |
12:05 | Avoid a dangling pointer comparison when renaming a table that has a trigger that itself contains a window function with an (illegal) column reference in a FOLLOWING expression. (check-in: d45bee36f2 user: dan tags: trunk) | |
11:38 | Fix a memory leak that could occur in fts3 when handling a corrupt database. (check-in: 65cebb06a0 user: dan tags: trunk) | |
2019-01-15
| ||
20:51 | Fix a problem with ALTER TABLE and vector assignments in UPDATE statements within triggers. (check-in: cc6cd7531f user: dan tags: trunk) | |
18:14 | Handle SQL NULL values without crashing in the fts5 snippet() and highlight() functions. (check-in: a5e9cc794f user: dan tags: trunk) | |
16:14 | Fix a buffer overread in fts3 caused by a corrupt record. (check-in: e54efd60c2 user: dan tags: trunk) | |
15:18 | Have fts5 reject attempts to insert a non-integer, non-null value into a rowid column with SQLITE_MISMATCH. (check-in: 4a9483f81e user: dan tags: trunk) | |
14:44 | Fix a harmless memory leak in the Lemon parser generator utility program. (check-in: 1caff0fb0b user: drh tags: trunk) | |
14:31 | Fix a problem triggered by DELETE statements with WHERE clauses that use the OR-optimization on some virtual tables. (check-in: ecf5caa7e9 user: dan tags: trunk) | |
2019-01-14
| ||
20:44 | Add scalar function "prefix_length()" to ext/misc/prefixes.c. (check-in: 0bf1550507 user: dan tags: trunk) | |
19:13 | Fix a problem causing some Tcl test cases to fail with errors like "expected: [<multiline-whitespace>], got: []". (check-in: 11b8a4cba7 user: dan tags: trunk) | |
16:16 | Add the "prefixes" table-valued function in the ext/misc folder. (check-in: f0fd2163fc user: drh tags: trunk) | |
15:35 | Fix a problem causing a crash if an fts5vocab table was created to query an fts3/4 FTS index. (check-in: 9cd64ce453 user: dan tags: trunk) | |
15:09 | Fix a problem with querying a corrupt fts3 database. (check-in: 65e50706a0 user: dan tags: trunk) | |
13:32 | When parsing a corrupt record using sqlite3VdbeRecordUnpack(), make sure that the resulting UnpackedRecord object is completely initialized, to avoid (harmless) MSAN warnings. (check-in: ddc3697efd user: drh tags: trunk) | |
11:56 | Have fts3 ignore empty sets of parenthesis if built with SQLITE_ENABLE_FTS3_PARENTHESIS. (check-in: c93c6b45a3 user: dan tags: trunk) | |
05:48 | Avoid reading off the front of a page buffer when balancing a corrupt btree page. (check-in: cb50509020 user: drh tags: trunk) | |
2019-01-13
| ||
20:23 | In dbfuzz2, avoid using a malloc in the LLVMFuzzerInitialize() initializer routine, so that no memory leaks are reported. Also, show the version of SQLite being used when the -v option is on. (check-in: 824f932469 user: drh tags: trunk) | |
20:17 | Relax the minimum size database file constraint on the dbtotxt utility program. (check-in: 97e723d746 user: drh tags: trunk) | |
20:17 | Improved detection of database corruption while balancing pages from an auto_vacuum database with overflow pages. Test cases in TH3. (check-in: 35f04235c4 user: drh tags: trunk) | |
00:58 | Move a local variable declaration into the outermost scope in which it is used. This fixes an ASAN warning. (check-in: ac3b6021d9 user: drh tags: trunk) | |
2019-01-12
| ||
21:30 | Prevent unsigned 32-bit integer overflow from leading to a buffer overread inside of an assert(). The problem fixed here is no reachable in production code. (check-in: 0f850a25d6 user: drh tags: trunk) | |
20:55 | Fix another problem with handling corrupt records in fts5_decode(). (check-in: 726e398b9d user: dan tags: trunk) | |
16:19 | Fix an off-by-one error on a Goto in the code generator, that only causes problems for a REPLACE on an INTEGER PRIMARY KEY in non-debug builds. Test case in TH3. (check-in: e35eb8776e user: drh tags: trunk) | |
14:58 | Fix a problem with corrupt database handling in the fts3 matchinfo() function. (check-in: 703646b1b5 user: dan tags: trunk) | |
00:45 | Improved detection of shadow table corruption in the fts5_decode() SQL function. (check-in: b74e5f3f30 user: drh tags: trunk) | |
00:12 | Improved shadow table corruption detection in the matchinfo() function of FTS3. (check-in: 567be3bb1e user: drh tags: trunk) | |
00:07 | Indicate that the database may be corrupt in the fts3corrupt4.test test script. (check-in: 473626d557 user: drh tags: trunk) | |
2019-01-11
| ||
23:08 | Use 64-bit APIs in the fileio.c extension. (check-in: b49d56a0fa user: drh tags: trunk) | |
21:34 | Fix a segfault in fts3 prompted by a corrupted database. (check-in: 2d7b1d1d41 user: dan tags: trunk) | |
19:27 | Fix another problem with corrupt databases in fts5. (check-in: 9bf811ea89 user: dan tags: trunk) | |
17:41 | Fix a problem handling a negative value in the "number-of-pages" database header field. Also a problem with running "REINDEX tbl" against a virtual table for which the SQL passed to sqlite3_declare_vtab() contains PRIMARY KEY or UNIQUE constraints. (check-in: 556dd8922f user: dan tags: trunk) | |
17:20 | Improved detection and reporting of errors in the readfile() extension SQL function. (check-in: d2f0b5a483 user: drh tags: trunk) | |
17:19 | Omit errors about missing SAVEPOINTs when aborting the .archive command in the CLI. (check-in: 2a47387ba6 user: drh tags: trunk) | |
16:44 | Improved detection of cell corruption in sqlite3VdbeRecordCompareWithSkip(). (check-in: fa47f4c658 user: drh tags: trunk) | |
15:26 | Merge documentation changes from branch begin-concurrent-wal2 into this branch. (check-in: cf8a0c71cf user: dan tags: begin-concurrent-pnu-wal2) | |
15:22 | Merge documentation changes from branch "begin-concurrent" into this branch. (Closed-Leaf check-in: 41e742bd0f user: dan tags: begin-concurrent-wal2) | |
15:06 | Add new documentation file begin_concurrent.md. (check-in: fdbf97e611 user: dan tags: begin-concurrent) | |
15:00 | Merge latest begin-concurrent-wal2 changes (documentation only). (check-in: d0ab56a3be user: dan tags: begin-concurrent-pnu-wal2) | |
14:59 | Merge latest wal2 changes (documentation only) into this branch. (check-in: 820ba1cc7e user: dan tags: begin-concurrent-wal2) | |
14:58 | Add new file doc/wal2.md to this branch. (check-in: b495dce153 user: dan tags: wal2) | |
14:46 | Fix PRAGMA integrity_check so that it does not cancel the PRAGMA vdbe_debug setting. (check-in: aaa3a19f8c user: drh tags: trunk) | |
14:38 | Add the --vdbe-debug command-line option to dbfuzz2. (check-in: 599b4df43c user: drh tags: trunk) | |
14:22 | Add the "dbfuzz2" target to main.mk. Remove an unused local variable from dbfuzz2.c. (check-in: 05c7609cfd user: drh tags: trunk) | |
13:32 | The keywords TRUE and FALSE should only act as boolean literal values if unquoted. (check-in: 5547f39de9 user: drh tags: trunk) | |
13:03 | Separate makefile targets for "dbfuzz2", "dbfuzz2-asan", and "dbfuzz2-msan". (check-in: ea119641a7 user: drh tags: trunk) | |
2019-01-10
| ||
19:26 | Fix a memory leak in fts5 that could occur if scalar function fts5_decode() was passed a corrupt record. (check-in: 240f1c0c92 user: dan tags: trunk) | |
18:35 | Fix a memory leak in fts5. (check-in: ff3b011f17 user: dan tags: trunk) | |
17:08 | Avoid use-after-free and double-free errors that could occur if an fts5 table is modified in certain ways while there are active cursors. (check-in: 3291b2a6fe user: dan tags: trunk) | |
15:17 | Fix further problems with fts5 handling corrupt databases. (check-in: 83c467d7af user: dan tags: trunk) | |
14:33 | More aggressive early detection of orphaned and malformed autoindexes when parsing the schema. (check-in: 10f9e39d6e user: drh tags: trunk) | |
13:56 | Use the new SQLITE_IDXTYPE_IPK values (3) on Index.idxType to indicate the fake INTEGER PRIMARY KEY index used during query planning. (check-in: e22d2f905f user: drh tags: trunk) | |
01:12 | Fix the fts3DecodeIntArray() function so that it will not read off the end of the buffer it is handed. Any unread integers are set to zero. (check-in: 666cf8f6b3 user: drh tags: trunk) | |
2019-01-09
| ||
21:12 | Fix an out-of-bounds read in SQL function fts5_decode() that could occur if it was passed a corrupt record. (check-in: 931278b257 user: dan tags: trunk) | |
14:49 | Adjustments to the page cache to try to avoid harmless TSAN warnings. (check-in: 383437be27 user: drh tags: trunk) | |
11:19 | Improved database corruption detection in the dbstat virtual table. (check-in: fc5ecc88f8 user: drh tags: trunk) | |
11:06 | Small simplification to the dbstat virtual table. (check-in: 9d4c156fa2 user: drh tags: trunk) | |
02:02 | Enhance the ALTER TABLE RENAME COLUMN feature so that it works on tables that have redundant UNIQUE and/or PRIMARY KEY constraints. Fix for ticket [bc8d94f0fbd633fd9a051e3] (check-in: f09aa3248e user: drh tags: trunk) | |
2019-01-08
| ||
20:02 | Use 64-bit math to compute the sizes of memory allocations in extensions. (check-in: ca67f2ec0e user: drh tags: trunk) | |
15:18 | Performance improvement on the instr() function, especially for large haystacks. (check-in: ce51f1a2b6 user: drh tags: trunk) | |
14:28 | Fix a possible memory leak when trying to UPDATE a corrupt RTREE index. (check-in: 63eb803dbc user: drh tags: trunk) | |
08:02 | Fix further problems with fts5 and corrupt databases causing integer overflow. (check-in: 673a7dd698 user: dan tags: trunk) | |
2019-01-07
| ||
16:52 | Fix another potential buffer overread in fts5. (check-in: d0e943d957 user: dan tags: trunk) | |
15:57 | For SQLITE_ENABLE_PREUPDATE_HOOK builds, disable the optimization for a REPLACE command on a WITHOUT ROWID table with no indexes, triggers or foreign key constraints that causes SQLite to clobber any existing row without separately checking for it. This optimization causes SQLite to omit the expected pre-update-hook callbacks. (check-in: 6281ef974c user: dan tags: trunk) | |
2019-01-06
| ||
02:06 | The page size becomes fixed as soon as page1 from a non-empty database file is read. (check-in: 6064584d57 user: drh tags: trunk) | |
01:46 | Give symbolic names to the special prepared statements used to implement incremental blob I/O. (Leaf check-in: 2a1cc6327a user: drh tags: named-blob-I/O) | |
2019-01-05
| ||
21:56 | Add the exprNodeCopy() routine that will safely memcpy() an Expr node that might be a size-reduced node. (check-in: a874c64996 user: drh tags: trunk) | |
21:09 | If the OP_ParseSchema opcode with a non-NULL P4 operand does not parse any rows out of the sqlite_master table, that indicates that the sqlite_master table is corrupt, so raise an SQLITE_CORRUPT error. (check-in: 598d7358e7 user: drh tags: trunk) | |
07:17 | Fix two more problems with corrupt database handling in fts5. (check-in: 444c7c99be user: dan tags: trunk) | |
2019-01-04
| ||
23:39 | Give a meaningful error on an attempt to use UPSERT on a virtual table, rather than simply not working. (check-in: fa98748f95 user: drh tags: trunk) | |
19:39 | Fix the experimental sqlite3_wal_info() API so that it works with wal2 mode. (check-in: 079daedbe4 user: dan tags: begin-concurrent-pnu-wal2) | |
19:25 | Add the "-external" option to the tserver_test.tcl program.. (check-in: ae7c5ed15d user: dan tags: begin-concurrent-pnu-wal2) | |
17:12 | Update the "tserver" program to report the amount of time spent in COMMIT commands. (check-in: cf494b412c user: dan tags: begin-concurrent-pnu-wal2) | |
14:29 | If the ".open" command fails in the CLI, reopen to :memory: so that at least some database is available for subsequent commands. (check-in: d7f55c62c3 user: drh tags: trunk) | |
12:46 | Remove debugging pragmas accidently left in by a prior change. (check-in: 26275fc7e8 user: drh tags: trunk) | |
12:35 | Fix a typo in a comment used to generate documentation. No functional changes. (check-in: ccc4f7424b user: drh tags: trunk) | |
11:20 | Fix another fts5 assert() that may fail if the database is corrupt. (check-in: 0888fc2e88 user: dan tags: trunk) | |
2019-01-03
| ||
19:12 | Fix some problems with sqlite3_interrupt() handling in fts5. (check-in: c0eb839a3c user: dan tags: trunk) | |
16:03 | Fix the csv01.test module so that it works on systems with \r\n line endings. (check-in: 36eaf5a5e5 user: drh tags: trunk) | |
16:00 | Fix the shmlock.test script so that it works with the Windows restriction that UnlockFile must exactly correspond to a prior LockFile. (check-in: df939c89fa user: drh tags: trunk) | |
15:22 | Fix a harmless compiler warning in the CSV extension. (check-in: fbcd72565f user: drh tags: trunk) | |
15:17 | Fix a problem with the second and subsequent sqlite3rbu_savestate() calls made on an RBU vacuum handle. (check-in: bef216dfa1 user: dan tags: trunk) | |
00:44 | Fix the geopoly_svg() function so that it returns NULL when given zero arguments. (check-in: 120cb17683 user: drh tags: trunk) | |
2019-01-02
| ||
23:49 | Fix harmless compiler warnings in the unicode2 logic of FTS3 and FTS5. (check-in: 703029ac6d user: drh tags: trunk) | |
19:46 | Merge changes from begin-concurrent-wal2. Also modify the unix-excl shm-locks on this branch so that it is not an error to attempt to release an EXCLUSIVE lock that is not held. (check-in: 86685679b4 user: dan tags: begin-concurrent-pnu-wal2) | |
17:08 | Reinstate assert() statements in os_unix.c that were removed to allow wal-mode SHARED locks to be taken over more than one locking slot (this branch no longer does that, so the assert() statements can go back in). (check-in: fcb7348ba1 user: dan tags: begin-concurrent-wal2) | |
17:00 | Reinstate assert() statements in os_unix.c that were removed to allow wal-mode SHARED locks to be taken over more than one locking slot (this branch no longer does that, so the assert() statements can go back in). (check-in: 8445fb6d5d user: dan tags: wal2) | |
16:30 | Merge latest wal2 changes into this branch. (check-in: c31eeb3c87 user: dan tags: begin-concurrent-wal2) | |
16:15 | Merge latest trunk changes into this branch. (check-in: 87ef68f917 user: dan tags: wal2) | |
16:08 | Merge latest trunk changes into this branch. (check-in: 5bf212f1a7 user: dan tags: begin-concurrent) | |
16:01 | Add another test case to vacuum-into.test. (check-in: 0465d2fc0d user: dan tags: trunk) | |
15:27 | Add test cases for commit [6c6fb1c6]. (check-in: 3961995cb5 user: dan tags: trunk) | |
14:34 | Raise an error if the argument to VACUUM INTO is an identifier. (check-in: 6c6fb1c6ea user: drh tags: trunk) | |
2019-01-01
| ||
19:17 | Fix another fts5 problem caused by a corrupt database. (check-in: 911342f751 user: dan tags: trunk) | |
18:00 | Ensure that when a new cursor is opened by OP_OpenDup, any existing cursor with the same id opened by a previous OP_OpenDup is closed first. (check-in: 5c188361a9 user: dan tags: trunk) | |
13:59 | Fix another case in fts5 where a corrupt database could cause a buffer overread. (check-in: f7e6cdc562 user: dan tags: trunk) | |
2018-12-31
| ||
21:43 | Fix harmless compiler warnings. (check-in: b57c545a38 user: drh tags: trunk) | |
20:39 | Additional steps to help ensure that scalar subqueries are only evaluated once even if they are used in multiple places within the query. This fixes a performance regression reported on the mailing list and caused by check-in [531eca6104e41e43] which was a fix for ticket [787fa716be3a7f650c]. Think of this check-in as an improved fix for that ticket. (check-in: e130319317 user: drh tags: trunk) | |
20:13 | Remove an optimization that can no longer occur, being superceded by the subquery-reuse optimization. Put an assert in place of the optimization to detect if the need for this optimization ever returns. (Closed-Leaf check-in: 4fcdc7a293 user: drh tags: reuse-subqueries) | |
18:30 | Merge enhancements and bug fixes from trunk. (check-in: 9fb646f29c user: drh tags: reuse-subqueries) | |
17:58 | Small changes to the OP_OpenEphemeral opcode to improve testability. (check-in: f856676c84 user: drh tags: trunk) | |
16:36 | Fix the OP_OpenEphemeral opcode in the bytecode engine so that if it is called a second or subsequent time, it merely clears the existing table rather than creating a new one. Proposed fix for ticket [d0866b26f83e9c55e30de0821f5d]. (check-in: 4678cb1044 user: drh tags: trunk) | |
2018-12-29
| ||
20:47 | Merge latest begin-concurrent changes with this branch. (check-in: 1625887c02 user: dan tags: begin-concurrent-wal2) | |
20:42 | Fix a problem causing a corrupt pager-cache if an OOM or IO error was encountered while committing a concurrent transacation. (check-in: 48ca30f9d7 user: dan tags: begin-concurrent) | |
16:34 | Increase coverage provided by permutation "coverage-wal" on this branch. (check-in: f664f940a7 user: dan tags: begin-concurrent-wal2) | |
14:23 | Additional small performance increase and size reduction to the sqlite3VdbeMakeLabel() mechanism. (check-in: 1bdee199a7 user: drh tags: trunk) | |
02:26 | A new implementation of sqlite3VdbeMakeLabel() is faster and makes fewer memory allocations by deferring memory allocation until sqlite3VdbeResolveLabel() is called, at which point the code generator has a better idea of how big the relocation table needs to be. The sqlite3VdbeMakeLabel() routine now takes a Parse* parameter instead of Vdbe*. (check-in: 4a0929ac76 user: drh tags: trunk) | |
2018-12-28
| ||
21:40 | Merge performance enhancements and other patches. (Leaf check-in: 44e1b55aab user: drh tags: factor-out-scalar-subselect) | |
21:32 | Merge performance enhancements from trunk. (check-in: 0f1b9ff9e1 user: drh tags: reuse-subqueries) | |
20:48 | Faster allocation of new sqlite3_stmt objects. (check-in: 891f1f7218 user: drh tags: trunk) | |
20:14 | Move the nOpAlloc field from Parse into Vdbe to avoid an extra pointer deference on the fast path in sqlite3VdbeAddOp3(). (check-in: 8f10efc29d user: drh tags: trunk) | |
19:23 | Merge patches and enhancements from trunk. (check-in: da53f3d3ff user: drh tags: reuse-subqueries) | |
18:09 | Fix another problem with loading the structure record from a corrupt fts5 database. (check-in: c4d44542d2 user: dan tags: trunk) | |
17:45 | Fix a problem causing the sqlite_master entry corresponding to a virtual table to be removed by a DROP TABLE even if the call to the vtabs xDestroy() method failed. (check-in: 0140f6dbfb user: dan tags: trunk) | |
16:20 | Add new test file wal2rollback.test to this branch. (check-in: 85a376fc6c user: dan tags: wal2) | |
14:33 | Avoid an undefined left-shift operation in fts5 caused by malformed utf-8 text. (check-in: c3a3a11194 user: dan tags: trunk) | |
13:57 | Fix a buffer overwrite in fts5 triggered by a corrupt database. (check-in: a385298df2 user: dan tags: trunk) | |
07:37 | Fix problems in fts5 found by ASAN. (check-in: c564bf8701 user: dan tags: trunk) | |
2018-12-27
| ||
20:12 | Fix another problem with corrupt database handling in fts5. (check-in: fb0d7fba07 user: dan tags: trunk) | |
17:11 | Merge latest wal2 changes with this branch. (check-in: ea96001e80 user: dan tags: begin-concurrent-wal2) | |
16:55 | Modify the btree02.test module so that it does not depend on whether TCL interprets a value as a string or as an integer. (check-in: d9e4f95746 user: drh tags: trunk) | |
16:49 | Increase test coverage of wal.c provided by permutation "coverage-wal" on this branch. (check-in: 2f7f893a70 user: dan tags: wal2) | |
15:36 | Fix another problem in fts5 caused by a corrupt schema record. (check-in: ff45f9dd69 user: dan tags: trunk) | |
13:52 | Fix the filename normalization in the dbtotxt utility program. (check-in: 29a00a05dd user: drh tags: trunk) | |
07:46 | Fix a null-pointer dereference that could occur in fts5 with a corrupt database. (check-in: b0a49d5e6d user: dan tags: trunk) | |
02:16 | Fix a problem in which nested CTEs with the same table name trick the code generator into thinking they are the same CTE, which then tries to use the manifest them both into the same transient table. (check-in: 202dd03301 user: drh tags: trunk) | |
00:30 | In the treeview debugging output, provide additional details for the Table object associated with each FROM clause term. (check-in: 11d4682d2e user: drh tags: trunk) | |
2018-12-26
| ||
20:42 | Improve testing of the wal.c changes on this branch. (check-in: 63483e22c7 user: dan tags: wal2) | |
19:10 | Merge latest trunk changes with this branch. (check-in: 404f9d99d3 user: dan tags: wal2) | |
18:34 | Update the coverage-wal permutation to cover branches enabled by SQLITE_ENABLE_SNAPSHOT. (check-in: 6821c61f1d user: dan tags: trunk) | |
17:49 | Add a few more test cases to walvfs.test. (check-in: 14a64df36e user: dan tags: trunk) | |
16:57 | Avoid an "uninitialized value read" problem in fts5 that can occur with a corrupt database. (check-in: 6cae81cd1f user: dan tags: trunk) | |
16:04 | Fix a problem with handling corrupt databases in the fts5 integrity-check function. (check-in: 9bad745d37 user: dan tags: trunk) | |
15:04 | Here is the beginnings of an effort to factor out the implementation of scalar subqueries to be implemented as subroutines at the end of the main body of bytecode, after the jump back to the OP_Init opcode. This is an incremental check-in that contains only preliminary changes. (check-in: 50e5f390fb user: drh tags: factor-out-scalar-subselect) | |
14:15 | Fix a problem with corrupt fts3 database detection introduced by [95a9a39f]. (check-in: c80bb15b44 user: dan tags: trunk) | |
12:50 | Fix a problem with the CSV extension when it uses the "header" option on a real file. (check-in: 7acaed08f9 user: drh tags: trunk) | |
2018-12-25
| ||
11:17 | Fix a typo in a comment that is used to generate session documentation. (check-in: a1f50d57ce user: drh tags: trunk) | |
00:15 | Increase the number of parameters to sqlite3WhereCodeOneLoopStart() to encourage the compiler to inline that routine within sqlite3WhereBegin(). This gives a performance improvement. (check-in: 3c2ad0e915 user: drh tags: reuse-subqueries) | |
2018-12-24
| ||
20:21 | Improvements to EXPLAIN QUERY PLAN formatting. The MULTI-INDEX OR now shows a separate "INDEX" subtree for each index. SCALAR SUBQUERY entries provide a subquery number that is related back to the .selecttrace output. (check-in: 7153552bac user: drh tags: reuse-subqueries) | |
20:00 | Add a few more test cases to walfault2.test and walvfs.test. (check-in: 3873941c4f user: dan tags: trunk) | |
19:59 | Change another fts5 assert() that may fail with a corrupt database to an assert_nc(). (check-in: d9653380c4 user: dan tags: trunk) | |
18:51 | Add test cases to test file walvfs.test. (check-in: d697f6f3ca user: dan tags: trunk) | |
18:10 | Always include OP_Explain opcodes in the bytecode when SQLITE_DEBUG is enabled. (check-in: d8c1235c02 user: drh tags: trunk) | |
15:22 | Copy some extra test infrastructure from the mutexfree-shmlock branch to trunk. (check-in: 883337ffdb user: dan tags: trunk) | |
15:15 | Merge latest trunk with this branch. (Leaf check-in: 606b1ead24 user: dan tags: mutexfree-shmlock) | |
14:30 | Restore a line of code that was previously commented out for debugging. (check-in: e53781f5bd user: drh tags: reuse-subqueries) | |
13:34 | Change the way a comparison used to detect corrupt databases in fts3 is done to avoid potential pointer overflow in 32-bit builds. (check-in: 95a9a39ff7 user: dan tags: trunk) | |
12:09 | Do not generate subroutines for non-static SELECT and EXISTS expressions. Fix up some test cases to account for the minor changes in EXPLAIN QUERY PLAN output. (check-in: 06de44ec9e user: drh tags: reuse-subqueries) | |
11:55 | Prevent the use of subroutines to implement the RHS of IN operators within CHECK constraints. (check-in: 6b24d3fb94 user: drh tags: reuse-subqueries) | |
02:34 | Experimental code that tries to put the computation of subqueries inside a subroutine, and reuse that subroutine if the same subquery is evaluated more than once. Current code does not work for CHECK constraints. (check-in: 6c44838adb user: drh tags: reuse-subqueries) | |
2018-12-23
| ||
21:27 | Split the code generation for the RHS of IN operators and for SELECT and EXISTS expressions into two separate subroutines, because there is now little commonality between those to functions. This is intended to help make the code easier to read and maintain. (check-in: 2b6494b150 user: drh tags: trunk) | |
2018-12-22
| ||
20:32 | Improve the coverage of wal.c provided by the "coverage-wal" test permutation. (check-in: 6231485114 user: dan tags: trunk) | |
19:54 | Catch another case of a corrupted structure record in fts5. (check-in: fb17fea4b9 user: dan tags: trunk) | |
19:09 | Change a faulty assert() in fts5 that could fail on a corrupt db to an if() statement. (check-in: 1daaba9fb4 user: dan tags: trunk) | |
16:14 | Fix one more instance of excess whitespace in test result strings. (check-in: c064964deb user: drh tags: trunk) | |
15:46 | Remove unnecessary whitespace from testcase result strings. (check-in: ad70f03ced user: drh tags: trunk) | |
15:21 | Fix an assert() in fts5 that could fail with a corrupt database. (check-in: 83935b9506 user: dan tags: trunk) | |
13:34 | Fix a "jump or move depends on uninitialized value" valgrind error that could occur in fts5 when dealing with a corrupt database. (check-in: 3518c09b94 user: dan tags: trunk) | |
09:39 | Avoid a left-shift of a negative value (undefined behaviour) when dealing with a corrupt database in fts3. (check-in: b851d12474 user: dan tags: trunk) | |
07:16 | Fix cut-and-paste error in test file fts4umlaut.test. (check-in: cad5da1bf5 user: dan tags: trunk) | |
01:13 | Do not all REPLACE to sneak a NULL value into a NOT NULL column. Detect that situation and ABORT instead. Fix for ticket [e6f1f2e34dceeb1ed61531c7e98]. (check-in: db8d1d12f5 user: drh tags: trunk) | |
00:34 | The OP_Eq and OP_Ne operators have a special P5 value SQLITE_NOTNULL that asserts that the values are not null. Except that is not always true for a corrupt database. Adjust the assert() and add a testcase() to make this point clear. (check-in: a3fdb2c78d user: drh tags: trunk) | |
2018-12-21
| ||
22:11 | Update the dbtotxt utility program so that it does not output characters that are special to TCL, thus making the output of dbtotxt easier to insert into test scripts. (check-in: 48438bb35b user: drh tags: trunk) | |
22:08 | Improved detection of shadow table corruption in RTREE. (check-in: b39bf4356e user: drh tags: trunk) | |
20:18 | Add new sqlite3_prepare_v3() flag SQLITE_PREPARE_NO_VTAB, for preparing statements that are not allowed to use any virtual tables. Use this to prevent circular references in triggers on virtual table shadow tables from causing resource leaks. (check-in: da587d1857 user: dan tags: trunk) | |
19:55 | Use SQLITE_PREPARE_NO_VTAB in rtree as well. (Closed-Leaf check-in: 82a2ae7132 user: dan tags: prepare-no-vtab) | |
19:30 | Add tests for the use of SQLITE_PREPARE_NO_VTAB in fts5. (check-in: 1999612010 user: dan tags: prepare-no-vtab) | |
19:29 | Add new sqlite3_prepare_v3() flag SQLITE_PREPARE_NO_VTAB, for preparing statements that are not allowed to use any virtual tables. Use this to prevent circular references in triggers on virtual table shadow tables from causing resource leaks. (check-in: 25666e3d03 user: dan tags: prepare-no-vtab) | |
18:51 | In FTS3, avoid calling memcpy() with a NULL source pointer, even if the transfer amount is zero bytes. (check-in: 1abb83d29a user: drh tags: trunk) | |
18:50 | The fts4umlaut.test module requires FTS5, so do not try to run it without that module. (check-in: dee3ae9001 user: drh tags: trunk) | |
18:17 | Fix the node hash function in RTREE to ensure that it always returns a non-negative value. (check-in: 0bf4c70a68 user: drh tags: trunk) | |
17:51 | Fix a potential NULL-pointer deference in RTREE due to corrupt shadow tables. (check-in: 1fdd3604ee user: drh tags: trunk) | |
16:53 | Prevent an infinite loop in rtree that can result from a corrupt shadow table. (check-in: 085667180b user: drh tags: trunk) | |
15:13 | Fix the RTree extension so that it correctly ignores constraints that it does not understand, even if they are against a dimension column. (check-in: ed8531e575 user: drh tags: trunk) | |
13:30 | In RTREE, turn an assert() into an if() because the condition can sometimes be false when dealing with a corrupt database. (check-in: 99bce4358b user: drh tags: trunk) | |
11:53 | Remove an assert() that could fail on a corrupt db from fts3_write.c. (check-in: 4cdcda408a user: dan tags: trunk) | |
2018-12-20
| ||
22:08 | Do not set the Cursor.seekOp test variable in the OP_NotExists opcode. This is a change to test logic only and does not affect deliverable builds. (check-in: 98f3430778 user: drh tags: trunk) | |
17:32 | Fix a bug in the code that detects self-referencing rows as part of foreign key processing. Fix for [50d2a6c2]. (check-in: 16fff05347 user: dan tags: trunk) | |
15:04 | Fix a segfault caused by using the RAISE function incorrectly (library now returns an error instead of crashing). (check-in: ddf06db702 user: dan tags: trunk) | |
2018-12-18
| ||
20:31 | In defensive mode, do not allow shadow tables to be renamed using ALTER TABLE. (check-in: 23e200da5c user: dan tags: trunk) | |
19:46 | Add the "mutexfree-shmlock" patch to this branch. (check-in: c3191fc8fc user: dan tags: begin-concurrent-pnu-wal2) | |
18:16 | Fix a test script problem in wal2savepoint.test. (check-in: e388d02940 user: dan tags: begin-concurrent-pnu-wal2) | |
18:01 | Merge the wal2 feature into this branch. (check-in: a2c618c0a4 user: dan tags: begin-concurrent-pnu-wal2) | |
17:47 | Merge latest begin-concurrent changes into this branch. (check-in: a93ca38b43 user: dan tags: begin-concurrent-pnu) | |
17:32 | Merge latest trunk changes into this branch. (check-in: b3a163b46c user: dan tags: begin-concurrent-wal2) | |
17:20 | Merge latest trunk changes into this branch. (check-in: 123cbb3312 user: dan tags: begin-concurrent) | |
2018-12-17
| ||
22:19 | Move variable declaration to address compilation issue (C89). (check-in: d64f248da3 user: mistachkin tags: trunk) | |
2018-12-15
| ||
07:12 | Fix errors in comments in fts5.h. (check-in: 4002790d94 user: dan tags: trunk) | |
2018-12-14
| ||
21:58 | Experimental support for new sqlite3_bind_blob() and sqlite3_bind_text() interfaces that take an extra void* argument that is passed into the destructor in front of the object that is to be destroyed. (Leaf check-in: 33a1924ebb user: drh tags: custom-destructors) | |
18:11 | Fix harmless compiler warnings. (check-in: 27221c6990 user: drh tags: trunk) | |