Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
100 check-ins occurring around 5c9a6c06871cb.
|
2021-07-07
| ||
| 18:43 | General cleanup of mkshellc.tcl, + help and options, decruft (check-in: 8bdd5fbf user: larrybr tags: cli_extension) | |
| 16:48 | Follow-up to [68db1ff9c44fa9c3]: The number of registers needed by PRAGMA foreign_key_check was increased too late for an assert() deep down inside of sqlite3ExprCode(). So move the size increase a little earlier. Forum post 79c9e4797d. (check-in: 7072404a user: drh tags: trunk) | |
| 15:52 | Fix a recently introduced segfault that might occur if a sub-select were used as a term on the RHS of an IN(...) operator for which the LHS is a row-value. (check-in: f586c06a user: dan tags: trunk) | |
| 13:53 | Improve the error message in cases where there is a row-value on the LHS of an IN() operator, the RHS is a list (not a sub-select) and at least one element of the list is not a row-value with the correct number of elements. (check-in: 6b22f4e7 user: dan tags: trunk) | |
| 11:51 | Instead of disallowing writes to fts5 tables if there are fts5vocab cursors open on them (commit [c49a6ed7]), abort any fts5vocab queries if the on-disk structure of the fts5 table changes. (check-in: 9dbdc900 user: dan tags: trunk) | |
|
2021-07-06
| ||
| 20:44 | Handle "<vector> IN (<vector>, <vector> ...)" in the same way as "<vector> IN (VALUES(<vector>, <vector>, ...)". (check-in: 981d230e user: dan tags: trunk) | |
| 15:47 | Create new branch named "branch-3.36" (check-in: 5eea151d user: larrybr tags: branch-3.36) | |
| 02:07 | More commands dispathed. Dispatch translation parameterized. (check-in: 372e3241 user: larrybr tags: cli_extension) | |
|
2021-07-05
| ||
| 19:01 | Do not allow writes to an fts5 table if there are any open fts5vocab cursors. (check-in: c49a6ed7 user: dan tags: trunk) | |
| 18:37 | Add the ATOMIC_INTRINSICS compile-time output output. Move sqlite3Int.h in front of ctime.c so that default values that are not overridden are shown in PRAGMA compile-time option output. (check-in: e3069526 user: drh tags: trunk) | |
| 17:15 | Revert [5204c2c4a7b73a64], restoring the old pre-processor logic for determining the availability of the __atomic_store_n()/__atomic_load_n() primitives. (check-in: e690abb9 user: dan tags: trunk) | |
| 11:27 | Allow sub-queries in the FROM clause of an UPDATE...FROM statement to access the object being updated without using an alias, as is required in the parent query. (check-in: 740cb430 user: dan tags: trunk) | |
| 02:40 | Improved rebustness in sqlite3ExprListDup() when it contains a vector assignment from an UPDATE where the initial term is omitted. This can happen during a UNION ALL query flattening while processing a virtual table update in which the first term of the vector is repeated. Forum post 16ca0e9f32. (check-in: 2547cfe3 user: drh tags: trunk) | |
| 01:11 | Ensure that TK_SELECT_COLUMN Expr nodes always have their iTable field set to to the number of columns in the vector. This is not strictly necessary. It just simplifies the state description and make the code easier to reason about. (check-in: 026f08d4 user: drh tags: trunk) | |
| 00:07 | Fix an incorrect comment on testcase misc1-27.0. No changes to code. (check-in: 49829ae3 user: drh tags: trunk) | |
|
2021-07-04
| ||
| 22:38 | Streamline most common command definition (check-in: dd76b41a user: larrybr tags: cli_extension) | |
| 22:33 | Enhance the treeview debugging mechanism so that it correctly deals with TK_ROW expression nodes. (check-in: e87fdb65 user: drh tags: trunk) | |
|
2021-07-03
| ||
| 19:20 | Create infrastructure for dynamic shell extension. (check-in: 5e7e0d4e user: larrybr tags: cli_extension) | |
| 18:57 | Fix a (harmless) typo in a comment. No code changes. (check-in: d2b9b8da user: drh tags: trunk) | |
| 02:55 | In the PRAGMA foreign_key_check, ensure that sufficient registers are allocated for the virtual machine, even if one or more foreign keys reuses the same column multiple times and has more columns than the table it is part of. Forum post a6b0c05277. (check-in: 68db1ff9 user: drh tags: trunk) | |
|
2021-07-02
| ||
| 12:25 | Put ALWAYS() on a branch that is always true due to [d4097364c511709b]. Fix a testcase precondition associated with the same check-in. (check-in: 55e2fbeb user: drh tags: trunk) | |
| 12:08 | Fix a harmless assertion fault discovered by OSSFuzz. The assertion fault is harmless because the assert() checks a condition that that is harmless if false and the assert() is disabled in production builds. (check-in: 79443aab user: drh tags: trunk) | |
|
2021-07-01
| ||
| 18:19 | Allow unqualified "rowid" references to be used in join queries, provided that only one of the source objects has a rowid. Forum post ab5b077c32636e49 (check-in: d4097364 user: dan tags: trunk) | |
| 16:57 | Import the updated nmakehlp.c file used by the TCL Extension Architecture (TEA) builds in the amalgamation tarballs. Forum post 5a5001f20a. (check-in: 595bf95b user: drh tags: trunk) | |
|
2021-06-30
| ||
| 14:04 | Fix a memory leak in fts5 that could occur when processing a query containing multiple classes of error. (check-in: 7c279670 user: dan tags: trunk) | |
| 12:11 | Fix an assert() in fts5 that might fail when dealing with corrupt records. (check-in: 9e760e25 user: dan tags: trunk) | |
| 11:53 | Do not enforce the SQLITE_LIMIT_FUNCTION_ARG limit for internally generated SQL. This prevents problems with statements like ALTER TABLE if the function argument limit is set too low. Forum post 17735aa21f3ddab2. (check-in: 763fdec5 user: drh tags: trunk) | |
|
2021-06-28
| ||
| 15:25 | When generating data for a RETURNING clause, avoid assuming cursor number 0 is available for use. (check-in: 4b2c59ac user: dan tags: trunk) | |
|
2021-06-25
| ||
| 14:48 | The OP_ReopenIdx opcode should clear the cursor if the cursor is being reused. This prevents leaving the cursor in an initialized state which could cause problems for a subsequent OP_SeekScan opcode. Fix for ticket [6dcbfd11cf666e21] and the OP_SeekScan optimization of check-in [4a43430fd23f8835]. (check-in: d4bf5243 user: drh tags: trunk) | |
| 11:21 | Remove a redundant "ANALYZE" line from the ".fullschema" output. (check-in: 7b47f420 user: drh tags: trunk) | |
| 11:14 | Fix a bug in the ".fullschema" dot-command of the CLI brought to light by check-in [a2ddb89b206c1387] - raising an error on at attempt to access the rowid of a subquery or view. (check-in: 36875827 user: drh tags: trunk) | |
| 00:25 | Document sqlite3_wal_hook() return. (check-in: 8acaf796 user: larrybr tags: trunk) | |
|
2021-06-24
| ||
| 18:23 | Allow the SQLITE_OPEN_EXCLUSIVE flag in sqlite3_open(). Return the SQLITE_CANTOPEN_EXISTS extended code to indicate that the open failed because the database already exists. (check-in: d091150f user: drh tags: trunk) | |
|
2021-06-23
| ||
| 17:41 | Make conditional nature of sqlite3_normalized_sql() better known. (check-in: 0e0ece71 user: larrybr tags: trunk) | |
| 16:07 | Fully incorporate *_changes64() API improvement. (check-in: 6699a2f6 user: larrybr tags: trunk) | |
| 11:39 | Show CROSS JOINs in the TreeView debuggingn output. (check-in: 72e09b6c user: drh tags: trunk) | |
| 11:12 | Fix a problem handling ORDER BY terms of the form "ORDER BY likely(<integer>)" within window frames. (check-in: 710f75b9 user: dan tags: trunk) | |
|
2021-06-22
| ||
| 23:30 | This is another alternative fix to the covering index on WHERE_MULTI_OR in a LEFT JOIN problem that is fixed on trunk nearby. In this alternative, covering indexes are simply disabled for WHERE_MULTI_OR on a LEFT JOIN. This might have run-time impact on some obscure queries. This patch is saved for historical reference only. (Closed-Leaf check-in: 66856410 user: drh tags: multi-or-covidx-fix3) | |
| 23:27 | This is an alternative fix to the covering index on WHERE_MULTI_OR in a LEFT JOIN problem that is fixed nearby. This one works by having the OP_NullRow opcode create the index if it does not already exist. That is slightly more complex. This patch is saved for historical reference only. (Closed-Leaf check-in: 956bafb6 user: drh tags: multi-or-covidx-fix2) | |
| 23:24 | When an index is used by all branches of the WHERE_MULTI_OR optimization and becomes a covering index, make sure the index has been created prior to NULLing it in the OP_NullRow opcode of a LEFT JOIN. See forum post 0575376e07. The covering-index for WHERE_MULTI_OR optimization was added by [62678be3df35cdcb]. Test cases are in the orindex01.test module of TH3. (check-in: 787c76a8 user: drh tags: trunk) | |
| 18:32 | Add the sqlite3_changes64() and sqlite3_total_changes64() API functions. (check-in: 48fdec22 user: dan tags: trunk) | |
| 18:06 | Fix a problem causing sqlite3_changes() to return an incorrect value following a "DELETE FROM tbl" command on an intkey table (because internal b+tree cells were being included in the count). (check-in: f662ff47 user: dan tags: trunk) | |
| 14:59 | Ensure that the cell array used in balance_nonroot() is always large enough, even if the pages being balanced are corrupt. (check-in: 53f64e83 user: dan tags: trunk) | |
|
2021-06-20
| ||
| 22:52 | Load recent dbsqlfuzz cases into test/fuzzdata8.db. (check-in: a766bba0 user: drh tags: trunk) | |
| 22:49 | Remove a NEVER() that is reachable by a corrupt database. dbsqlfuzz 8456a0f88e39a49fb675d8f77394979ebb2a03c1. Test case in TH3. (check-in: 1c41e86f user: drh tags: trunk) | |
| 22:17 | Do not allow an SQLITE_CORRUPT error to mask a prior SQLITE_IOERR in balance_nonroot(). dbsqlfuzz 5e1c86e7f18023f2b9abe3429977d7ea726f50d8 (check-in: 50e81876 user: drh tags: trunk) | |
|
2021-06-19
| ||
| 20:20 | Remove unnecessary (redundant) variable initialization from sqlite3BtreeInsert() for a small size reduction and performance increase. (check-in: 70d18364 user: drh tags: trunk) | |
| 18:35 | The previous check-in is a significant change to btree, so go ahead and increment the version number for the next development cycle. (check-in: 2eb66970 user: drh tags: trunk) | |
| 18:32 | Split the sqlite3BtreeMovetoUnpacked() routine into two separate routines sqlite3BtreeTableMoveto() and sqlite3BtreeIndexMoveto(), since we usually know the type of btree in advance. This results in less branching and better performance. (check-in: 3b0d34e5 user: drh tags: trunk) | |
|
2021-06-18
| ||
| 19:01 | Merge version 3.36.0 (check-in: 79221520 user: drh tags: reuse-schema) | |
| 18:36 | Version 3.36.0 (check-in: 5c9a6c06 user: drh tags: trunk, release, version-3.36.0) | |
|
2021-06-17
| ||
| 17:44 | Merge recent trunk changes into branch-concurrent-pnu-wal2 via that wal2 branch. (check-in: 2ff7e562 user: drh tags: begin-concurrent-pnu-wal2) | |
| 17:38 | Merge recent trunk enhancements into begin-concurrent-report via begin-concurrent-pnu. (check-in: 204dbc15 user: drh tags: begin-concurrent-report) | |
| 17:30 | Merge recent trunk enhancements into the begin-concurrent-pnu branch. (check-in: 229107b1 user: drh tags: begin-concurrent-pnu) | |
| 17:29 | Merge recent trunk enhancements into the begin-concurrent branch. (check-in: 59303e69 user: drh tags: begin-concurrent) | |
| 17:26 | Merge recent trunk enhancements into reuse-schema branch. (check-in: 92d8f967 user: drh tags: reuse-schema) | |
| 16:53 | Merge the latest trunk enhancements into the wal2 branch. (check-in: 41363fb9 user: drh tags: wal2) | |
|
2021-06-16
| ||
| 19:23 | Add an ALWAYS on a branch in STAT4-only logic that is now always true due to check-in [c21bc5a2353e660f]. (check-in: 7bd55eee user: drh tags: trunk) | |
| 18:58 | Omit a handful of tests from pager1.test when testing SQLITE_ENABLE_ZIPVFS builds. (check-in: c3036c28 user: dan tags: trunk) | |
| 16:56 | Do not run test file sessionbig.test on 32-bit platforms. It uses too much memory. (check-in: 75e53215 user: dan tags: trunk) | |
| 15:56 | Adjust the mkshellc.tcl script so that the generated code always uses unix-style line endings. (check-in: 9f69af4d user: drh tags: trunk) | |
| 13:26 | Avoid ominous warning in showdb.c (check-in: f8d7ac01 user: larrybr tags: trunk) | |
| 12:26 | Adjustments to requirement markings in comments. No changes to code. (check-in: be00e45b user: drh tags: trunk) | |
| 11:32 | Enhanced defense against integer overflow in ALTER TABLE. (check-in: fb7f094e user: drh tags: trunk) | |
|
2021-06-15
| ||
| 20:10 | Fix some #ifdefs so that builds work for all supported compile-time options. (check-in: ca7d78f7 user: drh tags: trunk) | |
| 20:07 | Fix an incorrect ifdef in the previous check-in. (Closed-Leaf check-in: 4b2bb467 user: drh tags: ifdef-update) | |
| 18:46 | Fix some minor #ifdef issues so that all of the supported compile-time options continue to build. (check-in: d8cf4070 user: drh tags: ifdef-update) | |
| 15:36 | Update the "nmakehlp.c" utility program sources associated with the TCL Extension Architecture, obtained from https://core.tcl-lang.org/sampleextension/file. Perhaps this new version will sit better with salmonx's static analyzer (ref1), (ref2). (check-in: 603b851d user: drh tags: trunk) | |
| 15:15 | Add NULL checks on all sqlite3_vfs_find(0) calls. This is not strictly necessary. There are no vulnerabilities here. However, adding these checks avoids unnecessary static analyzer complaints. Forum post ce1193be15. (check-in: 272a15b9 user: drh tags: trunk) | |
| 14:34 | Prevent an infinite loop in the trim() SQL function if the second argument is a carefully malformed UTF8 string. (check-in: 829343c2 user: drh tags: trunk) | |
|
2021-06-14
| ||
| 20:49 | Fix typos in testcase() macros from check-in [c09d90eec2a49b94]. (check-in: d4d9869c user: drh tags: trunk) | |
| 20:41 | In the authorizer, use SQLITE_DROP_INDEX, not SQLITE_DROP_TEMP_INDEX when dropping an index in an attached database. (check-in: e5a5acd6 user: drh tags: trunk) | |
| 20:08 | Fix an ALTER TABLE problem with handling infinite loops of CTEs in the database schema. (check-in: c09d90ee user: dan tags: trunk) | |
| 14:52 | Fix a test script problem in corruptN.test. (check-in: 3ddfe9ae user: dan tags: trunk) | |
| 14:35 | Update this branch with latest changes from trunk. (check-in: f21a53fb user: dan tags: reuse-schema) | |
| 14:25 | Update this branch with latest changes from wal2. (check-in: 0ab89901 user: dan tags: begin-concurrent-pnu-wal2) | |
| 14:12 | Update this branch with latest changes from trunk. (check-in: b38c2b8b user: dan tags: wal2) | |
| 14:01 | Update this branch with latest changes from begin-concurrent-pnu. (check-in: a485d1a1 user: dan tags: begin-concurrent-report) | |
| 14:00 | An index on the expression "likely(X)" should store the true value of X, even if X is a real number with no fractional part. The optimization of storing some REAL indexed values as INT (see [10d851353c2abead]) should be omitted for index columns that are expressions. Forum post 3b940c437a. (check-in: 791473bc user: drh tags: trunk) | |
| 13:49 | Update this branch with latest changes from trunk. (check-in: 74b42754 user: dan tags: begin-concurrent-pnu) | |
| 12:23 | Update this branch with latest trunk changes. (check-in: d8199634 user: dan tags: begin-concurrent) | |
| 11:20 | Avoid allocating oversized column names in the sqllimit1.test module. (check-in: 7068f1f6 user: drh tags: trunk) | |
|
2021-06-13
| ||
| 19:14 | The opcode that preceeds OP_SeekScan is usually OP_IdxGT but can sometimes be OP_IdxGE. Adjust asserts accordingly. Follow-up to [9a2ab6092d644fc3]. dbsqlfuzz e41762333a4d6e90a49e628f488d0873b2dba4c5 (check-in: d6803ed8 user: drh tags: trunk) | |
| 17:55 | Improved defense against OOM errors in sqliteInitOne(). dbsqlfuzz 429db39cd5777c46479641e8637c48c62890042d (check-in: b6c57e0e user: drh tags: trunk) | |
| 08:23 | Correct link in .ar help. (check-in: de9d0f95 user: larrybr tags: trunk) | |
|
2021-06-12
| ||
| 18:12 | The sqlite3WithPush() routine might destroy its second argument if there was a prior OOM. Do not use the second argument of sqlite3WithPush() if this happens. (check-in: baa3f962 user: drh tags: trunk) | |
| 17:45 | An alternative approach to the use-after-free problem. (Closed-Leaf check-in: 6796b7a2 user: drh tags: alternative-oom-fix) | |
| 16:17 | Fix a use-after-free that could follow an OOM in ALTER TABLE. (Edit:) Still contains problems. (Closed-Leaf check-in: 193b14a5 user: dan tags: original-uaf-fix) | |
|
2021-06-11
| ||
| 13:18 | Reapply two recent ALTER TABLE error checks that turned out to be necessary after all. dbsqlfuzz fc5a9deefda00dda914748985155a6d4c44174e5. (check-in: 230fedd9 user: drh tags: trunk) | |
| 12:41 | Avoid bugs in some implementations of dlopen() by avoiding calls to dlopen() with a filename argument that exceeds FILENAME_MAX bytes. Forum post 08a0d6d9bf. (check-in: 01f3877c user: drh tags: trunk) | |
| 12:14 | Fix a crash in new ALTER TABLE code that could follow an OOM. (check-in: 6f1f2a0a user: dan tags: trunk) | |
| 12:02 | Remove an unreachable branch added by the previous commit. (check-in: 852ee0e9 user: dan tags: trunk) | |
| 11:14 | Fix problems with ALTER TABLE and schemas that contain views with some recursive CTEs, or CTEs that reference other CTEs. Forum post 9d965f2956. (check-in: 8b1f9a51 user: dan tags: trunk) | |
|
2021-06-10
| ||
| 19:37 | Fix an assert() added by the previous commit that could fail with some database schemas. (Closed-Leaf check-in: 5079b5d5 user: dan tags: alter-table-cte-fix-1) | |
| 19:08 | Fix problems with ALTER TABLE and schemas that contain views with complex CTEs. (check-in: 32e3ecfc user: dan tags: alter-table-cte-fix-1) | |
| 14:36 | Improved robustness following OOM in the constant propagation optimization. dbsqlfuzz 001a20255c0df7495c21df62a20ea5b51e22c390. (check-in: 8658a64d user: drh tags: trunk) | |
|
2021-06-09
| ||
| 14:45 | Set the database connection error code for an oversize argument to sqlite3_bind_blob64() or sqlite3_bind_text64(). Forum post a636276f0d451667 and 33821c8db543c. (check-in: 56ff58c0 user: drh tags: trunk) | |
| 13:39 | Add defensive code to better handle corrupt records in fts5. (check-in: a99c29b5 user: dan tags: trunk) | |
|
2021-06-08
| ||
| 23:46 | Remove two assert()s in FTS5 that checks for database corruption right before actual code does the same check. dbsqlfuzz 9680db8f5338cb9b6060eb3a02d984555d4472e3. (check-in: f3f82967 user: drh tags: trunk) | |
| 17:56 | Try to avoid a double-free in the showdb utility program when running pgidx on a corrupt database. See forum post a60ea5a4d4 for context (check-in: e3188a86 user: drh tags: trunk) | |