Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
470 check-ins using file src/func.c version ed33e38cd6
|
2021-08-11
| ||
| 18:56 | Fix an RBU problem with restarting an update after the *-oal file is already larger than 4GiB. (Leaf check-in: 56869f54f3 user: dan tags: rbu-replace-hack) | |
| 18:44 | If the special "sqlite_rbu_replace_hack" table is present in an RBU database, use REPLACE instead of INSERT when writing index entries to imposter tables. (check-in: 4b73e151cd user: dan tags: rbu-replace-hack) | |
|
2019-12-31
| ||
| 18:12 | Set the SQLITE_DIRECTONLY flag on the fts3_tokenizer() function of FTS3, thus preventing that function from being called from within a trigger or view. (check-in: f3171dc22e user: drh tags: trunk) | |
| 15:12 | Refactor names. Use "small" instead of "mini" to describe the smaller of the two lookaside memory allocation sizes. (check-in: 88d2449838 user: drh tags: two-size-lookaside) | |
| 14:49 | Merge recent enhancements from trunk. (check-in: 39d5557937 user: drh tags: two-size-lookaside) | |
| 12:18 | The OP_ResultRow opcode releases the SCopy dependences on all its registers, as the values in those registers will not be reused. (check-in: 1dc83c5d54 user: drh tags: trunk) | |
|
2019-12-30
| ||
| 23:50 | Back out the NEVER() from check-in [40d10e7aad5b8992] because it is reachable after all. Ticket [892575cdba4e1e36] (check-in: f481636f1a user: drh tags: trunk) | |
| 23:41 | Add an ALWAYS() to a conditional which is apparently always true. (check-in: ea44c607b8 user: drh tags: trunk) | |
| 20:42 | Put a NEVER() on a defensive branch. (check-in: 40d10e7aad user: drh tags: trunk) | |
| 20:42 | Fix a typo in a comment. No changes to code. (check-in: d4813a8d81 user: drh tags: trunk) | |
| 14:32 | Do not use HIDDEN columns for NATURAL joins. Fix for [7c0e06b16]. (check-in: ab09ef4271 user: dan tags: trunk) | |
| 06:55 | In ALTER TABLE, rename columns and tables in expressions that are optimized out by the "AND 0" optimization. Doing this also fixes an otherwise harmless assert() failure. (check-in: a9e0354c99 user: dan tags: trunk) | |
|
2019-12-29
| ||
| 22:08 | Do not allow triggers that run as part of REPLACE conflict resolution during an UPDATE to modify the the table being updated. Otherwise, those triggers might delete content out from under the update operation, leading to all kinds of problems. Ticket [314cc133e5ada126] (check-in: db4b7e1dc3 user: drh tags: trunk) | |
| 00:52 | Add the OP_FinishSeek opcode which completes an OP_DeferredSeek if the seek has not already completed. Also add the sqlite3WhereUsesDeferredSeek() interface to the query planner. The UPDATE implementation adds an OP_FinishSeek before running the final OP_Insert if one is needed. Ticket [ec8abb025e78f40c] and also an assertion fault reported by Yongheng. (check-in: 21ef6e9933 user: drh tags: trunk) | |
|
2019-12-28
| ||
| 18:25 | Do not attempt to flatten compound sub-queries in a FROM clause into the parent if any component of the sub-query uses a window function. (check-in: eeb76f621d user: dan tags: trunk) | |
| 18:08 | Change an assert() in where.c to a testcase() macro, since the condition may be false. This was a problem with the assert() only, there is no bug in release builds that omit assert(). (check-in: 82be135dee user: dan tags: trunk) | |
| 16:20 | Disable early coding of transitive constraints at the end of each loop in the WHERE clause processing if the loop being coded is for a LEFT JOIN, even if the loop is part of an OR-clause optimization for virtual tables. Test cases in TH3. (check-in: 9421b442ca user: drh tags: trunk) | |
| 15:24 | Fix an instance where the planner might choose to use the OR-optimization when it adds no benefit. The same quirk causes an assert() to fail. This is not a bug in released versions - without the assert() the library still gets the right answer, it just does so less efficiently than it should. (check-in: f4bed1d7af user: dan tags: trunk) | |
| 14:33 | Further improvements to .wheretrace during loop code generation. (check-in: c4d5b75c93 user: drh tags: trunk) | |
| 14:07 | New enhancements to .wheretrace. The 0x20000 flag shows the WHERE clause before and after coding each loop. The 0x800 flag shows status at the start and at the end of each loop. An extra "C" tag is shown on coded terms. (check-in: 59cc46e5a6 user: drh tags: trunk) | |
| 13:39 | Expose some of the Where data structure debug printing routines to the entire WHERE-clause processing module. (check-in: 85e7688776 user: drh tags: trunk) | |
| 13:17 | Add the --enable-all option to the main configure script as a short-hand to enable FTS4, FTS5, Geopoly/Rtree, JSON, and Sessions. In the amalgamation-autoconf, the --enable-rtree option (which is enabled by default) also now activates Geopoly. (check-in: 52ea0672d7 user: drh tags: trunk) | |
| 13:01 | Simplifications to the initialization of the sqlite3_index_info structure that is used to communicate with virtual table modules. Avoid adding unused constraints to the sqlite3_index_info structure. Extra constraints are harmless, but might be confusing to people trying to understand the code. (check-in: 5e6357fc95 user: drh tags: trunk) | |
| 12:33 | If the ".wheretrace 0x10000" bit is set, print WhereTerm.prereqAll and .prereqRight fields in the WhereTerm trace output. (check-in: a4f330b133 user: drh tags: trunk) | |
| 11:55 | Two new test cases added to test/fuzzdata8.db. (check-in: 1be2c18f40 user: drh tags: trunk) | |
| 09:08 | Second attempt to fix a problem with unwinding the WITH stack of the Parse object following an error. (check-in: 315d1f1a50 user: dan tags: trunk) | |
| 08:33 | Merge latest trunk changes with this branch. (Closed-Leaf check-in: d693be3753 user: dan tags: better-error-handling-1) | |
| 08:26 | Fix an assert() failure in altertable3-22.4. (check-in: c566a91d5c user: dan tags: better-error-handling-1) | |
| 03:55 | Fix a faulty assert() associated with query search limiting query plans. (check-in: 0d743585c2 user: drh tags: trunk) | |
| 02:40 | Convert an assert() back into a conditional. The conditional was converted into an assert() by check-in [6ae4ad6ebee4db88] (2009-05-28) because we were unable to find a test case for it. Yongheng's fuzzer just now found that test case. (check-in: 4d0b9109f7 user: drh tags: trunk) | |
| 01:52 | When an INSERT is receiving content from a SELECT, run an OP_ReleaseReg opcode at the top of each iteration of the loop in order to prevent spurious OP_SCopy misuse complaints. Ticket [de4b04149b9fdeae] (check-in: 6afadd3b3a user: drh tags: trunk) | |
| 00:36 | Recompute the values for all generated columns after NOT NULL ON CONFLICT REPLACE constraints fire. Tickets [37823501c68a09f9] and [5fbc159eeb092130]. (check-in: 4cc12c1886 user: drh tags: trunk) | |
|
2019-12-27
| ||
| 20:54 | Do not attempt to unwind the WITH stack in the Parse object following an error. This fixes a separate case to [de6e6d68], but also causes an assertion fault at select.c:4666 for test case altertab3-22.4. (check-in: d29edef934 user: dan tags: better-error-handling-1) | |
| 20:06 | Remove a NEVER() that is no longer true. Fix for [36ffedcb9]. (check-in: 597896ed0a user: dan tags: trunk) | |
| 19:46 | Fix a problem involving window function aliases being referenced from sub-selects. (check-in: e3b5fc05c0 user: dan tags: trunk) | |
| 18:15 | Add a missing "ifcapable rtree {...}" line to the window1.test test module. (check-in: 4e6fbb1a51 user: drh tags: trunk) | |
| 16:25 | Fix a problem with window functions in aggregate queries that do not have GROUP BY clauses. (check-in: 99609786f4 user: dan tags: trunk) | |
| 15:31 | Do not mistake constant integers in a PARTITION BY expression for references to ORDER BY style references to values returned by the SELECT statement. (check-in: 45c64d39d5 user: dan tags: trunk) | |
| 13:30 | Follow-up to check-in [c8c6dd0e6582ec91] - change the xAccess() method to return true if the file exists and it is anything other than a regular file, or if it is a regular file with a non-zero file size. (check-in: 8a39803ef8 user: drh tags: trunk) | |
| 08:57 | When determining if an aggregate within a sub-query should be processed as part of the sub-query or an outer query, consider any FILTER clause in the same way as the arguments to the aggregate. (check-in: 1ffc045d2b user: dan tags: trunk) | |
| 01:50 | Ensure that the Pager.changeCountDone flag is cleared whenever dropping the write lock, even when transitioning from EXCLUSIVE locking mode into NORMAL locking mode while in WAL mode. Ticket [fb3b3024ea238d5c]. (check-in: 846b1de6e5 user: drh tags: trunk) | |
| 00:19 | Fix buffer underflows in the zipfile extension associated with zero-length or NULL filename in the ZIP archive. But report on the mailing list by Yongheng and Rui. (check-in: 465a15c5c2 user: drh tags: trunk) | |
|
2019-12-26
| ||
| 23:40 | If an UPSERT can cause an Abort due to a constraint failure, make sure the query planner knows this. Ticket [7c13db5c3bf74001]. (check-in: f14ce94866 user: drh tags: trunk) | |
| 23:16 | An UPDATE of a table that is indexed by a constant virtual column that uses the one-pass optimization might cause the table seek to be omitted before reaching row DELETE/INSERT. Fix this by coding an extra OP_Column in that circumstance. Ticket [ec8abb025e78f40c] (check-in: e545604959 user: drh tags: trunk) | |
| 14:36 | Fix an assert() in fts5 that could fail if an xSavepoint() call on another vtab fails. Fix for [167b2aac] . (check-in: a5d7f5d24a user: dan tags: trunk) | |
| 01:10 | Makefile.in fix so that it works on systems that require a .EXE suffix on executables. (check-in: f482a4cdfa user: drh tags: trunk) | |
| 01:02 | Add two new fuzzer test cases from dbsqlfuzz. (check-in: f8d5586a8c user: drh tags: trunk) | |
| 00:56 | In the xAccess() method of the unix VFS, return true if the named object is a directory, regardless of what stat() reports as the st_size for the object. Different filesystems report st_size differently for directories. Problem reported on the mailing list by Stefan Brüns. (check-in: c8c6dd0e65 user: drh tags: trunk) | |
| 00:54 | Fix an FTS3 test case that depends on the ICU extension so that it only runs if SQLite is compiled with ICU. (check-in: 19c6240bdb user: drh tags: trunk) | |
| 00:53 | Fix an incorrect assert() added by check-in [fa58aad48a788802]. Problem found by dbsqlfuzz. (check-in: a17b29f08c user: drh tags: trunk) | |
| 00:20 | Disable the optimization that tries to pull the value of an expression from an index on that expression if the expression is a constant. (check-in: e5fd8b5050 user: drh tags: trunk) | |
|
2019-12-25
| ||
| 23:54 | When the sqlite3WindowRewrite() routine detects and error, have it convert the SELECT statement into just "SELECT null" so that it does not leave the parse tree in a goofy state that can cause problems with subsequent code before the stack has a chance to unwind and report the error. Ticket [d87336c81c7d0873] (check-in: fa58aad48a user: drh tags: trunk) | |
|
2019-12-24
| ||
| 21:42 | Fix a minor performance regression from check-in [401c9d30e06191d9] (check-in: 76f54ee867 user: drh tags: trunk) | |
| 21:01 | Backout the early VTable cursor close change from yesterday, as Yongheng and Rui have found a test case for which it fails. The new test case is added to fuzzcheck. (check-in: ddb10f0374 user: drh tags: trunk) | |
| 20:51 | Fix a problem causing fts5 integrity-check failures if numeric values are inserted into a table within a utf-16 database. Fix for [752fdbf6]. (check-in: bae060f382 user: dan tags: trunk) | |
| 18:53 | Make the zipfile() extension function more robust against zero-length filenames. (check-in: b9c2005f02 user: drh tags: trunk) | |
| 16:20 | Fix another case where malformed utf-8 was being mishandled in fts5. Fix for [df46a6f3]. (check-in: 1c0a05b09a user: dan tags: trunk) | |
| 15:35 | Fix an assert() added as part of commit [a11b393dc] that can fail if fts5 database records are corrupt. (check-in: 4630c1ec01 user: dan tags: trunk) | |
| 15:01 | Extra defense against problems following an OOM. dbsqlfuzz find. Also import the latest dbsqlfuzz test cases. (check-in: 0a70f5ddaf user: drh tags: trunk) | |
| 14:27 | Fix a spurious report of corruption that could be made by the fts5 integrity-check in SQLITE_DEBUG builds if the fts5 index contains malformed utf text. Ticket [d62981b76de521e3] (check-in: a11b393dc2 user: dan tags: trunk) | |
| 13:41 | Convert an ALWAYS() into an assert() with an extra error term. Dbsqlfuzz find, with test case in TH3. (check-in: b473ad35c5 user: drh tags: trunk) | |
| 01:53 | Allow comparison operators of a register against itself. Ticket [188f912b51cd802a], (check-in: 401c9d30e0 user: drh tags: trunk) | |
|
2019-12-23
| ||
| 21:11 | Test case for the zipfile-extension bug fix of the previous check-in. (check-in: bc8bfc7fcd user: drh tags: trunk) | |
| 21:04 | Fix the zipfile() function in the zipfile extension so that it is able to deal with goofy filenames that contain embedded zeros. (check-in: cc0fb00a12 user: drh tags: trunk) | |
| 20:41 | Thoroughly reset the rtree cursor at the start of each VFilter operation, including clearing its cache. This prevents left over pages in the cache which can cause problems on shutdown after a LEFT JOIN. Ticket [5eadca17c4dde90c] (check-in: 4c50afafce user: drh tags: trunk) | |
| 20:07 | Remove an extra deflateInit2() call accidently left in check-in [f5ee30426e8876e7] (check-in: 953e6aa6d9 user: drh tags: trunk) | |
| 19:28 | Do an early close of virtual table cursors to avoid unnecessary cursor contention in UPDATE for some virtual table implementations. Ticket [56a74875be799b85] (check-in: eb95dac7f6 user: drh tags: trunk) | |
| 18:02 | Early detection of database corruption in balance_deeper(). (check-in: 61c2233654 user: drh tags: trunk) | |
| 15:17 | Fix a case in which SQLite could fail to identify "x BETWEEN ? AND ?" being true as implying that x is not null. Ticket [dfd66334]. (check-in: 2f17974912 user: dan tags: trunk) | |
| 14:20 | For expressions like (x, y) IN (SELECT ...) where the SELECT uses window-functions, require that all columns on the LHS be indexed before an index can be used. Fix for [d9ed4ebe]. (check-in: 0b1dbd60f5 user: dan tags: trunk) | |
| 13:24 | Fix a shift-overflow problem in yesterday's check-in [36fdeb4f0a66970a] that OSSFuzz helpfully discovered overnight. Thanks Google. (check-in: bff38e2b53 user: drh tags: trunk) | |
| 03:37 | Fix the OP_Cast operator so that when casting to TEXT, it always leaves the result in the encoding of the database. Ticket [0911b5d161b039c6]. Test cases in TH3. (check-in: f347744e0d user: drh tags: trunk) | |
| 02:43 | Change the assert() back into a testcase(). See also check-ins [9ab985a9c8160b90] and [ddb17d92df194337] and other check-ins that those reference. Fix for ticket [9d708e474201c001] (check-in: 2c44c73499 user: drh tags: trunk) | |
| 02:18 | Enhance the sqlite3VdbeMemAboutToChange() shallow-copy validation mechanism by adding the new OP_ReleaseReg opcode to tell MemAboutToChange() that a range of registers is no longer needed so that the source register can be freely changed. This is a change to debugging and test builds only and does not impact release builds. Fix for ticket [c62c5e58524b204d] and [5ad2aa6921faa1ee]. The previous fix to ticket [5ad2aa6921faa1ee] is backed out by this change since this change is a better fix. (check-in: 36fdeb4f0a user: drh tags: trunk) | |
|
2019-12-22
| ||
| 23:48 | Change the code generator for the IN operator so that it avoids creating OP_Eq and OP_Ne opcode with the same P1 and P3 arguments. This enables us to back out check-in [ddb17d92df194337] and also fix ticket [188f912b51cd802]. (check-in: 9ab985a9c8 user: drh tags: trunk) | |
| 20:29 | Make a hard copy of strings in constraint checks prior to applying OP_RealAffinity, to avoid problems with a pointer accounting assert. This change is not strictly necessary - the correct answer is obtained without it and no UB occurs - however the pointer accounting asserts are useful to prevent other problems so it is a simple matter to bring this piece into compliance. Ticket [5ad2aa6921faa1ee] (check-in: 89a9dad633 user: drh tags: trunk) | |
| 20:03 | When constructing the virtual MATCH term of the WHERE clause for a virtual table that is in a LEFT JOIN, be sure to set the correct Expr.iRightJoinTable value. This value does not appear to ever be used, except inside of a single assert(). But it is good to set it correctly, nevertheless. This fixes ticket [7929c1efb2d67e98], which as far as I can tell is completely harmless. (check-in: ef604882a2 user: drh tags: trunk) | |
| 19:41 | In the WHERE clause debugging output (the .wheretrace output) show the parent index of any WhereTerm that is a child. (check-in: 7fc733328c user: drh tags: trunk) | |
| 18:55 | Fix to the optimization of check-in [a47efb7c8520a011] that reads the values of expressions used in an index-on-expression directly from the index rather than recomputing the value. If the expression has a top-level COLLATE or unlikely() operator, be sure to clear the corresponding flags prior to converting it into a TK_COLUMN expression. Failure to do this is most likely harmless in production, but might cause an assertion fault in debugging builds. Ticket [b0cb8aff561a6dcd]. Test cases in TH3. (check-in: 56539e1c13 user: drh tags: trunk) | |
| 18:23 | Back out the asserts of check-ins [a500893b6f64aced] and [d9c9fe9f5ad3fc91] as ticket [1b06916e01c82b66] demonstrates a case that refutes them. (check-in: ddb17d92df user: drh tags: trunk) | |
| 18:06 | When parsing a CREATE TABLE from the sqlite_master table, delete the CHECK constraints if there are any errors, since there might otherwise be attempts to use those CHECK constraints if PRAGMA writable_schema=ON is set. This undoes the fix in check-in [ea721b34477ab8b4] for a more general solution. (check-in: a982e6434c user: drh tags: trunk) | |
| 17:32 | Ensure sqlite3WindowRewrite() is called on a SELECT statement before any terms aremoved from it as part of IN() clause processing. Fix for [f00d096ca]. (check-in: 8c856404b4 user: dan tags: trunk) | |
| 14:29 | Fix "PRAGMA data_version" so that it works the same way with locking_mode=PERSIST and journal_mode=PERSIST configured. Fix for 7a458c2a5f. (check-in: 45748e2db0 user: dan tags: trunk) | |
|
2019-12-21
| ||
| 20:43 | Fix a spurious report of corruption from the fts3/4 integrity-check triggered by using the languageid option. (check-in: 70815e273f user: dan tags: trunk) | |
| 19:37 | When creating a new virtual table, ensure that the OP_ParseSchema opcode processes the correct entry in the sqlite_master table even if there is a second entry with the same name and table values due to database corruption and the use of writable_schema=ON. Dbsqlfuzz find. (check-in: 4dbb6e1cb0 user: drh tags: trunk) | |
| 14:09 | When a corrupt schema is loaded using writable_schema=ON, the CHECK constraints (or other expressions in the table definition) might not be fully resolved. Ensure that the code generator can deal with this if the table is subsequently used in a DML statement. dbsqlfuzz find. (check-in: ea721b3447 user: drh tags: trunk) | |
|
2019-12-20
| ||
| 22:46 | Do not try to access a generated column through an index if the collating sequence for the generated column is non-standard. Part 2 of ticket [e0a8120553f4b082] (check-in: 056bb8dcbd user: drh tags: trunk) | |
| 20:45 | Apply real affinity to generated columns of type REAL that are extract from an index. Ticket [e0a8120553f4b082] (check-in: 728ad39e3b user: drh tags: trunk) | |
| 20:08 | Debugging enhancment: Show the Expr.y.pTab pointer on TK_COLUMN nodes of an expression tree in the treeview. (check-in: 64154ac450 user: drh tags: trunk) | |
| 20:03 | Fix a bad interaction between RBU and [df51ae19]. (check-in: 0b9d8a1202 user: dan tags: trunk) | |
| 19:41 | Fix two more cases in fts5 where sqlite3_value_bytes() was being called before sqlite3_value_text(). Fix for e431c355. (check-in: a1ba9a37d7 user: dan tags: trunk) | |
| 17:41 | Export the public RBU entry points from the Win32 DLL. (check-in: e62d1791f4 user: mistachkin tags: trunk) | |
| 17:25 | Debugging improvement: when tracing the VDBE, output the register value for OP_Cast after the cast is completed. (check-in: dc5f1d282d user: drh tags: trunk) | |
| 15:35 | When computing dependencies on WHERE-clause terms, be sure to take into account the FILTER clause of aggregate functions. Problem reported by Manuel Rigger. (check-in: 3cc2b5709e user: drh tags: trunk) | |
| 14:37 | Improved assert() statements on the OP_Column opcode. (check-in: c538601e4c user: drh tags: trunk) | |
| 14:24 | Load new dbsqlfuzz test cases into test/fuzzdata8.db. (check-in: 01aeeb5c4c user: drh tags: trunk) | |
| 14:18 | Fix a potential problem with ALTER TABLE commands on schemas that contains a triggers or view featuring a sub-query that uses an ON clause with a join that is not a LEFT JOIN. (check-in: 365dd8b663 user: dan tags: trunk) | |
| 14:08 | Fix harmless compiler warnings. (check-in: 2bbd014c90 user: drh tags: trunk) | |
| 13:24 | Ensure fts5 consistently calls sqlite3_value_text() before sqlite3_value_bytes(). This matters when storing blobs in fts5 tables within utf-16 databases. Fix for [cf36cb4e02]. (check-in: c16305eba0 user: dan tags: trunk) | |
| 12:55 | Do not continue an ALTER TABLE tree walk of a subquery if a prior error is seen in the tree. Report the problem immediately, without continuing. (check-in: 01ca865f31 user: drh tags: trunk) | |
| 12:33 | Fix an assert() in btree.c that can fail if an INSERT is attempted on a corrupted database while in PRAGMA writable_schema=ON. (check-in: 07beb3629e user: drh tags: trunk) | |
|
2019-12-19
| ||
| 22:08 | Continuation of [e2bddcd4c55ba3cb]: Add another spot where it is necessary to abort early due to prior errors in sqlite3WindowRewrite(). (check-in: cba2a2a44c user: drh tags: trunk) | |
| 21:11 | Fix the regexp extension so that it correctly translates all over-length 3-byte UTF8 sequences into 0xfffd. (check-in: 3d4c0bf890 user: drh tags: trunk) | |
| 20:37 | When an error occurs while rewriting the parser tree for window functions in the sqlite3WindowRewrite() routine, make sure that pParse->nErr is set, and make sure that this shuts down any subsequent code generation that might depend on the transformations that were implemented. This fixes a problem discovered by the Yongheng and Rui fuzzer. (check-in: e2bddcd4c5 user: drh tags: trunk) | |
| 17:42 | Detect the invalid use of NULLS LAST on an INTEGER PRIMARY KEY definition of a WITHOUT ROWID table. (check-in: 4417c5bf0a user: drh tags: trunk) | |
| 15:15 | Fix the zipfile extension so that INSERT works even if the pathname of the file being inserted is a NULL. Bug discovered by the Yongheng and Rui fuzzer. (check-in: a80f84b511 user: drh tags: trunk) | |
| 13:17 | Fix incorrect SQL generated by the Lemon change of the previous check-in. (check-in: fccfb8a9ed user: drh tags: trunk) | |
| 12:29 | Modify the parse.sql output file from Lemon so that the RULE table contains a column with the complete text of the rule. (check-in: 329cbb372d user: drh tags: trunk) | |
| 03:14 | More restrictions on changes to shadow tables when in defensive mode. (Leaf check-in: 4146c629c6 user: drh tags: branch-3.30) | |
| 00:59 | Minor enhancement to Win32 locking semantics. (Leaf check-in: ed66ffce57 user: mistachkin tags: win32LockFlags) | |
|
2019-12-18
| ||
| 21:22 | Change an unreachable testcase() into an assert() (check-in: 062e00c4d7 user: drh tags: trunk) | |
| 20:51 | Continue to back away from the LEFT JOIN optimization of check-in [41c27bc0ff1d3135] by disallowing query flattening if the outer query is DISTINCT. Without this fix, if an index scan is run on the table within the view on the right-hand side of the LEFT JOIN, stale result registers might be accessed yielding incorrect results, and/or an OP_IfNullRow opcode might be invoked on the un-opened table, resulting in a NULL-pointer dereference. This problem was found by the Yongheng and Rui fuzzer. (check-in: 862974312e user: drh tags: trunk) | |
| 13:42 | Add the --vdbe-debug option to fuzzcheck. (check-in: 289158aa24 user: drh tags: trunk) | |
| 13:02 | Enhancements to fuzzcheck on unix so that it captures signals and prints the current test file and test number prior to terminating. (check-in: ae7cbb246b user: drh tags: trunk) | |
| 09:17 | Add a test case for fix in check-in [1ca0bd982ab1183b] (check-in: 519864da8b user: drh tags: trunk) | |
| 02:12 | Correctly unwind the savepoint stack if the database goes read-only in the middle of a savepoint. (check-in: a96d02b4ab user: drh tags: trunk) | |
| 01:31 | Remove an assert() in the code generator that can fail due to an invalid INSERT statement. The same assert() appears again in the VDBE and will fire there if the problem persists in an SQL statement that does not contain errors. Gramfuzz find. (check-in: 9bdd57cb9a user: drh tags: trunk) | |
| 00:05 | When processing constant integer values in ORDER BY clauses of window definitions (see check-in [7e4809eadfe99ebf]) be sure to fully disable the constant value to avoid an invalid pointer dereference if the expression is ever duplicated. This fixes a crash report from Yongheng and Rui. (check-in: 1ca0bd982a user: drh tags: trunk) | |
|
2019-12-17
| ||
| 12:03 | Clean up the ExprList that holds the names of columns in a CTE before checking for unused references in the ALTER TABLE implementation. (check-in: 8223e79f98 user: drh tags: trunk) | |
|
2019-12-16
| ||
| 16:52 | When a table is part of a LEFT JOIN and should be a completely NULL row due to the semantics of a LEFT JOIN, make sure any generated columns on that row evaluate to NULL. Ticket [3b84b42943644d6f] (check-in: 0271491438 user: drh tags: trunk) | |
|
2019-12-15
| ||
| 02:49 | Ensure that all ON CONFLICT REPLACE indexes are sorted to the end of the list of indexes for a table, even for weird cases where the same UNIQUE constraint occurs twice with the ON CONFLICT REPLACE clause only on the second one. This avoids an out-of-order contraint processing problem that can arise due to the optimization of check-in [469a62ca33081854]. (check-in: 1e3918ca2f user: drh tags: trunk) | |
| 00:36 | Abort the OP_Savepoint opcode early if a virtual table xSync method fails, perhaps due to an interrupt. (check-in: 672e749aef user: drh tags: trunk) | |
|
2019-12-14
| ||
| 19:55 | When a generated column expression is initially parsed, if it contains errors convert the expression to a simple NULL. This prevents additional pieces of the expression tree that are allocated from lookaside from leaking into the schema, where lookaside allocations are not allowed. (check-in: a89b386056 user: drh tags: trunk) | |
| 18:08 | Due to the previous change, the p5 parameter to OP_VColumn no longer ever contains extraneous bits, so change a testcase() into an assert() to show as much. (check-in: 5b4a88cd3b user: drh tags: trunk) | |
| 17:43 | Do not set OP_Column flags on the instructions generated by sqlite3ExprCodeGetColumn() if the opcode generated is not really an OP_Column, which might happen if the column is virtual. Fix for ticket [b439bfcfb7deedc6] (check-in: 2401e04730 user: drh tags: trunk) | |
| 15:01 | Make the sqlite3ExprCodeTarget() expression code generator routine robust in the face of unknown opcodes - it simply generates a NULL value. (check-in: f8e876c82a user: drh tags: trunk) | |
| 14:15 | Fix problems in fts3 with processing corrupt records and undefined integer overflows. (check-in: 3b873029ef user: dan tags: trunk) | |
|
2019-12-13
| ||
| 23:38 | Fix an utterly harmless "applying zero offset to null pointer" UB warning in sqlite3AtoF(): (check-in: 052fdf5e58 user: drh tags: trunk) | |
| 22:59 | Initial exploration of blocky APIs, starting with a variant of sqlite3_exec. (Leaf check-in: fd849d5b6a user: numist tags: blocky-exec) | |
| 21:24 | Patch to the page cache to avoid harmless pointer arithmetic that due to bugs in the STD-C spec is technically UB. This is to fix a harmless UBSAN complaint that OSSFuzz is hitting. (check-in: c29fc21288 user: drh tags: trunk) | |
| 16:04 | Change the default lookaside configuration to 40 slots of 1200-bytes each. This actually works out to 30 big slots and 93 small slots using the mini-lookaside allocator. We get the same (or better) lookaside coverage but with 72KB less memory per connection. (check-in: 47b71a84d1 user: drh tags: two-size-lookaside) | |
| 15:48 | Cleanup and performance enhancements for mini-lookaside. (check-in: 7480566843 user: drh tags: two-size-lookaside) | |
| 12:14 | Merge fixes from trunk. (check-in: 9c471195f6 user: drh tags: two-size-lookaside) | |
| 11:42 | Ensure that there is a containing SELECT statement when processing a normal aggregate function as if it were a window function. (check-in: c1014e80b2 user: drh tags: trunk) | |
| 11:32 | *Remove* the macros, not just comment them out. (check-in: e5dc2939d3 user: drh tags: trunk) | |
| 11:31 | Remove unused macros formerly used to define built-in aggregate functions. (check-in: 3ef0d44edd user: drh tags: trunk) | |
| 01:12 | Reduce the size of ExprList to 24-bytes per entry, down from 32-bytes per entry. This helps keep lookaside allocations in the mini-size. (check-in: bda92d92bf user: drh tags: two-size-lookaside) | |
| 00:49 | Tie up the loose ends in the ExprList size reduction. (Leaf check-in: 59d0f3afe5 user: drh tags: two-size-lookaside, exprlist-size-reduction) | |
|
2019-12-12
| ||
| 22:11 | Work toward reducing the incremental size of an ExprList object to 24-byte per entry, from 32-bytes (on a 64-bit machine). This helps the new mini-lookaside allocator to run better by avoiding excessive reallocs. The current change mostly works, but still has a few loose ends to tie up. This check-in is merely a snapshot to save my work. (check-in: fdda76cfb0 user: drh tags: two-size-lookaside, exprlist-size-reduction) | |
| 20:58 | Reduce the number of call sites to dbMallocRawFinish in the case where n > lookaside.sz (check-in: e392e7f228 user: numist tags: two-size-lookaside) | |
| 20:39 | Fix an issue where malloc could be used to fulfill a small allocation when a large lookaside slot could have beeen used instead. (check-in: 611020e337 user: numist tags: two-size-lookaside) | |
| 20:22 | Change the name of the Expr.a.zName field to zEName, so that it has a name that is distinct from other fields and variables and is hence easier to grep for. (check-in: d3783357f8 user: drh tags: two-size-lookaside) | |
| 17:17 | Change the size of a mini-lookaside allocation to a macro (MINI_SZ) rather than a magic number (128). (check-in: 5e1949bca9 user: drh tags: two-size-lookaside) | |
| 16:49 | Merge enhancements from trunk. (check-in: 6cb053f99b user: drh tags: two-size-lookaside) | |
| 15:19 | Factor out the conditional (which is only true for rare errors) from the notValid() function in resolve.c, for a performance improvement and size reduction. Also cause failures to set the Expr node to a NULL operator so that it does not cause problems later in case PRAGMA writable_schema=ON has been set. Test cases in TH3. (check-in: c6af9f655b user: drh tags: trunk) | |
| 02:50 | More efficient implementation of a lookaside allocator that supports mini (in this case, harcoded to 128B) slots. (check-in: b02fdc09c8 user: numist tags: two-size-lookaside) | |
| 00:20 | Fix goofy string formatting in lemon.c that dates from the K&R-C days. (check-in: 48ba5e5a22 user: drh tags: trunk) | |
|
2019-12-11
| ||
| 18:53 | Improved tracing output from the LEMON-generated parser. (check-in: 4d6d2fc046 user: drh tags: trunk) | |
| 16:22 | Strengthen check-in [83da4d4104ee1870] by ignoring *all* WHERE-clause constraints for a virtual table that is the right table of a LEFT JOIN, as such constraints are never useful. This fixes an issue discovered by Manuel Rigger. (check-in: 840de36df1 user: drh tags: trunk) | |
| 15:07 | When trying to drop a virtual table that has no xDestroy method, invoke the xDisconnect method rather than doing nothing, to avoid a memory leak. (check-in: 1fa29a5f2a user: drh tags: trunk) | |
| 14:25 | The xDestroy method of a module might be NULL if the schema is corrupt. (check-in: 0457e7d196 user: drh tags: trunk) | |
|
2019-12-10
| ||
| 20:41 | Enhance LEMON to provide the ability to mark a rules as one that will never reduce due to intervention of actions. Use this new capability to designate the EXPLAIN rule as NEVER-REDUCES. (check-in: 136cdefb2f user: drh tags: trunk) | |
| 18:10 | Use a statement journal on a CREATE VIRTUAL TABLE statement in case the VCreate opcode fails. (check-in: aa3b0eb8c3 user: drh tags: trunk) | |
| 15:05 | Fix parser bugs: require a semicolon after an EXPLAIN command. (check-in: 707a058a22 user: drh tags: trunk) | |
| 03:40 | Avoid a buffer overread in fts5 that could occur when parsing corrupt configuration records. (check-in: 355afd77df user: dan tags: trunk) | |
| 02:48 | Set the affinity of regular columns prior to computing the values of generated columns. Ticket [d7c3f125c925c522] (check-in: d47d66e3d3 user: drh tags: trunk) | |
|
2019-12-09
| ||
| 19:44 | Rename a local variable to avoid masking a function parameter and thus causing a harmless compiler warning. (check-in: f065cf003b user: drh tags: trunk) | |
| 19:29 | Add a NEVER() to the ALTER TABLE fix in check-in [1d2e53a39b87e364] (check-in: c7309ed3c7 user: drh tags: trunk) | |
| 18:22 | The previous check-in was not quite correct, and introduced a new problem with the USING clause. Use this version instead. (check-in: ed28aaa485 user: drh tags: trunk) | |
| 17:14 | Ensure that the SrcList_item.colUsed field is set correctly (set to have a 1 for all columns of the table) when a generated column appears in the USING clause of a join. (check-in: 1923efb283 user: drh tags: trunk) | |
| 15:52 | Fix the NOT NULL verification logic in PRAGMA integrity_check so that it works for generated columns whose value is the result of a comparison operator. Ticket [bd8c280671ba44a7] [https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2019-19646|CVE-2019-19646]. (check-in: f3b39c71b8 user: drh tags: trunk) | |
| 14:34 | Fix a gramfuzz find. If a partial index that does not reference any column of its table is used by an UPDATE statement in one-pass mode, then avoid the use of OP_DeferredSeek since the seek might not be resolved prior to the OP_Delete and OP_Insert that implement the UPDATE. (check-in: e3398c5ffb user: drh tags: trunk) | |
| 08:13 | Avoid infinite recursion in the ALTER TABLE code when a view contains an unused CTE that references, directly or indirectly, the view itself. (check-in: 1d2e53a39b user: dan tags: trunk) | |
| 02:20 | Fix possible null pointer dereferences in the fts5_expr() scalar function. (check-in: c5d4414359 user: dan tags: trunk) | |
|
2019-12-08
| ||
| 00:06 | Fix incorrect column-usage accounting associated with generated columns and added by check-in [6601da58032d18ae]. Fix for ticket [b92e5e8ec2cdbaa1]. (check-in: 9d75e1ccc7 user: drh tags: trunk) | |
|
2019-12-07
| ||
| 13:42 | Correctly deal with multi-row VALUES clauses that contain window functions. (check-in: 26d991f214 user: drh tags: trunk) | |
| 12:13 | Ignore type affinity for auxiliary columns in RTREE as it is too error-prone to try to parse out the type from other constraints. Ticket [8bf76328ac940d52] (check-in: 7fa664ea0e user: drh tags: trunk) | |
| 00:22 | Fix a problem with foreign keys and generated columns discovered by Manuel Rigger. (check-in: 27c0fdab1b user: drh tags: trunk) | |
|
2019-12-06
| ||
| 19:07 | Early detection and reporting of too few columns in an rtree CREATE VIRTUAL TABLE statement. (check-in: ef73107f47 user: drh tags: trunk) | |
| 13:10 | In the rtree extension, actively suppress extra tokens on the column names, and be pedantic about column affinities. Ticket [367a86e28859f1ff] (check-in: 97fb5a72f9 user: drh tags: trunk) | |
| 11:48 | The check-in [b7810062ec2489e1] was not quite right in that it allowed an oversized shift operation (which is UB in C) on some obscure inputs. OSSFuzz found the problem for us overnight. (check-in: 62f2235adf user: drh tags: trunk) | |
| 01:38 | Further enhancements to RTREE, following up to check-in [f898d04cf272ef01] so that constraints are allowed to be compared to strings that look like numbers. Ticket [aa573a787b8f1fdf] (check-in: 32772dfd50 user: drh tags: trunk) | |
| 01:23 | Do not allow UPSERT of a view. (check-in: ebf897e861 user: drh tags: trunk) | |
|
2019-12-05
| ||
| 21:46 | Fix to the rowvalue logic that avoids disabling rowvalue inequality contraints on a virtual table when the virtual table sets the omit flag. The logic has been incorrect since row-values were added. However, this does not come up often as very few virtual tables implement inequality constraints using the omit flag. Ticket [f096d191b6641daa] (check-in: b7810062ec user: drh tags: trunk) | |
| 18:29 | Documentation update: Clarify that the sqlite3_index_info.aConstraintUsage.omit flag is a hint only and that the code generator is not obligated to follow the hint. (check-in: 7d3e3f6e7e user: drh tags: trunk) | |
| 17:31 | Additional debugging information printed with the ".wheretrace 0x100" option. No changes to normally delivered code. (check-in: fc72ec52c9 user: drh tags: trunk) | |
| 14:42 | Make sure the rowid of an RTREE virtual table has integer affinity. Ticket [9fe487ba3c064b4e] (check-in: 7ae8c0d52f user: drh tags: trunk) | |
| 14:31 | Fix a problem in ALTER TABLE that could occur if an index, view or trigger in the schema features an expression of the form ((SELECT ...) IN ()) where the SELECT statement uses a CTE. (check-in: 7e5ad8e0ab user: dan tags: trunk) | |
| 13:34 | Be strict about type enforcement on rowid look-ups in the RTREE extension. Ticket [30e2c183b6b356e4] (check-in: d43e0efb96 user: drh tags: trunk) | |
| 00:44 | Enhance the RTREE extension so that it give correct query results even if the query uses non-numeric constraints. Ticket [a55ab6d97d01ecbc] (check-in: f898d04cf2 user: drh tags: trunk) | |
|
2019-12-04
| ||
| 19:45 | Window functions are never constant. (check-in: 35f0b5a8c7 user: drh tags: trunk) | |
| 15:08 | Ensure that an ALWAYS() in the rename logic really is always true, even for faulty inputs. (check-in: 54410f0e77 user: drh tags: trunk) | |
| 14:26 | Fix an assert() failure that could occur in ALTER TABLE code when the schema contains a view that uses a CTE. (check-in: 75b04a4b0d user: dan tags: trunk) | |
| 03:46 | Fix a buffer overread that could occur in fts3 with corrupt %_stat records. (check-in: e01fdbf9f7 user: dan tags: trunk) | |
| 03:31 | Fix an incorrect NEVER() macro. (check-in: 96b6a76da0 user: drh tags: trunk) | |
| 01:42 | Fix a double-free that could occur when a component of a compound SELECT with an ORDER BY clause uses named window definitions. (check-in: 92893b7980 user: dan tags: trunk) | |
|
2019-12-03
| ||
| 03:34 | Avoid a crash that could occur when a database containing a table with a temp trigger that has the same name as a temp table is detached. (check-in: c4cb9708d4 user: dan tags: trunk) | |
| 03:31 | Avoid a harmless zero offset of a null pointer in FTS3, so as to not provoke unnecessary warnings from run-time checkers. (check-in: 85d95abec4 user: drh tags: trunk) | |
| 02:51 | Avoid computing a zero offset of a null pointer, which though this is technically harmless, is upsetting to pedantic run-time checkers. (check-in: 3ce804e99b user: drh tags: trunk) | |
|
2019-11-30
| ||
| 19:29 | Do not allow a term in the WHERE clause of the query to qualify a partial index on the right table of a LEFT JOIN. Ticket [7f39060a24b47353] (check-in: 4066a34da7 user: drh tags: trunk) | |
|
2019-11-29
| ||
| 13:01 | Add comments in the parse.sql file to help demonstrate how rules are encoded. (check-in: 2c4f714892 user: drh tags: trunk) | |
| 12:51 | Fix the parse.sql output file so that it is readable into a database where foreign keys are enforced. (check-in: 3a82c554c3 user: drh tags: trunk) | |
|
2019-11-27
| ||
| 20:40 | More details on the comment justifying the aleged data race in the WAL-mode checkpoint logic. (check-in: 0fab65c79f user: drh tags: trunk) | |
| 12:03 | Fix obsolete comment. No changes to code. (Leaf check-in: 0ff59edbbb user: drh tags: reuse-shm) | |
|
2019-11-26
| ||
| 14:24 | Fix various documentation typos suggested by Mark Benningfield. No changes to code. (check-in: 08f09dc4f9 user: drh tags: trunk) | |
| 11:10 | Fix to check-in [5d9a369301a65f32] so that it compiles without SQLITE_DEBUG. (check-in: fefe2d0467 user: drh tags: trunk) | |
| 02:22 | Add the new -S option to the lemon parser generator to cause it to output SQL that describes the input grammar. (check-in: 4dbd398d64 user: drh tags: trunk) | |
| 02:03 | Fix an assert() in fts3 that could fail due to corrupt database records. (check-in: 5d9a369301 user: dan tags: trunk) | |
|
2019-11-25
| ||
| 23:55 | Experimental file-controls for controlling the use of the -shm file. (check-in: 12b8fa2336 user: drh tags: reuse-shm) | |
| 04:15 | Modify journal3.test so that it can handle Tcl command "file attr <file> -perm" returning octal values in either "00755" or "0o755" format. (check-in: b0b655625c user: dan tags: trunk) | |
| 00:07 | Minor enhancements to the TclKit download tool. (check-in: 75e31b1d56 user: mistachkin tags: trunk) | |
|
2019-11-23
| ||
| 16:34 | Abort the window function tree rewrite early following an OOM. (check-in: d66f95a515 user: drh tags: trunk) | |
| 15:10 | Avoid evaluating PARTITION BY or ORDER BY expressions multiple times for window function queries that use multiple window functions with the same window-definition. (check-in: 57070c68bb user: dan tags: trunk) | |
| 08:51 | Enhance the path arguments in JSON functions to access "#-N" array indexes. (check-in: ffeec62cb5 user: drh tags: trunk) | |
| 08:36 | Fix a memory leak in fts5 error handling code. (check-in: de6637e15d user: dan tags: trunk) | |
|
2019-11-22
| ||
| 17:37 | Extend the json-path mechanism with array indexes of the form "#" or "#-n" for some positive number "n", to reference the end of an array. (Closed-Leaf check-in: 35ed68a651 user: drh tags: json-path-enhancement) | |
| 11:49 | Fix a harmless compiler warning. (check-in: 34343c4b06 user: drh tags: trunk) | |
| 11:47 | Fix an indentation error and improve the placement of an assert() based on a tweet from Yuri Kotov. (check-in: e6314e386c user: drh tags: trunk) | |
| 11:38 | Add more test cases from Yongheng Chen and Rui Zhong. (check-in: f24e6ca4f2 user: drh tags: trunk) | |
| 10:14 | Fix a crash that could occur if a sub-select that uses both DISTINCT and window functions also used an ORDER BY that is the same as its select list. (check-in: bcdd66c169 user: dan tags: trunk) | |
| 00:42 | Revise the SQLITE_OPEN_NOFOLLOW so that it actually uses O_NOFOLLOW in the open() system call. This backs out the SQLITE_ACCESS_SYMLINK value but adds the new SQLITE_OK_SYMLINK return code from the xFullPathname method of sqlite3_vfs when that routine resolves symbolic links. O_NOFOLLOW is always included in open() system calls for journal files. (check-in: 6a64fb6a2d user: drh tags: trunk) | |
|
2019-11-21
| ||
| 20:24 | Add all fuzz-test cases received from Yongheng Chen and Rui Zhong. Also fix a minor problem in gencol1.test cases. (check-in: ac080432b4 user: drh tags: trunk) | |
| 20:10 | Fix an out-of-bounds array reference in the generated column logic. Problem discovered by valgrind. (check-in: a0ab42f779 user: drh tags: trunk) | |
| 19:37 | Whenever a generated column is used, assume that all columns are used. (check-in: 6601da5803 user: drh tags: trunk) | |
| 18:28 | Fix a problem that comes up when using generated columns that evaluate to a constant in an index and then making use of that index in a join. (check-in: 8b12e95fec user: drh tags: trunk) | |
| 17:14 | Fix a recently introduced memory leak in the test code in test_vfs.c. (check-in: 2d53a30cc2 user: dan tags: trunk) | |
| 14:20 | Prevent direct and indirect recursive content= options in fts3/4 and fts5. (check-in: 2eb997327c user: dan tags: trunk) | |
|
2019-11-20
| ||
| 16:10 | Fix a use-after-free problem in the test suite. No changes to production code. (check-in: 0d1055a5da user: dan tags: trunk) | |
| 13:31 | Further improve detection of corrupt records in fts3. Also fix an error with check-in [dfcf081d842629a0] (check-in: a0f6d526ba user: dan tags: trunk) | |
| 12:07 | Changes to extensions and test logic so that the build works with gcc and with -std=iso9899:1999 (check-in: 2575a68c39 user: drh tags: trunk) | |
|
2019-11-19
| ||
| 21:22 | Enhancements to DBSTAT: (1) Make name=? queries efficient. (2) Add the ability to aggregate results across each individual btree using the aggregate=TRUE constraint in the WHERE clause. (check-in: a48f6e1759 user: drh tags: trunk) | |
| 18:48 | Get the aggregate=TRUE feature working on the DBSTAT virtual table. (Closed-Leaf check-in: 16fef3db06 user: drh tags: dbstat-enhancements) | |
| 14:01 | Begin an enhancement effort for the built-in DBSTAT virtual table. (check-in: 9b5722f0fe user: drh tags: dbstat-enhancements) | |
| 00:13 | Make the result of sqlite3_normalized_sql() survive its statement being reprepared. (check-in: 4330f0795d user: mistachkin tags: trunk) | |
|
2019-11-18
| ||
| 23:48 | Add support for the SQLITE_ACCESS_SYMLINK flag in the Win32 VFS. (check-in: 175c15008e user: mistachkin tags: trunk) | |
| 22:34 | Ensure all file names passed to the VFS layer are double-zero terminated. (check-in: 251230cf43 user: mistachkin tags: trunk) | |
| 18:43 | In the SQLITE_OPEN_NOFOLLOW processing, distinguish between an I/O error on the xAccess() call and an actual symlink encounter. (check-in: 2e98b42fcb user: drh tags: trunk) | |
| 17:46 | Add support for SQLITE_OPEN_NOFOLLOW. (check-in: cb79c82849 user: drh tags: trunk) | |
| 14:04 | Improvements to detection of corruption in the %_stat shadow table of FTS4. (check-in: 6b67eba54e user: drh tags: trunk) | |
| 12:04 | Further improvements to shadow table corruption detection in FTS3. (check-in: e35d8c76aa user: drh tags: trunk) | |
| 11:14 | Detect and prevent infinite recursion in fts3SelectLeaf() due to a malformed FTS3 btree. (check-in: dfcf081d84 user: drh tags: trunk) | |
| 10:37 | Improved detection of corruption in the %_stat table of FTS4. Chromium ticket 1025467. (check-in: 10f8a3b718 user: drh tags: trunk) | |
|
2019-11-17
| ||
| 11:47 | Fix an assert that can fail if the schema is corrupt. (check-in: ed57c48e4b user: drh tags: trunk) | |
| 02:41 | Better detection of corruption in the %_stat and %_docsize shadow tables of FTS3. (check-in: 1e44968788 user: drh tags: trunk) | |
| 00:08 | Remove a reachable NEVER() in FTS3. (check-in: 8bd75bf636 user: drh tags: trunk) | |
|
2019-11-16
| ||
| 23:47 | More improvements to shadow table corruption detection in FTS3. (check-in: 51525f9c32 user: drh tags: trunk) | |
| 21:40 | Improved detection of corrupt shadow tables in FTS3. Enable the debugging special-inserts for FTS3 for both SQLITE_DEBUG and SQLITE_TEST. (check-in: 04b2873be5 user: drh tags: trunk) | |
| 18:36 | Do not allow shadow tables to be dropped in defensive mode. (check-in: 0a988ce340 user: drh tags: apple-osx) | |
| 18:30 | Modify three test cases so that they work even with unusual versions of the library printf(). (check-in: c25289075a user: drh tags: apple-osx) | |
| 16:54 | More restrictions on changes to shadow tables when in defensive mode. (check-in: bae76a5c40 user: drh tags: trunk) | |
| 14:15 | Do not allow CREATE TABLE or CREATE VIEW of an object with a name that looks like a shadow table name. (Closed-Leaf check-in: 6aef58b629 user: drh tags: defensive-improvements) | |
| 13:51 | Break out the test for writable shadow tables into a separate subroutine. (check-in: 8ad34d36a1 user: drh tags: defensive-improvements) | |
| 12:04 | Do not allow shadow tables to be dropped in defensive mode. (check-in: 70390bbca4 user: drh tags: trunk) | |
| 11:33 | Fix a potential NULL pointer dereference on a RENAME TABLE that references a VIEW with a logic error in a window function in the ORDER BY clause. (check-in: 0adb273f7e user: drh tags: trunk) | |
|
2019-11-15
| ||
| 21:16 | Modify three test cases so that they work even with unusual versions of the library printf(). (check-in: 8f4a3750b7 user: drh tags: trunk) | |
| 02:52 | Merge the version 3.30.1 changes into the apple-osx branch. (check-in: 2c8af35206 user: drh tags: apple-osx) | |
| 00:52 | Fix table-valued functions so that they will work as the right table in a LEFT JOIN. Ticket [2ae0c599b735d59e] (check-in: 2c35d3f67b user: drh tags: trunk) | |
|
2019-11-14
| ||
| 23:08 | Backport support for the sqlite3_hard_heap_limit64() interface and the hard_heap_limit pragma to the 3.30 branch. (check-in: ba27012d43 user: drh tags: branch-3.30) | |
| 18:07 | Add support for the sqlite3_hard_heap_limit64() interface and the hard_heap_limit pragma. (check-in: 6399c47ea8 user: drh tags: trunk) | |
| 17:46 | Fix a bug in the hard_heap_limit pragma so that it returns the new value of the hard_heap_limit, not the soft_heap_limit. Change SQLITE_MAX_MEMORY so that it works by setting the default hard_heap_limit value. (Closed-Leaf check-in: 33fd0c3abc user: drh tags: hard-heap-limit) | |
| 15:21 | Merge recent enhancements from trunk. (check-in: b8a631fd30 user: drh tags: hard-heap-limit) | |
| 13:57 | New test cases added to fuzzdata8.db. (check-in: 5baffcda7d user: drh tags: trunk) | |
| 13:24 | Ensure that the same subquery does not go through the window-function rewrite more than once, even when that subquery is part of a virtual table constraint that lacks the omit flag. (check-in: d0bc7db6b0 user: drh tags: trunk) | |
|
2019-11-13
| ||
| 18:50 | Add header guard to the expert extension. (check-in: 7e3151855f user: mistachkin tags: trunk) | |
| 16:50 | Fix the sqlite3TreeView() output to consistently use a colon and not a comma after the table number in the AST dumps. (check-in: fa2416f623 user: drh tags: trunk) | |
|
2019-11-12
| ||
| 16:21 | For for the previous check-in: Always enable the nUri variable, even when debugging is turned off. (check-in: f84a15394c user: drh tags: trunk) | |
| 14:43 | Ensure that the main filename and the journal filenames in the pager object are all correctly double-zero terminated. (check-in: df51ae19c1 user: drh tags: trunk) | |
| 03:45 | Remove a harmless unused variable. (check-in: 28091a48aa user: drh tags: trunk) | |
|
2019-11-11
| ||
| 15:13 | Remove an asm() block from build tool mksourceid.c, as it causes build failures on some systems and performance is not important at build-time. (check-in: 8e100e6c35 user: dan tags: trunk) | |
|
2019-11-10
| ||
| 11:09 | Proper surrogate pair decoding added to JSON functions. See the mailing list bug report and [https://bugs.python.org/issue38749]. More test cases needed here, but it seems to work so far. (check-in: 51027f08c0 user: drh tags: trunk) | |
| 10:08 | Remove an incorrect ALWAYS() macro. (check-in: f7a74f89db user: drh tags: trunk) | |
|
2019-11-09
| ||
| 15:31 | Change the COLFLAG_NOTAVAIL bitmask so that it fits in the single-byte Walker.eCode field. (check-in: 53847f5c28 user: drh tags: trunk) | |
| 14:44 | New dbsqlfuzz test cases added. (check-in: 18c3a17dc8 user: drh tags: trunk) | |
| 14:38 | Make sure the WITH stack in the Parse object is disabled following an error. (check-in: de6e6d6846 user: drh tags: trunk) | |
|
2019-11-08
| ||
| 20:13 | When a virtual table reports orderByConsumed, that tells us nothing about whether or not the SQLITE_DISTINCTBY constraint is met. Fix for ticket [7e59041f9c4e5102]. Test cases in TH3. (check-in: b59f94e4da user: drh tags: trunk) | |
|
2019-11-07
| ||
| 19:59 | Experimental branch to provide SQLITE_LIMIT_HEAP_K to limit the amount of allocated memory used by each database connection, individually. (Leaf check-in: 4ab9dbb939 user: drh tags: limit_heap_k) | |
| 14:51 | Fix the xferCompatibleIndex() function so that it recognizes that a PRIMARY KEY index for a WITHOUT ROWID table is different from a UNIQUE constraint index on the primary key. Ticket [302027baf1374498] (check-in: 34f64f11ca user: drh tags: trunk) | |
| 02:32 | Restore generated column loop detection logic that was incorrectly removed from the previous check-in [9e07b48934e9a972]. This fixes ticket [299b50ba812d8e54] (check-in: 104a2beb57 user: drh tags: trunk) | |
|
2019-11-06
| ||
| 22:19 | Change the way generated columns are computed so that no column is computed inside branch code that might not be taken. Ticket [4fc08501f4e56692] (check-in: 9e07b48934 user: drh tags: trunk) | |
| 19:25 | Minor simplification to the changes from check-in [36c11ad51f]. (check-in: 7bc8205dd9 user: mistachkin tags: trunk) | |
| 17:31 | Fix the OP_DeferredSeek index-to-table column map in P4 so that it works with generated columns. Ticket [ce22a07731530118] (check-in: 36c11ad51f user: drh tags: trunk) | |
| 14:49 | Fix the handling of NOT NULL constraint violations for generated columns in a REPLACE statement. Ticket [2399f5986134f79c] (check-in: 77b1c90add user: drh tags: trunk) | |
|
2019-11-04
| ||
| 12:49 | Changes an unreachable testcase() into an assert(). (check-in: 5710845b63 user: drh tags: trunk) | |
| 02:05 | Fix a false-postive in the sqlite3ExprImpliesNonNullRow() decision routine, that resulted in an incorrect LEFT JOIN strength reduction when the WHERE clause contained a row-value comparison. Ticket [02aa2bd02f97d0f2] (check-in: ea20068e6d user: drh tags: trunk) | |
|
2019-11-03
| ||
| 00:07 | The optimization of check-in [9b2879629c34fc0a] is incorrectly reasoned. The WHERE clause of the partial index might not be true if the table of the partial index is the right table of a left join. So disable the optimization in that case. Ticket [623eff57e76d45f6] (check-in: 3be19e1151 user: drh tags: trunk) | |
|
2019-11-02
| ||
| 17:59 | Fix a problem in VIEW creation that was introduced by the generated columns feature. (check-in: 9c795c4d2b user: drh tags: trunk) | |
| 13:45 | "STORED" is not actually a keyword. The parser looks for STORED as an ordinary identifier. (check-in: 167cd574d6 user: drh tags: trunk) | |
| 13:32 | In sqlite3GenerateIndexKey(), do not attempt to reuse column values from the previous index if the current index is a partial index as the partial index test may have changed those values. Ticket [a9efb42811fa41ee] (check-in: 17e9f65814 user: drh tags: trunk) | |
| 00:00 | Restore the on-line ".help" for the ".explain" command in the CLI. (check-in: 28b8f1036d user: drh tags: trunk) | |
|
2019-11-01
| ||
| 18:52 | Fix a harmless compiler warning. (check-in: f0ebea35d9 user: drh tags: trunk) | |
| 17:31 | The SET DEFAULT and SET NULL conflict resolution actions for foreign key constraints should cause an error when they appear on a generated column. (check-in: b47513d2b3 user: drh tags: trunk) | |
| 16:37 | Fix a potential use-after-free bug that follows an OOM error in code added two days ago by check-in [84e02d773d60cffe]. Problem discovered by OSSFuzz. (check-in: 0a2eb949f8 user: drh tags: trunk) | |
| 16:08 | Add another missing column number translation to the foreign key logic. (check-in: 32df5edcfe user: drh tags: trunk) | |
| 15:19 | Omit the optimization that reduces the column-count on rowid-table cursors when the table has generated columns, because we do not know what columns the generator expressions might try to access. (check-in: e6c96ed91e user: drh tags: trunk) | |
| 13:37 | Fix an fts3 assert() that could fail when operating on a database containing corrupt records. (check-in: 7c52f5478f user: dan tags: trunk) | |
| 12:14 | Add missing column translations to foreign key logic. Ticket [c28a01da72f8957c] (check-in: bc6a43e7ee user: drh tags: trunk) | |
| 10:49 | Fix a potential array bounds overflow in the mkkeywordhash.c code generator. Also add marks to omit keywords specific to generated columns when building with -DSQLITE_OMIT_GENERATED_COLUMNS. (check-in: cc6a408183 user: drh tags: trunk) | |
| 02:30 | Slightly faster keyword hash table. (check-in: f12e743e19 user: drh tags: trunk) | |
|
2019-10-31
| ||
| 20:54 | Correctly generate pre-UPDATE content for virtual columns that are used by foreign key constraints. Ticket [b9befa4b83a660cc] (check-in: 40d3282ec2 user: drh tags: trunk) | |
| 17:13 | Add a few simple TCL test cases for generated columns. (Full test coverage of the generated column logic is provided separately by TH3.) (check-in: acedb5c7f7 user: drh tags: trunk) | |
| 13:16 | Ignore differences in Expr.op2 in sqlite3ExprCompare() in cases where it does not matter. Ticket [1d2a8efc6c3a595a]. (check-in: 329820673a user: drh tags: trunk) | |
| 12:30 | Enhance the TreeView logic to show information about Expr.op2 for FUNCTION and COLUMN nodes. (check-in: aceeaf9e28 user: drh tags: trunk) | |
|
2019-10-30
| ||
| 18:50 | Always disallow the use of non-deterministic functions in CHECK constraints, even date/time functions that use the 'now' or similar keywords. Provide improved error messages when this requirement is not met. Ticket [830277d9db6c3ba1] (check-in: 2978b65ebe user: drh tags: trunk) | |
| 16:29 | Simplify the bytecode generation for SQL function calls such that the OP_Function or OP_PureFunc opcodes are coded directly, rather than using the intermediate OP_Function0 or OP_PureFunc0 - opcodes that are now removed. (check-in: 84e02d773d user: drh tags: trunk) | |
| 13:00 | New tokens ALWAYS, GENERATED, and STORED used by generated columns should all be fallback tokens. (check-in: 13fe6978b7 user: drh tags: trunk) | |
|
2019-10-29
| ||
| 16:18 | Remove the legacy_file_format PRAGMA. In its place, provide the SQLITE_DBCONFIG_LEGACY_FILE_FORMAT option to sqlite3_db_config(). Fix for ticket [6484e6ce678fffab] (check-in: 4d424f3047 user: drh tags: trunk) | |
| 03:39 | Tighten the generated column requirement such that every table must have at least one non-generated column. Ticket [166347c6fc994155]. (check-in: 4fba090e67 user: drh tags: trunk) | |
| 03:30 | Add a missing translation from table column numbers to storage table numbers while processing constraints on an UPDATE statement. Ticket [9621dd78a024d07a] (check-in: 361ea81ae8 user: drh tags: trunk) | |
| 01:26 | Disallow tables that have only virtual columns. (check-in: 591973217f user: drh tags: trunk) | |
|
2019-10-28
| ||
| 13:54 | Fix a problem in fts3 causing it to report corruption if a doclist contains consectutive rowid entries that differ by more than 2^63. (check-in: 0b0a3048f9 user: dan tags: trunk) | |
| 13:07 | Remove a NEVER() comparison from sqlite3ExprCompare(). (check-in: f4285297e1 user: drh tags: trunk) | |
| 04:20 | ALWAYS macro added for coverage. (check-in: a1e1ba9145 user: drh tags: trunk) | |
|
2019-10-27
| ||
| 22:22 | Improved detection of attempts to use a generated column as part of the primary key. Ticket [91e86951016a6802] (check-in: 6d1bbba9a0 user: drh tags: trunk) | |
|
2019-10-26
| ||
| 23:51 | Remove code from the constraint checker that generates virtual column values with the incorrect time. Turns out the the regular code for all other columns works correctly and so the incorrect special-case code is not actually needed. Fix for ticket [3ea175512444b0d1]. (check-in: 5b4c0f2ddc user: drh tags: trunk) | |
| 18:56 | Remove redundant code resulting from a merge error on the previous check-in. (check-in: 713fe86b8c user: drh tags: trunk) | |
| 18:47 | Add support for generated columns. (check-in: b855acf183 user: drh tags: trunk) | |
| 17:08 | Performance optimization on sqlite3GenerateConstraintChecks() - bypass the loop that checks each column for NOT NULL constraints if it is known in advance that the table has no NOT NULL constraints. (check-in: e3c3f4d787 user: drh tags: trunk) | |
| 16:38 | Clarify some comments and add assert() and testcase() macros to the replace-trigger recheck logic for ticket [c1e19e12046d23fe] (check-in: 8c0042bd5c user: drh tags: trunk) | |
| 16:02 | Fix a problem in the legacy ".explain on" formatting when it is used on a query with more than 8 output columns. (check-in: 070b49825c user: drh tags: trunk) | |
| 15:40 | Minor changes to help ensure the pointer returned by sqlite3VdbeGetOp() is not used after it becomes invalid. (check-in: 346bdd49fb user: drh tags: trunk) | |
| 12:27 | Overnight, OSSFuzz helpfully pointed out a potential use-after-free bug in yesterdays changes, involving continued use of a pointer after the memory pointed to had been realloc()-ed. Thanks Google. (check-in: c422afb507 user: drh tags: trunk) | |
| 01:43 | Add missing VdbeCoverage() macro. Fix an off-by-one error in partial index handling. New test cases. Ticket [c1e19e12046d23fe] (check-in: 41cc8e3dab user: drh tags: trunk) | |
| 00:04 | If replace triggers are run during uniqueness checking, then rerun all uniqueness checks a second time using the ABORT algorithm. Fix for ticket [c1e19e12046d23fe] (check-in: fbac0c65d8 user: drh tags: trunk) | |
|
2019-10-25
| ||
| 14:46 | Performance optimization in sqlite3BtreeCursor(). (check-in: ea068b099c user: drh tags: trunk) | |
|
2019-10-24
| ||
| 23:43 | Merge fixes from trunk. (Closed-Leaf check-in: 4ec57d8841 user: drh tags: generated-columns) | |
| 21:02 | The previous fix was incomplete. It is also necessary to disable the Expr.y.pTab field when making the translation. (check-in: b99d570131 user: drh tags: generated-columns) | |
| 20:35 | In fts5, fix a case of overreading a buffer by 1 byte when counting characters in malformed utf-8. Fix for [dd1f67bf]. (check-in: 8d964e1c21 user: dan tags: trunk) | |
| 20:29 | Fix handling of covering indexes that use virtual columns. (check-in: e0f7e321ec user: drh tags: generated-columns) | |
| 19:35 | Correction to check-in [bec5b6d4d083556d] so that it detects *all* triggers that might perturb the insertion cursor. Ticket [50c09fc2cf0d91ce]. (check-in: 521f1d3628 user: drh tags: trunk) | |
| 01:04 | Do not allow ALTER TABLE ADD COLUMN for a STORED column. (check-in: 42fc08bc15 user: drh tags: generated-columns) | |
|
2019-10-23
| ||
| 21:00 | Add an experimental set of UUID processing functions as the uuid.c extension in ext/misc/uuid.c. (check-in: 1a6e009372 user: drh tags: trunk) | |
| 18:09 | When a vector comparison appears in the WHERE clause and the constraint side has a COLLATE clause on the first term of the vector, be sure to honor that COLLATE clause. Ticket [135c9da7513e5a97]. (check-in: 978b2d20cf user: drh tags: trunk) | |
| 18:01 | Create new branch named "uuid-funcs" (check-in: 815dc0437c user: numist tags: uuid-funcs) | |
| 15:47 | Minor adjustments for clarity and test coverage. (check-in: 3006571687 user: drh tags: generated-columns) | |
| 03:53 | Fix incorrect arguments to testcase() macros. (check-in: 812467fbf0 user: drh tags: generated-columns) | |
| 00:31 | Fix the foreign key logic to be compatible with generated columns. (check-in: 3de57145a1 user: drh tags: generated-columns) | |
|
2019-10-22
| ||
| 21:01 | Take the declared column time into account when computing the values for generated columns, and apply appropriate affinity. (check-in: 9e04ba22df user: drh tags: generated-columns) | |
| 20:16 | Merge the row-value fix from trunk. (check-in: 1fbd743861 user: drh tags: generated-columns) | |
| 19:51 | Disqualify row-value comparisons for use by an index if the right-hand side has an affinity that does not match the index. Fix for ticket [6ef984af8972c2eb] (check-in: 5c118617cf user: drh tags: trunk) | |
| 15:45 | Do not allow generated columns in the PRIMARY KEY. (check-in: 1a54743a3d user: drh tags: generated-columns) | |
| 13:59 | In UPDATE processing, include generated columns in the set of columns being updated if and only if their generator expressions reference some other column that is being updated. (check-in: d38176e93a user: drh tags: generated-columns) | |
| 13:01 | New testcase() macros. Fix a problem with INSERT when the IPK is to the right of generated columns. (check-in: 412799fc55 user: drh tags: generated-columns) | |
| 12:02 | Merge changes on trunk into the generated-columns branch. (check-in: ba123b8c20 user: drh tags: generated-columns) | |
| 11:29 | Previous check-in to fix [b47e3627ecaadbde] was incomplete. This check-in completes the fix and adds a test cases. (check-in: c7da1c01f1 user: drh tags: trunk) | |
| 01:00 | Move the sqlite3LocateCollSeq(), sqlite3FindCollSeq(), and sqlite3GetCollSeq() routines so that they are all beside one another in the callback.c source file. No logic changes. (check-in: 9ff80a9bf8 user: drh tags: trunk) | |
| 00:03 | When flipping ("commuting") a comparison operator, set the new EP_Commuted bit rather than adding extra EP_Collate bits, to single later processing that the order of precedence for COLLATE operators is inverted. Fix for ticket [b47e3627ecaadbde] (check-in: 90f7c47735 user: drh tags: trunk) | |
|
2019-10-21
| ||
| 23:41 | Add a VdbeCoverage() macro that was omitted from check-in [eea1e7aa57e74c43]. (check-in: cd2317d04c user: drh tags: trunk) | |
| 16:15 | Try to fix a harmless compiler warning reported by ICC. (check-in: 7f41b44ca2 user: drh tags: trunk) | |
| 01:04 | Changes to the INSERT logic to make it simpler and faster and so that it works with generated columns and BEFORE triggers. (check-in: bc368cb090 user: drh tags: generated-columns) | |
|
2019-10-20
| ||
| 08:26 | Fix a resource leak in fts5 that could occur if an auxiliary function is called from within a query that does not use the full-text index. (check-in: b528bdcd45 user: dan tags: trunk) | |
|
2019-10-19
| ||
| 18:47 | Work toward getting generated columns to work with triggers. Still more work to do in this area. (check-in: 932a37275d user: drh tags: generated-columns) | |
| 15:01 | Add testcase macros. (check-in: fb9c9bb284 user: drh tags: generated-columns) | |
| 13:29 | Refactor names of column index transformation functions, for clarity. Get generated columns working with ALTER TABLE RENAME COLUMN. (check-in: 27ab41c910 user: drh tags: generated-columns) | |
|
2019-10-18
| ||
| 22:54 | Use an allocation count and freelist instead of a membership bitfield in the mini lookaside allocator (Leaf check-in: 9496b4d378 user: numist tags: 2-size-lookaside) | |
| 18:33 | Fixes for WITHOUT ROWID tables with VIRTUAL columns and an INTEGER PRIMARY KEY. (check-in: 86074da0fd user: drh tags: generated-columns) | |
| 17:47 | Merge trunk enhancements into the generated-columns branch. (check-in: 85bc4524d7 user: drh tags: generated-columns) | |
| 17:42 | Get generated columns working for WITHOUT ROWID tables. (check-in: 9f409649ec user: drh tags: generated-columns) | |
| 15:58 | Enhance the ".imposter" command in the CLI so that the first argument can be an existing WITHOUT ROWID table instead of an index. The resulting imposter is the same table, but with columns in storage order and with all constraints removed. (check-in: 9dc0d34586 user: drh tags: trunk) | |
| 12:52 | Claw back some performance from the sqlite3ExprGetColumnOfTable() routine. (check-in: e8426acb94 user: drh tags: generated-columns) | |
| 12:14 | Fix sqlite3ColumnOfIndex() to account for virtual columns. (check-in: 447271123e user: drh tags: generated-columns) | |
| 10:05 | Get indexes working on virtual columns. (check-in: 450c48766c user: drh tags: generated-columns) | |
| 02:19 | STORED columns can now reference other STORED columns, in any order, as long as there are not loops. (check-in: 0d236698e6 user: drh tags: generated-columns) | |
|
2019-10-17
| ||
| 18:35 | In the Table object, change the nVCol field to nNVCol - the number of non-virtual columns, as that is the quantity that we need most. (check-in: 4ad66af04a user: drh tags: generated-columns) | |
| 18:07 | Bug fix with INSERT using an explicit column list on a table with a non-final STORED column. (check-in: 61b4459ae6 user: drh tags: generated-columns) | |
| 17:54 | Some (but not all) INSERT and UPDATE statements now work for STORED columns. (check-in: fe7517bf4d user: drh tags: generated-columns) | |
| 16:16 | Fix the xfer optimization for generated columns, so that VACUUM works again. (check-in: 8f67b89b04 user: drh tags: generated-columns) | |
| 15:59 | Basic UPDATE functionality working for VIRTUAL columns. (check-in: c21959d4eb user: drh tags: generated-columns) | |
| 15:41 | Avoid an infinite loop in fts3/4 incremental-merge in the case where the lowest level in the database contains segments but no data (because there is a delete-marker for each valid entry). Fix for [bf1aab89]. (check-in: 35beaee059 user: dan tags: trunk) | |
| 14:21 | Bug fixes so that "make test" once against runs with no errors. (check-in: 7bfe0f679d user: drh tags: generated-columns) | |
| 13:15 | Fix the table_info and table_xinfo pragmas so that they work with virtual columns. Table_info omits virtual columns. Table_xinfo gives them a "hidden" flag of 2, and 3 for STORED columns. (check-in: 069351b85f user: drh tags: generated-columns) | |
|
2019-10-16
| ||
| 22:01 | INSERT with named columns for a table with generated columns. (check-in: 64db39f92d user: drh tags: generated-columns) | |
| 20:05 | ALTER TABLE is able to add a VIRTUAL column. (check-in: 120c6b78cb user: drh tags: generated-columns) | |
| 19:31 | Simple INSERT and SELECT operations working with VIRTUAL columns. (check-in: 7f9f90b1b8 user: drh tags: generated-columns) | |
| 17:46 | Enhancements to SQL query normalization for UPDATE statements. (check-in: bba975c7af user: mistachkin tags: trunk) | |
| 14:56 | If an AFTER DELETE trigger fires when a conflict row is deleted by REPLACE conflict resolution, make sure the conflict really has been resolved and that the trigger did not recreate the row before continuing. Ticket [a8a4847a2d96f5de] (check-in: eea1e7aa57 user: drh tags: trunk) | |
| 12:18 | Initial experimental code for generated column support. Non-functional. (check-in: 11d472c1df user: drh tags: generated-columns) | |
|
2019-10-15
| ||
| 19:01 | Formatting change on a multi-line conditional, for improved clarity. No logic changes. (check-in: 7248e34765 user: drh tags: trunk) | |
|
2019-10-14
| ||
| 20:32 | Futher improvements to the IN operator for row-values on virtual tables. (check-in: bc751fb64d user: drh tags: trunk) | |
| 15:24 | New test cases in test/fuzzdata8.db. (check-in: 344d9cb0c7 user: drh tags: trunk) | |
| 15:15 | Fix a problem with row-value IN(...) operators and virtual tables. (check-in: aa57d7abac user: dan tags: trunk) | |
|
2019-10-12
| ||
| 23:38 | When Select-Trace is enabled (in debugging builds only) do not show the result of Window function tree rewrites if there are no window functions. (check-in: d1acf72ae1 user: drh tags: trunk) | |
|
2019-10-11
| ||
| 18:55 | Update the zipfile extension to use deflateBound(), instead of compressBound(), to learn the maximum possible size of a deflate()d buffer. (check-in: f5ee30426e user: dan tags: trunk) | |
| 17:14 | Futher improvements to LEFT JOIN strength reduction. (check-in: 8a39167bd2 user: drh tags: trunk) | |
| 16:01 | Improvements to the LEFT JOIN strength reduction optimization. (check-in: 548082dfab user: drh tags: trunk) | |
| 15:33 | Ensure fts3/4 prefix indexes are flushed to disk before an 'optimize' command. Fix for [745f1abc]. (check-in: 4ed905b188 user: dan tags: trunk) | |
| 14:27 | Fix the fts3/4 integrity-check command so that it works with "ORDER=DESC" tables. Fix for [8a6fa2bb]. (check-in: 5863546df9 user: dan tags: trunk) | |
| 14:25 | Increase the version number to 3.31.0 for the next release cycle. (check-in: ffd4c30620 user: drh tags: trunk) | |
| 14:21 | Faster response to sqlite3_interrupt() in the OP_IntegrityCk and OP_Count opcodes. (check-in: bf875dc599 user: drh tags: trunk) | |
| 11:21 | Merge the 3.30.1 changes into reuse-schema. (Leaf check-in: eff7cd7f12 user: drh tags: reuse-schema-3.30) | |
|
2019-10-10
| ||
| 23:58 | Update the autoconf makefile for MSVC. (check-in: 9455643eec user: mistachkin tags: trunk) | |
| 23:58 | Fix harmless compiler warning seen with MSVC. (check-in: 73a8211473 user: mistachkin tags: trunk) | |
| 20:19 | Version 3.30.1 (check-in: 18db032d05 user: drh tags: release, version-3.30.1, branch-3.30) | |
| 17:09 | Prevent SQLite from assuming that if ((? IS NOT NULL) IS NOT NULL) is true, ? may not be NULL. Fix for [c31034044bb72c89]. (check-in: 7833feecfe user: dan tags: trunk) | |
| 16:41 | Fix a problem in the fts3 integrity-check routine causing it to report as corrupt tables that contain values with embedded 0x00 bytes. Ticket [278ac142625e3dcd] (check-in: 629e20c988 user: dan tags: trunk) | |
| 16:21 | Avoid assuming that an expression that contains the sub-expression (? IS FALSE) or (? IS TRUE) may only be true if ? is non-null. Fix for [a976c487]. (check-in: c6cc2390e9 user: drh tags: branch-3.30) | |
| 16:10 | Fix a problem with running ALTER TABLE ADD COLUMN statements within a transaction that writes to one or more virtual tables. Fix for [8fe768e9]. (check-in: a4974a0f95 user: drh tags: branch-3.30) | |
| 15:57 | Fix the OP_SeekRowid opcode so that it works correctly with a Real argument without damaging the value in the register that is the argument. Ticket [b2d4edaffdc156cc]. Test cases in TH3. (check-in: b02630fe6e user: drh tags: branch-3.30) | |
| 15:48 | Be sure to rewrite column references inside FILTER clauses and window frame definitions when flattening queries. Fix for [1079ad19]. (check-in: df2060f34f user: drh tags: branch-3.30) | |
| 15:42 | Version number to 3.30.1. (check-in: 12e28cc7d9 user: drh tags: branch-3.30) | |
| 15:17 | Be sure to rewrite column references inside FILTER clauses and window frame definitions when flattening queries. Fix for [1079ad19]. (check-in: ccba7202b6 user: dan tags: trunk) | |
|
2019-10-09
| ||
| 21:14 | Avoid assuming that an expression that contains the sub-expression (? IS FALSE) or (? IS TRUE) may only be true if ? is non-null. Fix for [a976c487]. (check-in: eb7ed90b8a user: dan tags: trunk) | |
| 18:36 | Do not allow users to effectively disable fts5 crisismerge operations by setting the crisismerge threshold to higher than the maximum allowable segment b-trees on a single level. Fix for [d392017c]. (check-in: 86e4972092 user: dan tags: trunk) | |
| 17:38 | Merge recent fixes and enhancements from trunk. (check-in: 553258c2ed user: drh tags: 2-size-lookaside) | |
| 17:06 | An alternative, experimental lookaside memory allocator that uses two different slot sizes. (check-in: 5ba8cee8f7 user: sperry tags: 2-size-lookaside) | |
| 15:37 | An improved fix for the dbsqlfuzz-discovered ALWAYS() failure following OOM in sqlite3ExprCollSeq(). This time with a test case (engineered by Dan). (check-in: 907f7965b3 user: drh tags: trunk) | |
| 15:26 | Disallow fts5 page sizes greater than 65536 bytes - as there are 16-bit offsets used in the page header. Fix for [81a7f7b9]. (check-in: 75775c5ab4 user: dan tags: trunk) | |
| 15:08 | Remove the "fastfuzztest" makefile target. Use only "fuzztest". (check-in: e17571c789 user: drh tags: trunk) | |
| 15:03 | New test cases added to test/fuzzdata8.db. (check-in: b68d21e93a user: drh tags: trunk) | |
| 15:00 | Change sqlite3SelectDup() to always return NULL if an OOM has occurred. (check-in: 01ba4641ab user: drh tags: trunk) | |
| 13:52 | Use the "fuzztest" target in place of "fastfuzztest". (Closed-Leaf check-in: 7129e468fd user: drh tags: makefile-cleanup) | |
| 13:12 | Remove the obsolete "fastfuzztest" target from the makefiles. (check-in: 3d44f1ee0f user: drh tags: makefile-cleanup) | |
| 01:19 | Save CPU cycles in sqlite3Prepare() by, among other things, shifting the EXPLAIN column label logic over to sqlite3VdbeMakeReady(). (check-in: 1ca5ef8450 user: drh tags: trunk) | |
|
2019-10-08
| ||
| 19:45 | Store 64-bit offset values in osinst log files. (Leaf check-in: 74d975c69a user: dan tags: osinst) | |
| 13:34 | Fix problems with using the fts5 'rebuild' command inside a transaction that contains other updates of the same table. Fix for [e258f008]. (check-in: 238e083571 user: dan tags: trunk) | |
|
2019-10-07
| ||
| 20:36 | Disallow page-sizes smaller than 32 bytes in fts5. Also ensure the fts5 integrity-check works even when "PRAGMA reverse_unordered_selects" is true. Fix for [265e935b26]. (check-in: 8ab0aebdb3 user: dan tags: trunk) | |
| 20:33 | Omit a pointless memory allocation in vdbeSorterSort(). (check-in: 5d76dbc5b0 user: drh tags: trunk) | |
| 18:43 | Fix a problem with running ALTER TABLE ADD COLUMN statements within a transaction that writes to one or more virtual tables. Fix for [8fe768e9]. (check-in: 31e85fbbc4 user: dan tags: trunk) | |
| 13:46 | Add a test case for the assert() fix in the previous commit. (check-in: 32fba11ab7 user: dan tags: trunk) | |
| 13:26 | Fix a faulty assert() statement in the sqlite3VdbeMemExpandBlob() routine. (check-in: 69a26eade2 user: drh tags: trunk) | |
| 10:29 | Avoid attempting to read 0 bytes from a file in the osinst vtab code. (check-in: 1fb76c3e7d user: dan tags: osinst) | |
| 01:05 | Fix the OP_SeekRowid opcode so that it works correctly with a Real argument without damaging the value in the register that is the argument. Ticket [b2d4edaffdc156cc]. Test cases in TH3. (check-in: 3cde82c86b user: drh tags: trunk) | |
|
2019-10-05
| ||
| 19:53 | Omit the check for conflicting shared-cache locks in sqlite3Prepare() if the database connection uses no shared cache. We might be able to go back and remove this code completely, due to the newer Schema.iGeneration logic, but that will take more analysis. This check-in gives the speed benefit but not the reduction in code size. (check-in: 0b73a09270 user: drh tags: trunk) | |
| 19:37 | Fix a problem in Makefile.msc on this branch. (check-in: ed3499a1a5 user: dan tags: osinst) | |
| 19:24 | Optimization to sqlite3VtabUnlockList() for the common case when there is no work to do. (check-in: fc8d45086d user: drh tags: trunk) | |
| 19:17 | Add a timestamp to each osinst log entry, to facilitate merging multiple log files. (check-in: c9519c0cec user: dan tags: osinst) | |
| 18:33 | Call sqlite3Prepare() directly, rather than going through sqlite3_prepare(), when processing the sqlite_master.sql column for a schema. (check-in: cc49380ea7 user: drh tags: trunk) | |
| 17:29 | Have sqlite3.c automatically turn on osinst logging for all connections. (check-in: eea231f81b user: dan tags: osinst) | |
| 15:28 | Reorder two comparisons for a small performance gain in OP_Transaction. (check-in: d7667f6560 user: drh tags: trunk) | |
| 14:39 | Performance optimization to the lookaside-memory disabling mechanism. (check-in: 17ce1c49cb user: drh tags: trunk) | |
|
2019-10-04
| ||
| 16:15 | Bring the wal2 branch up to date with version 3.30.0 (check-in: 16e1dced8b user: drh tags: wal2) | |
| 15:41 | Bring the apple-osx branch up-to-date with the 3.30.0 release. (check-in: 76b26acaba user: drh tags: apple-osx) | |
| 15:31 | Bring the reuse-schema branch up-to-date with version 3.30.0. (check-in: 6e09afecd6 user: drh tags: reuse-schema) | |
| 15:03 | Version 3.30.0 (check-in: c20a353364 user: drh tags: trunk, release, version-3.30.0) | |
|
2019-10-03
| ||
| 16:02 | Avoid running a couple of tests in affinity2.test as part of the valgrind permutations, as the way floating point computations are simulated by valgrind causes them to fail. (check-in: 7f9a4b6015 user: dan tags: trunk) | |
| 14:51 | Increase the precision of floating point value display in VDBE debugging output. No changes to normally deployed code. (check-in: a561a656ff user: drh tags: trunk) | |
| 14:36 | Fix some test script issues that come up with SQLITE_OMIT_VIRTUALTABLE builds. (check-in: 3934d2d08e user: dan tags: trunk) | |
| 13:44 | Fix issues with command line invocation of wapptest.tcl. (check-in: 9e0d5d2640 user: dan tags: trunk) | |
|
2019-10-02
| ||
| 19:44 | Update error detection logic in releasetest.tcl to account for new output formats in USAN. (check-in: 3f36b98836 user: drh tags: trunk) | |
| 19:43 | Update corruptM.test to account for the fact that the database schema may be loaded from within the "sqlite3" command for some test permutations. (check-in: cb9470fc06 user: dan tags: trunk) | |
| 19:33 | Fix a long-standing problem in fts4 incrmental merge. (check-in: 67da31e24e user: dan tags: trunk) | |
| 00:25 | Disable sqlite3_drop_module() tests when virtual tables are omitted from the build. (check-in: 9f4035f91a user: drh tags: trunk) | |
|
2019-10-01
| ||
| 21:31 | Remove references to test scripts analyzeA.test and analyzeB.test from permutations.test, since those test modules no longer exist. (check-in: 2394879698 user: drh tags: trunk) | |
|
2019-09-30
| ||
| 19:13 | The nodeReaderInit() function in FTS3 may not assume that the node is non-empty. (check-in: 361eb2f682 user: drh tags: trunk) | |
| 16:57 | Merge in the 3.30.0 beta 1 changes. (check-in: 0679612412 user: drh tags: reuse-schema) | |
| 16:44 | Merge 3.30.0 beta 1 changes from trunk. (check-in: 918bd97d29 user: drh tags: wal2) | |
| 16:13 | Merge the 3.30.0 beta 1 changes from trunk. (check-in: 9ce4c79171 user: drh tags: apple-osx) | |
|
2019-09-28
| ||
| 18:28 | Add missing comment to constant definitions. (check-in: 661a3789eb user: drh tags: trunk) | |
| 16:14 | Improvements to a comment. No code changes. (check-in: 1a3671c700 user: drh tags: trunk) | |
| 11:19 | In FTS3/4, the poslist end marker must be larger than any other possible poslist value, even on a corrupt poslist. (check-in: 752679aea5 user: drh tags: trunk) | |
|
2019-09-27
| ||
| 17:36 | Update requirement marks due to changes in documentation wording. (check-in: 326cdc16f4 user: drh tags: trunk) | |
| 16:33 | Allow compile-time options -DHAVE_GETHOSTUUID=0 and -DHAVE_GETHOSTUUID=1. (check-in: 3bcb9cc104 user: drh tags: trunk) | |
| 15:15 | Alternative implementation of the previous check-in which is testable. (check-in: 102ef64462 user: drh tags: trunk) | |
| 15:01 | Fix sqlite3ExprCompare() so that it ignores differences in the Expr.iTable field for IN operators, as otherwise it can lead to false negatives, which is usually harmless, but can cause problems for an assert() in the window function logic. (check-in: 6a204b192a user: drh tags: trunk) | |
|
2019-09-26
| ||
| 20:57 | More minor test file fixes. (check-in: 352878fe1c user: dan tags: trunk) | |
| 20:05 | Allow DROP TABLE to work on tables name "sqlite_parameters" just as it does with tables named "sqlite_stat%". Fix for the ".parameter clear" command in the shell. (check-in: e768179baa user: drh tags: trunk) | |
| 19:53 | Minor fix for test file indexexpr1.test. (check-in: 0ceab26f15 user: dan tags: trunk) | |
| 16:57 | New dbsqlfuzz cases added to test/fuzzdata8.db. (check-in: 49073b7003 user: drh tags: trunk) | |
| 16:08 | Test for an OOM condition in resolveAlias(). (check-in: 322eca7f6a user: drh tags: trunk) | |
| 15:53 | Fix a window-functions problem that could occur if an ORDER BY clause contains an alias for a window-function that is not a top-level expression. (check-in: 1cc6cf6407 user: dan tags: trunk) | |
|
2019-09-25
| ||
| 18:44 | Add a missing VdbeCoverage() macro. (check-in: 36d35dbd5a user: drh tags: trunk) | |
| 17:47 | Remove an unused variable. (check-in: a19884455b user: drh tags: trunk) | |
| 16:41 | Further simplifications to window-function code. (check-in: 5fe15c1d83 user: dan tags: trunk) | |
| 11:49 | In the previous check-in, the variable should be openMode, not openFlags. (check-in: 77b0db22d6 user: drh tags: trunk) | |
| 10:36 | In the unix VFS layer, do not attempt to chown() the journal to be the same as the database if running in 8+3 filename mode. Also, update the comments on the chown() attempt to be more precise. (check-in: ab853724a7 user: drh tags: trunk) | |
| 02:07 | Simplifications to the window-function code. (check-in: 489a1eb3aa user: drh tags: trunk) | |
|
2019-09-24
| ||
| 20:20 | Bug fixes for window frames of the form (... RANGE BETWEEN b PRECEDING AND a PRECEDING) or (... RANGE a FOLLOWING AND b FOLLOWING) where (a > b). (check-in: 040e196a8b user: dan tags: trunk) | |
|
2019-09-23
| ||
| 12:38 | Omit the sqlite3IntTokens array constant for a code simplification. (check-in: f907395ef5 user: drh tags: trunk) | |
| 11:55 | When a scalar subquery has a pre-existing "LIMIT X" then change it to "LIMIT X<>0" rather than just "LIMIT 1" so that if X is 0 the limit will still be zero. Ticket [99cd4807dc03f178]. Test cases in TH3. (check-in: 82e5dcf5c1 user: drh tags: trunk) | |
|
2019-09-21
| ||
| 18:49 | Fix a harmless unused variable warning in the test logic. (check-in: d7673a445f user: drh tags: trunk) | |
| 17:31 | Fix harmless compiler warnings. (check-in: 8ea1dc727d user: drh tags: trunk) | |
| 15:44 | Allow SQLite to omit redundant ORDER BY sorts in the case where a SELECT statement has GROUP BY and ORDER BY clauses that use the same expressions, even when the ORDER BY expressions are marked "DESC". (check-in: 20f7951bb2 user: dan tags: trunk) | |
| 13:34 | Add --multithread, --serialized, and --singlethread options to the speed-check.sh test script. (check-in: c17078af60 user: drh tags: trunk) | |
|
2019-09-20
| ||
| 21:12 | Simplify the fix in the previous commit. (check-in: 5ef64b0f55 user: dan tags: trunk) | |
| 20:52 | Fix a problem with queries containing a min() or max() function for which the FILTER clause excludes all rows. (check-in: b1d7e104e0 user: dan tags: trunk) | |
|
2019-09-19
| ||
| 13:51 | Fix an assert() in fts3 that could fail when accessing a corrupt database. (check-in: 601ce9532d user: dan tags: trunk) | |
|
2019-09-18
| ||
| 20:34 | Always clear the temporary register cache after coding a subroutine. (check-in: b6f2a7f9cd user: drh tags: trunk) | |
| 17:22 | Add extra tests for the handling of corrupt records in fts3. (check-in: 40e29a47d1 user: dan tags: trunk) | |
| 12:49 | Fix another potential "jump depends on uninitialized value" warning. (check-in: 633b214e9b user: drh tags: trunk) | |
| 11:46 | Fix a "jump depends on unititialized value" valgrind error in fts5 triggered by corrupt database records. (check-in: 6b6751cd90 user: dan tags: trunk) | |
| 11:16 | Fix an OOB read in the INSTR() function introduced yesterday by check-in [3fb40f518086c1e8] and detected by OSSFuzz. The test case is in TH3. (check-in: d49047c1b5 user: drh tags: trunk) | |