Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
89 check-ins for the month beginning 2012-12-01
Following month ↑|
2012-12-31
| ||
| 20:16 | More test cases for the REGEXP operator. Fix minor bugs uncovered by these test cases. (check-in: a611c75061 user: drh tags: trunk) | |
| 19:18 | Add the test_regexp.c module containing a cross-platform implementation of the REGEXP operator. (check-in: 46c8c01b75 user: drh tags: trunk) | |
|
2012-12-21
| ||
| 16:15 | Ensure the database size field in the db header of a backup database is set correctly. Fix for [0cfd98ee201]. (check-in: ff6857b6ed user: dan tags: trunk) | |
|
2012-12-20
| ||
| 01:15 | Add PRAGMA foreign_key_check. Name the child and parent tables as part of a "foreign key mismatch" error. (check-in: e408dc9080 user: drh tags: trunk) | |
| 00:32 | Remove an unreachable branch operation from the foreign_key_check pragma. (Closed-Leaf check-in: b5a8f3160b user: drh tags: foreign-key-check) | |
|
2012-12-19
| ||
| 20:39 | Version 3.7.15.1 (check-in: 6b85b767d0 user: drh tags: release, version-3.7.15.1, branch-3.7.15) | |
| 17:10 | Backport to the 3.7.15 branch the fix to the segfault problem of ticket [a7b7803e8d1e869] which involved the use of "AS" named result columns as logical terms of the WHERE clause. Also, change the version number to 3.7.15.1. (check-in: bae528f486 user: drh tags: branch-3.7.15) | |
| 16:58 | Fix the segfault problem of ticket [a7b7803e8d1e869] which involved the use of "AS" named result columns as logical terms of the WHERE clause. (check-in: b3c9e8f812 user: drh tags: trunk) | |
| 16:32 | A couple more test cases. (Closed-Leaf check-in: 0ae142178d user: drh tags: tkt-a7b7803e) | |
| 16:25 | Additional test cases involving the use of a result column specified by its "AS" name as a logical term of the WHERE clause. Ticket [a7b7803e8d1e8699cd]. (check-in: d7cc364e87 user: drh tags: tkt-a7b7803e) | |
| 15:53 | Proposed fix for ticket [a7b7803e8d1e8699cd8a]. (check-in: 3d0e00c731 user: drh tags: tkt-a7b7803e) | |
| 13:41 | Add commentary to the ExprList object to explain how zSpan is overloaded. Add test cases for the new name resolution functionality. (check-in: 3e7d84db78 user: drh tags: name-resolution-fix) | |
| 02:36 | Better resolution of table and column names in joins where some of the terms of the FROM clause are parenthesized. (check-in: 7344e791b9 user: drh tags: name-resolution-fix) | |
|
2012-12-18
| ||
| 19:36 | Optimize the degenerate case of a FROM clause table name enclosed all by itself inside parentheses. Generate code as if the parentheses did not exist, rather than the old behavior of manifesting the parenthesized table into a transient table. Also, tag every FROM-clause SELECT subquery that is generated by a parenthesized FROM-clause expression using the SF_NestedFrom flag. The new SF_NestedFrom flag is not yet used for anything. (check-in: 7fecced466 user: drh tags: trunk) | |
| 16:07 | Resolve names in FROM-clause subqueries prior to resolving names in the result set expressions of a SELECT statement. (check-in: 9b67c633d9 user: drh tags: trunk) | |
| 13:12 | Remove old commented-out code from parse.y. Fix some over-length source lines in parse.y. No logical changes. (check-in: 7e30c021ab user: drh tags: trunk) | |
| 11:59 | On atomic-write capable systems, if copying the contents of an in-memory journal to disk fails, close the (on disk) journal file before returning the error to the caller. This causes the subsequent rollback operation to use the in-memory journal. Fix for [df678d738adb]. (check-in: 8183d8d7ae user: dan tags: trunk) | |
|
2012-12-17
| ||
| 22:32 | Added test cases for PRAGMA foreign_key_check. Fixed a bug that appears when the column order of the child and parent differ. (check-in: 25411f83f9 user: drh tags: foreign-key-check) | |
| 20:57 | Add to the foreign_key_check pragma an extra output column "parent" that contains the name of the parent table for the constraint that failed. (check-in: 97f7f73772 user: drh tags: foreign-key-check) | |
| 20:40 | Enhance the error message for "foreign key mismatch" to include the names of the child and parent tables. Begin adding test cases for PRAGMA foreign_key_check. Make sure PRAGMA foreign_key_check gets all necessary table locks. (check-in: 0f9963526c user: drh tags: foreign-key-check) | |
| 18:43 | "PRAGMA foreign_key_check" with no argument checks the foreign keys on all tables. (check-in: 57d47423a6 user: drh tags: foreign-key-check) | |
| 18:05 | Fix an issue with child keys that are not INTEGER PRIMARY KEY referencing INTEGER PRIMARY KEY in the parent. (check-in: 603b695bf0 user: drh tags: foreign-key-check) | |
| 16:46 | Prototype for PRAGMA that checks all foreign key constraints on a table. (check-in: 01c980e9bb user: drh tags: foreign-key-check) | |
|
2012-12-14
| ||
| 17:54 | Optimize IN operators in the WHERE clause of queries using virtual tables. (check-in: 3d65c70343 user: drh tags: trunk) | |
| 17:48 | Remove an unreachable branch. Improvements to comments. (Closed-Leaf check-in: d2fb7619b0 user: drh tags: vtab-IN-opt) | |
| 15:54 | Merge in all the trunk changes that have occurred since this branch was opened. (check-in: 6d507e4de9 user: drh tags: vtab-IN-opt) | |
| 15:36 | Fix the virtual table IN optimizer so that it work even if the virtual table implementation leaves the sqlite3_index_info.aConstraintUsage[].omit flag clear for an equality constraint that it intends to use. (check-in: d6e045f89c user: drh tags: vtab-IN-opt) | |
|
2012-12-13
| ||
| 18:57 | Generalize the min/max optimization so that if an appropriate index exists, the index it can be used by any aggregate query that contains only a single max() or min() and does not contain a GROUP BY clause. (check-in: 52e755943f user: drh tags: trunk) | |
| 18:51 | Increase the version number to 3.7.16 in advance of adding new features for the next release. (check-in: 8bcf5f51d8 user: drh tags: trunk) | |
| 16:37 | Attempt to further generalize the min/max optimization so that, if an appropriate index exists, it can be used by any aggregate query that contains only a single aggregate of the form max(colname) or min(colname) and does not contain a GROUP BY clause. (Closed-Leaf check-in: 7280e14cd8 user: dan tags: minmax-opt) | |
|
2012-12-12
| ||
| 14:30 | Allow the error message from "PRAGMA integrity_check" to be longer than 20,000 bytes. (check-in: 120c82d56e user: drh tags: trunk) | |
| 13:36 | Version 3.7.15 (check-in: cd0b37c526 user: dan tags: trunk, release, version-3.7.15) | |
|
2012-12-11
| ||
| 19:40 | Update requirements marks and test cases for multi-VALUE INSERT and to fix typos in requirements text. (check-in: 81d9ee0f0d user: drh tags: trunk) | |
| 19:10 | Fix a typo in a comment that is used to generate documentation. No changes to code. (check-in: e38adb60a7 user: drh tags: trunk) | |
|
2012-12-10
| ||
| 22:19 | When an arithmetic operation with two integer operands must give a floating-point answer due to overflow, make sure the answer is not rounded back to integer by affinity. (check-in: bd7aeeb691 user: drh tags: trunk) | |
| 10:22 | Modify releasetest.tcl so that it runs the "checksymbols" test on a build without SQLITE_DEBUG defined. If SQLITE_DEBUG is defined, the sqlite3WhereTrace variable causes the test to fail. (check-in: 75e545a9e2 user: dan tags: trunk) | |
| 09:08 | Remove a reference to JumpOnce from a comment in vdbe.c. No code changes. (check-in: ee662c039d user: dan tags: trunk) | |
|
2012-12-08
| ||
| 23:37 | Pull the latest trunk changes into the sessions branch, and in particular the collating-sequence refactorization. (check-in: 4f6d69ae94 user: drh tags: sessions) | |
| 22:14 | Factor some work out of the index loop of the bestBtreeIndex() routine for a small performance increase. (check-in: 92c9ab56b1 user: drh tags: trunk) | |
| 21:51 | Refactor collating-sequence handling as a fix for ticket [71e333e7d2e642]. The Expr.pColl field is removed from the Expr object. The COLLATE operator now becomes a separate instance of Expr in the expression tree. The code generator looks up the correct collating function as needed, rather than referring to Expr.pColl. (check-in: 8542e6180d user: drh tags: trunk) | |
| 21:36 | Adjustments to the collating-sequence refactoring to facilitate full-coverage testing and to fix some minor issues found by TH3. (Closed-Leaf check-in: cdbfa66483 user: drh tags: ticket-71e333e7) | |
| 14:16 | Make sure WHERE clause constraints A=B and B=A work the same even with COLLATE clauses. (check-in: b3f5366811 user: drh tags: ticket-71e333e7) | |
| 13:26 | Fix the header comment on the OP_Compare operator. (check-in: 2722f4074f user: drh tags: ticket-71e333e7) | |
| 06:46 | Merge updates from trunk. (check-in: e65db42c9f user: mistachkin tags: configReadOnly) | |
| 04:10 | Duplicating check constraint expressions prior to code generation is no longer required, and so it is removed. (check-in: 1a9db8dc2f user: drh tags: ticket-71e333e7) | |
| 03:34 | Veryquick passes all tests now. (check-in: 7b96115e81 user: drh tags: ticket-71e333e7) | |
| 00:52 | Recognize TK_COLLATE operators that have been transformed into TK_REGISTER. Skip both TK_COLLATE and TK_AS operators when looking for the top of an expression. (check-in: f66c1db296 user: drh tags: ticket-71e333e7) | |
|
2012-12-07
| ||
| 23:23 | Improved error messages when column integers in an ORDER BY clause are out of range. (check-in: bd960d937f user: drh tags: ticket-71e333e7) | |
| 23:10 | For an ORDER BY on a compound SELECT, take the collating sequence from the left-most term of the compound. (check-in: 8e724b383d user: drh tags: ticket-71e333e7) | |
| 22:54 | Do not report missing collating functions while parsing the schema. Also remove a stray comment. (check-in: e313edca04 user: drh tags: ticket-71e333e7) | |
| 22:18 | Fix a long-standing issue with the distinct-as-aggregate optimization that only expressed when the new collating-sequence logic is turned on. (check-in: 0aaf52a339 user: drh tags: ticket-71e333e7) | |
| 21:02 | Fix the processing of ORDER BY clauses with COLLATE terms on compound queries. 52 veryquick test failures remain. (check-in: 49654453ad user: drh tags: ticket-71e333e7) | |
| 20:31 | Many more tests are passing. Only about 100 failures remain in veryquick. (check-in: f9fa7581c0 user: drh tags: ticket-71e333e7) | |
| 19:28 | Ensure the VerifyCookie sub-routine has been run before the database is accessed in an obscure case. Fix for ticket [d6b36be38]. (check-in: 2d5f37c99a user: dan tags: trunk) | |
| 18:38 | Some errors in veryquick resolved. Many more to go. (check-in: 972443b4eb user: drh tags: ticket-71e333e7) | |
| 14:02 | Veryquick now runs to completion without segfaulting or asserting. But there are still lots of errors. (check-in: 7fafab12e4 user: drh tags: ticket-71e333e7) | |
| 10:55 | Do not run test file malloc3.test as part of the inmemory_journal permutation. Explanation is in a comment at the top of malloc3.test. (check-in: df1827b19e user: dan tags: trunk) | |
| 09:08 | Spelling and header fixes for the async extension. (check-in: 930ba85262 user: mistachkin tags: trunk) | |
| 04:46 | Backport VSIX tooling and packaging fixes and enhancements. (check-in: e890bcde34 user: mistachkin tags: branch-3.7.14) | |
|
2012-12-06
| ||
| 23:16 | In the async VFS, treat the new error code SQLITE_IOERR_DELETE_NOENT as success for xDelete. (check-in: c507ca4a8e user: mistachkin tags: trunk) | |
| 21:16 | Remove the Expr.pColl field and compute the collating sequence as it is needed. This fixes the test script "shared9.test", though there is still a memory leak. And there are other problems. Consider this a work-in-progress. (check-in: fd011cb22f user: drh tags: ticket-71e333e7) | |
| 20:27 | Cherrypick [79ef8e3c77] (Apply the same restrictions on constant refactoring to statements within a trigger program as top-level statements). (Leaf check-in: 34aafb7436 user: dan tags: branch-3.7.9) | |
| 20:19 | Cherrypick [557c69055a3] and [0064bab7714] (OP_Once-related fixes for triggers). (check-in: 0d7b5d455c user: dan tags: branch-3.7.9) | |
| 19:41 | Merge the constant-refactoring-in-triggers branch with the trunk. (check-in: 79ef8e3c77 user: dan tags: trunk) | |
| 19:37 | Apply the same restrictions on constant refactoring to statements within a trigger program as top-level statements. Candidate fix for [ae3c5670b6]. (Closed-Leaf check-in: 0c31a46801 user: dan tags: constant-refactoring-in-triggers) | |
| 19:01 | Add the SQLITE_FCNTL_TEMPFILENAME file control that asks the underlying VFS to return a new temporary filename. Per request from NSS team at Mozilla. (check-in: 1a63b1d5fa user: drh tags: trunk) | |
| 15:15 | Add a test case that demonstrates ticket [71e333e7d2e642]. (check-in: cc6e0785df user: drh tags: ticket-71e333e7) | |
| 04:33 | For the sqlite3-all.c target, use backslashes when calling the splitter script via the MSVC makefile. (check-in: d507648d82 user: mistachkin tags: trunk) | |
| 04:19 | Bring the ancient malloc3.test file closer into relevance with the latest core code. (check-in: 9793a21c13 user: mistachkin tags: trunk) | |
| 02:56 | Stop using the TCL_LIBS configuration variable when linking with Tcl. Remove superfluous 'rm' command from the "clean" target. (check-in: 1551277208 user: drh tags: trunk) | |
| 01:03 | Remove obsolete comments from the definition of the CollSeq object. (check-in: 7b6951016f user: drh tags: trunk) | |
| 00:16 | Stop using the TCL_LIBS configuration variable when linking with Tcl. Remove superfluous 'rm' command. (Closed-Leaf check-in: e1feab9834 user: mistachkin tags: noTclLibs) | |
|
2012-12-05
| ||
| 19:04 | Disable a couple of tests in where.test when running the "no_optimization" permutation. (check-in: 5e961e37c5 user: dan tags: trunk) | |
| 16:44 | Modify a query in test/thread001.test to ensure that results are returned in the order expected by the test. (check-in: 51cbddd51d user: dan tags: trunk) | |
| 14:37 | Fixes to a test case in shared_err.test. The fix is to prevent sqlite3_prepare() from being called when the test intends to test the response of sqlite3_step() to an OOM condition. (check-in: 40c143955e user: dan tags: trunk) | |
| 14:31 | Make sure memory is freed using a routine that is compatible with the routine used to allocate the memory. The SQLITE_MEMDEBUG compile-time option enforces this. (check-in: 9f6c68856b user: drh tags: trunk) | |
| 11:30 | Since errors are ignored anyway, use sqlite3OsFileControlHint() instead of OsFileControl() to invoke the SQLITE_FCNTL_BUSY_HANDLER file-control. (check-in: bdceca2738 user: dan tags: trunk) | |
| 10:01 | If the atomic-write property is enabled, a transaction may be committed in journal_mode=DELETE mode without ever actually creating a journal file on disk. In this case, do not attempt to unlink() the journal file when committing the transaction. (check-in: 11aa47b0a8 user: dan tags: trunk) | |
| 09:12 | Update a couple of test scripts so that they work in auto-vacuum mode. (check-in: 04ab2dc691 user: dan tags: trunk) | |
|
2012-12-04
| ||
| 11:03 | Fix a harmless compiler warning. (check-in: 12693deba9 user: drh tags: trunk) | |
| 05:24 | Add a test to ensure that the fix in [b0c1ba655d] has worked. (check-in: 41806de5c8 user: dan tags: trunk) | |
| 05:21 | Add a test to ensure that the fix in [b0c1ba655d] has worked. (Closed-Leaf check-in: 59c233942e user: dan tags: mistake) | |
| 00:59 | Improvements to the 'tcl' shell output mode. Escape doublequotes, set separator to space when mode is set, and skip separator after final column. (check-in: 487ba75313 user: drh tags: trunk) | |
| 00:53 | Fix an uninitialized variable detected by valgrind. (check-in: ffd1e51490 user: drh tags: trunk) | |
| 00:37 | Fix an out-of-order function declaration when compiled with MEMDEBUG. (check-in: 6d31557837 user: mistachkin tags: trunk) | |
| 00:23 | Improvements to the 'tcl' shell output mode. Escape double quotes, set separator to space when mode is set, and skip separator after final column. (Closed-Leaf check-in: 41fd9dd290 user: mistachkin tags: tclMode) | |
|
2012-12-03
| ||
| 19:42 | Remove an unreachable condition. Replace it with an assert(). (check-in: 7d5fc1a339 user: drh tags: trunk) | |
| 17:04 | Make sure that the optimization that set the maximum column that will be used on a particular query does not mistakenly change an opcode other than OP_OpenRead or OP_OpenWrite. In particular, make sure it does not overwrite the P4 field of an OP_SorterOpen. (check-in: b0c1ba655d user: drh tags: trunk) | |
|
2012-12-01
| ||
| 20:21 | Do not exit the command-line shell if a ".read" command invokes a script that contains two errors. But reported on the mailing list. (check-in: 2e5e651527 user: drh tags: trunk) | |