Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
100 events occurring around 7f2246a17be9915b.
2019-08-15
| ||
21:27 | Add the SQLITE_DBCONFIG_ENABLE_VIEW option, together with a "db config" command in the TCL interface that can access that option as well as all the other sqlite3_db_config() boolean options. (check-in: 61b4bccd user: drh tags: trunk) | |
20:04 | Provide the SQLITE_DIRECTONLY flag for app-defined functions that prohibits the use of those functions within triggers or views. (Closed-Leaf check-in: fc745845 user: drh tags: directonly) | |
14:36 | • Fixed ticket [256741a1]: null pointer dereference caused by window functions in result-set of EXISTS(SELECT ...) plus 3 other changes (artifact: cf72f220 user: drh) | |
14:35 | • Fixed ticket [e4598ecb]: Division by zero in the query planner. plus 5 other changes (artifact: b83367a9 user: drh) | |
14:35 | Ensure that the optional "sz=N" parameter that can be manually added to the end of an sqlite_stat1 entry does not have an N value that is too small. Ticket [e4598ecbdd18bd82] (check-in: 98357d8c user: drh tags: trunk) | |
14:28 | • New ticket [e4598ecb] Division by zero in the query planner.. (artifact: d93508fc user: drh) | |
14:09 | • Ticket [256741a1] null pointer dereference caused by window functions in result-set of EXISTS(SELECT ...) status still Open with 6 other changes (artifact: 6fef4d64 user: dan) | |
13:53 | Ensure that SQLite does not attempt to process incompatible window functions in a single scan. Fix for [256741a1]. (check-in: 4f5b2d93 user: dan tags: trunk) | |
13:51 | • New ticket [256741a1] null pointer dereference caused by window functions in result-set of EXISTS(SELECT ...). (artifact: ef08d208 user: dan) | |
13:46 | Avoid downgrading SQLITE_CORRUPT errors detected by the schema parser into SQLITE_NOMEM or SQLITE_ERROR errors. (check-in: b2e79f8f user: drh tags: trunk) | |
13:17 | Ensure that the cell overwrite optimization does not overwrite the header of the b-tree page. (check-in: 4cc5694c user: drh tags: trunk) | |
00:04 | Early detection out-of-bounds page numbers on the direct-overflow-read optimization gives consistent error messages regardless of whether or not the optimization is enabled. (check-in: b517a52f user: drh tags: trunk) | |
2019-08-14
| ||
19:55 | • Closed ticket [17db54d7]: Unexpected error in DELETE with existing trigger plus 6 other changes (artifact: dd9f8d3e user: dan) ... 1 similar event omitted. | |
15:17 | Always check for reads past the end of the file on the in-memory journal driver. This used to be an assert(). (check-in: 4d41ca7d user: drh tags: trunk) | |
2019-08-13
| ||
20:34 | Add some static to the 'regexp' extension. (check-in: a21d1dde user: mistachkin tags: trunk) | |
17:56 | Update test cases so that they work with ICU enabled. (check-in: 408144a1 user: dan tags: trunk) | |
17:27 | Update some corruption test cases to take [724f4df9c] into account. (check-in: 927cd7b4 user: dan tags: trunk) | |
15:22 | Merge latest trunk changes with this branch. (check-in: 5b4689d8 user: dan tags: reuse-schema) | |
15:11 | Fix a problem with RBU function sqlite3rbu_bp_progress() when used during an RBU vacuum. (check-in: 8c44b02f user: dan tags: trunk) | |
2019-08-12
| ||
18:26 | • Fixed ticket [3cf9bb22]: Database corruption caused by REPLACE plus 5 other changes (artifact: 927499fa user: drh) | |
18:26 | Make sure the btree cursor overflow cache is cleared when overwriting a cell in sqlite3BtreeInsert(). Ticket [3cf9bb227e9a5d32] (check-in: 7dae7b96 user: drh tags: trunk) | |
18:25 | • New ticket [3cf9bb22] Database corruption caused by REPLACE. (artifact: 4c0dd574 user: drh) | |
16:36 | Experimental implementation of NULLS FIRST/LAST. This branch still has problems - the most significant of which is that ORDER BY clauses with a non-default NULLS FIRST/LAST qualifier can never use an index. (check-in: 07babb0f user: dan tags: nulls-last) | |
16:25 | For the ".testctrl prng_seed" command in the CLI, if the argument is "random" then select a random integer seed and print the seed value on stdout. (check-in: 636ca447 user: drh tags: trunk) | |
15:37 | Fix the build for compilation using separate files rather than the amalgamation. (Leaf check-in: 1e0f4869 user: drh tags: sha3) | |
15:19 | Make the sha3 extension part of the amalgamation, enabled using SQLITE_ENABLE_SHA3. (check-in: a48a7b78 user: drh tags: sha3) | |
14:17 | Add a TESTCTRL that disables the verification of type, name, and tbl_name in the sqlite_master table. (check-in: 48d9b48a user: drh tags: trunk) | |
00:08 | Validate the type, name, and tbl_name fields of the sqlite_master table when loading the schema, unless writable_schema is engaged. (check-in: 724f4df9 user: drh tags: trunk) | |
2019-08-10
| ||
15:07 | • Edit [da01ba4fa47c6508|da01ba4f]: Edit check-in comment. (artifact: d9c628b3 user: drh) | |
15:06 | • Fixed ticket [9080b622]: Constant expression in partial index results in row not being fetched plus 6 other changes (artifact: 53eb33a8 user: drh) | |
15:06 | Fix the sqliteExprImpliesExpr() routine so that it recognizes that "(NULL IS FALSE) IS FALSE" doe not imply "NULL NOT NULL". Ticket [9080b6227fabb466] (check-in: da01ba4f user: drh tags: trunk) | |
14:35 | Add extra test cases related to the previous commit. (check-in: 3c690b2b user: dan tags: trunk) | |
14:16 | Remove two incorrect assert() statements from the logic used to derive column names and types from subqueries. (check-in: 712e4771 user: drh tags: trunk) | |
2019-08-09
| ||
23:00 | • New ticket [9080b622] Constant expression in partial index results in row not being fetched. (artifact: af49a21f user: mrigger) | |
20:26 | Allow the RHS of a row-value IN operator to use a compound query with an ORDER BY clause. (check-in: eaf15d9b user: drh tags: trunk) | |
14:54 | In shared-schema mode, handle the case where a connection has created a virtual-table object, but is later assigned a different shared-schema object for which the virtual-table schema has not yet been initialized. (check-in: e30c7414 user: dan tags: reuse-schema) | |
01:59 | Modify signature of sqlite3VdbeChangeOpcode() to accept int instead of u32. (check-in: 68b7f2ac user: mistachkin tags: trunk) | |
01:11 | Fix more compiler warnings. (check-in: 1b85442b user: drh tags: trunk) | |
2019-08-08
| ||
19:40 | Fix harmless compiler warnings. (check-in: 1eef4dde user: drh tags: trunk) | |
19:19 | Remove support for SQLITE_ENABLE_STAT3. The sqlite_stat3 table is now ignored, if it exists. Run ANALYZE using STAT4 to get the equivalent functionality, which presumably everybody has been doing for a long time now. (check-in: f1cd234c user: drh tags: trunk) | |
18:49 | Fix a case of the Expr.affinity to Expr.affExpr refactor that was missed in the [a29f2a7d07beff64] check-in. (check-in: 83450d10 user: drh tags: trunk) | |
16:23 | More legacy STAT3 code removed. (Closed-Leaf check-in: 845d2f17 user: drh tags: omit-stat3) | |
15:47 | Fix spurious "no such table: x1" errors in shared-schema mode that could occur when a query that is a join across two or more databases needs to call the xConnect() methods of a virtual table. (check-in: edf45cd7 user: dan tags: reuse-schema) | |
15:24 | Remove support for STAT3. The sqlite_stat3 tables are ignored, if they exist. STAT4 continues to work as it always has, and as it is a superset of STAT3 is the recommended replacement. (check-in: 1e17ea2f user: drh tags: omit-stat3) | |
11:44 | Fix a bug causing "no such table" and other similar sqlite3_prepare*() errors to return SQLITE_SCHEMA instead of SQLITE_ERROR in shared-schema mode when there is an attached database for which the schema has never been loaded. (check-in: 111e2c73 user: dan tags: reuse-schema) | |
08:31 | • Ticket [f8a7060e] Incorrect result for query that uses MIN() and a CAST on rowid status still Open with 3 other changes (artifact: 64317f51 user: dan) | |
01:39 | Remove a NEVER() that is reachable from a corrupt database. (check-in: 30e6ee27 user: drh tags: trunk) | |
2019-08-07
| ||
21:30 | • Ticket [f8a7060e] Incorrect result for query that uses MIN() and a CAST on rowid status still Open with 7 other changes (artifact: 3bee902b user: drh) ... 1 similar event omitted. | |
19:57 | Remove use of the affinity() function from view.test, as it is only available in SQLITE_DEBUG builds. (check-in: 7f2246a1 user: dan tags: trunk) | |
18:34 | Add "set TMP=%CD%" to the start of each msvc script output by releasetest_data.tcl. Otherwise, since binaries compiled with SQLITE_TEST all choose the same sequence of pseudo-random numbers, collisions between temp file names cause errors when running multiple tests in parallel. (check-in: f5d0436d user: dan tags: trunk) | |
17:45 | Eliminate some more cases of redundant sorting in window-function queries. (check-in: 8158d2ac user: dan tags: trunk) | |
13:25 | Do not make SQLITE_READ authorizer calls for tables without names, as all such tables will be internal-use-only tables for subqueries and whatnot. (check-in: 193c87fc user: drh tags: trunk) | |
2019-08-06
| ||
23:28 | • Fixed ticket [0a5e2c1d]: Unexpected affinity conversion for view column in IN operator plus 6 other changes (artifact: d65bc0de user: drh) | |
21:16 | Ensure that when the col in an operator like "val IN(col)" is a column of a view, its affinity is not used to coerce val. Fix for [0a5e2c1d]. (check-in: 17b3d221 user: dan tags: trunk) | |
20:55 | Improved reuse of file descriptors for which close() is delayed to prevent clearing of posix advisory locks. (check-in: 509c1ba2 user: drh tags: trunk) | |
20:43 | • New ticket [0a5e2c1d] Unexpected affinity conversion for view column in IN operator. (artifact: 4bce8b02 user: mrigger) | |
20:26 | Enhance the ".recover" output in the shell to use double-quotes around table and column identifiers. (check-in: 846d2d2d user: drh tags: trunk) | |
18:40 | Add "PRAGMA foreign_keys=OFF;" to the start of the script output by ".recover", just as is done for ".dump". (check-in: bfc29e62 user: dan tags: trunk) | |
17:05 | • Fixed ticket [9f66f887]: EXPLAIN PRAGMA executes PRAGMA plus 8 other changes (artifact: f8525f61 user: drh) | |
15:33 | • Fixed ticket [61c85385]: Row is not fetched in SELECT from VIEW plus 6 other changes (artifact: 505de8a7 user: drh) | |
15:32 | Ensure that columns of views and sub-queries that are expressions with no affinity are comparied without any type conversions, as required in the documentation. Tickets [61c853857f40da49] and [d52a29a9e6bc55c5]. (check-in: 9c8c1092 user: drh tags: trunk) | |
15:18 | Performance optimization to the new affinity handling logic. (Closed-Leaf check-in: c9724e76 user: drh tags: pending) | |
15:02 | Add a multi-threaded test to threadtest3.c for the feature on this branch. (check-in: 62557fab user: dan tags: reuse-schema) | |
14:37 | Use 0x40 (ASCII '@') instead of 0x00 to mean "no affinity" so that columns with no affinity can appear in a zero-terminated string. Use the new SQLITE_AFF_NONE macro for this new magic number. (check-in: e8234f69 user: drh tags: pending) | |
11:45 | Add an assert() to the code on this branch. (check-in: fdd44bbb user: dan tags: reuse-schema) | |
2019-08-05
| ||
23:31 | • Edit [e15a0977ddfad3d0|e15a0977]: Move to branch pending. Edit check-in comment. (artifact: 37715ac2 user: drh) | |
21:56 | • New ticket [9f66f887] EXPLAIN PRAGMA executes PRAGMA. (artifact: 6c982a95 user: mrigger) | |
21:04 | • Edit [470ac8d50ce2f7cc|470ac8d5]: Mark "Closed". (artifact: 8029b648 user: drh) | |
20:53 | Ensure that columns of views and sub-queries that are expressions with no affinity are not assigned BLOB affinity. This matches the documentation. Fix for [61c853857f40da49]. (On a branch because there are still subtle issues.) (check-in: e15a0977 user: dan tags: pending) | |
20:45 | Add test cases to this branch. (Closed-Leaf check-in: f37317d8 user: dan tags: tkt-61c853-A) | |
20:41 | • Ticket [d52a29a9] Unexpected affinity conversion in view status still Fixed with 4 other changes (artifact: 7c070101 user: drh) | |
19:44 | Fix the patch on this branch so that it works with sub-queries, as well as views. (check-in: 7480db30 user: dan tags: tkt-61c853-A) | |
19:34 | The second option for [61c853857f40da49]: In this mode, columns of VIEWs and subqueries that are formed by expressions have affinity BLOB rather than affinity none, as has usually been the case for a while. But this mode fixes a couple of corner cases involving query flattening and the push-down optimization where that rule was violated. (Closed-Leaf check-in: 470ac8d5 user: drh tags: tkt-61c853-B) | |
19:32 | One of two options on how to address ticket [61c853857f40da49]. In this mode, we back out the documentation change of [07b7749da88d54e5] and change the core to work as it has been documented to work since 2017, rather than how it has actually worked since 2009. (check-in: 09cd0c0c user: drh tags: tkt-61c853-A) | |
18:01 | Refactor field Expr.affinity into Expr.affExpr to avoid confusion with other fields and variables named "affinity" and display affExpr it in sqlite3TreeViewExpr() output. (check-in: a29f2a7d user: drh tags: trunk) | |
16:22 | Improved detection of corruption on the freeblock list of a btree page. (check-in: 4b00799b user: drh tags: trunk) | |
15:09 | • New ticket [61c85385] Row is not fetched in SELECT from VIEW. (artifact: 82f66d84 user: mrigger) | |
13:19 | Fix a problem with renaming a table when a view or trigger within the schema uses a FILTER with an aggregate function that is not currently registered with the database. (check-in: 2ac0e42f user: dan tags: trunk) | |
13:06 | • Fixed ticket [d52a29a9]: Unexpected affinity conversion in view plus 8 other changes (artifact: ac513e2e user: drh) | |
12:55 | Prevent an fts5 table from being its own content table, or part of a view that is the content table. (check-in: b6d52c93 user: dan tags: trunk) | |
11:27 | • New ticket [d52a29a9] Unexpected affinity conversion in view. (artifact: c86e9f56 user: mrigger) | |
2019-08-04
| ||
09:26 | • Edit [1bd4b97d6405c9b0|1bd4b97d]: Edit check-in comment. (artifact: aba6b05a user: drh) | |
2019-08-03
| ||
19:06 | Strengthen an assert() in the WHERE clause code generator for the min/max optimization. (check-in: 1bd4b97d user: drh tags: trunk) | |
16:51 | • Closed ticket [71e183ca]: MIN() malfunctions for a query with ISNULL condition plus 4 other changes (artifact: 3dfda9a4 user: dan) | |
16:37 | Fix a problem with queries of the form "SELECT min(<expr>) ... WHERE <expr>=?" where there is an index on <expr>. Fix for [71e183ca]. (check-in: d465c3ee user: dan tags: trunk) | |
16:17 | In the ".wheretrace 0x100" debugging mode, show the structure of the main parameters to sqlite3WhereBegin() calls. (check-in: fd598e47 user: drh tags: trunk) | |
14:30 | • Ticket [71e183ca] MIN() malfunctions for a query with ISNULL condition status still Open with 4 other changes (artifact: afd8a0b2 user: drh) ... 1 similar event omitted. | |
01:40 | Add the SQLITE_TESTCTRL_PRNG_SEED which can control the PRNG seed either directly or through the schema cookie of a supplied database connection. (check-in: 2660e929 user: drh tags: trunk) | |
01:39 | Give the SQLITE_TESTCTRL_PRNG_SEED two arguments. The second argument if not NULL is a pointer to a database connection which seeds the connection from its schema cookie. In this way, fuzzers can control the PRNG seed. (Closed-Leaf check-in: 49aa3448 user: drh tags: prng-seed-test-control) | |
2019-08-02
| ||
22:32 | • New ticket [71e183ca] MIN() malfunctions for a query with ISNULL condition. (artifact: cb71fa4f user: mrigger) | |
21:03 | Fixes to the PRNG_SEED pragma idea. (check-in: c7109840 user: drh tags: prng-seed-test-control) | |
20:45 | Add the SQLITE_TESTCTRL_PRNG_SEED test control. (check-in: 3ac57231 user: drh tags: prng-seed-test-control) | |
19:40 | • Closed ticket [41866dc3]: MIN() malfunctions for UNIQUE column plus 6 other changes (artifact: 7206ad61 user: dan) | |
19:40 | If a query like "SELECT min(a), b FROM t1" visits no rows where "a" is not null, extract a value for "b" from one of the rows where "a" is null. Fix for ticket [41866dc37]. (check-in: faaaae49 user: dan tags: trunk) | |
18:43 | If a query like "SELECT min(a), b FROM t1" visits no rows where "a" is not null, extract a value for "b" from one of the rows where "a" is null. Possible fix for ticket [41866dc37]. (Closed-Leaf check-in: a7277ed0 user: dan tags: tkt-41866dc37) | |
16:11 | • New ticket [41866dc3] MIN() malfunctions for UNIQUE column. (artifact: 467863cb user: mrigger) | |
2019-08-01
| ||
22:48 | The sqlite3_set_authorizer() interface should only expire prepared statements when it is setting a new authorizer, not when clearing the authorizer. And statements that are running when sqlite3_set_authorizer() is invoked should be allowed to continue running to completion. (check-in: 961e2f08 user: drh tags: trunk) | |