SQLite

Timeline
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

651 check-ins using file tool/speed-check.sh version c24c30cd

2023-10-12
20:03
Fix a problem with an fts5 secure-delete on a rowid/term pair that follows a legacy delete of the same pair. (Leaf check-in: f5913e76 user: dan tags: branch-3.42)
2023-09-08
19:32
Fix a problem with fts5 secure-delete mode causing fts5 to corrupt its records. (check-in: cc0f82a4 user: dan tags: branch-3.42)
2023-09-02
22:07
Eliminate a redundant NK_DbAboutToClose notification. (Leaf check-in: eccffaa7 user: larrybr tags: cli_extension)
22:00
Issue NK_DbAboutToClose notifications for any shell DB about to be closed, whether presently visible in ShellExState or not. (check-in: 8e6e54e4 user: larrybr tags: cli_extension)
2023-07-26
16:41
Improve the output when ".scanstats vm" is enabled. (Closed-Leaf check-in: 7df08fd3 user: dan tags: scanstatus-exp)
2023-07-03
18:18
Enhance sqlite3AtoF() so that it honors the USELONGDOUBLE test-control. Enable the test/atof1.test module on non-intel platforms. (Closed-Leaf check-in: b7850fd4 user: drh tags: fp-to-decimal-refactor)
15:03
Improved precision of binary-to-decimal conversion of floating-point values on all systems but especially non-x86 systems that lake extended precision hardward floating point. (check-in: 32befb22 user: drh tags: trunk)
14:56
Whether or not to use the long-double floating point conversion technique is now a test-control setting. (check-in: 02ae6caf user: drh tags: fp-to-decimal-refactor)
12:58
New test cases for floating-point conversions. (check-in: 9435ed76 user: drh tags: fp-to-decimal-refactor)
11:44
Use Dekker double-precision floating point to do the value normalization step in sqlite3FpDecode() for systems on which "long double" is 8 bytes or less. (check-in: 51f492b4 user: drh tags: fp-to-decimal-refactor)
11:39
Final clean-up of the idea of using Dekker double-precision floating computations to improve accuracy. (Closed-Leaf check-in: 40285d63 user: drh tags: fp-to-decimal-branch1)
10:35
Squeeze as many bits as we can out of normalization. (check-in: 25d1ad6d user: drh tags: fp-to-decimal-branch1)
10:18
Faster Dekker multiplication that removes the restriction on input magnitude. (check-in: 2994caf5 user: drh tags: fp-to-decimal-branch1)
10:00
Fix one constant in the normalization logic. Improved error output from atof1.test. (check-in: d3c48807 user: drh tags: fp-to-decimal-branch1)
09:07
Replace the dodgy error estimating logic in the previous check-in with full-up Dekker double-double multiplication, and this idea works much better. There are still a few inaccuracies, but it is much closer. (check-in: 4fa6938d user: drh tags: fp-to-decimal-branch1)
00:51
The fptest(X) routine used for exploring another idea for improving fp-to-decimal conversion without the use of long double. (Leaf check-in: d9a883d7 user: drh tags: fp-to-decimal-branch2)
00:40
An idea for improving accuracy of fp-to-decimal conversion for systems that do not have "long double". Does not quite work. Retained only for documentation purposes. (check-in: 70224333 user: drh tags: fp-to-decimal-branch1)
2023-07-01
20:21
This check-in adds the use of "long double" to sqlite3FpDecode() which when run on x86 using a compiler that translates "long double" into binary80, causes all legacy tests to pass. But on a system where "long double" is just an alias for "double", some of the tests are still failing. (check-in: ce06982f user: drh tags: fp-to-decimal-refactor)
18:33
Reinstate the fpdecode() SQL function for testing, but only when compiled with SQLITE_DEBUG. (check-in: 07eab52e user: drh tags: fp-to-decimal-refactor)
17:56
Improved rounding policy. (check-in: 6f1122e9 user: drh tags: fp-to-decimal-refactor)
15:23
Improved response to Infinity and NaN. Remove the termporary fpdecode() function. (check-in: 76ab8ae8 user: drh tags: fp-to-decimal-refactor)
14:03
Progress towards fixing corner-case floating-point formatting issues. (check-in: 1138b8d1 user: drh tags: fp-to-decimal-refactor)
2023-06-30
23:42
Fix an issue with the "+" attribute in printf(). (check-in: bdfd1eb1 user: drh tags: fp-to-decimal-refactor)
23:18
Begin cutting over sqlite3FpDecode() into printf(). The code in this check-in gets a lot right, but there are still some formatting discrepencies. (check-in: c134d423 user: drh tags: fp-to-decimal-refactor)
19:41
Add decimal rounding to the sqlite3FpDecode() routine. (check-in: 27871140 user: drh tags: fp-to-decimal-refactor)
19:14
Add experimental ".scanstats vm" command to the shell tool. (check-in: e727640f user: dan tags: scanstatus-exp)
19:13
Fix an error in the previous commit on this branch. (check-in: 47c11ca9 user: dan tags: scanstatus-exp)
18:59
Add the "nexec" and "ncycle" columns to the bytecode virtual table. For accessing counters collected when SQLITE_DBCONFIG_STMT_SCANSTATUS is enabled. (check-in: f7b163a3 user: dan tags: scanstatus-exp)
18:35
Experiments with a new algorithm for converting ieee-754 binary64 numbers into decimal. (check-in: e923405e user: drh tags: fp-to-decimal-refactor)
18:31
Add support for sqlite3_stmt_scanstatus_v2() profiling of GROUP BY clauses that use a temp b-tree, and for sub-queries implemented as co-routines. (check-in: 7afad1f7 user: dan tags: trunk)
18:23
Improve support for sqlite3_stmt_scanstatus_v2() profiling of sub-queries implemented as co-routines. (Closed-Leaf check-in: 4e8718dc user: dan tags: scanstatus-exp)
17:14
Add support for sqlite3_stmt_scanstatus_v2() profiling of GROUP BY clauses that use a temp b-tree. (check-in: 796eadcc user: dan tags: scanstatus-exp)
16:16
Add extra debugging code for sqlite3_stmt_scanstatus_v2() to test1.c. (check-in: f936f101 user: dan tags: trunk)
14:01
Completely unwind the enhanced precision sum() from [a915f15a916af698] so as not to offend UBSAN and OSS-Fuzz. (check-in: 85be05b6 user: drh tags: trunk)
12:59
Omit the doubleToReal() function in vdbemem.c. Use the equivalent sqlite3RealToI64() function in its place. (check-in: 625820e8 user: drh tags: trunk)
11:51
Make the sum() function less precise and slower in order to avoid harmless signed integer overflow UBSAN warnings from OSS-Fuzz. (check-in: 1be0646a user: drh tags: trunk)
2023-06-29
23:03
Cleanup and commenting of the new DECIMAL extension code. No functional changes. (check-in: 51244816 user: drh tags: trunk)
20:28
Enhancements to the DECIMAL extension: (1) If the argument to decimal(X) is a floating point value (or an 8-byte blob), the floating point value is expanded into its exact decimal representation. (2) Function decimal_sci(X) works the same except it returns the result in scientific notation. (3) New function decimal_pow2(N) returns the full decimal expansion of the N-th integer power of 2. (check-in: 8baf8c10 user: drh tags: trunk)
17:58
Sync w/trunk (check-in: fe9aa2e9 user: larrybr tags: cli_extension)
17:48
Fix harmless compiler warnings about unused function arguments. (check-in: 24927c13 user: drh tags: trunk)
17:36
Improve the range of floating-point numbers that can be handled as integers for numeric-to-text conversion. (check-in: 41580ba4 user: drh tags: trunk)
17:26
Further refine the dtostr() testing function in the CLI so that it takes an optional second parameter which is the number of significant digits to display. (check-in: 2f9d4444 user: drh tags: trunk)
16:48
Change the (undocumented) dtostr() SQL function in the CLI so that it only shows the first 26 significant digits - 10 more digits than are available in a 64-bit double. (check-in: d758859f user: drh tags: trunk)
14:49
Add the decimal_sci(X) function to the Decimal extension for showing a decimal value in scientific notation: +D.DDDDDe+DD (check-in: 61d49239 user: drh tags: trunk)
12:14
CLI enhancements to facilitate SQLite core testing: (1) Add built-in functions strtod() and dtostr() that convert text to floating point and back using C-library routines. (2) Do not disable all of ".testctrl" without --unsafe-testing, but only those subcommands of .testctrl that are actually dangerous. (check-in: 669996a8 user: drh tags: trunk)
2023-06-28
12:02
Enhance the SUM() aggregate (and related AVG() and TOTAL()) so that the running sum is accurate to about 100 bits. (check-in: a915f15a user: drh tags: trunk)
11:46
Improved accuracy of floating-point conversion constants as suggested by forum post 598d32f6135c41c1. (check-in: 4943e8a1 user: drh tags: trunk)
2023-06-26
19:57
Simple test cases for the enhanced SUM() function. (Closed-Leaf check-in: c66ef244 user: drh tags: extended-precision-fp)
19:35
Use ideas from T. J. Dekker in "A Floating-Point Technique for Extending the Available Precision" (1971-07-26) to enhance the accuracy of the SUM() aggregate function in cases where input magnitudes vary wildly. (check-in: 439fc00f user: drh tags: extended-precision-fp)
17:23
Fix CLI .sha3sum error message for keyword identifiers, reported in forum post 7cbe081746dd. (check-in: bc27bda1 user: larrybr tags: branch-3.42)
17:08
Fix CLI .sha3sum error message for keyword identifiers, reported in forum post 7cbe081746dd. (check-in: 9d7813dd user: larrybr tags: trunk)
07:20
Replace some JS 'self' references with 'globalThis', as reported via the npm subproject. (check-in: b6ae8491 user: stephan tags: branch-3.42)
07:05
Replace some JS 'self' references with 'globalThis', as reported via the npm subproject. (check-in: d4e66dfc user: stephan tags: trunk)
2023-06-25
20:23
Fix typo in the README.md file. (check-in: 6d435d1d user: drh tags: trunk)
2023-06-23
11:10
Add support for the octet_length() SQL function, with low-level optimizations that avoid loading large string content from disk when the argument to octet_length() is a table column. (check-in: fa5f7786 user: drh tags: trunk)
2023-06-22
21:19
Optimize the argument to the octet_length() function so that it does not attempt to read content from disk. (Closed-Leaf check-in: 8b8ea4e3 user: drh tags: octet_length)
15:26
basic test cases for octet_length(). (check-in: 034a5fe2 user: drh tags: octet_length)
14:50
Merge the latest trunk enhancements into the reuse-schema branch. (check-in: 905c6c9d user: drh tags: reuse-schema)
14:40
Merge the latest trunk enhancements into the bedrock branch. (check-in: 0551b79b user: drh tags: bedrock)
14:20
Merge the latest trunk enhancements into the wal2 branch. (check-in: 80ac494f user: drh tags: wal2)
13:28
Merge recent trunk enhancements into the begin-concurrent branch. (check-in: 5bd26fea user: drh tags: begin-concurrent)
13:01
Merge recent trunk enhancements into the octet_length branch. (check-in: d35c2148 user: drh tags: octet_length)
01:03
Simplification to the getAndInitPage() routine that results in improved performance. (check-in: 2e9734c2 user: drh tags: trunk)
2023-06-21
21:47
Improved comment on the getAndInitPage() in btree.c. No code changes. (check-in: dc468cfd user: drh tags: trunk)
18:12
This is a failed attempt to optimize the usage of the page cache in sqlite3BtreeIndexMoveto() by deferring calls to sqlite3PagerUnref() for child pages of the btree until we are sure those pages will not be reused. The code does not work in two senses: (1) Many tests still fail. (2) The performance gain is less than 2 million cycles and does not seem worth the extra risk and complication. (Closed-Leaf check-in: 84731a1f user: drh tags: pcache-opt)
14:48
Sync w/trunk (for CLI -cachetrace) (check-in: a61f9377 user: larrybr tags: cli_extension)
14:11
Add the --pcachetrace option to the CLI. (check-in: 61dfa92b user: drh tags: trunk)
2023-06-20
17:45
Omit unnecessary calls to table locking routines in the common case when there is no shared cache. (check-in: f94f3021 user: drh tags: trunk)
16:26
Avoid unnecessary calls to sqlite3DeleteTable() with NULL pointers. (check-in: 4a35f391 user: drh tags: trunk)
15:48
Avoid unnecessary calls to sqlite3CodeRowTrigger(). (check-in: 27c174db user: drh tags: trunk)
11:46
Performance optimization in sqlite3VdbeAddOp4Int(). (check-in: 456a2432 user: drh tags: trunk)
2023-06-19
23:27
Avoid unnecessary NULL pointer checks on calls to sqlite3WalkExpr(). (check-in: 73d86a12 user: drh tags: trunk)
20:54
Small performance improvement and size reduction by recognizing that no SQL keywords have less than 2 characters. (check-in: 6b3d25b7 user: drh tags: trunk)
18:16
Add the sqlite3_commit_status() API. For querying a connection for (a) the ranges of wal/wal2 frames written after a successful commit, and (b) the conflicting frame following a failed commit of a BEGIN CONCURRENT transaction. (check-in: 4b08d4da user: dan tags: bedrock)
13:09
Fix large integer constants so that they work on older C compilers. (check-in: 3c94f878 user: drh tags: trunk)
2023-06-17
23:53
Cure tclshext build warning. Sync w/trunk. (check-in: e9f21191 user: larrybr tags: cli_extension)
15:42
Fix harmless compiler warnings that show up on 32-bit RaspberryPI builds. (check-in: bc4d20f3 user: drh tags: trunk)
15:22
In date/time functions, fix the rendering of "subsecond" and the computeHMS() routine to better deal with floating point rounding errors that arise on some 32-bit systems. (check-in: e46a00ae user: drh tags: trunk)
01:56
Sync w/trunk. (check-in: 746733c1 user: larrybr tags: cli_extension)
2023-06-16
19:45
Clear MSVC warnings and fix a typedef issue with MSVC build. (check-in: b0cb4cfc user: larrybr tags: cli_extension)
14:39
Address various harmless compiler warnings from forum post d526da8ee4. (check-in: 365caf2f user: drh tags: trunk)
11:05
Fix the pager such that if the xOpen() message sends back the SQLITE_OPEN_MEMORY flag, the pager is opened in journal-mode MEMORY, even if compiled with SQLITE_OMIT_DESERIALIZE. No changes to the logic as long as that OMIT flag is omitted. We need to better document the behavior of xOpen to describe this. (check-in: da1252b2 user: drh tags: trunk)
02:18
More doc tweaks, one teensy enhancement, and a new feature bug fix (check-in: 1073ef06 user: larrybr tags: cli_extension)
2023-06-15
16:02
Updates to the README.md file. (check-in: fa5cb6da user: drh tags: trunk)
15:43
Fix harmless typo in a comment. (check-in: 6937986a user: drh tags: trunk)
2023-06-14
13:45
Add SQLITE_DBCONFIG_RANDOM_SCANORDER. This option causes unconstrained table and index scan to happen in a random order, in order to help detect under-constrained queries in applications. Closed because this feature does not add any new capabilities that are not already available with SQLITE_DBCONFIG_REVERSE_SCANORDER, but it does make the behavior non-deterministic, and hence more difficult to test. (Closed-Leaf check-in: 33021986 user: drh tags: random-scanorder)
12:19
Work around what appears to be a GCC 32-bit optimization problem in the computeHMS() routine of the date/time logic. (check-in: aebdbcbe user: drh tags: trunk)
2023-06-13
23:04
Improve and add docs for extensible shell. (check-in: 1a57af86 user: larrybr tags: cli_extension)
18:10
Improved sqlite3_error_offset() values for bare column errors on '*' and 'table.*' expressions in SELECT statements. (check-in: 118fe600 user: drh tags: trunk)
16:55
Extended the time-diff format so that it can have a 5-digit year. This enables time differences to span the full range of dates supported by SQLite. Also fix some inaccuracies in the computation of time differences and add new test cases. (check-in: d57ddbf4 user: drh tags: trunk)
2023-06-12
18:22
Initialize a local variable in FTS3 to avoid a GCC compiler warning. (check-in: 7ec4ab32 user: drh tags: trunk)
18:11
Finish .parameter improvements. (no set <cast_op> anymore, not useful enough for its apparent complexity) (check-in: bbbdb864 user: larrybr tags: cli_extension)
14:26
The decision was made to halt work on this enhancement. See the discussion beginning with forum post 4c8a673560d7999a. The changes on this branch work, but they add unnecessary user complexity with no corresponding benefit. (Closed-Leaf check-in: c8ef3b78 user: drh tags: newbie-safe-aggregates)
14:16
Fix a buffer over-read that could occur as part of an fts5 secure-delete operation on a database containing corrupt fts5 records. dbsqlfuzz a463eac437783bcd246cc675b87842618a569cae (check-in: 44fdc798 user: drh tags: branch-3.42)
14:12
Make sure temporary sqlite3_value objects used by STAT4 are reset to NULL prior to being reused, to prevent a memory leak associated with the custom xDel string value destructor. dbsqlfuzz 65209ab5d5730e4fd14fff9d024325a77239036e. (check-in: 2cf197a2 user: drh tags: branch-3.42)
14:09
Defer setting the BTS_PAGESIZE_FIXED flag until after we are certain there are no errors and the page size has been set. Forum post b901808856. (check-in: cefaf128 user: drh tags: branch-3.42)
14:03
Extra space to prevent a buffer overread on corrupt STAT4 records. dbsqlfuzz 7128d1b41ce9df2c007f9c24c1e89e2f1b2590ca. (check-in: 566c4c14 user: drh tags: branch-3.42)
13:57
Simplify a memcpy() in defragmentPage(). It now might copy more content than is strictly necessary, but runs faster and uses less code space. Possible reasons for the improved performance: (1) the copy is now always 8-byte aligned, (2) fewer intermediate results are required which means less register pressure which helps the compiler to optimize the subroutine. (check-in: 1292d676 user: drh tags: branch-3.42)
13:53
Fix CLI fault on missing -nonce reported by forum post f8c14a1134. (check-in: 67cf7c40 user: drh tags: branch-3.42)
13:49
Fix a problem causing bloom filters to be used less efficiently than they might be. Problem reported by forum post 06c3f01da6. (check-in: 56d9bb7a user: dan tags: trunk)
13:46
Fix a problem causing an infinite loop in the recovery extension when processing a corrupt free-block list. (check-in: d21c06b7 user: drh tags: branch-3.42)
13:43
Fix an issue with sqldiff reported by forum post 9bd2155bdfae25a7. (check-in: 50721074 user: drh tags: branch-3.42)
13:31
Check-in [70e12efd1b189560] left a path through lockBtree() that failed to set the BTS_PAGESIZE_FIXED flag. That omission is fixed by this check-in. dbsqlfuzz 02c9ab74eb0619c9ae3b894ce157fae1d7f94645 (check-in: 30dcc872 user: drh tags: trunk)
11:06
Fix a buffer over-read that could occur as part of an fts5 secure-delete operation on a database containing corrupt fts5 records. dbsqlfuzz a463eac437783bcd246cc675b87842618a569cae (check-in: a35c50ec user: dan tags: trunk)
2023-06-11
22:05
Merge the latest trunk enhancements into the newbie-safe-aggregates branch. (check-in: 75a877a9 user: drh tags: newbie-safe-aggregates)
21:58
Make sure temporary sqlite3_value objects used by STAT4 are reset to NULL prior to being reused, to prevent a memory leak associated with the custom xDel string value destructor. dbsqlfuzz 65209ab5d5730e4fd14fff9d024325a77239036e. (check-in: 88be7075 user: drh tags: trunk)
2023-06-10
19:33
Defer setting the BTS_PAGESIZE_FIXED flag until after we are certain there are no errors and the page size has been set. (check-in: 70e12efd user: drh tags: trunk)
18:40
Fix an assert that can go bad if STAT4 content is corrupt. This is a follow-up to the previous check-in. (check-in: ac1d3860 user: drh tags: trunk)
17:27
Add documentation for SQLITE_DBCONFIG_STRICT_AGGREGATE. (check-in: 76fb9151 user: drh tags: newbie-safe-aggregates)
17:15
Merge the latest trunk enhancements into the newbie-safe-aggregates branch. (check-in: e19e20fb user: drh tags: newbie-safe-aggregates)
17:05
Extra space to prevent a buffer overread on corrupt STAT4 records. dbsqlfuzz 7128d1b41ce9df2c007f9c24c1e89e2f1b2590ca. (check-in: b9913528 user: drh tags: trunk)
10:37
Enhance fuzzcheck so that it understand the sqlite3_db_config() and optimization settings from unused bits in the database header, which dbsqlfuzz is now using. (check-in: 918708c6 user: drh tags: trunk)
09:24
Fix a potential UAF in the parser caused by the changes on this branch. (check-in: c29ec769 user: drh tags: newbie-safe-aggregates)
08:19
Add CFLAGS_FIDDLE to ext/wasm/fiddle.make as a way to pass on additional compilation flags to the fiddle build from the CLI, e.g. CFLAGS_FIDDLE=-DSQLITE_STRICT_AGGREGATE. Expose SQLITE_DBCONFIG_STRICT_AGGREGATE to JS. (check-in: 0b017208 user: stephan tags: newbie-safe-aggregates)
2023-06-09
19:24
Naming consistency: Always use the singular "STRICT AGGREGATE", not the plural. (check-in: 275d056e user: drh tags: newbie-safe-aggregates)
18:58
Improved sqlite3_error_offset() values for bare column errors on '*' and 'table.*' expressions in SELECT statements. (check-in: 446ad162 user: drh tags: newbie-safe-aggregates)
18:30
The -DSQLITE_STRICT_AGGREGATES compile-time option enables SQLITE_DBCONFIG_STRICT_AGGREGATE by default. (check-in: cd1f6b27 user: drh tags: newbie-safe-aggregates)
15:54
Simplify a memcpy() in defragmentPage(). It now might copy more content than is strictly necessary, but runs faster and uses less code space. Possible reasons for the improved performance: (1) the copy is now always 8-byte aligned, (2) fewer intermediate results are required which means less register pressure which helps the compiler to optimize the subroutine. (check-in: 6e5607ae user: drh tags: trunk)
13:08
Experimental change to add a per-connection option that raises an error if a bare column appears in an aggregate query. (check-in: f5878911 user: drh tags: newbie-safe-aggregates)
06:04
More .parameter improvements, a WIP for ".parameter set <cast_op> ..." (check-in: 2b51dcda user: larrybr tags: cli_extension)
01:11
Sync w/trunk. Fix spelling nits. Add ".parameter edit ..." improvements. (check-in: d73f065b user: larrybr tags: cli_extension)
2023-06-08
20:49
Remove SQLITE_EXTENSION_INIT macros from dbdata.c. (check-in: 106ec745 user: dan tags: trunk)
16:19
Improvements to the src-verify.c utility program so that it continues to muddle through a corrupt manifest file, making the most sense of it that it can and reporting "manifest" has having changed at the end. (check-in: 157b5d25 user: drh tags: trunk)
15:30
Fix the amagamation generator so that it works one a clean source tree. (check-in: f502610d user: drh tags: trunk)
15:27
Include source provenance in the header comment of the amalgamation. (check-in: e3eb5af9 user: drh tags: trunk)
12:52
Enhance Lemon so that it remembers which -D command-line options are actually used in the grammar and includes a list of all such options in the header of the generated output file. (check-in: c47a4dbd user: drh tags: trunk)
2023-06-07
23:10
Add spell-checking for all comments in main sources. (./src/ .c .h .in) make target "misspell" tells of misspelled words for which no exceptions have been put into tool/custom.txt . (check-in: 5332abf0 user: larrybr tags: trunk)
17:03
Fix straggler misspellings and tidy the custom dictionary. Also include pickups from forum post c61fb09afd. (Closed-Leaf check-in: 8c291d99 user: larrybr tags: spell-check)
15:00
Remove additional traces of FTS1 and FTS2 that were missed by [2bb50d5aedef0fd2] (check-in: 9a12bac5 user: drh tags: trunk)
14:47
Remove unused, legacy shell script "mkso.sh" from the root of the repo. (check-in: 57f75e48 user: drh tags: trunk)
08:40
Add a C-source spell-checking facility. make misspell (on Nix) (check-in: 26c1bb4b user: larrybr tags: spell-check)
2023-06-06
01:07
More ExportHandler use. Sync w/trunk. Fix .archive UB. (check-in: 4fd4240f user: larrybr tags: cli_extension)
2023-06-05
17:12
Fix compiler warning in src-verify.c. (check-in: 2d58c774 user: drh tags: trunk)
13:57
Update the README.md file to talk about the verify-source makefile target. (check-in: aa8e79e1 user: drh tags: trunk)
11:30
Remove extraneous console trace output from previous check-in. (check-in: c9f6e36d user: stephan tags: trunk)
11:26
Add a JS test confirming that binding of statement parameters in a subquery works. (check-in: 5dfaf0bc user: stephan tags: trunk)
01:05
Generalize the src-verify.c utility program and provide additional documentation in the header comment. (check-in: 7ed84046 user: drh tags: trunk)
00:57
Minor cosmetic normalizations for the MSVC makefile. (check-in: c9fda8d6 user: mistachkin tags: trunk)
2023-06-04
22:59
Fix harmless compiler warnings in tool/src-verify.c (check-in: 0e79ee97 user: drh tags: trunk)
22:56
Enhancement to tool/src-verify.c to also check the manifest.uuid file. (check-in: a39a5694 user: drh tags: trunk)
22:41
Add a prototype implementation for the "verify-source" Makefile target. (check-in: ed876ff6 user: drh tags: trunk)
2023-06-03
11:22
Prototype implementation of the octet_length() SQL function. (check-in: 2db989c8 user: drh tags: octet_length)
2023-06-02
21:31
Clarify sqlite3_is_interrupted() return. No code change. (check-in: b6c2d3d2 user: larrybr tags: trunk)
20:27
Get built-in ExportHandler instances to do more of .mode work. Replace atexit() with something that works for (not quite) "main" returns. Sync w/trunk. (check-in: a588dd9f user: larrybr tags: cli_extension)
18:05
Restore the LEFT JOIN strength reduction that was partially lost in [d747afda5683ca5f] when strength reduction was generalized to all OUTER JOINs. (check-in: 2d962b84 user: drh tags: trunk)
13:31
Do that last CLI fix more simply, using established means. (check-in: a1823b29 user: larrybr tags: trunk)
12:56
Fix CLI fault on missing -nonce reported by forum post f8c14a1134. (check-in: cd24178b user: larrybr tags: trunk)
00:03
Fix the OUTER JOIN strength reduction theorem prover's handling of the CASE operator. (check-in: 73d7b14b user: drh tags: trunk)
2023-06-01
20:38
Generalize the LEFT JOIN strength reduction optimization so that it works for RIGHT and FULL JOIN as well. Rename it to the "OUTER JOIN strength reduction" optimization. (check-in: d747afda user: drh tags: trunk)
16:40
If the filename argument to sqlite3_load_extension() is an empty string, then raise an error. It turns out that if dlopen() is called with an empty filename, it tries to load the current executable. And then if the caller requests some other function from the current executable, mischief can result. (check-in: d0168855 user: drh tags: trunk)
00:28
Add an ALWAYS() on an unreachable branch. (check-in: a00928d4 user: drh tags: trunk)
00:01
Fix the LEFT JOIN strength reduction for IN operators in the WHERE clause. Further simplifications and refinement of the algorithm. (check-in: 96c72dde user: drh tags: trunk)
2023-05-31
18:52
Add a comment to explain why both sides of an AND must be non-null-row in order for the overall expression to be non-null-row. No code changes. (check-in: 8396032c user: drh tags: trunk)
18:35
Improved detection of when the LEFT JOIN strength reduction optimization can be applied. (check-in: f544a8e4 user: drh tags: trunk)
12:42
Minor enhancement to the sqlite3_trace_v2() documentation. No code changes. (check-in: fba97230 user: drh tags: trunk)
11:03
Fix a problem causing an infinite loop in the recovery extension when processing a corrupt free-block list. (check-in: 5a516c6c user: dan tags: trunk)
2023-05-30
16:40
Enhancements to date/time functions: (1) Add support for time-shift modifiers of the form (+|-)YYYY-MM-DD HH:MM:SS.SSS (2) Add the timediff(A,B) SQL function that returns the time-shift modifier sufficient to move B to A. (check-in: 574bf370 user: drh tags: trunk)
14:46
Simplifications to the new timediff() logic. (Closed-Leaf check-in: d6954259 user: drh tags: timediff)
14:10
Minor tweak to the OPFS async proxy to attempt to account for a spurious high-load console error message which was reported via the sqlite-wasm npm project. (check-in: a61467d1 user: stephan tags: trunk)
13:27
Minor internal JS doc fix. No code changes. (check-in: 68abee3c user: stephan tags: trunk)
11:51
New test cases for the timediff enhancements. (check-in: f14633ce user: drh tags: timediff)
11:13
Allow date/time modifiers of the form (+/-)YYYY-MM-DD without the following HH:MM:SS.SSS. (check-in: fa9237a8 user: drh tags: timediff)
02:50
Problems fixed. All tests are passing now. (check-in: cff293b0 user: drh tags: timediff)
02:16
Improvements to the timediff() function. Test cases added. Not all test cases are passing. (check-in: 2cf65181 user: drh tags: timediff)
2023-05-29
20:33
New date/time modifier: (+|-)YYYY-MM-DD HH:MM(:SS.SSS)? (check-in: 3832b8a9 user: drh tags: timediff)
18:01
Prototype implementation of a proposed "timediff(X,Y)" SQL function. (check-in: 054a1951 user: drh tags: timediff)
2023-05-27
20:08
Fix an issue with sqldiff reported by forum post 9bd2155bdfae25a7. (check-in: 736a79c1 user: drh tags: trunk)
19:34
Add extra tests for fts5, large rowids and "ORDER BY rowid DESC". No changes to code. (check-in: b5b5eaee user: dan tags: trunk)
00:03
CLI much closer to being callable. (fewer exit() calls) ^C interrupt response improved. Sync w/trunk. Some diagnostic code removed. Some refactoring for sake of FIDDLE build. Streamline main(), by moving code to more dedicated routines and otherwise. Make -quiet more useful. (check-in: bff3a9a8 user: larrybr tags: cli_extension)
2023-05-25
16:49
Add ability for the JS Worker1.exec() API to report the number of changes made to the caller, per request in forum post d0b19483642e20dd. (check-in: 6e79505d user: stephan tags: trunk)
2023-05-23
19:11
Minor cleanups in demo JS code, per forum feedback. (check-in: 80c7c036 user: stephan tags: trunk)
17:43
Avoid performing aggregate processing more than once for a single expression. (check-in: 1bc54971 user: drh tags: branch-3.42)
17:35
Avoid performing aggregate processing more than once for a single expression. Forum post 36ff78b2a3. (check-in: 367b4cc5 user: dan tags: trunk)
15:22
When converting an aggregate expression to AGG_COLUMN, be sure to remove the EP_Unlikely flag if it is present. (check-in: 20c8a924 user: drh tags: branch-3.42)
15:16
When converting an aggregate expression to AGG_COLUMN, be sure to remove the EP_Unlikely flag if it is present. Forum post fbfe330a2009d036. (check-in: 48e65f18 user: drh tags: trunk)
14:05
Fix a buffer overread in the recovery extension that might occur on 32-bit platforms. (check-in: ff4a9a2b user: dan tags: trunk)
11:47
Add test cases for the recovery extension. No changes to code. (check-in: cec49c7d user: dan tags: trunk)
2023-05-22
20:36
Avoid duplicate WIN32_LEAN_AND_MEAN #define. Fix a trivial nit likely never observed. (check-in: 81ffcf41 user: larrybr tags: trunk)
16:35
When a floating-point RTREE is presented with large integer constraints - integers that are too big to be represented exactly by a float - then take extra steps to ensure that all possibly relevant entries in the RTREE are returned, even in boundary cases. Fix for the problem identified by forum post da70ee0d0d. (check-in: bfd8d910 user: drh tags: trunk)
11:02
Fix a buffer overrun that could occur in fts5 when processing corrupt records. dbsqlfuzz 0c6d3451d115974bc27ebed9b14b7a8e13ea05f3. (check-in: 4891dbd9 user: dan tags: trunk)
2023-05-21
05:06
First serious stab at FIDDLE-build-ready. Fix a flub with absent -A option for some builds. (check-in: bfc36f0f user: larrybr tags: cli_extension)
02:48
CLI code cleanup and shuffle for clarity. Moderate refactoring to exploit commmon patterns. Extend OOM simulation for debug builds. Plug one leak. Fix trivial nit copied from trunk. (check-in: b64ef7f9 user: larrybr tags: cli_extension)
2023-05-19
18:26
Extend detection of the INSERT...RETURNING locking case to the DB.selectValue(s)/selectArray/selectObject() family of functions. Add tests for INSERT/UPDATE...RETURNING with those functions. (check-in: 58be212d user: stephan tags: branch-3.42)
18:23
Extend detection of the INSERT...RETURNING locking case to the DB.selectValue(s)/selectArray/selectObject() family of functions. Add tests for INSERT/UPDATE...RETURNING with those functions. (check-in: 3181c505 user: stephan tags: trunk)
17:59
Minor doc typo fix. (check-in: 4533be08 user: stephan tags: trunk)
17:52
oo1.Stmt.finalize() no longer throws, but instead returns the same as the C API. oo1.DB.exec() now triggers the INSERT...RETURNING locking failure as an exception via reset() instead of finalize(). Some code-adjacent internal API renaming for clarity's sake. (check-in: 15f105c7 user: stephan tags: branch-3.42)
17:50
oo1.Stmt.finalize() no longer throws, but instead returns the same as the C API. oo1.DB.exec() now triggers the INSERT...RETURNING locking failure as an exception via reset() instead of finalize(). Some code-adjacent internal API renaming for clarity's sake. (check-in: 4ee6b3aa user: stephan tags: trunk)
16:42
Improved documentation for sqlite3_reset(), in response to forum post a72bab3dea. (check-in: 1dfeb3dc user: drh tags: trunk)
16:39
sqlite3.oo1.Stmt.reset() and finalize() now throw if their underlying C-level APIs return non-0, in order to avoid silent failure in certain locking-related cases. (check-in: db36a9ef user: stephan tags: branch-3.42)
16:34
sqlite3.oo1.Stmt.reset() now throws if sqlite3_reset() returns non-zero, analog to [f23eb5c6d365]. (check-in: 487ae12c user: stephan tags: trunk)
16:25
CLI resmanage used wherever resources should not be leaked upon abrupt (OOM) termination. (check-in: 1687d123 user: larrybr tags: cli_extension)
16:12
Add a basic INSERT RETURNING test for the oo1.DB.exec() API. (check-in: d29d62cf user: stephan tags: trunk)
15:54
sqlite3.oo1.Stmt.finalize() now throws if sqlite3_finalize() returns non-zero. This is intended to address the INSERT RETURNING case covered in forum post 36f7a2e7494897df. (check-in: f23eb5c6 user: stephan tags: trunk)
12:46
Correct typo in an 'extern' decl name, reported in forum post 1d4342156439233b. (check-in: 3bb1d7b3 user: stephan tags: branch-3.42)
12:41
Correct typo in an 'extern' decl name, reported in forum post 1d4342156439233b. (check-in: 6ac18827 user: stephan tags: trunk)
11:48
Change the RTREE engine so that equality comparisons of floating point values are duplicated in the byte-code so that the comparisons are done using 64-bit floats and consistent results are always obtained. Forum post da61c4a1b5b4af19. (check-in: 4ec0f0a3 user: drh tags: trunk)
2023-05-18
23:42
CLI resmanage use to 100% by commands, 84% by LOC. WIP (check-in: 48f81d8f user: larrybr tags: cli_extension)
23:28
Fix incorrect b-tree root page reports from the enhanced PRAGMA integrity_check output introduced in version 3.42.0. (check-in: 0a0b7a2d user: drh tags: branch-3.42)
23:25
Fix incorrect b-tree root page reports from the enhanced PRAGMA integrity_check output introduced in version 3.42.0. (check-in: 9b538f62 user: drh tags: trunk)
23:16
Incorrect fix. (Closed-Leaf check-in: 81b1bce7 user: drh tags: mistake)
2023-05-17
19:23
Fast-path the btree transaction start for the common case where a transaction is already active. (check-in: 798c3ff2 user: drh tags: trunk)
16:13
Simplification, performance improvement, and size reduction in the codeAllEqualityTerms() function of the code generator. (check-in: 4fbe0265 user: drh tags: trunk)
15:46
New assert() statements to verify that Expr.iColumn is never used as an array index when its value is negative. (check-in: 6084c5fb user: drh tags: trunk)
15:22
More CLI resmanage use (75% LOC) (check-in: 3428ce82 user: larrybr tags: cli_extension)
13:21
Improved assertions on substExpr(). (check-in: 6dbd5b3c user: drh tags: trunk)
11:45
Avoid an assert() failure that could occur in SQLITE_ENABLE_UPDATE_DELETE_LIMIT builds when a WITH clause attached to an UPDATE or DELETE statement created a CTE of the same name as the table being modified. (check-in: 60f03e32 user: drh tags: branch-3.42)
11:31
Avoid an assert() failure that could occur in SQLITE_ENABLE_UPDATE_DELETE_LIMIT builds when a WITH clause attached to an UPDATE or DELETE statement created a CTE of the same name as the table being modified. (check-in: 8edf56d1 user: dan tags: trunk)
03:57
Further CLI resmanage use. (69% by LOC, 53% by command count) Some renames for clarity. (check-in: c010e1a7 user: larrybr tags: cli_extension)
00:47
Resolve the JS-side corner case reported in forum post 7774b773937cbe0a by not caching oo1.Stmt.columnCount. (check-in: adee1440 user: stephan tags: branch-3.42)
00:32
Increase the version number to 3.42.1 (check-in: da0c662b user: drh tags: branch-3.42)
00:29
Fix two assert() statements that failed to take into account the likely() built-in function. This does not impact production code. (check-in: 634fe4c2 user: drh tags: branch-3.42)
00:26
Fix two assert() statements that failed to take into account the likely() built-in function. This does not impact production code. (check-in: 8504fe81 user: drh tags: trunk)
2023-05-16
20:37
Further extend use of CLI resmanage. (40% of dot commands) (check-in: 0cdab9f1 user: larrybr tags: cli_extension)
14:45
Use CLI resmanage to ensure input redirection is unwound on abort. Remove unused resmanage feature. Extend no-leak-on-abort fraction. Detect OOM in .cd . (check-in: 24308527 user: larrybr tags: cli_extension)
14:20
Increase the version number to 3.43.0 in order to begin the next development cycle. (check-in: 7874023c user: drh tags: trunk)
14:18
Resolve the JS-side corner case reported in forum post 7774b773937cbe0a by not caching oo1.Stmt.columnCount. (check-in: 37d44e8c user: stephan tags: trunk)
13:11
Merge all 3.42.0 release enhancements into the reuse-schema branch. (check-in: d67b6c39 user: drh tags: reuse-schema)
13:04
Merge all 3.42.0 release changes into the bedrock branch. (check-in: 1348c2a5 user: drh tags: bedrock)
13:00
Merge all 3.42.0 release changes into the wal2 branch. (check-in: 137057f9 user: drh tags: wal2)
12:56
Merge all 3.42.0 release changes into the begin-concurrent branch. (check-in: 804bb138 user: drh tags: begin-concurrent)
12:36
Version 3.42.0 (check-in: 831d0fb2 user: drh tags: trunk, release, version-3.42.0)
12:34
Update the tool/mkmsvcmin.tcl script so that it prints a one-line message to explain what it has done. (check-in: 4254f086 user: drh tags: trunk)
2023-05-15
23:43
CLI debug build ready for testing whether OOM handling is well-behaved. (check-in: 3cec1488 user: larrybr tags: cli_extension)
21:33
Fix CLI memory leak upon loading any shell extension. (check-in: b91cec47 user: larrybr tags: cli_extension)
19:17
Fix a C++-style variable declaration in the generate_series() extension. (check-in: 1d3e0089 user: drh tags: trunk)
17:24
Limit the number of nested NOT nodes in an fts5 expression to 256. (check-in: 01219e69 user: dan tags: trunk)
17:20
Increase the stack size available to fuzzcheck to 8MB on Windows. (check-in: 2290dc68 user: drh tags: trunk)
17:14
Limit the number of nested NOT nodes in an fts5 expression to 256. (Closed-Leaf check-in: 0e5c1ee4 user: dan tags: fts5-expr-limit)
10:55
Add the --wait command-line option to fuzzcheck. (check-in: 5a479a2c user: drh tags: trunk)
03:48
Make generate_series() correct on ones complement ALUs and acceptable to UBSAN. (check-in: 4c5cd3e6 user: larrybr tags: trunk)
02:06
As evidenced by forum post f3f546025a, the new RIGHT JOIN related restriction on the push-down optimization implemented by [da3fba18742b6e0b] also needs to apply to the automatic index (a.k.a. hash-join) optimization and to the Bloom filter optimization. Computation of the restriction is now moved into the sqlite3ExprIsSingleTableConstraint() routine. (check-in: 4902015d user: drh tags: trunk)
01:02
Simplify the interface to constructAutomaticIndex(). (check-in: c5da1655 user: drh tags: trunk)
2023-05-13
20:01
Fix windows-specific problems in test scripts sessionnoop2.test and sessionrowid.test. (check-in: e4989c11 user: dan tags: trunk)
19:38
Remove unnecessary call to sqlite3_dbdata_init() from shell.c. (check-in: 84ca2000 user: dan tags: trunk)
19:13
Add an "#ifndef SQLITE_HAVE_SQLITE3R" directive to shell.c.in, to avoid including the recovery code a second time if the shell is being compiled with sqlite3r.c. (check-in: 6f1a60fb user: dan tags: trunk)
16:54
Extend CLI resource manager usage. Improve interface to it for clarity and simplicity. (check-in: fd379f22 user: larrybr tags: cli_extension)
15:55
Add the "--unsafe-testing" option to invocations of the shell tool made by test script reuse5.test. (check-in: d6971af7 user: dan tags: reuse-schema)
15:34
Mention that sqlite3_mutex_free() accepts NULL as a no-op. (no code change) (check-in: 603d9ad5 user: larrybr tags: trunk)
15:21
Merge the latest 3.42.0 changes into the reuse-schema branch. (check-in: 515bb26e user: drh tags: reuse-schema)
15:17
Merge the latest 3.42.0 changes from trunk into the begin-concurrent branch. (check-in: 18c5e16b user: drh tags: begin-concurrent)
15:10
Merge the latest 3.42.0 beta changes into the bedrock branch. (check-in: d55ba8bb user: drh tags: bedrock)
15:07
Merge the latest 3.42.0 beta from trunk into the wal2 branch. For some reason, the previous merge didn't work. Probably a PEBKAC. (check-in: 3bbfbdcd user: drh tags: wal2)
15:00
Merge the latest 3.42.0 beta changes into the wal2 branch. (check-in: 771361e1 user: drh tags: wal2)
07:23
CLI closer to doing full cleanup on error exits. Needs testing with OOM simulation. (check-in: 8751f93f user: larrybr tags: cli_extension)
05:13
Fix compilation issue seen with MSVC. (check-in: 67001887 user: mistachkin tags: trunk)
2023-05-12
21:21
CLI resource management applied to about 30% of shell source. A few minor flaws fixed. Comments begin to describe OOM responses. (check-in: ce25a079 user: larrybr tags: cli_extension)
19:06
Fix harmless compiler warning caused by [0772ddf56713d013]. (check-in: f06c16a8 user: drh tags: trunk)
19:02
Add multi-threaded test for "INSERT ... RETURNING ..." and SQLITE_BUSY errors. (check-in: 559cb1c6 user: dan tags: trunk)
15:45
Make the regexp extension more robust against OOM errors during compilation of the recognizer engine. Forum post f50aecd5e8. (check-in: 0772ddf5 user: drh tags: trunk)
14:51
Fix unlikely strange result of CLI edit(). (check-in: ab3797e8 user: larrybr tags: trunk)
13:20
Ensure the _O_U16TEXT is defined for Windows builds of the CLI. Forum post c80aa942dc6bf2. (check-in: 504effa8 user: drh tags: trunk)
12:57
Remove a stray CR from end of line in a source file. (check-in: 48139fb9 user: drh tags: trunk)
12:45
Fix harmless typos in comments. (check-in: 607c3efb user: drh tags: trunk)
10:52
Ensure that all prepared statements are expired and forced to reprepare whenever the database text encoding changes. Forum post 43cbfc4d23. (check-in: 469718f1 user: drh tags: trunk)
2023-05-11
21:47
More CLI resource manager use. Catch more OOMs. Eliminate some opportunities to lose resources under OOM conditions. Conform to coding guidelines. (check-in: efdcf109 user: larrybr tags: cli_extension)
21:15
Add restriction (5) to the omit-noop-join optimization. Fix for the issue reported by [forum/forumpost/49f2c7f690|forum post 49f2c7f690]. (check-in: 3e9c9bbd user: drh tags: trunk)
18:03
Fix a minor typo in a comment. No code changes. (check-in: 607cfb5b user: drh tags: trunk)
17:25
Fix the rbu_exclusive_checkpoint=1 option so that it works with a zipvfs database in pass-through mode. (check-in: caca67d3 user: dan tags: trunk)
17:04
For CLI: Cure an old leak and one segfault. Get all shell tests to pass (by altering expected error output in some cases.) Ready now for full use of CLI resource manager. (check-in: 0f55868e user: larrybr tags: cli_extension)
03:11
WIP, CLI resource manager worked in, barely used yet. (Leaking one object.) (check-in: dbd00bca user: larrybr tags: cli_extension)
2023-05-10
21:08
Doc typo fix. No code changes. (Closed-Leaf check-in: e841b142 user: stephan tags: oo1-no-cache-Stmt.columnCount)
21:06
Resolve the JS-side corner case reported in forum post 7774b773937cbe0a by not caching oo1.Stmt.columnCount. (check-in: c3d25c3a user: stephan tags: oo1-no-cache-Stmt.columnCount)
16:04
Enhance the split-sqlite3c.tcl script so that it uses single-character extensions on subordinate source files, even when the number of subordinates exceeds 9. This is not yet needed, but might be soon. (check-in: 4150e416 user: drh tags: trunk)
13:56
Fix harmless compiler warnings from the CBMC compiler. (check-in: 5bc17cbc user: drh tags: trunk)
11:05
Rename sqlite3IsTableConstraint() to sqlite3IsSingleTableConstraint() and improve its header comment, in an attempt to make the code easier to reason about. No functional changes - should generate identical machine code. (check-in: 5dae8974 user: drh tags: trunk)
10:03
Add a missing comma to the documentation for the OP_Jump opcode. No changes to code. (check-in: 7e2c2b1c user: drh tags: trunk)
09:51
Begin use of a resource manager for CLI. (check-in: 1527d429 user: larrybr tags: cli_extension)
00:45
Get TCL extension to build and extensible shell tests to pass. (check-in: 05ecd8a5 user: larrybr tags: cli_extension)
00:33
New restriction on the push-down optimization to prevent an ON or USING clause on the left side of a RIGHT JOIN from being pushed down into a subquery that is on the right side of that RIGHT JOIN. Forum post a7d4be7fb6. This is similar to an earlier problem fixed at [1783655ea422185e]. Also add comments to describe previously undocumented push-down optimization restrictions. (check-in: da3fba18 user: drh tags: trunk)
00:20
Fix bug in fall-back logic of CLI .clone (check-in: 0dcd9785 user: larrybr tags: trunk)
00:15
Passes shell{1-8}.test . (check-in: 76734a4b user: larrybr tags: cli_extension)
2023-05-09
21:53
sqlite3x builds too. Fails a few shell2.test TCL tests. (check-in: 2749aea2 user: larrybr tags: cli_extension)
21:20
Made to build (again). (check-in: 3eb19495 user: larrybr tags: cli_extension)
2023-05-08
21:35
WIP, sync with trunk to pickup 22 weeks of shell library enhancements. (check-in: 00d0d5b1 user: larrybr tags: cli_extension)
16:26
Fix recover1.test so that it works with DEFAULT_AUTOVACUUM builds. (check-in: 4ffae48e user: dan tags: trunk)
13:47
Improved error handling when ".open" fails in the CLI. (check-in: 38544b11 user: drh tags: trunk)
13:38
In editPage(), if an attempt is made to edit an empty 64KB page, then take the slow path through rebuildPage(), in order to maintain strict invariants in the editPage() code. Forum post 0f37727c0d. (check-in: a865e574 user: drh tags: trunk)
11:11
Ensure that the label for a JSON object really is a string and not an array or object that just happens to end with a string. dbsqfuzz 32f62ed8676d0a153dbe756aeae3202d3670fe50. (check-in: c335737c user: drh tags: trunk)
10:51
Add tests for recovering data to a lost-and-found table from a utf-16 db. No code changes. (check-in: 832f6e9f user: dan tags: trunk)
2023-05-07
03:23
It turns out that pagerExclusiveLock() can be called with the lock state already set to RESERVED if the SQLITE_FCNTL_PERSIST_WAL setting is set and a specific sequence of multiple journal mode changes occur. Enhance pagerExclusiveLock() to deal with this. Forum post 8130545bc6 (check-in: 2bb8d977 user: drh tags: trunk)
2023-05-05
20:22
Fix a comment in the JSON logic. No code changes. (check-in: 3f30ea5e user: drh tags: trunk)
19:36
Fix a buffer overrun that could occur when using the format() function to format a very small real value with the "," modifier. (check-in: 910e770a user: dan tags: trunk)
15:52
Reduce the maximum depth of nesting in json objects to 1000. (check-in: c7697a0d user: dan tags: trunk)
15:28
Do not attempt to run json502.test with SQLITE_OMIT_VIRTUALTABLE builds. (check-in: 66648506 user: dan tags: trunk)
14:16
Updates to compile-time option usage and especially to the compile-time option test script in tool/omittest.tcl. (check-in: fece588b user: drh tags: trunk)
14:14
Fix the pager2-3.1 test case so that it does not run if a required dependence (shared-cache) is omitted from the build. Fix for forum post 7a028538a7. (check-in: a77c8a78 user: drh tags: trunk)
11:47
Allow virtual tables to prepare writes to shadow tables in defensive mode from with xRelease() callbacks - even if the xRelease() is invoked from within sqlite3_finalize(). (check-in: a1710649 user: dan tags: trunk)
11:33
Improved formatting of comments on SQLITE_CONFIG_xxx defines as those comments are used by documentation. (check-in: 2cb840bc user: drh tags: trunk)
2023-05-04
20:28
Add the "subsecond" modifier to the date/time functions. (check-in: 6fcb5a94 user: drh tags: trunk)
20:19
Add some tests of subsecond modifier for date/time functions. Sync with trunk. (Closed-Leaf check-in: 6499ebff user: larrybr tags: subsec-modifier)
14:44
Enhance the format() function so that the "," modifier works for floating point numbers in addition to integers. (check-in: 83683e10 user: drh tags: trunk)
14:41
Merge latest wal2 changes into this branch. (check-in: b2e0800b user: dan tags: bedrock)
14:31
Update this branch with latest changes from trunk. (check-in: 49777032 user: dan tags: wal2)
13:07
Add support for the comma (,) modifier to %f formats in the format() function. (Closed-Leaf check-in: 7080e196 user: drh tags: comma-format)
11:29
Fix a bug in cursor hints that can cause references to tables that have not been opened. Cursor hints are intended for use by COMDB2 only and should not appear in production builds, so this should not be a factor for the vast majority of users. (check-in: d3370d59 user: drh tags: trunk)
2023-05-03
14:10
Add another assert() to FTS5 to demonstrate that a pointer is never NULL. (check-in: b8a84a1b user: drh tags: trunk)
13:57
Add an assert() to fts5_config to ensure that a potential OOM is being handled correctly. (check-in: fe9c2076 user: dan tags: trunk)
13:48
Protect a macro argument with parentheses in FTS5. (check-in: bc07fe51 user: drh tags: trunk)
13:33
Add a comment to the [0512f82a2cde7447] change to link it to its TH3 test case. (check-in: 1281d4d9 user: drh tags: trunk)
13:03
Enhancement to showwal such that it shows what the page hash should be on a hash mismatch. (check-in: 960dc970 user: drh tags: trunk)
11:27
Conform CLI double-quoted escaping to what its doc says. (check-in: 910535ba user: larrybr tags: trunk)
10:56
Fix a problem in test script changebatch1.test. (check-in: 98ecffdf user: dan tags: begin-concurrent)
07:30
In the zipfile extension, leave the catalog list in a consistent state when the last entry is removed from the list. forum post f03f1e4c5a5c9959. (check-in: 1b489d00 user: drh tags: trunk)
06:48
If the page size is wrong on the page1 content inside a WAL file, then running VACUUM could trigger an assert(). This check-in changes that assert() into a branch that returns SQLITE_CORRUPT. Forum post 4ef9b37d74d2d2e8 (check-in: 0512f82a user: drh tags: trunk)
06:38
Improved detection of excess recursion on arrays and objects in the JSON parser. Fixes a problem detected by dbsqlfuzz. (check-in: d40fd592 user: drh tags: trunk)
05:00
Do not overflow the Index.aSample[] array if the same index appears in the sqlite_stat4 table under multiple names because it is a WITHOUT ROWID primary key index. Forum post 537d8ab118df7edd (check-in: 9350a25a user: drh tags: trunk)
04:21
Do not even attempt to load the sqlite_stat4 table if the use of STAT4 is disabled using sqlite3_test_control(). (check-in: 0bf94c77 user: drh tags: trunk)
02:32
Fix CLI mishandling of OOM during .import _of_ a DB. Reported by forum post 7ed4a250d3. (check-in: 30da3f0e user: larrybr tags: trunk)
01:25
In the CLI, ensure that the argument to quoteChar() is not a NULL pointer due to a prior OOM. (check-in: 776d1a47 user: drh tags: trunk)
2023-05-02
21:26
Merge the latest trunk enhancements into the reuse-schema branch. (check-in: d08fa7d4 user: drh tags: reuse-schema)
21:10
Update this branch with latest changes from trunk. (check-in: 6f22c767 user: dan tags: begin-concurrent)
20:34
Fix a problem with the sqlite3changeset_size() API and rowid tables. (check-in: 92ade220 user: dan tags: trunk)
19:58
Merge the latest trunk enhancements into the wal2 branch. (check-in: 0215dcc7 user: drh tags: wal2)
19:33
Give CLI quoted args hex escapes (as documented.) (Closed-Leaf check-in: 90e434a6 user: larrybr tags: cli-hex-escape)
18:03
Cure CLI hang on .import with high-ASCII column separator. (check-in: aac8ac63 user: larrybr tags: trunk)
17:33
Allow for lower-case "nan" to mean NaN in JSON. (check-in: 6bee0a19 user: drh tags: trunk)
16:34
Interpret negative arguments to sqlite3_sleep() as zero. (check-in: 2b542326 user: drh tags: trunk)
11:12
Improved handling of NULL arguments to json_valid() and json_error_position(). Forum post 06c6334412. (check-in: ab78e694 user: drh tags: trunk)
10:22
Fix typo in comment. Forum post 3da7d9c445. (check-in: 41a8a15c user: drh tags: trunk)
00:33
Minor code change in the CLI to make a UAF warning from scan-build go away. (check-in: ad6aae76 user: drh tags: trunk)
2023-05-01
20:42
Fix harmless compiler warnings. (check-in: 342af5b4 user: drh tags: trunk)
20:09
The gcc definition of the SQLITE_INLINE macro is not compatible with string ANSI, so disable it if the -std=c89 command-line option is used. (check-in: 62d703d8 user: drh tags: trunk)
19:59
Fix harmless compiler warnings in FTS5. (check-in: f6210017 user: drh tags: trunk)
19:28
Remove xFileControl() handling from the OPFS VFS altogether, re-routing all syncing through xSync() instead. This eliminates superfluous syncing introduced in [a371374148a2], as discussed in forum thread 647d2f811dbc2dfe. (check-in: f809de7f user: stephan tags: trunk)
18:52
Fix harmless compiler warnings. (check-in: 0a926994 user: drh tags: trunk)
18:28
Add support for JSON5. (check-in: f8c3ed23 user: drh tags: trunk)
15:59
Fix typos in comments in sqlite3session.h preventing documentation from being correctly generated. (check-in: ab75170d user: dan tags: trunk)
15:57
New enhancements to the sqlite3_db_config() constants documentation. (check-in: d74011a3 user: drh tags: trunk)
15:42
Expose the SQLITE_DBCONFIG_STMT_SCANSTATUS and SQLITE_DBCONFIG_REVERSE_SCANORDER sqlite3_db_config() options to JS. (check-in: 0a7024af user: stephan tags: trunk)
15:15
Implement full xSync() for the OPFS VFS. The previous implementation was not correct for journal files. Reported in forum post 647d2f811dbc2dfe. (check-in: a3713741 user: stephan tags: trunk)
14:50
Fix typos in documentation comments in sqlite3.h. (check-in: 6a8b0078 user: drh tags: trunk)
11:24
Do not apply the "AND false" optimization if either operand comes from the ON clause of a join. Fix for the problem identified by forum post 96cd4a7e9e. (check-in: d095da0e user: drh tags: trunk)
03:56
Fix abbreviated paths for objects such that they work even if the object key begins with '$'. (Closed-Leaf check-in: 1b991c78 user: drh tags: json5)
2023-04-30
23:52
Fix a problem with json_patch() when one side or the other is JSON5. dbsqlfuzz bc10593a4ba8e7a7862593532285be31f00f8e41 (check-in: e18c0899 user: drh tags: json5)
20:37
Accept the full ECMAScript 5.1 IdentifyName syntax for keys in objects. (check-in: 9be2c875 user: drh tags: json5)
19:45
All JSON to understand floating point literals "Inf" and "QNaN" and "SNaN" in any case, without the SQLITE_EXTENDED_NAN_INF compile-time option. This extension is always available. (check-in: fb551145 user: drh tags: json5)
19:34
Omit the json_valid() function. Change the name of json_error() to json_error_position(). Use "NOT json_error_position(X)" as a substitute for "json_valid5(X)". (check-in: 34c4e900 user: drh tags: json5)
2023-04-29
18:40
Merge all the latest trunk enhancements into the JSON5 branch to facilitate testing of the JSON5 branch. (check-in: 30d12ede user: drh tags: json5)
18:31
Fix a crash that could occur in fts5 'secure-delete' mode when operating on corrupt records. (check-in: 2e85b0e3 user: dan tags: trunk)
17:35
More ALWAYS() macros. (check-in: 770b09f7 user: drh tags: json5)
17:13
Add ALWAYS macros on unreachable branches in the new JSON5 logic. (check-in: 91e15ed9 user: drh tags: json5)
16:31
Simplification of the logic that normalizes JSON5 integer literals into canonical JSON integer literals. Improved reporting of OOM. (check-in: 01ee613c user: drh tags: json5)
16:00
Do not allow leading zeros on non-zero numeric literals in JSON. (check-in: 3e914943 user: drh tags: json5)
15:42
Cure CLI generate_series() overflow bug (noted by forum post #754e2d4db2a5) and bring behavior with negative step arguments closer to as-documented and eponymous function in PostgreSQL. (check-in: 07383758 user: larrybr tags: trunk)
15:29
Replace duped generate_series() test with another corner case. (Closed-Leaf check-in: fb2f0878 user: larrybr tags: generate_series-revamp)
12:29
Sync w/trunk, improve generate_series() variable names. (check-in: ecbf7e13 user: larrybr tags: generate_series-revamp)
12:13
Minor fixes to to the JSON% parser. (check-in: 2fe684cd user: drh tags: json5)
00:59
Correctly recognize an isolated U+feff as a space character in JSON5. (check-in: 4473dc8e user: drh tags: json5)
2023-04-28
23:39
Add tests for generate_series(), fix a so bug found, and change an existing test that enforced varying rowid-to-value mapping with query ordering. (That no longer varies.) (check-in: 9605db4a user: larrybr tags: generate_series-revamp)
23:38
Bug fixes in the logic to skip over JSON5 comments. (check-in: c736b77a user: drh tags: json5)
23:19
Bug fix in the offset computation for json_error(). (check-in: 8f7ab5d9 user: drh tags: json5)
21:25
Revise generate_series() extension (in CLI) to address overflow reported in forum post #754e2d4db2a5 and to make behavior better match the like-named PostgreSQL function. (check-in: beeea3e1 user: larrybr tags: generate_series-revamp)
17:38
Improvements to the accuracy of json_error(). Add the extension SQL functions random_json(SEED) and random_json5(SEED). (check-in: 8d09dc1c user: drh tags: json5)
14:48
Add the json_error(X) function that returns the 1-based character offset to the first syntax error in JSON5 string X, or 0 if there are no errors. (check-in: 901ad995 user: drh tags: json5)
13:25
Fix indentation and omit trailing whitespace in the random JSON generator script. (check-in: 629db09f user: drh tags: json5)
11:02
Fix json_tree() so that it is able to deal with the JNODE_RAW labels of a JSON5 object. (check-in: f56528d4 user: drh tags: json5)
10:23
Merge all the latest trunk fixes and enhancements into the json5 branch. (check-in: b5ca15cf user: drh tags: json5)
10:10
Do not assert() a bad string representation in an sqlite3_value after an OOM. dbsqlfuzz c822a17a23c524a0ac7cfb203c7198209da15de8. (check-in: 91fee79a user: drh tags: trunk)
00:28
Make CLI .load slightly harder to abuse. (check-in: 2779f927 user: larrybr tags: trunk)
00:19
In the CLI, early out from the output_quoted_string() routine if the input string is NULL. forum post 8a22c0bfa7. (check-in: 2881978d user: drh tags: trunk)
2023-04-27
23:59
Use a new technique to detect fresh OOM faults in columnName() that does not rely on there being no OOMs prior to entry into columnName(), as [forum/forumpost/fb6811c2f9|forum post fb6811c2f9] demonstrates a technique which could cause an OOM prior to entry into columnName(). (check-in: a63346d6 user: drh tags: trunk)
23:44
Remove a faulty assert() from the CLI. Forum post 726c4f7db0. (check-in: 4bbebb6b user: drh tags: trunk)
23:40
Remove an ALWAYS() macro that can in fact sometimes be false. Forum post f0e872fcee. (check-in: d85880e4 user: drh tags: trunk)
23:29
Test cases added, and some bugs fixed. (check-in: bc84a82e user: drh tags: json5)
22:17
Fix compilation issues seen with older versions of MSVC. (check-in: 4ba2a651 user: mistachkin tags: trunk)
21:31
Adjust OSTRACE usage so that it works with the changes in the previous check-in. (check-in: 1eb4573e user: mistachkin tags: trunk)
19:30
Further optimizations to the JSON parser. (check-in: bb8f1c16 user: drh tags: json5)
19:27
Dynamically resize the node hash table used by the rtree module. (Leaf check-in: f94f3da5 user: dan tags: rtree-batch-insert)
19:13
Faster implementation of numberic literal parsing in JSON. (check-in: 58398292 user: drh tags: json5)
18:28
Add support for hexadecimal integer literals in JSON. (check-in: 85e00c9e user: drh tags: json5)
17:32
All floating point literals "NaN" and "Infinity". Additional variants of these literals are available if compiled with SQLITE_EXTENDED_NAN_INF. (check-in: c13346af user: drh tags: json5)
16:57
Fix handling of reverse solidus in string literals. Allow decimal points in floating point literals to occurs and the beginning or end of the mantissa. (check-in: d92a6ab2 user: drh tags: json5)
16:24
Fix the handling of escape solidus in the JSON routines. (check-in: 676877ac user: drh tags: json5)
15:48
JSON string literals may span multiple lines by escaping new line characters. (check-in: 66da4bd4 user: drh tags: json5)
15:11
Allow the labels on JSON objects to be unquoted identifier names. (check-in: fb428db3 user: drh tags: json5)
14:38
Performance optimization in the JSON parser. (check-in: 5a88ba74 user: drh tags: json5)
13:55
Fix a problem on windows sometimes causing SQLITE_NOMEM to be returned if an error occurred while opening a temp file. (check-in: c22a39a6 user: dan tags: trunk)
13:44
Permit JSON5 whitespace in all contexts of objects and arrays. (check-in: 93f3ab26 user: drh tags: json5)
12:24
Translate JSON5-only string literal escape sequences into the JSON equivalents. (check-in: 14e82f36 user: drh tags: json5)
2023-04-26
20:41
Fix some problems with the rtree module and existing tests on this branch. (check-in: 50e8a30b user: dan tags: rtree-batch-insert)
20:26
Implement some of the JSON5 enhancements to string and numeric literals. This is an incremental check-in of work in progress. (check-in: 9508efa9 user: drh tags: json5)
18:50
Fix a typo in a printf() format string the sessions module. (check-in: 14d1c4a8 user: dan tags: trunk)
18:23
When changing a large integer into a floating point value, cancel the string representation. Fix for forum post 5c74a3bc4a. (check-in: 3e2da8a7 user: drh tags: trunk)
17:30
Partial implementation of JSON5 numeric literal extensions. Use a switch() statement in the parser for better performance. (check-in: 78404dc3 user: drh tags: json5)
15:58
Improvement to the way the JSON performance measure scripts work → keep the test database in the directory of the test, not in the source tree. (check-in: ac411dbd user: drh tags: json5)
15:19
Work toward implementing JSON5 whitespace. Untested and incomplete. (check-in: d262c059 user: drh tags: json5)
13:52
Fix to the json-speed-check.sh script. (check-in: d839c954 user: drh tags: json5)
13:25
Add scripts for JSON performance testing. (check-in: 3051d6a7 user: drh tags: json5)
2023-04-25
21:24
The json_valid() function only returns true for pure JSON. JSON5 (or at least that subset of JSON5 that has been so far implemented) is accepted by all routines, but json_valid() still returns false for JSON5 inputs. The new json_valid5(X) routine returns true or false if X is or is not valid JSON5. All of this is experimental and subject to change. (check-in: 5d33ab77 user: drh tags: json5)
20:41
Allow the sessions module to be configured to capture changes from tables with no explicit PRIMARY KEY. (check-in: 5d4431bc user: dan tags: trunk)
14:54
Remove unused variable ltoj accidentally added by [1783655e]. (check-in: faa1575c user: dan tags: trunk)
14:37
Avoid assuming that an expression in an ON() clause that evaluates to zero implies that the query will return zero rows when the query contains a RIGHT JOIN. Forum Post 95849acbe1. (check-in: 1783655e user: dan tags: trunk)
04:28
Handle newline-trimmed input TEXT correctly in base64, base85 UDFs, an issue exposed in forum post 8d6ed893c0. (check-in: 8f637aae user: larrybr tags: trunk)
02:44
Check for OOM sqlite_value_x() returns in base64, base85 extensions, a partial response to forum post 74dd86263e. (check-in: e6f9c0b1 user: larrybr tags: trunk)
2023-04-24
23:14
Allow trailing commas in objects and arrays of JSON. (check-in: 4031b231 user: drh tags: json5)
21:08
Hold the RTREE node cache open and defer writes until the very end of the INSERT statement, for a 20x performance improvement on the test script identified at forum post 6b8bca17bb884ef3. However, there are still bugs and tests will crash. Follow-up: Performance improvements were not real. The bugs caused incorrect answers. In reality, no performance improvement seen. (check-in: b35cb745 user: drh tags: rtree-batch-insert)
19:23
Update the compile-time detection of architecture byte-order in the RTREE extension so that it is aligned with the latest enhancements in the core. (check-in: 122431d3 user: drh tags: trunk)
19:22
Allow the sessions module to be configured to capture changes from tables with no explicit PRIMARY KEY. (Closed-Leaf check-in: 8a612f08 user: dan tags: sessions-rowid-tables)
04:25
Add a note about the journaling mode in the OPFS VFS. No code changes. (check-in: e79c95fc user: stephan tags: trunk)
2023-04-23
00:12
CLI to have "undocumented" dot-commands, not usually shown by .help (check-in: 17f5dd2d user: larrybr tags: trunk)
2023-04-22
22:32
The sqlite_stat4.idx field is case insensitive. It should work even if some entries use a different case than others. Fix for forum post 6c118daad0f1f5ef. (check-in: f097ca70 user: drh tags: trunk)
20:47
Avoid a crash in fts5 when combining an empty expression with another. (check-in: 33681ff3 user: dan tags: trunk)
20:22
Enhance the ".tables" command in the CLI so that it is able to deal gracefully with an OOM situation. (check-in: 5157fd1c user: drh tags: trunk)
20:07
The assertion-fault fix in [53a61f7423a7f057] was not quite complete. This additional change enhances it to deal with WITHOUT ROWID tables. No changes to deliverable code. (check-in: 4d5af42c user: drh tags: trunk)
17:41
Add the --unsafe-testing command-line option to the CLI. Without this option, features of the CLI that are intended for testing and that might result in incorrect answers, assertion faults, and/or corrupt database files if misused are disabled. Fix for the unfounded concerns raised by 13 separate forum posts last night. (check-in: f28256a9 user: drh tags: trunk)
16:46
Add --unsafe-testing invocation option to CLI. Needs some tests added and changed. (Closed-Leaf check-in: b3d9ac05 user: larrybr tags: shell-for-test)
12:47
Fix an incorrect assert() statement in btree.c as found by forum post d03345d572713fe6. (check-in: cd485b30 user: drh tags: trunk)
12:37
Fix a incorrect assert() statement in the pre-update hook logic. No changes to production code. Fix for the problem reported by forum post 19b217bfe709a072. (check-in: 53a61f74 user: drh tags: trunk)
12:11
Fix harmless compiler warning in [1489e7f53a4863b3]. (check-in: da907dbc user: drh tags: trunk)
11:33
Also fix no-length string intolerance for CLI json mode. (check-in: 3ac19840 user: larrybr tags: trunk)
11:29
Early out from sqlite3Prepare() following an OOM to avoid possible problems further along in the parse. Fix for the NULL pointer dereference reported by forum post 2e5131839365682a. (check-in: f35ce7c1 user: drh tags: trunk)
11:24
CLI to handle absurd string length limit better. forum post 5180af725f1cc375 (check-in: 1489e7f5 user: larrybr tags: trunk)
09:26
Avoid need for cast in CLI shell_check_oom() calls. (check-in: c88550ef user: larrybr tags: trunk)
09:14
Fix unchecked return in CLI .dbi command. (Forum post 79e2cd1bc) (check-in: 81a74ee9 user: larrybr tags: trunk)
2023-04-21
15:37
Change the return type of the ts_read/write() family of functions from int to ssize_t, per report in forum post 947169d5e7. (check-in: 7809e7ce user: stephan tags: trunk)
15:30
Add a new modifier to date/time functions: "subsecond". May be abbreviated as just "subsec". This modifier causes functions to try to show fractional seconds if they do not already. (check-in: 03f2a15e user: drh tags: subsec-modifier)
2023-04-20
20:07
Increase CLI tolerance of artificially low memory. (check-in: 13f26386 user: larrybr tags: trunk)
16:15
CLI .ar made to enforce getting a proper subcommand. (check-in: 104ab105 user: larrybr tags: trunk)
10:40
Fix a function declaration so that it correctly appears as private in the amalgamation. Forum post ec8c77516c. (check-in: 1864569b user: drh tags: trunk)
10:26
Fix a broken assert() in the recovery extension. Forum post 82a88f5812. (check-in: 8a1ad137 user: dan tags: trunk)
2023-04-19
20:29
Add tests for the progress-handler callback with the fts5 'secure-delete' feature. (check-in: 48505ad9 user: dan tags: trunk)
18:36
Improved rebustness of the pcache tracing logic. (check-in: 0a43235b user: drh tags: trunk)
18:32
Remove unreachable legacy code. (check-in: e3e7fb87 user: drh tags: trunk)
17:07
Fix a segfault that could occur if a non-empty in-memory database was the destination of a backup operation from a database with a smaller page size. Forum post 679b4de86e763d52. (check-in: 020968f8 user: dan tags: trunk)
15:35
Do not remove pages from the cache of an in-memory database due to a failure to acquire the page due to it being larger than the maximum page size. Fix for forum post a19bb49140. (check-in: 982b3556 user: drh tags: trunk)
14:26
Fix ambiguity in the header comment to the sqlite3PagerUnrefNotNull() routine. No code changes. (check-in: d419e65e user: drh tags: trunk)
13:56
Fix a debugging printf() in pcache.c. This only comes up on custom builds. (check-in: f2d67567 user: drh tags: trunk)
13:39
Correct use of an undefined symbol in one of the OPFS VFS loading failure error reporting cases (copy/paste error). (check-in: c81c968b user: stephan tags: trunk)
13:30
Add pager debug tracing of truncate operations. (check-in: 0daadf36 user: drh tags: trunk)
12:08
Attempt to suppress harmless compiler warnings reported by forum post fc98845c06. (check-in: f6e6c5f9 user: drh tags: trunk)
2023-04-18
23:05
CLI to cease dumping back to OS shell on single ^C when interactive. (check-in: ba8e4378 user: larrybr tags: trunk)
22:53
Add comment on CLI ^C fix (because it is a bit subtle.) (Leaf check-in: 19c062f1 user: larrybr tags: cli-no-dump)
22:28
CLI to cease dumping back to OS shell on single ^C when interactive. (check-in: 7b1f6352 user: larrybr tags: cli-no-dump)
20:02
Make use of F_BARRIERFSYNC as an alternative to F_FULLFSYNC when it is available. (Leaf check-in: 3c517ba5 user: drh tags: barrier-fsync)
15:21
Ensure that the VACUUM command is not confused by alternative encodings when it is the first command run on new database connection. Forum post 09503b4d33. Problem introduced by check-in [a02da71f3a80dd8e]. (check-in: 8b0fe63f user: drh tags: trunk)
14:13
Another fix to the indexed expressions in aggregate queries with GROUP BY enhancement of ticket [99378177930f87bd] and implemented by check-in [b9190d3da70c4171] to address a problem described by forum post f34e32d120, (check-in: 5acc3ef8 user: drh tags: trunk)
11:35
Add the --fullsync option to speedtest1. (check-in: 960fe54c user: drh tags: trunk)
11:33
Fix an assert() in fts5 that could fail following an OOM or IO error. (check-in: 49595083 user: dan tags: trunk)
2023-04-17
23:59
When CLI -utf8 active, do no translation to MBCS for output to Windows console. plus code style improvements (check-in: 25edf608 user: larrybr tags: trunk)
22:28
Minor coding style changes. (Leaf check-in: 481aa3cc user: mistachkin tags: cli-utf8)
21:18
When CLI -utf8 active, do no translation to MBCS for output to Windows console. (check-in: cc1d4296 user: larrybr tags: cli-utf8)
18:32
Fix some problems with using fts5 options 'secure-delete' and detail=none together. (check-in: 4d3f27ba user: dan tags: trunk)
11:41
Fix problems handling corrupt records in new fts5 code. (check-in: 2f63d4a8 user: dan tags: trunk)
10:39
Fix an assert() in fts5 that may be true when dealing with corrupt records. (check-in: e8de2774 user: dan tags: trunk)
2023-04-15
21:00
Change a memcpy() to memmove() in new fts5 code. (check-in: 8de45425 user: dan tags: trunk)
19:58
Fix a use-after-free that could follow an OOM or IO error in fts5 when using 'secure-delete' mode. (check-in: 904be83d user: dan tags: trunk)
19:13
Avoid dropping the error code following an OOM in fts5 when using 'secure-delete' mode. (check-in: 9ee24a49 user: dan tags: trunk)
17:56
Add -utf8 option to CLI, for improved console behavior on Windows. (check-in: 414010d2 user: larrybr tags: trunk)
17:47
Add the 'secure-delete' option to the fts5 extension. For configuring fts5 to delete old entries directly from the full-text index instead of using delete keys. (check-in: 394980e4 user: dan tags: trunk)
16:12
For CLI -utf8, set output codepage too. Adjust PP vars so that the code can be entirely omitted for targets pretending to be "WIN32" but not quite doing so. (check-in: 543594a7 user: larrybr tags: cli-utf8)
2023-04-14
21:34
Fix stray edits in comments in last check-in. (check-in: fa3ce7f0 user: larrybr tags: cli-utf8)
21:23
Set CLI -utf8 option and build with line-editing package to be mutually exclusive. Integration of console-invasive UTF-8 handling with line-editing takeover of console may come later. (check-in: 047344a9 user: larrybr tags: cli-utf8)
19:56
Cure CLI double-prompting (by ditching gcc fgetws()), general cleanup. Work remaining is to avoid effect of -utf8 when a line editor is linked/used as part of CLI. (check-in: 73a5f542 user: larrybr tags: cli-utf8)
18:00
Further tests for fts5 secure-delete mode. (Closed-Leaf check-in: c5a47063 user: dan tags: fts5-secure-delete)
17:00
Add extra OOM test for the new code on this branch. (check-in: 846ae7e0 user: dan tags: fts5-secure-delete)
16:11
Add tests for the new feature on this branch. (check-in: 0268d1a5 user: dan tags: fts5-secure-delete)
14:36
Fix SQLITE_OMIT_VIRTUALTABLE builds of testfixture. (check-in: 430ec1a6 user: dan tags: trunk)
13:34
Update test script recoverbuild.test so that it may be run using old test harnesses that run more than one test script in a process. (check-in: 5135aab0 user: dan tags: trunk)
10:40
When setting the column types on a subquery, ensure that the COLFLAG_HASCOLL flag is cleared from column names from when the collating sequence name has been removed. Forum post 6916dacf83. (check-in: 8d9dcd7c user: drh tags: trunk)
10:35
Fix a cosmetic indentation issue. (check-in: 90deb844 user: drh tags: trunk)
00:20
Avoid double de-quoting of table names when processing RESTRICT actions in foreign key constraints. Chromium 1405220. Problem introduced by [9e503e2d0428c9e8] 2009-09-28 for version 3.6.19. (check-in: bb2b5ab1 user: drh tags: trunk)
2023-04-13
18:44
Fix an obscure issue with ALTER TABLE RENAME that comes up with triggers that have UPDATE statements that contain errors. Forum post ff3840145a. (check-in: c4845a7c user: drh tags: trunk)
14:53
Add extra test cases to window1.test. (check-in: ebc844fb user: dan tags: trunk)
14:50
Fix a code-generator issue associated with very unusual use of window functions. Both the expr.c or the window.c changes will each independently fix the problem. They are both included in this patch for defense in depth. Forum post 0d48347967. (check-in: 1ba22631 user: drh tags: trunk)
14:14
Get CLI utf8_fgets() to not consume more input than it returns. Get console setup restoration to happen for all non-crash exits. (check-in: b4fa233d user: larrybr tags: cli-utf8)
2023-04-12
20:23
Unwrap the loop in the WAL hash function. (check-in: eb94ae13 user: drh tags: trunk)
19:40
Automatically set HAVE_PREAD and HAVE_PWRITE on linux, as has been done in MacOS for a long time now. (check-in: 2f7a36d2 user: drh tags: trunk)
18:57
Small performance enhancement to integer-to-text conversion. (check-in: cfb3dba9 user: drh tags: trunk)
18:18
Avoid leaking a database handle in test script rbuexlock.test. (check-in: 2edf98d1 user: dan tags: trunk)
18:06
Remove temporary debugging code accidentally left on this branch. (check-in: 0a0f6487 user: dan tags: fts5-secure-delete)
17:54
WIP: CLI option to take control of console on Windows and make it support UTF-8 input pasting (or typing). Needs work to become robust per "ToDo:". (check-in: 82438239 user: larrybr tags: cli-utf8)
17:40
Add the 'secure-delete' option to fts5. Used to configure fts5 to aggressively remove old full-text-index entries belonging to deleted or updated rows. (check-in: 4240fd09 user: dan tags: fts5-secure-delete)
2023-04-11
19:38
New #ifdef to enable building with -DSQLITE_OMIT_WINDOWFUNC. (check-in: e1ff83fa user: drh tags: trunk)
15:06
Remove an ALWAYS() that can sometimes be false. Add a test case that makes the test false. Forum post 6c5678e3da. (check-in: c8fb143d user: drh tags: trunk)
02:10
Better handling of OOM errors in the cursor-hint logic. dbsqlfuzz 60cd5fff91974af91c2c3692beb4a2d7fdafef46 (check-in: 68fcfb58 user: drh tags: trunk)
2023-04-10
23:21
For sha3 extension, mention NIST standard implemented. (check-in: 529ab138 user: larrybr tags: trunk)
18:44
New assert() statements to validate the parameters to sqlite3BtreeCursorHint(). Fix a problem with the construction of those parameters discovered by forum post 0b53708c95. (check-in: 4c5a3c5f user: drh tags: trunk)
13:20
Sync the vt02.c test virtual table with TH3, in order to pull in the fix for long delays when there are huge OFFSET values. (check-in: 49ba0300 user: drh tags: trunk)
2023-04-09
20:44
Fix an assertion fault that can occur when compiling with both SQLITE_ENABLE_UNKNOWN_SQL_FUNCTIONS and SQLITE_ENABLE_STAT4. (check-in: 9ff69e59 user: drh tags: trunk)
10:09
Fix a faulty assert() in whereRangeScanEst() that should only apply if there are not prior errors. dbsqlfuzz 567ad91132879cbab8172b38c6a68ad40fa8d650. (check-in: 8ba9b08b user: drh tags: trunk)
2023-04-08
19:27
Fix a couple harmless compiler warnings seen with MSVC. (check-in: 1b864a37 user: mistachkin tags: trunk)
19:23
Fix a typo in [83e84531b46814ae] that changed the value of SQLITE_DBCONFIG_STMT_SCANSTATUS. (check-in: b7177653 user: drh tags: trunk)
16:51
Faster implementation of keywordCode() - the routine that determines if an identifier is really a keyword and if so, which keyword. (check-in: 0ff3d3d5 user: drh tags: trunk)
13:31
Fix a harmless compiler warning. (check-in: c9559ba6 user: drh tags: trunk)
13:01
Optimizations to btree.c save about 4.5 million CPU cycles: (1) Clone insertCell() into a separate insertCellFast() routine for use by sqlite3BtreeInsert(). (2) Mark allocateSpace() as always-inline. (3) Improved coalesence of adjacent free blocks in pageFreeArray(). (check-in: 5c12c400 user: drh tags: trunk)
2023-04-07
18:27
Guard against oversized cells in the newly enhanced pageFreeArray(). (Closed-Leaf check-in: 2dcdbb50 user: drh tags: btree-freespace-opt)
16:30
Add NEVER on an unreachable branch. (check-in: 9b3febbd user: drh tags: btree-freespace-opt)
15:49
Fix an assert(). (check-in: 7eff46ba user: drh tags: btree-freespace-opt)
15:07
Tweaks to the new insertCellFast(). (check-in: 203a581a user: drh tags: btree-freespace-opt)
14:33
Clone insertCell() into insertCellFast() for use by sqlite3BtreeInsert() for a substantial performance increase. (check-in: f225afd9 user: drh tags: btree-freespace-opt)
14:03
Fix a problem causing the rbu_exclusive_checkpoint=1 to be ignored with zipvfs databases. (check-in: d8f50b31 user: dan tags: trunk)
13:21
Small performance improvement in freeSpace(). (check-in: 8dc5292e user: drh tags: btree-freespace-opt)
11:55
Try to use a heap to make coalescence of adjacent free slots faster when freeing space on a btree page. Turns out that the overhead of managing the heap overwhelms any performance gain and the result is slower. (Closed-Leaf check-in: 5d7e833f user: drh tags: deadend)
11:18
Add further tests for the rbu_exclusive_lock=1 URI option. (check-in: c07b62be user: dan tags: trunk)
2023-04-06
20:14
Increase the size of the cache of free blocks inside of pageFreeArray() to reduce the number of calls to freeSpace(). (check-in: 27c59f1e user: drh tags: btree-freespace-opt)
17:29
Work around a harmless assertion fault associated with sqlite3VdbeMemAboutToChange() such that the detection of stale values in registers is preserved in debugging builds, but we avoid a false-positive assertion fault in cases involving a virtual table with a LIMIT clause in an IN-operator loop. dbsqlfuzz 3fd70d4ab4950acf1deb8f610a7a7c67cd38713b (check-in: 56ea2c2f user: drh tags: trunk)
13:35
A prepared statement that aborts due to SQLITE_SCHEMA should not invalidate cursors in other prepared statements that are already running. See forum post cae4367d9b for the original trouble report. (check-in: 857d0f5e user: drh tags: trunk)
01:05
In the CLI, during error processing while looking for a word boundary, avoid being deceived by malformed input that has a very long sequence of 0x80 characters. forum post ab93a23ba1. (check-in: 82609d5a user: drh tags: trunk)
00:59
In the zipfile extension, defend against corrupt ZIP files that contain a zero-length filename. Forum post b15f5e3ad8. (check-in: 46db2e42 user: drh tags: trunk)
00:18
In the new .scanstatus command in the CLI, make sure the database is opened before invoking sqlite3_db_config(). Forum post 6e26dcf544. (check-in: 1cd993c4 user: drh tags: trunk)
2023-04-05
02:55
Add a test case for the ALWAYS() macro removed by the previous check-in. (check-in: 68a1a837 user: drh tags: trunk)
02:50
Remove an ALWAYS() that might now be false due to the prior check-in. (check-in: fc689935 user: drh tags: trunk)
02:21
Fix the function that determines the collating function for an expression tree to handle new cases that arise as a result of the recently added ability to use indexed expressions in aggregate queries. [forum/forumpost/0713a16a44|Forum post 0713a16a44]. (check-in: cc5041f3 user: drh tags: trunk)
2023-04-04
19:56
Shell to use SQLITE_SHELL_HAVE_RECOVER consistently (correcting check-in 0421cc03e0efa8f1) (check-in: 5b980d72 user: larrybr tags: trunk)
18:55
Fix an incorrect entry in the array that maps sqlite3_value values into actual datatype numbers. dbsqlfuzz f660c659bcec48577a43d3bab37f46baaa22f59e (check-in: fa8537dc user: drh tags: trunk)
18:10
Remove an assert() statement that is no longer valid due to enhancements to query planner for improved use of indexes. Forum post dc16ec63d3. (check-in: 2b23dd24 user: drh tags: trunk)
17:35
Expose the new SQLITE_VTAB_USES_ALL_SCHEMAS to JS. (check-in: b7ef09be user: stephan tags: trunk)
08:49
Omit shell call to sqlite3_dbdata_init() when it is omitted. (check-in: 0421cc03 user: larrybr tags: trunk)
2023-04-03
23:49
When translating arguments of aggregate functions into references to expression indexes, make sure to only translate them for the current aggregate when there are nested aggregates. Forum post 409ebc7368. (check-in: 898bfa1a user: drh tags: trunk)
20:11
Improved diagnostic output from PRAGMA vdbe_addoptrace. (check-in: 050958c1 user: drh tags: trunk)
17:46
When changing a COLLATE expression node into TK_AGG_COLUMN because the nodes value is contained in an indexed expression, be sure to clear the EP_Collate property from the expression node. Fix for the assertion faults reported by forum post e45108732c and forum post 44270909bb. (check-in: cf6454ce user: drh tags: trunk)
15:01
Add the SQLITE_VTAB_USES_ALL_SCHEMAS option to sqlite3_vtab_config(). Update the sqlite_dbpage, sqlite_dbdata, and sqlite_dbptr virtual tables to make use of that interface. This was formerly handled by the internal sqlite3VtabUsesAllSchemas() routine that was called directly from sqlite_dbpage. But since sqlite_dbdata and sqlite_dbptr are an extension, an external interface to that functionality had to be provided. dbsqlfuzz 1a29c245175a63393b6a78c5b8cab5199939d6a8 (check-in: bcd51abe user: drh tags: trunk)
12:48
Make the sqlite_dbdata and sqlite3_dbptr virtual tables accessible to the CLI. (check-in: c0eff028 user: drh tags: trunk)
12:33
Stronger constraint checking in allocateSpace(). dbsqlfuzz 93d4c9ff5ef7cd29f16e767af1ee71c29ec5a4c0 (check-in: 9e968f4f user: drh tags: trunk)
2023-04-02
20:56
With the -DSQLITE_ENABLE_JSON_NAN_INF compile-time option, non-standard JSON numeric values "Inf", "Infinity", "-Inf", "-Infinity", "NaN", "QNaN", and "SNaN" are all accepted. SQLite should never generate these values, but it will accept that with the appropriate compile-time option. (check-in: 0a050e90 user: drh tags: trunk)
20:46
More off-by-one errors in the new JSON parsing. (Closed-Leaf check-in: dbc99662 user: drh tags: json-nan-inf)
20:27
Fix an off-by-one error in the recognition of -Infinity. (check-in: f7ebf3e6 user: drh tags: json-nan-inf)
16:43
Earlier detection of corruption in sqlite3BtreeDelete(). dbsqlfuzz a4c48c291d6e40157a1b749a05eaa7c7faf5a625. (check-in: 978dc71c user: drh tags: trunk)
14:53
Clear executable bit on base64.c (check-in: ec1ddbce user: larrybr tags: trunk)
2023-04-01
23:29
Allow special floating-point value names in JSON: "inf", "-inf", "infinity", "-infinity", "nan", "qnan", and "snan". All are converted into valid JSON values: 9e999, -9e999, or null. Requires the SQLITE_ENABLE_JSON_NAN_INF compile-time option to operate. (check-in: fc8793e5 user: drh tags: json-nan-inf)
16:14
Swat grammar nit on README.md (check-in: 715c00e5 user: larrybr tags: trunk)
15:51
Fix harmless compiler warnings. (check-in: a4fb2864 user: drh tags: trunk)
13:14
Improved error messages from PRAGMA integrity_check. Identify the root of the tree when a problem is found in a b-tree, making it easier to track the problem to a specific table or index. (check-in: a1cb152e user: drh tags: trunk)
12:22
In the b-tree module use %u instead of %d to print unsigned quantities such as page numbers and offsets. (check-in: 33ac62d8 user: drh tags: trunk)
2023-03-31
23:48
Omit the the count-of-view optimization if there is a HAVING clause. This fixes a problem that originated with check-in [9322a7c21f1c22ba]. dbsqlfuzz 6a107e3055bd22afab31cfddabc2d9d54fcbaf69 (check-in: babe2b5e user: drh tags: trunk)
16:27
Merge recent trunk fixes into the nan-inf branch. (Leaf check-in: a51fb9d7 user: drh tags: nan-inf)
2023-03-30
19:05
Earlier error detection for index expression usage by aggregate functions. dbsqlfuzz 29214ace4e25c98d2ddff8fbcf97afdda23f28b9 (check-in: 8e841e7f user: drh tags: trunk)
16:08
Omit a branch that is no longer needed following [c9c4f287652933eb]. (check-in: 960a488a user: drh tags: trunk)
12:19
Fix an error in new test script test/aggfault.test. (check-in: 8724fe74 user: dan tags: trunk)
11:05
Fix a crash that could follow an OOM error while processing aggregate functions. (check-in: 804435a2 user: dan tags: trunk)
2023-03-29
21:58
Avoid having OP_SeekScan jump over an OP_IdxGT or OP_IdxGE that follows the OP_SeekGE opcode. Fix for [b50528af4468237c]. (check-in: c9c4f287 user: dan tags: trunk)
18:54
Fix a problem with sqlite3_stmt_scanstatus() calls made from within an SQLITE_TRACE_STMT callback made from within a trigger. (check-in: 1fa78faf user: dan tags: trunk)
17:26
Back out elements (1) and (2) from [96ec8306457eebf5]. (check-in: d8b17615 user: drh tags: nan-inf)
16:28
Merge recent trunk fixes into the nan-inf branch. (check-in: 248bf629 user: drh tags: nan-inf)
15:16
Another #ifdef to omit code that is only used by STAT4. (check-in: 445c7556 user: drh tags: trunk)
14:42
New #ifdefs to omit code that is unused except under STAT4. (check-in: 09a9b30b user: drh tags: trunk)
11:36
Enhance PRAGMA integrity_check so that it can detect that a NOT NULL column contains a NaN value and report that as an error. dbsqlfuzz f144b642fe6f1a1c196f258ac6e60118a0cb59b2. (check-in: 7638d975 user: drh tags: trunk)
00:05
Modify the OP_IsType opcode so that it does not need to distinguish between NaN and other floating point values. Later: This branch causes an unnecessary slowdown of PRAGMA integrity_check. Check-in [7638d9755dc90fd3] does a better job of detecting when a NaN value occurs in a NOT NULL column. (Closed-Leaf check-in: 242cb36c user: drh tags: istype-opcode-refactor)
2023-03-28
16:02
Fix a weird corner case in aggregate function processing that results from the recent addition of support for index expressions on aggregate queries. Forum post bad532820c. (check-in: c34fd9fe user: drh tags: trunk)
11:18
Fix multiple problems with RETURNING on a DML statement against a view, all inspired by forum post dc3b92cfa0. (1) Do not allow a RETURNING clause to trick the code generator into thinking that the view being updated has an INSTEAD OF trigger. (2) Generate all result columns for a view in a DML statement. (3) The automatic covering index for a view should cover all result columns of the view. (check-in: c8bedef0 user: drh tags: trunk)
2023-03-27
13:57
Remove a meaningless JS test. Add a timer to the OPFS async-side worker loader in an attempt to catch a browser-specific quirk in which the worker loading silently fails, per discussion in/around [forum post a708c98dcb3ef|forum:a708c98dcb3ef]. (check-in: 4fc1904b user: stephan tags: trunk)
13:24
Do not allow constant factoring during PRAGMA integrity_check, since the constants might be stored in registers that are later reused for other purposes. dbsqlfuzz dc9ab26037cf5ef797d28cd1ae0855ade584216d. Problem discovered by a new assert() statement added in [6f8b97f31a4c8552]. (check-in: 0bba27b7 user: drh tags: trunk)
2023-03-26
16:36
Improvements to register allocation, especially in the ANALYZE command. New assert() statements added to help verify that memory allocation is correct, and to help fuzzer find lingering errors. (check-in: 6f8b97f3 user: drh tags: trunk)
11:54
Disable factoring of constant values during ANALYZE. This is a temporary fix for forum post 07de5f6216. The register allocation logic in ANALYZE needs to be completely refactored, but that will take longer. This check-in will serve to resolve the issue until a better fix can be devised. (check-in: c3967d12 user: drh tags: trunk)
2023-03-25
23:52
When the left table of a RIGHT JOIN is used inside an aggregate function and the left table employs an index on expressions, then make sure the expressions evaluate to NULL for the cases where the left table should be NULL. Fix for forum post 9b491e1deb. (check-in: ffe23af7 user: drh tags: trunk)
23:40
Add usage detection to the NULL value generator for the left table of a RIGHT JOIN inside of an aggregate. (Closed-Leaf check-in: 4d05a009 user: drh tags: rightjoin-agg-idxexpr)
22:37
When reading sqlite_stat4 data during query planning, be sure to expand zeroblobs prior to running comparisons. Fix for the issue identified by forum post 5275207102. (check-in: 5c8dd8df user: drh tags: trunk)
21:01
When the left table of a RIGHT JOIN is used inside an aggregate function and the left table employs an index on expressions, then make sure the expressions evaluate to NULL for the cases where the left table should be NULL. Proposed fix for forum post 9b491e1deb. More testing an analysis needed - there is a FIXME in this check-in. (check-in: 3572b40a user: drh tags: rightjoin-agg-idxexpr)
19:44
In the byte-code generator, when the result of an expression needs to be in a particular register, always use the sqlite3ExprCode() routine because it has the smarts to know whether to use OP_Copy or OP_SCopy. Do not try to OP_SCopy inline because an OP_Copy might be required. Fix for the problem identified by forum post 5522082cfc. (check-in: c104e5c6 user: drh tags: trunk)
18:41
Add the tag-20230325-1 comment that was omitted from the prior check-in. (check-in: a13c01d0 user: drh tags: trunk)
18:31
The fix at [2bf5413dc2c19d5f] was incomplete in that it failed to clear the reusable register cache that might contain registers in the STAT4 buffer region. This additional change corrects the problem. Forum post 83cb4a95a0. Test case in TH3. (check-in: 5d554e4d user: drh tags: trunk)
12:27
Even tighter bounds on the maximum length of the filename for sqlite3_load_extension(). (check-in: 78729141 user: drh tags: trunk)
03:17
Tighter constraints on the maximum length of the filename handed over to sqlite3_load_extension(), due to forum post a43074729e. This is a follow-on to [01f3877c7172d522] and forum post 08a0d6d9bf. (check-in: 9f351bde user: drh tags: trunk)
02:07
New test case to further validate the fix at [221fdcec964f8317]. Forum post d34ad68c36. (check-in: a6e218a6 user: drh tags: trunk)
01:29
Fix CLI non-handling of OOM reported at Forum post 6872514e04. (check-in: 6f6a0fd6 user: larrybr tags: trunk)
2023-03-24
22:17
Remove undocumented, vestigial SQL functions in the CLI that were once used for the ".recover" command but are now no longer needed. (check-in: 1ef461aa user: drh tags: trunk)
21:35
Fix possible integer overflow in bounds checking for the debugging function "shell_int32()" found in the CLI. This change does not affect the core SQLite. Forum post be9c294ee0. (check-in: 62114711 user: drh tags: trunk)
21:24
Fix an error in the OP_SeekScan opcode added by check-in [4a43430fd23f8835]. Problem reported by forum post 8cc1dc0fe9. (check-in: 651a13fc user: drh tags: trunk)
20:35
Fix the handling of indexed expressions in an outer query that appear as corelated values inside an aggregate function within a subquery. Forum post 79cf371080. (check-in: d8259877 user: drh tags: trunk)
19:17
Fix a problem in cursor-hints for WITHOUT ROWID tables used in a RIGHT JOIN. Forum post 591006b1cc. (check-in: 221fdcec user: drh tags: trunk)
17:15
Add a JS test which checks for the issue described in forum post 895425b49a. (check-in: 98d30400 user: stephan tags: trunk)
16:57
Fix byte-code register allocation in ANALYZE for STAT4 when there multiple indexes with differing numbers of columns. forum post bc39e531e5. This problem arose when expression indexes were added by check-in [2131a5ca53f0e9b0]. (check-in: 2bf5413d user: drh tags: trunk)
2023-03-23
19:22
Fix test cases so that they work when SQLITE_ENABLE_NAN_INF is defined. (check-in: 95c7af79 user: drh tags: nan-inf)
12:00
Fix #ifdefs that use the wrong preprocessor macro. (check-in: 0aecf360 user: drh tags: nan-inf)
10:54
The attempt to bring STAT4 up to 100% MC/DC at [55a26c67ed4a3a93] and at [168fa2fb22b8c1ad] are incorrect. Back them out and replace them with a simple NEVER() macro. Error reported by forum post dc4854437b. (check-in: 5992370a user: drh tags: trunk)
2023-03-22
20:21
Add the SQLITE_ENABLE_NAN_INF compile-time option which makes the following behavior changes: (1) sqlite3_value_double(NULL) returns NaN, (2) SQLite preserves NaN values rather than converting them to NULL. (3) CAST statements understand "NaN" and "Inf" and make the right conversions. (4) Non-standard JSON is never generated by SQLite JSON routines, but those routines will accept floating point literals "NaN", "Inf", and "-Inf". (check-in: 96ec8306 user: drh tags: nan-inf)
19:57
Internal cleanups in JS code. No functional changes. (check-in: 8fbdf7d1 user: stephan tags: trunk)
16:55
The floating-point-to-text conversion with the zero-padding option now renders NaN as "null". (check-in: ad59fa17 user: drh tags: trunk)
16:37
For consistency, the ".mode json" output from the CLI now renders infinity in the same format as the JSON functions. (check-in: abee339d user: drh tags: trunk)
16:24
The double-to-text conversion renders infinity as 9e999, so that JSON output is compliant and so that values can be round-tripped. (check-in: b52081d0 user: drh tags: trunk)
16:01
In the CLI, the magic parameter :inf and :nan bind floating point values Infinity and NaN, respectively, as an aid to testing SQLite's handling of those quantities. (check-in: c70a61d8 user: drh tags: trunk)
14:51
Update the version number for the TEA tarball to 3.42.0, to match the core. (check-in: 03e6918e user: drh tags: trunk)
2023-03-21
14:20
Add ALWAYS() on a branch this is always true now due to [84417bbd144b2197]. (check-in: badf7d0e user: drh tags: trunk)
12:29
Add the fuzzcheck-asan.exe target to the MSVC makefile. (check-in: 0901bc02 user: drh tags: trunk)
11:56
Add the fuzzcheck-asan target to the main posix makefile. (check-in: 55e94add user: drh tags: trunk)
11:13
Fix a valgrind error and potential buffer overread when handling a corrupt database. (check-in: b1e0cd64 user: dan tags: trunk)
2023-03-20
20:22
Reinsert two NEVER() macros for b-tree branches that were previously needed for [b6a82f3c3b9d89fd] but which are now obsolete due to [73f0036f045bf371]. Later: dbsqlfuzz quickly found new cases for which those two branches are needed. The new test cases have been added to TH3. (Closed-Leaf check-in: 3065dadb user: drh tags: backout)
18:35
Minor change to btreeNext() to facilitate coverage testing. (check-in: 20b3ef04 user: drh tags: trunk)
14:59
Fix a problem causing a cursor to retain an out-of-date cell-info cache when processing a DISTINCT query on values that are identical according to their collation sequence, but different on disk. Forum post e123e6cde4. (check-in: 1b3abc1d user: dan tags: trunk)
10:43
Back out the extra margin added to the input buffer of the CLI, as it is not needed. (check-in: ac8d1e5d user: drh tags: trunk)
01:55
A better fix for the sqlite3_error_offset() problem on generated columns. (check-in: 770b3e67 user: drh tags: trunk)
00:53
Expression errors in generated columns should not generate non-negative sqlite3_error_offset() returns. Second of two defenses against [33aa4c0de8a62e33]. (check-in: 2adb4e0d user: drh tags: trunk)
00:48
When reporting errors in the CLI, ignore the output of sqlite3_error_offset() if the value returned is clearly out-of-range. One of two lines of defense against [33aa4c0de8a62e33]. (check-in: 26adbb80 user: drh tags: trunk)
2023-03-19
21:48
Increase the size of ref-count values in the pager layer to 64-bits, to avoid any reasonable possiblity of overflowing the counters. There is a performance and memory penality for this. Forum post b741f15a35. (check-in: 6c5d99a8 user: drh tags: trunk)
2023-03-18
16:12
Avoid a buffer overread in fts3 that could occur when processing a corrupt record. (check-in: 02ac2297 user: dan tags: trunk)
2023-03-17
20:31
Fix json rendering so that it shows positive and negative infinity as 9.0e+999 and -9.0e+999 respectively. (Closed-Leaf check-in: efce4690 user: drh tags: numeric-only-json)
19:18
Add the ability to name functions using one of the join keywords like CROSS FULL INNER LEFT NATURAL OUTER RIGHT. (check-in: 0910b192 user: drh tags: trunk)
19:07
Add test cases for functions named the same as join keywords. (Closed-Leaf check-in: 94944b23 user: drh tags: functions-named-left)
14:18
Fix a potential buffer overread in the recovery extension. (check-in: 0b3b5bf9 user: dan tags: trunk)
10:43
Ensure that an error does not delete the Table object out from under the xConstruct method of a virtual table. dbsqlfuzz 7cc8804a1c6d4e3d554d79096e6ea75a7c1c7d2d (check-in: df4928c9 user: drh tags: trunk)
00:42
Add safety margin on the CLI input buffer for tickets [33aa4c0de8a62e33], [b97e6c5e6a91d97f], [2971fbe3f993e95a], and [2971fbe3f993e95a]. (check-in: 741af08a user: drh tags: trunk)
00:01
Fix assert() statements that would (incorrectly) fire if an IF NOT EXISTS trigger that already exists contained two or more RETURNING clauses. Tickets [89d259d45b855a0d] and [d15b3a4ea901ef0d]. (check-in: 648899e4 user: drh tags: trunk)
2023-03-16
20:54
Correctly handle SELECT DISTINCT ... ORDER BY when all of the result set terms are constant and there are more result set terms than ORDER BY terms. Fix for these tickets: [c36cdb4afd504dc1], [4051a7f931d9ba24], [d6fd512f50513ab7]. (check-in: 12ad822d user: drh tags: trunk)
12:28
Additional debug/test output from the query invariant checker showing the row-number that is being checked. (check-in: e4b6eb58 user: drh tags: trunk)
11:50
Update the tracing output for the query-invariant checker such that it shows the SQL that is run to verify that a found query-invariant discrepency is valid. Changes to testing logic only. (check-in: 8f45ad27 user: drh tags: trunk)
10:17
Do not use the one-pass optimization on an UPDATE if there is a subquery in the WHERE clause, since if the subquery is hidden behind a short-circuit operator, the subquery might not be evaluated until after one or more rows have been updated. Fix for the problem reported by forum post 0007d1fdb1. This is the same problem that was fixed by [73f0036f045bf371] only for UPDATE instead of DELETE. (check-in: 2c56b984 user: drh tags: trunk)
09:07
Remove a NEVER() from btreeNext() that dbsqlfuzz 460aa158f9a2c41145831cc924296cde1f312b3f found could sometimes be reached. I will find a way to test that branch later. (check-in: 1dffeffe user: drh tags: trunk)
02:30
Another approach at attempting to contain the damage caused by corruption that leaves MemPage.isInit clear. Works better than the previous but is still not perfect. (Closed-Leaf check-in: ba964eb0 user: drh tags: corruption-in-btree-init)
01:20
When the btreeInitPage() routine detects database corruption, it should continue to the end and set MemPage.isInit before it returns SQLITE_CORRUPT, because if it leaves MemPage.isInit unset, then can cause difficulty later. dbsqlfuzz 460aa158f9a2c41145831cc924296cde1f312b3f (check-in: 44e83f8b user: drh tags: corruption-in-btree-init)
2023-03-15
17:58
Disallow the one-pass optimization for DELETE if the WHERE clause contains a subquery. Fix for the problem reported by forum post e61252062c9d286d. This fix is more restrictive than necessary. It could be relaxed if the subquery does not involve the table that is the subject of the DELETE. (check-in: 73f0036f user: drh tags: trunk)
13:53
Fix a broken assert() in the recovery extension. (check-in: 4c4e66f2 user: dan tags: trunk)
2023-03-14
20:08
Fix Bloom filters on an expression index. forum post 2e427099d5 and forum post d47a0e8e3a. This problem goes back to the original introduction of Bloom filters (check-in [633bfeeea2bccdd4]) for SQLite version 3.38.0. (check-in: c028fb66 user: drh tags: trunk)
2023-03-13
16:08
Include CLI's tip for -- in all builds. Better show optionality of its non-option arguments. (check-in: 9e2c771d user: larrybr tags: trunk)
2023-03-11
23:21
The check-in at [198b3e33dcfd74c7] caused a performance regression for some queries, which is here fixed. Problem reported by forum post b405033490fa56d9. (check-in: dc9f025d user: drh tags: trunk)
12:27
Allow functions named using keywords "CROSS", "FULL", "INNER", "LEFT", "NATURAL", "OUTER", and "RIGHT". (check-in: eeac3d5e user: drh tags: functions-named-left)
00:15
CLI help to reflect no-more-options option (check-in: 30d95a12 user: larrybr tags: trunk)
2023-03-10
21:27
Fix a typo in a comment. No code changes. (check-in: 76acc075 user: drh tags: trunk)
20:54
Give CLI a no-more-options option. (--) (check-in: 08227887 user: larrybr tags: trunk)
13:36
Fix a problem with the fts5 snippet() function that shows up when snippets just 1 token in length are requested. (check-in: 96d5116d user: dan tags: trunk)
11:57
Export SQLITE_FCNTL_RESET_CACHE to JS. (check-in: 6195cfc8 user: stephan tags: trunk)
2023-03-09
22:09
Replace a lingering use of 'self' with 'globalThis' in JS code, for node compatibility. (check-in: 7e3782b5 user: stephan tags: trunk)
16:11
Reinstate some test cases accidentally removed by [cb023fe28560ce0f]. (check-in: 870de61f user: dan tags: trunk)
15:08
Fix countofview.test so that it works with SQLITE_OMIT_PROGRESS_CALLBACK builds. (check-in: 2fc7c3fc user: dan tags: trunk)
08:51
Experimental addition of sqlite3-node.mjs, for node.js, based on feedback from forum post ac7a94d4f77db235 and related off-list discussions. Build changes only - no code changes. (check-in: a5db97fa user: stephan tags: trunk)
2023-03-08
23:05
Fix a possible NULL pointer dereference due to the sqlite3_interrupt() enhancement at [bd8fa10e59f58886]. Reported by forum post f5a2b1db87. (check-in: 84417bbd user: drh tags: trunk)
22:48
Backout the OP_MakeRecord optimization as it does not work. (check-in: 25017312 user: drh tags: trunk)
18:05
Export the new SQLITE_CHANGESETAPPLY_IGNORENOOP flag to JS. (check-in: ac7359b2 user: stephan tags: trunk)
18:03
Add the SQLITE_CHANGESETAPPLY_IGNORENOOP flag, which may be passed to sqlite3changeset_apply_v2() to have it ignore changes that would be no-ops if applied to the database (e.g. deleting a row that has already been deleted), instead of considering them conflicts. (check-in: cb023fe2 user: dan tags: trunk)
17:09
Small performance improvement in the OP_MakeRecord opcode. (check-in: ca89daef user: drh tags: trunk)
14:28
Change to [44135d6ea84f7ba6] that retains the historical datatype ("INT", not "NUM") for a table created as follows: "CREATE TABLE t1 AS SELECT CAST(123 AS INT) AS value;". The use of FLEXNUM only occurs on compound queries. (check-in: 6d5b5896 user: drh tags: trunk)
10:05
Extend wasm build to support a custom sqlite3.c to support building against sqlite3-see.c. The JS code now binds the SEE-specific functions if it detects an SEE build. (check-in: dd8612c8 user: stephan tags: trunk)
00:47
Fix a problem in the count-of-view optimization that can lead to incorrect bytecode. dbsqlfuzz 23d782160b71c3f8f535ccb2da313dfc8eb8c631. (check-in: f4500953 user: drh tags: trunk)
2023-03-07
23:47
Fix a bug introduced 4 days ago by [e95439119ac200cb]: do not set the Expr.affExpr field of a generated column expression if the expression is a RAISE() function, as affExpr has a different meaning for RAISE. Forum post b312e075b5. (check-in: 1096b5a7 user: drh tags: trunk)
19:39
A proposed change to [44135d6ea84f7ba6] that retains the historical datatype ("INT", not "NUM") for a table created as follows: "CREATE TABLE t1 AS SELECT CAST(123 AS INT) AS value;". (Closed-Leaf check-in: a0e54fe2 user: drh tags: flexnum-proposed-fix)
19:23
Improve how sqlite3.initWorker1API() determines whether it's running in a Worker thread. Based on feedback in forum post ac7a94d4f77db235. (check-in: 2f712b83 user: stephan tags: trunk)
19:12
Replace use of 'self' in JS code with 'globalThis', as that works in browsers and node environments. Avoid using globalThis.location if it's not set (e.g. in node). Based on feedback in forum post ac7a94d4f77db235. Minor JS build tweaks. (check-in: dbbe8f25 user: stephan tags: trunk)
12:59
In the JS sqlite3.vfs/vtab utility APIs, use a local reference to StructBinder instead of sqlite3.StructBinder, as that object is removed from the sqlite3 namespace during the final steps of API initialization. Based on feedback from forum post d19d96183badca70. (check-in: 0d89885d user: stephan tags: trunk)
02:24
Fix a couple minor spacing issues in the MSVC makefile. (check-in: 46b3ac6d user: mistachkin tags: trunk)
2023-03-06
23:39
Repair an unintential fork. (check-in: 8b524c84 user: drh tags: trunk)
23:38
Improvements to query invariant testing such that it uses the new SQLITE_DBCONFIG_REVERSE_SCANORDER opcode to sqlite3_db_config() to make more accurate judgements about when a query is ambiguous, and hence when query invariant testing is approprate. (check-in: be9ab292 user: drh tags: trunk)
21:38
Cause CLI to fail noisily when deserialize option used for non-seekable "file". (check-in: 24bd7e82 user: larrybr tags: trunk)
19:04
Add SQLITE_DBCONFIG_REVERSE_SCANORDER for direct C-language access to the "PRAGMA reverse_unordered_selects" setting. (check-in: 83e84531 user: drh tags: trunk)
2023-03-05
07:44
Rename sqlite3-worker1-bundler-friendly.js to sqlite3-worker1-bundler-friendly.mjs and refactor it to work as an ES6 module, based on feedback in forum post a255f89c2eadf4c4. (check-in: af312b13 user: stephan tags: trunk)
07:33
Correct rendering of error messages in demo-worker1.js. Remove some stray EOL whitespace. (check-in: 1d5d515a user: stephan tags: trunk)
2023-03-04
15:36
Fix to check-in [b9190d3da70c4171] - the agg-with-indexed-expr optimization requested by ticket [99378177930f87bd] - that can cause an incorrect answer if an aggregate subquery has a GROUP BY clause, and that GROUP BY contains a term that is not in the result set, and the outer query makes use of expression indexes. Problem reported by forum post a68313d054. (check-in: e0697387 user: drh tags: trunk)
12:57
Show the output value from OP_AggFinal when doing byte-code tracing. (check-in: 35f10a06 user: drh tags: trunk)
2023-03-03
21:17
Make the SQLITE_DBCONFIG_STMT_SCANSTATUS option on by default in the CLI. (check-in: 5a091911 user: dan tags: trunk)
19:56
Fix a vdbe-coverage macro added by [f418bdd627e84e7d]. (check-in: 77f559d2 user: drh tags: trunk)
19:43
Follow-up to [e95439119ac200cb] to fix a case where a pointer is NULL due to OOM. (check-in: 2535bc8c user: drh tags: trunk)
18:35
Enhance PRAGMA integrity_check so that it can detect when there are extra bytes at the end of an index record, which might cause OP_IdxRowid to malfunction. dbsqlfuzz c1aa3986534d5feab8d21f28b3c1712df2ef358ba. Test case in TH3. (check-in: f418bdd6 user: drh tags: trunk)
16:25
When it is known when preparing a statement that X cannot be NULL or is always NULL, transform the expression (X IS NULL) to integer value 1 or 0 instead of 'true' or 'false'. This is because under some circumstances, "Y IS TRUE" or "Y IS FALSE" may not be equivalent to "Y IS 1" of "Y IS 0". This problem was introduced by [de9c86c9e4cdb34f] and was reported by forum post 2cd11c2d37. (check-in: cc4bb05b user: dan tags: trunk)
15:12
Do not use an expression index on a generated column if generated column has the wrong affinity. dbsqlfuzz 65f5eb57f8859344d5f1f33e08c77ee12960ed83 (check-in: e9543911 user: drh tags: trunk)
10:42
Remove unnecessary call to sqlite3_dbdata_init() from shell.c.in. (check-in: c4d083a3 user: dan tags: trunk)
2023-03-02
13:49
When flattening the right operand of a LEFT JOIN (check-in [41c27bc0ff1d3135]), ensure that the OP_IfNullRow opcode does not NULL-out a subquery result that was computed within OP_Once. This fixes the problem problem reported by forum post 402f05296d. (check-in: 8fe13f7a user: drh tags: trunk)
06:58
Resolve a parallel build timing issue when building sqlite3.c/h from ext/wasm. For the time being, do not add sqlite3_wasm_extra_init.c to fiddle.wasm because it can cause duplicate definitions of extensions which are already built into the shell (a better resolution for this conflict is pending). No longer add sqlite3_wasm_extra_init.c to speedtest1.wasm because it's useless there. (check-in: 75fdd5b8 user: stephan tags: trunk)
05:51
Rename some vars in the ext/wasm makefiles for consistency's sake. (check-in: c23589d9 user: stephan tags: trunk)
2023-03-01
20:23
When flattening a view that is the right operand of a LEFT JOIN, using the optimization of check-in [41c27bc0ff1d3135], always insert the TK_IF_NULL_ROW expression nodes, even for TK_COLUMN expressions, as the TK_COLUMN might be a column from an outer query and hence still need to be NULLed out. This fixes the problem described by forum post 26387ea7ef. (check-in: 198b3e33 user: drh tags: trunk)
15:21
Follow-up to [bbaf1f2eb1e1637b]: Make sure subtypes do not cross a subquery boundary even if the function that returned the value with a subtype is buried down inside a larger expression. This fixes a problem identified by forum post 37dd14a538. (check-in: e72661eb user: drh tags: trunk)
13:54
Do not attempt to apply the count-of-view optimization to a CTE. dbsqlfuzz ef8623915d843b150c159166ee4548c78cc6895a (check-in: abc3a383 user: drh tags: trunk)
2023-02-28
21:23
Activate SQLITE_DBCONFIG_STMT_SCANSTATUS in fuzzcheck. (check-in: 4fe1419a user: drh tags: trunk)
20:06
Updates to speedtest1.c and the speed-check.sh test script so that they work with the new SQLITE_DBCONFIG_STMT_SCANSTATUS control. (check-in: bd02df05 user: drh tags: trunk)