Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
200 most recent check-ins using file sqlite3.1 version fc7ad8990f
|
2024-04-02
| ||
| 18:48 | Fix typos in comments. Provided ".wheretrace" debugging output for the interstage heuristic module. Do omit automatic index loops in the interstage heuristic. (Leaf check-in: 357d9513d2 user: drh tags: branch-3.28) | |
| 18:31 | Fix table-valued functions so that they will work as the right table in a LEFT JOIN. Ticket [2ae0c599b735d59e] (check-in: 1f97086d62 user: drh tags: branch-3.28) | |
| 11:54 | Add a heuristic in between the two solver() passes of the query planner that tries to prevent a very slow query plan in cases where the output row count estimate is imprecise. (check-in: 74b247d958 user: drh tags: branch-3.28) | |
|
2023-11-29
| ||
| 16:26 | Fix a duplicate assert() caused by the second cherrypick in the previous check-in. (check-in: f10d4fc4a8 user: drh tags: branch-3.28) | |
| 16:07 | Remove Window objects from the corresponding Select.pWin list when they are deleted, as they are, for example, when the ORDER BY clause is optimized out. (check-in: f9c6e6a710 user: drh tags: branch-3.28) | |
|
2023-10-19
| ||
| 21:05 | Ensure that when an ephemeral cursor is reopened with a second invocation of to OP_OpenEphemeral, the sequence counter is reset and the cache marked as stale. Fix for [9cdc5c46]. (check-in: d4bfa8d21a user: drh tags: branch-3.28) | |
|
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: f5913e7632 user: dan tags: branch-3.42) | |
|
2023-09-15
| ||
| 20:04 | Drop support for the view-scan optimization as it was causing multiple performance regressions. In its place, reduce the estimated row count for DISTINCT subsqueries by a factor of 8. (check-in: 796a65fa61 user: drh tags: branch-3.28) | |
|
2023-09-08
| ||
| 19:32 | Fix a problem with fts5 secure-delete mode causing fts5 to corrupt its records. (check-in: cc0f82a480 user: dan tags: branch-3.42) | |
|
2023-09-02
| ||
| 22:07 | Eliminate a redundant NK_DbAboutToClose notification. (Leaf check-in: eccffaa7ca 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: 8e6e54e48c user: larrybr tags: cli_extension) | |
|
2023-08-21
| ||
| 18:25 | Fix problems related to structured-exception-handling on this branch. (check-in: d3d77e35ea user: dan tags: bedrock) | |
| 17:49 | Fix problems related to structured-exception-handling on this branch. (check-in: dd86c2729b user: dan tags: begin-concurrent) | |
| 17:26 | Fix MSVC testfixture.exe builds. (check-in: 7004c9027e user: dan tags: begin-concurrent) | |
|
2023-08-12
| ||
| 20:50 | Correct JNI .jar rules to only include *.java/class files, not *.* (*~ files). (Closed-Leaf check-in: 1ba7754045 user: stephan tags: jni) | |
| 15:37 | More JNI docs. (check-in: 290028d3ca user: stephan tags: jni) | |
| 15:09 | Further simplifications in the interface of the OutputPointer family of Java classes. (check-in: 962c3e0de2 user: stephan tags: jni) | |
| 10:39 | Tweaks and docs for the OutputPointer family of Java classes. (check-in: 265c8fd0d4 user: stephan tags: jni) | |
| 10:27 | Bind sqlite3_db_status() to JNI. (check-in: b79477a0af user: stephan tags: jni) | |
| 10:06 | Bind sqlite3_status(64)() to JNI. (check-in: cefb6614e6 user: stephan tags: jni) | |
|
2023-08-11
| ||
| 21:25 | Add timing info to SQLTester. (check-in: b69b5facbf user: stephan tags: jni) | |
| 21:24 | Merge trunk into jni branch. (check-in: 4f0aeeba02 user: stephan tags: jni) | |
| 20:32 | Fix a makefile deps problem which caused ext/jni build to fail if sqlite3.c/h were not created beforehand. (check-in: 101de67077 user: stephan tags: jni) | |
| 19:38 | Merge latest trunk changes, including support for handling of structured-exceptions in MSVC builds, with this branch. (check-in: cb3cf9bab0 user: dan tags: begin-concurrent) | |
| 19:36 | Merge latest wal2 changes, including support for handling of structured-exceptions in MSVC builds, with this branch. (check-in: 331f2f3e5d user: dan tags: bedrock) | |
| 19:33 | Merge latest trunk changes, including support for handling of structured-exceptions in MSVC builds, with this branch. (check-in: 00bfae9ffa user: dan tags: wal2) | |
| 19:31 | If SQLITE_USE_SEH is defined, handle structured-exceptions thrown by MSVC builds if the *-shm file mapping is accessed after it becomes invalid for some reason. (check-in: 8a6b0c2493 user: dan tags: trunk) | |
| 18:59 | Minor SQLTester test tweaks and have 'make tester' include the out-of-tree tests if they are found. (check-in: 6c8538d834 user: stephan tags: jni) | |
| 18:31 | Fix a problem with handling savepoints in BEGIN CONCURRENT transactions. (check-in: ca422d00d7 user: dan tags: bedrock) | |
| 18:25 | Fix a problem with handling savepoints in BEGIN CONCURRENT transactions. (check-in: 4a5c0439ae user: dan tags: begin-concurrent) | |
| 18:04 | Bind sqlite3_sql() and sqlite3_expanded_sql() to JNI. Start marking C-side functions which would need explicit mutex support if we remove 'synchronized' from their Java entry points (but there are many more left to mark). (check-in: c7fb32d1ef user: stephan tags: jni) | |
| 17:45 | Add a doc link for the new sqlite3_js_posix_create_file(). (check-in: 0d7aac45b8 user: stephan tags: trunk) | |
| 17:38 | Add sqlite3.capi.sqlite3_js_posix_create_file() and oo1.OpfsDb.importDb() as alternatives for the newly-deprecated sqlite3_js_vfs_create_file(). (check-in: da6eaf8d82 user: stephan tags: trunk) | |
| 14:31 | Deprecate sqlite3_js_vfs_create_file() because, it was discovered today, its out-of-scope use of the sqlite3_vfs, sqlite3_file, and sqlite3_io_methods APIs triggers unresolvable assertions in the core when built with SQLITE_DEBUG. (check-in: f3647a3ac8 user: stephan tags: trunk) | |
| 11:46 | Merge all the latest trunk enhancements into the bedrock branch. (check-in: 859932ebce user: drh tags: bedrock) | |
| 11:40 | Merge the latest trunk enhancements into the wal2 branch. (check-in: fe1bf30ca0 user: drh tags: wal2) | |
| 11:34 | Merge the latest trunk enhancements into the begin-concurrent branch. (check-in: 7cb2b4e35f user: drh tags: begin-concurrent) | |
| 11:30 | Provide the -DSQLITE_LEGACY_JSON_VALID compile-time option to restore the (incorrect) legacy behavior of json_valid(NULL). (check-in: 00bc9f1b57 user: drh tags: trunk) | |
| 11:24 | Fix bug in the test case for the previous check-in. (Closed-Leaf check-in: c5daae8861 user: drh tags: legacy-json-valid) | |
| 11:23 | Update test scripts specific to this branch so that they work on windows. (check-in: f0ca13edad user: dan tags: wal2) | |
| 11:12 | Up until version 3.42.0, there was a bug in json_valid() such that it would return False (0) for a NULL input. That bug is fixed in 3.42.0. This check-in adds a compile-time option -DSQLITE_LEGACY_JSON_VALID that restores the old buggy behavior for applications that depend on it. (check-in: 15c2eadbff user: drh tags: legacy-json-valid) | |
|
2023-08-10
| ||
| 21:50 | Mark _all_ JNI binding funcs as synchronized so that Java can lock them and protect our global-state access. The alternative is writing a mountain of C-side code to do the same thing. (check-in: afe190a940 user: stephan tags: jni) | |
| 21:29 | Work around jdk8 and jdk19 mangling the C name of sqlite3_db_config() differently. Correct the variadic arg handling of the JNI-side subset of sqlite3_db_config() options. (check-in: 746894c3c0 user: stephan tags: jni) | |
| 21:20 | Add tests for structured-exception-handling and wal2. Fix a problem with the same. (Leaf check-in: 87242777ec user: dan tags: wal2-shm-exceptions) | |
| 20:52 | Add a working dist zip file for the JNI bits. (check-in: d6a4d212ce user: stephan tags: jni) | |
| 18:58 | Experimental merge of the wal-shm-exceptions and wal2 branches. (check-in: fd58c61624 user: dan tags: wal2-shm-exceptions) | |
| 18:57 | Initial pass at 'make dist' rules for the JNI bundle, but they still need a bare-bones, posix-make-compatible makefile to include in the bundle. (check-in: ff54e66a4d user: stephan tags: jni) | |
| 18:50 | New testcase() macro in the tokenizer, to better document its behavior. (check-in: b2fdac0b15 user: drh tags: trunk) | |
| 17:39 | Merge trunk into jni branch for the newly-relocated version-info tool. (check-in: cc8e8cba67 user: stephan tags: jni) | |
| 17:32 | Move ext/wasm/version-info.c to tool/ for re-use in build other dist bundles. (check-in: 4b0871fd36 user: stephan tags: trunk) | |
| 17:07 | Merge latest trunk changes into this branch. (Leaf check-in: 3ed89c344f user: dan tags: wal-shm-exceptions) | |
| 17:04 | Mark test script wal2big.test as "TESTRUNNER: slow". (check-in: ae7f3bcb5b user: dan tags: wal2) | |
| 17:03 | Merge latest trunk changes into this branch. (check-in: 72f7807b34 user: dan tags: wal2) | |
| 16:42 | Add SQLTester --keep-going flag to allow it to continue to the next script after an error. (check-in: 4d635f781b user: stephan tags: jni) | |
| 13:10 | Document that auto-extensions registered via JNI will fail if they open a db (to avoid triggering an endless loop in the auto-extensions). Discover the hard way that JDKv19 creates different mangled JNI names for some functions than JDKv8. Start reformatting much of the JNI API decls in prep for making them even longer. (check-in: 99c0941f1c user: stephan tags: jni) | |
| 12:36 | Resolve two assertions in the auto-extension JNI which were triggered via new SQLTester infrastructure. Move SQLTester's db-init SQL injection into an auto-extension. (check-in: 2952906c30 user: stephan tags: jni) | |
| 11:15 | Make test completion status more visible at a glance on modern terminals. (check-in: a4e96c306c user: stephan tags: jni) | |
| 11:04 | Disable REQUIRED_PROPERTIES handling in SQLTester, per /chat. Scripts with that directive are now skipped. (check-in: ddc534cb25 user: stephan tags: jni) | |
| 10:58 | Add a visual indicator (emoji) to TestScript verbose messages, dependent on the message's level of verbosity. (check-in: dc323d3894 user: stephan tags: jni) | |
| 10:44 | Add the current --testcase name to SQLTester --verbose output. (check-in: f87367402b user: stephan tags: jni) | |
| 10:34 | More SQLTester docs. Add --verbosity command to help zoom in on script areas while debugging. Spice up test-start/end output with some emoji. (check-in: 8dd0802149 user: stephan tags: jni) | |
| 05:25 | Give DbException the option of closing the db to simplify error handling in one case. (check-in: 908c9a4450 user: stephan tags: jni) | |
| 05:14 | Initial version of REQUIRED_PROPERTIES support for SQLTester, with TEMPSTORE_(FILE/MEM) and RECURSIVE_TRIGGERS options. (check-in: 48d16c9d2f user: stephan tags: jni) | |
| 04:24 | Generic cleanups and fixes in SQLTester. (check-in: fc5d3cc30d user: stephan tags: jni) | |
| 02:09 | Change the SQLite3Jni API annotations to use SOURCE retention (used only at compile-time). (check-in: 3c3fea6bf2 user: stephan tags: jni) | |
| 01:44 | Move all of the SQLTester code into a single file, since it's only got 1 public class. Remove 'public' from many methods which don't need it. Add more documentation to it. (check-in: 2815d67695 user: stephan tags: jni) | |
| 01:19 | Defer static JNI-side init of SQLTester until main() is called so that its auto-extensions do not leak over to clients of the main library. (check-in: e461fdd53b user: stephan tags: jni) | |
| 01:05 | Merge trunk into jni branch. (check-in: 52fa6f7841 user: stephan tags: jni) | |
| 01:03 | Treat all args to --glob as a single glob. (check-in: 4737bd3457 user: stephan tags: jni) | |
| 00:58 | Remove --print's ability to read ahead. (check-in: 15cfee972b user: stephan tags: jni) | |
| 00:34 | Replace the SQLTester infrastructure with a line-oriented, non-regex-heavy parser. Add --column-names command. (check-in: 88863908ee user: stephan tags: jni) | |
|
2023-08-09
| ||
| 23:47 | Port the SQLTester 'v1' commands to the 'v2' evaluation bits. Still TODO is swapping out v1 with these separate impls. (check-in: 0cf57e5b0f user: stephan tags: jni) | |
| 22:30 | Correct REQUIRED_PROPERTIES handling to not fail if there are no properties. (check-in: 7a19bef4f5 user: stephan tags: jni) | |
| 22:18 | More for the SQLTester rework. Can read input and dispatch commands, but only --print is currently implemented. (check-in: 4fa2ad33ed user: stephan tags: jni) | |
| 19:51 | Initial sketches of a line-by-line parser for SQLTester to overcome its compatibility shortcomings. Far from complete. (check-in: 43534cd042 user: stephan tags: jni) | |
| 18:25 | Adapt TestScript to skip REQUIRED_PROPERTIES, per [7a07863e082664da], and improve a couple of adject reasons-for-skipping messages. (check-in: f937097e9b user: stephan tags: jni) | |
| 17:47 | Add an extra constraint to the test-script-interpreter.md spec. (check-in: 7a07863e08 user: drh tags: jni) | |
| 17:23 | SQLTester --result command: do not double-{}-wrap error messages and do not throw on db error. (check-in: f7be20f5a6 user: stephan tags: jni) | |
| 17:04 | Move the test command body's trim() back into the TestScript class (to simplify Command-level code) but do it after double-verbose has had the chance to emit it as-is (so that debug output is not mangled by the trim). (check-in: f15ecb68f7 user: stephan tags: jni) | |
| 16:56 | Tweak the SQLTester --verbose and double-verbose output a bit for legibility. (check-in: 46b79afaaf user: stephan tags: jni) | |
| 16:29 | Strip --oom commands from SQLTester input since (A) we can't currently do anything with them and (B) they can appear as body content of --testcase commands and the current parser cannot deal with that. If --verbose is provided once, emit the name and args of each command as it's run. If --verbose is used twice or more, also emit the command's body text, if any. (check-in: 0770e8467d user: stephan tags: jni) | |
| 16:03 | Update the definition of "special characters" for the SQLTester and correct the code to match it. (check-in: 217f6e0c9c user: stephan tags: jni) | |
| 15:46 | Extend TestScript to be able to report why it should be skipped. Expand the test-skipping rules to account for the current spec doc. Add the {} empty-string case to the spec doc. (check-in: 4fcc8cb0cc user: stephan tags: jni) | |
| 14:47 | Add a test for the --null command. (check-in: 83ac815deb user: stephan tags: jni) | |
| 14:43 | Extend SQLTester glob support with '#'. (check-in: 756ef83f45 user: stephan tags: jni) | |
| 14:24 | Implement the SQLTester's result escaping rules. (check-in: 61bb950873 user: stephan tags: jni) | |
| 13:51 | Add --json and --json-block SQLTester commands. (check-in: 478129d901 user: stephan tags: jni) | |
| 13:16 | Add SQLTester --tableresult command. (check-in: 8c5b6d893d user: stephan tags: jni) | |
| 12:05 | Clean up the SQLTester output a bit by using the module name, instead of filename, where appropriate. (check-in: 5323e4fd25 user: stephan tags: jni) | |
| 11:10 | Document SQLTester's --print command and add some argument validation to it. (check-in: ab9c945bb0 user: stephan tags: jni) | |
| 11:05 | Rework how SQLTester's Command objects are dispatched and how TestScript stores its command entries. (check-in: f929f1f7f7 user: stephan tags: jni) | |
| 09:56 | Some minor tweaks in SQLTester in prep for larger refactoring. (check-in: 1d93f93ac9 user: stephan tags: jni) | |
|
2023-08-08
| ||
| 22:10 | Remove the current-statement tracking from the JNI internals because it will break down in the face of client-side mixed-mode native/java code, e.g. in cases like SQLTester. This makes tracing of sqlite3_stmt a micron slower but also reliably correct. (check-in: 4c0ec89dca user: stephan tags: jni) | |
| 21:22 | Adapt JNI build to be buildable with or without SQLTester. (check-in: adae7d7869 user: stephan tags: jni) | |
| 21:05 | Add SQLTester dup() and dup_count() UDFs. Correct arg handling of the --run command. (check-in: 0dba3073f4 user: stephan tags: jni) | |
| 20:41 | Correct --result arg count check and add infrastructure to let us add custom C-side behavior to SQLTester via an auto extension. (check-in: bb8321702e user: stephan tags: jni) | |
| 20:15 | If SQLTester now treats a no-args --result as comparing against an empty string. (check-in: 49005ca5cc user: stephan tags: jni) | |
| 20:02 | Add glob/notglob commands to SQLTester and complete the interrupted-midway impls of the strglob() and strlike() JNI bindings. (check-in: 4ba98ec0bf user: stephan tags: jni) | |
| 19:20 | When SQLTester hits an unknown command, emit a warning and skip the rest of that script instead of aborting the whole run, per /chat discussion. Reduce verbosity a bit. (check-in: 3e78d22d04 user: stephan tags: jni) | |
| 17:36 | Minor cleanups in sqlite3Int.h. By reordering some fields in the Parse object, it packs more tightly and uses less memory and less CPU to initialize. (check-in: aa6de539c0 user: drh tags: trunk) | |
| 16:53 | Fix an issue in the amalgamation generator in which it was not correctly expanding the SQLite version in the header comment. (check-in: 293f6191e9 user: drh tags: trunk) | |
| 14:58 | Correct the spacing output of multi-select SQL blocks for SQLTester --result. (check-in: 8d98645a9e user: stephan tags: jni) | |
| 14:49 | Updates to the test script interpreter spec: Rather than failing immediately upon encountering an incompatibility, simply abandon the rest of that particular input file. (check-in: d2c99b96f4 user: drh tags: jni) | |
| 14:40 | A number of baby steps for SQLTester.java. It can now handle basic --result cases. (check-in: 0404f688f6 user: stephan tags: jni) | |
| 14:25 | Add a few words of clarification on how the SQLite Test Script Interpreter should be initialized to start each test script. (check-in: 3aa2b5a5ca user: drh tags: jni) | |
| 13:05 | Correct JNI sqlite3_prepare() to emit a null stmt handle when the C counterpart succeeds but results in a NULL pointer. (check-in: 94628f88b5 user: stephan tags: jni) | |
| 11:46 | Implement the new/open/close SQLTester commands. (check-in: dc823bf00f user: stephan tags: jni) | |
| 11:46 | Add a JNI-layer sqlite3.toString() for debugging. (check-in: 456691649a user: stephan tags: jni) | |
| 10:58 | Bind sqlite3_strlike/strglob() to JNI. (check-in: eb5440f71b user: stephan tags: jni) | |
| 09:45 | SQLTester: add print command and improve argument error reporting infrastructure. (check-in: 1b6e84f6aa user: stephan tags: jni) | |
| 00:59 | Add missing license header. Minor cleanups in SQLTester. (check-in: 5be50fd588 user: stephan tags: jni) | |
| 00:37 | Rework SQLTester dispatching and add stub impls for several commmands. (check-in: 9e61af75ac user: stephan tags: jni) | |
|
2023-08-07
| ||
| 23:59 | Add command dispatcher to SQLTester. (check-in: e0a06931e9 user: stephan tags: jni) | |
| 23:04 | SQLTester now ignores tests which contain constructs specified in the spec doc. (check-in: ecaeee652a user: stephan tags: jni) | |
| 22:32 | SQLTester can now split a test script into a series of individual commands. (check-in: d3d1accc8b user: stephan tags: jni) | |
| 22:02 | SQLTester can now read a script and strip it of all noise content. (check-in: 59bd392817 user: stephan tags: jni) | |
| 21:04 | Initial skeleton for adding an SQL-driven test script interpreter for the JNI bindings. (check-in: 2aa8f0edec user: stephan tags: jni) | |
| 17:09 | Enhance sqlite3_vtab_nochange() so that it works with "UPDATE ... FROM ..." statements. Use this to allow some updates on fts5 contentless-delete tables. (check-in: 27ff86e4d8 user: dan tags: trunk) | |
| 16:15 | Changes so that sqlite3_vtab_nochange() works with "UPDATE ... FROM...". Use this to allow UPDATE on a contentless fts5 table if new values are supplied for all indexed columns. (Closed-Leaf check-in: 16cd2161e3 user: dan tags: fts5-contentless-delete) | |
| 11:18 | When converting a Java exception to a db error message, use Throwable.toString() instead of getMessage() so that the exception type's name is included. More internal API renaming for consistency. (check-in: 2d44720d06 user: stephan tags: jni) | |
| 10:59 | Lots of JNI internal API renaming, for consistency, and moving-around of utility functions. Make it safe for more callback types to throw. (check-in: 9a494394b9 user: stephan tags: jni) | |
| 10:01 | JNI API renaming to better match the C API. (check-in: 6e0bd03d0b user: stephan tags: jni) | |
| 09:44 | Rename fts5_api pContext parameters to pUserData, per /chat discussion. This is a cosmetic change made to reduce confusion between those parameters and the two other context-type parameters in that API. (check-in: 2ca064d8eb user: stephan tags: trunk) | |
| 01:06 | Make sqlite3_stmt() Java ctor private - it's only constructed from JNI code. (check-in: ce82c42f15 user: stephan tags: jni) | |
| 00:29 | Minor internal cleanups and additional test metrics. (check-in: fa0a6b6e8e user: stephan tags: jni) | |
| 00:06 | Bind the auto-extension APIs to JNI. (check-in: 746a5fa079 user: stephan tags: jni) | |
|
2023-08-06
| ||
| 22:09 | Rework the sqlite3_open(_v2)() order of operations so that pending auto-extension support can get ahold of the open-time Java state despite the Java/C (sqlite3*) binding not having yet been established. (check-in: 34da294ab5 user: stephan tags: jni) | |
| 21:29 | Completely rework how the JNI sqlite3_open(_v2) and sqlite3_prepare(_vN)() bindings deal with output pointers to give the JNI side full control over the origin of db and stmt handles (necessary for solving chicken/egg situations in auto-extensions and prepare-time trace). Lots of adjacent internal API renaming. (check-in: 644999caff user: stephan tags: jni) | |
| 20:01 | Start adding JNI-side support for auto extensions, but hit a brick wall which requires slightly awkward semantics changes in the JNI bindings for sqlite3_open(_v2)() to resolve, so stash this #if'd out for the time being. (check-in: 77a32d238e user: stephan tags: jni) | |
| 15:01 | Doc cleanups. (check-in: 5f56b00770 user: stephan tags: jni) | |
| 14:19 | Bind sqlite3_config() to JNI for call variants taking (int,int*) and (int,const char *) variadic arguments. (check-in: 6119289da8 user: stephan tags: jni) | |
| 13:02 | JNI-internal cleanups and API renaming. Add a C-side java-string-to-utf8 conversion. (check-in: 672d85795d user: stephan tags: jni) | |
| 11:05 | Remove doc outdated warning about sqlite3_trace_v2() JNI binding being incompatible with MUTF-8. Use new to-string capability to simplify Fts5ExtensionApi::xColumnText() JNI binding. (check-in: ebcfc2379b user: stephan tags: jni) | |
| 10:49 | Add a way to convert from standard UTF-8 to a Java string (JNI lacks this capability). (check-in: 07dd082c9e user: stephan tags: jni) | |
| 10:14 | Bind sqlite3_db_filename() and (closely related) (A) add many more docs about the UTF-8/MUTF-8 discrepancy (B) start adding internals to enable us to perform the standard-UTF-8-to-Java conversion from C. (check-in: 586720fa71 user: stephan tags: jni) | |
|
2023-08-05
| ||
| 22:41 | Bind sqlite3_set_authorizer() to JNI. (check-in: e0fa031359 user: stephan tags: jni) | |
| 21:35 | Refactor the per-JNIEnv cache from a fixed-size static array to a linked list of dynamically-allocated entries. Uncache all per-db state (which is necessarily JNIEnv-specific) when the corresponding JNIEnv is uncached. (check-in: 9dd8b78419 user: stephan tags: jni) | |
| 20:19 | Add SQLite3Jni.uncacheJniEnv(), a way for Java threads to clear their thread-specific cached state from the JNI bindings when they're about to terminate (or are otherwise done using the library). (check-in: 7468f8761b user: stephan tags: jni) | |
| 19:20 | In the JNI docs, note that the sizeof SQLITE_TRANSIENT and SQLITE_STATIC should ideally be the same as the platform's pointer size. (check-in: 7d4ac44ec4 user: stephan tags: jni) | |
| 19:15 | Tiny doc typo fixes from [forum:31eb707250|forum post 31eb707250]. No code changes. (check-in: 8ede50082e user: stephan tags: trunk) | |
| 12:48 | JNI internal refactoring and cleanups. (check-in: 7397973a2f user: stephan tags: jni) | |
| 11:16 | Bind Fts5ExtensionApi::xUserData() to JNI and extend xCreateFunction() to accept that argument. In test code, use assert() instead of exceptions if assert() is enabled so that test failures (exceptions) thrown via callbacks do not get suppressed (which they otherwise necessarily are to avoid crashing the host app). (check-in: e438373776 user: stephan tags: jni) | |
| 04:30 | Add missing Fts5PhraseIter.java to the build. (check-in: 96281ad0d5 user: stephan tags: jni) | |
| 04:23 | Bind fts5_api::xCreateFunction() to JNI and demonstrate it with a test. (check-in: c653bf16cb user: stephan tags: jni) | |
| 01:28 | Start adding fts5_api to JNI. (check-in: 14d18fe983 user: stephan tags: jni) | |
| 00:40 | Bind the remaining Fts5ExtensionApi methods to JNI, noting that all are completely untested because the higher-level bits needed to do so are still missing. (check-in: 23383c1dfd user: stephan tags: jni) | |
|
2023-08-04
| ||
| 16:01 | In the opfs-sahpool VFS's importDb() and exportFile() methods, throw if the actually-wrote/read amounts differ from the expected-to-write/read amounts, per feedback in [forum:a4122e986f|forum post a4122e986f]. (check-in: a617ebf4e5 user: stephan tags: trunk) | |
| 15:38 | More work on the JNI binding of fts5 customization (still a long ways to go). (check-in: 1a246fd216 user: stephan tags: jni) | |
| 13:29 | Merge the json_remove() bug fix from trunk into the bedrock branch. (check-in: a8872aa588 user: drh tags: bedrock) | |
| 13:27 | Eliminate code duplication in the two JNI tester classes. (check-in: 63e7bbe3d5 user: stephan tags: jni) | |
| 13:24 | Merge the json_remove() bug fix from trunk into the wal2 branch. (check-in: e6c066cae4 user: drh tags: wal2) | |
| 13:19 | Fix a bug in the jsonLookup() routine used to search JSON. Bug was introduced by the recent (unreleased) JSON caching enhancements check-in [a4c1af616e672a0d] merged into trunk at [df099ad713011b67]. (check-in: ba7a66db13 user: drh tags: trunk) | |
| 13:03 | FTS-related JNI refactoring. Move FTS-specific tests into their own class and dynamically load it, if possible, from the main test app. (check-in: b7a8428fcd user: stephan tags: jni) | |
| 12:44 | More work towards binding FTS5 customization to JNI. Add Fts*.java files missing from previous checkin. (check-in: 91263178f4 user: stephan tags: jni) | |
| 11:10 | Change comment on Fts5ExtensionApi.iVersion from "always set to 3" to "always set to 2". (check-in: b7de399c27 user: dan tags: trunk) | |
| 11:08 | Start including fts5 customization bits into JNI, but it's far from functional. (check-in: abaf5edd04 user: stephan tags: jni) | |
| 09:53 | Improve internal error handling in the JNI create_function() impl. (check-in: 2c88390faa user: stephan tags: jni) | |
| 08:45 | Resolve the timing/ordering issue of a JS-to-WASM-converted xDestroy() function being uninstalled from WASM right before the underlying native call tries to call it. This has been a long-unnoticed bug which appears only when removing such functions or replacing them. (check-in: 031c9a76b6 user: stephan tags: trunk) | |
| 08:41 | Part 2 of the fix from [a0f808363318c00fd1db78b]. (check-in: 5c3104228d user: stephan tags: trunk) | |
| 08:39 | More work towards fts5 customzation in JS. (Leaf check-in: ce2a65d80f user: stephan tags: wasm-fts5) | |
|
2023-08-03
| ||
| 23:26 | Merge the latest trunk enhancements into the bedrock branch. (check-in: 877ffece05 user: drh tags: bedrock) | |
| 23:22 | Merge the latest trunk enhancements into the wal2 branch. (check-in: 6fa71e1119 user: drh tags: wal2) | |
| 22:43 | Minor internal cleanups in the JS-side fts5 cleanup steps. (check-in: 2666f52e88 user: stephan tags: wasm-fts5) | |
| 21:14 | Retain all xDestroy method WASM proxies created for fts5_api::xCreateFunction() until the db is closed, then free them all at once. (check-in: f4b9743abd user: stephan tags: wasm-fts5) | |
| 20:45 | Rework the fts5 db-close cleanup to avoid that fts5 internally calls a just-unbound-from-wasm xDestroy() method. There is still work to do here to cover the case of fts5 functions being replaced (in which all xDestroy methods have to be retained until the db is closed). (check-in: 795f22421b user: stephan tags: wasm-fts5) | |
| 20:02 | Part 2 of the fix from [a0f808363318c00fd1db78b]. (check-in: ce0674b192 user: stephan tags: wasm-fts5) | |
| 18:46 | Resolve the timing/ordering issue of a JS-to-WASM-converted xDestroy() function being uninstalled from WASM right before the underlying native call tries to call it. This has been a long-unnoticed bug which appears only when removing such functions or replacing then. (check-in: a0f8083633 user: stephan tags: wasm-fts5) | |
| 18:14 | Fix harmless scan-build warnings. (check-in: 65a8716d8a user: drh tags: trunk) | |
| 16:22 | Fix a discrepancy between sqlite3_values_to_js() and its API docs/intended behavior. (check-in: 7b85c68fe2 user: stephan tags: trunk) | |
| 16:21 | More infrastructure towards binding FTS5 customization support to WASM. (check-in: 163e3e44b1 user: stephan tags: wasm-fts5) | |
| 16:02 | Fix a discrepancy between sqlite3_values_to_js() and its API docs/intended behavior. (check-in: 16658b3c64 user: stephan tags: wasm-fts5) | |
| 13:30 | Fix build problems associated with SQLITE_OMIT_WSD and SQLITE_OMIT_TRIGGER. (check-in: de031a5ec8 user: drh tags: trunk) | |
| 13:07 | Fix a test case whose value changes under the no_optimization permutation due to the recent change to add the ability for SQLITE_TESTCTRL_OPTIMIZATION to disable one-pass. (check-in: 976e39c632 user: drh tags: trunk) | |
| 12:41 | Unix builds now assume the presence of nanosleep() in the standard library. The -DHAVE_NANOSLEEP=0 compile-time option can be used to build on systems (if any still exist) where this is not the case. (check-in: 779d5dc879 user: drh tags: trunk) | |
| 07:20 | Initial work on exposing the FTS5 APIs to wasm, per [forum:28402061cb|request in the forum]. This builds and loads the structs into JS but is completely untested. (check-in: 52c8b73ae3 user: stephan tags: wasm-fts5) | |
|
2023-08-02
| ||
| 18:20 | If a query has an ORDER BY clause that only refers to result columns of the left-most table and the left most table is a MATERIALIZED common table expresion, then attempt to push the ORDER BY clause down into the subquery. (Leaf check-in: 8e7a70b2bb user: drh tags: order-by-push-down) | |
| 17:29 | More JNI-internal docs. Correct handling of a Java exception if a trace callback throws. (check-in: 306b269a01 user: stephan tags: jni) | |
| 17:20 | Internal doc additions in the JNI bits. (check-in: b56643644e user: stephan tags: jni) | |
| 16:06 | Performance optimization for JSON rendering logic. (check-in: ea0b9aecba user: drh tags: trunk) | |
| 13:45 | Remove an unreachable branch in the ascii-to-floating-point conversion that was added by [e989a37ff9d5b52e]. (check-in: c4347e4400 user: drh tags: trunk) | |
| 13:38 | Stricter enforcement of the idea that a MATERIALIZED common table expression is an optimization fence. (check-in: 354425f8d9 user: drh tags: trunk) | |
| 13:29 | Remove a condition from query flattening that is now taken care of by the caller. Factor out the reverse_unordered_selects processing from the main loop of sqlite3WhereBegin() for performance. (Closed-Leaf check-in: f068f105fb user: drh tags: materialize-order-by) | |
| 12:39 | Simplify the select.c logic that enforces the optimization fence around a MATERIALIZED common table expression. Do not allow the reverse_unordered_selects setting to affect a fenced CTE that itself has an ORDER BY clause. (check-in: 165b9bda24 user: drh tags: materialize-order-by) | |
| 11:06 | Never flatten a CTE that is labeled MATERIALIZED. This is really something of a bug fix because the documentation says that a MATERIALIZED CTE is an optimization barrier. (check-in: b7ef9796f5 user: drh tags: materialize-order-by) | |
| 00:09 | When a query is inside an AS MATERIALIZED CTE, do not attempt the omit-ORDER-BY optimization. If the developer specifies MATERIALIZED, that means he wants the sort to actually happen. (check-in: aa769ee747 user: drh tags: materialize-order-by) | |
|
2023-08-01
| ||
| 19:10 | Avoid ASAN warnings when converting over-sized long double values into double. (check-in: e989a37ff9 user: drh tags: trunk) | |
| 16:41 | Filter the wasmfs-specific JS module result type check out of non-wasmfs builds. (check-in: ef7c121049 user: stephan tags: trunk) | |
| 16:38 | Fix part 2 of 2 for /fiddle: a make deps problem which caused fiddle to load but fail to run when sqlite3.c was missing. (check-in: 6011bc2676 user: stephan tags: trunk) | |
| 16:17 | Fix part 1 of 2 for /fiddle not running: build flags which were renamed during refactoring. (check-in: 214ffeeff7 user: stephan tags: trunk) | |
| 11:03 | Fix an assert() in fts5_index.c that might fail when dealing with corrupt records. (check-in: aa55c83f35 user: dan tags: trunk) | |
| 10:19 | More docs for the Java side of the JNI bindings. (check-in: d8e9bcee96 user: stephan tags: jni) | |
| 09:44 | Internal exception-handling cleanups in the JNI bindings. (check-in: 057b1d4f6f user: stephan tags: jni) | |
| 08:48 | Minor JNI-related doc tweaks and code re-ordering. (check-in: b663b27e42 user: stephan tags: jni) | |
| 01:07 | Remove redundant comparison operator. (check-in: 0189843298 user: drh tags: trunk) | |
| 00:34 | Restore part of the UPDATE one-pass optimization that was removed by check-in [2c56b984a0bd3be5]: only disable one-pass if the WHERE clause contains a subquery. Allow subqueries in the SET expressions. Fix for performance problem reported by [forum:/forumpost/8ab195fd44e75ed0|forum post 8ab195fd44e75ed0]. (check-in: 42916af9fc user: drh tags: trunk) | |
|
2023-07-31
| ||
| 22:03 | Restrict the new column cache to table-btree, which is the common case anyhow. That way, writes to indexes do not need to clear the column cache. (check-in: 659284ab0e user: drh tags: trunk) | |
| 20:02 | The original column-cache implementation from check-in [ab1edcc7fedcf279] (merged to trunk at [771fe35074b50b8d]) is unsound. This check-in fixes the issue. Had to give back a little performance, the optimization is still a overall win. (check-in: ec95e970fb user: drh tags: trunk) | |
| 17:39 | Add bit to sqlite3_test_control(SQLITE_TESTCTRL_OPTIMIZATIONS) that will disable the one-pass optimizating in DELETE and UPDATE. (check-in: 22f6dd5cd3 user: drh tags: trunk) | |
| 17:20 | Improved debug tracking of the Parse.nQueryLoop value. (check-in: 4a8a17498e user: drh tags: trunk) | |
| 13:52 | Experimentally change the JNI sqlite3_trace_v2() callback type to have more convenient access to the current Java-side sqlite3_stmt at the cost of some uncomfortably fiddly current-statement tracking in the JNI layer. Subject to change. (check-in: 459db332af user: stephan tags: jni) | |