11007 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Pavan-Nambi	8024bb3bde	add tests	2025-11-25 07:51:41 +05:30
Pavan-Nambi	06c06ac8d2	drop triggers if table drops	2025-11-23 10:56:14 +05:30
Preston Thorpe	a97081eafe	Merge 'translate/planner: Implement Index creation on arbitrary expressions' from Preston Thorpe ... This PR implements the functionality to create an index on an expression instead of just a column, and makes an attempt for the planner to be able to use these indexes. https://sqlite.org/expridx.html This adds the field: `expr: Option<Box<Expr>>` to `IndexedColumn` which is `None` for regular indexed columns. This also adds the sentinel value `EXPR_INDEX_SENTINEL` for the `position_in_table` field of indexed col expressions, since they are not, actually in the table.. and similarly: `Constraint` now has the `table_col_position` as an Option now, because it is no longer a guarantee that it refers to a col expression. Reviewed-by: Jussi Saurio <jussi.saurio@gmail.com> Closes #3990	2025-11-22 06:52:20 -05:00
Pekka Enberg	b5c5666a49	Turso 0.4.0-pre.3	2025-11-22 09:56:18 +02:00
Pekka Enberg	bb2df133ea	Merge 'Kill unwrap in incremental module' from Pekka Enberg ... As preparation for #3992 which prohibits use of unwrap, remove uses of it in the incremental module. Closes #4001	2025-11-22 09:55:35 +02:00
Jussi Saurio	06a01c4366	Merge 'mvcc: refactor RowID.row_id to be either i64 or a record' from Jussi Saurio ... to make interoperability with VDBE opcodes easier, we are going to store versioned index rows (unlike in Hekaton). for this reason refactor RowID.row_id to either be an i64 or an index key (record). this lets us store a SkipMap from index key to a list of index row versions like normal table row versions. Reviewed-by: Pere Diaz Bou <pere-altea@homail.com> Closes #4002	2025-11-21 22:07:15 +02:00
Jussi Saurio	c1a9f5c951	mvcc: refactor RowID.row_id to be either i64 or a record ... to make interoperability with VDBE opcodes easier, we are going to store versioned index rows (unlike in Hekaton). for this reason refactor RowID.row_id to either be an i64 or an index key (record). this lets us store a SkipMap from index key to a list of index row versions like normal table row versions.	2025-11-21 17:05:56 +02:00
Pekka Enberg	f1401ab007	core/incremental: Improve error handling	2025-11-21 15:08:01 +02:00
Pekka Enberg	15f1619c42	Merge 'Kill unwrap() calls in extensions' from Pekka Enberg ... As preparation for #3992 that actually enforces no unwraps, remove them from extensions. Reviewed-by: Preston Thorpe <preston@turso.tech> Closes #4000	2025-11-21 14:19:18 +02:00
Pekka Enberg	227c979882	extensions/tests: Allow use of unwrap() ... They're tests, it's fine.	2025-11-21 14:11:53 +02:00
Pekka Enberg	dcaf6f88d0	extensions: Replace unwrap() with expect() in fuzzy extension	2025-11-21 14:11:53 +02:00
Pekka Enberg	f044e39d43	extensions: Eliminate unwrap() in fuzzy extension	2025-11-21 14:11:53 +02:00
Pekka Enberg	cae17e7a36	extensions: Eliminate unwrap() call in csv extension	2025-11-21 14:11:53 +02:00
Pekka Enberg	deeb79c80e	extensions: Eliminate unwrap() in crypto extension	2025-11-21 14:11:53 +02:00
Pekka Enberg	eb07a5164b	extensions: Improve error handling in percentile extension ... Eliminate unwrap() where we can, replace with expect() where we can't.	2025-11-21 14:11:53 +02:00
Pekka Enberg	96e63b4556	extensions: Replace unwrap() with expect() in get_column_names() ... ...error handling would be better here.	2025-11-21 14:11:53 +02:00
Pekka Enberg	582cc057ce	extensions: Eliminate unwrap() in IndexInfo::to_ffi()	2025-11-21 14:11:53 +02:00
Pekka Enberg	23859ec6ef	extensions: Replace unwrap() with expect() in generate_random_number() ... Propagating the error would be even better, of course.	2025-11-21 14:11:53 +02:00
Pekka Enberg	d2fc52834a	Merge 'SQLite C API improvements' from Nikita Sivukhin ... This PR adds rich error codes for sqlite3 C bridge layer and also do few minor adjustments: 1. Enable indices by default 2. Return debug logs to the stderr Closes #3999	2025-11-21 13:20:44 +02:00
Nikita Sivukhin	cb3ae4d7fd	remove unnecessary constant ... - it is not exported as a symbol - so no need to have it	2025-11-21 14:27:04 +04:00
Nikita Sivukhin	78b9fa54a5	fix clippy	2025-11-21 14:04:04 +04:00
Nikita Sivukhin	daaa91259b	remove debug logs	2025-11-21 13:58:20 +04:00
Nikita Sivukhin	228f2c1be9	return proper error code in case of certain LimboErrors	2025-11-21 13:58:07 +04:00
Nikita Sivukhin	f5e8eaa6c1	enable indices by default	2025-11-21 13:57:34 +04:00
Nikita Sivukhin	139ec0035e	copy errors codes from SQLite	2025-11-21 13:56:56 +04:00
Pekka Enberg	2f078f13e6	Merge 'simulator: only check all tables if we have any tables to check' from Pedro Muniz ... Fixes: #3996 Closes #3997	2025-11-21 09:39:55 +02:00
pedrocarlo	03f38ae72c	simulator: only check all tables if we have any tables to check	2025-11-21 01:10:55 -03:00
Preston Thorpe	e40b07122d	Merge 'Support table xinfo' from Nikita Sivukhin ... Support extended pragma `table_xinfo` which in addition to all `table_info` columns report extra column about if the column is hidden/generated/dynamic (while usually all such columns filtered out from the output of `table_info` pragma). Reviewed-by: Preston Thorpe <preston@turso.tech> Closes #3994	2025-11-20 19:51:52 -05:00
PThorpe92	5a801f5405	Fix clippy error	2025-11-20 17:38:18 -05:00
PThorpe92	9223dfdbff	Add integration test to ensure covering expr index is used	2025-11-20 12:47:50 -05:00
PThorpe92	94a0f7f17a	Use expression indexes as covering when possible	2025-11-20 12:47:50 -05:00
PThorpe92	2259e62fae	Add expression_index.rs to translate module with some helpers	2025-11-20 12:47:50 -05:00
PThorpe92	1719bbbc2f	Rewrite delete rowset_plan to replace covering idx cursor w/ table cursor for idx expressions	2025-11-20 12:47:49 -05:00
PThorpe92	a8e22d4253	Clippy fix	2025-11-20 12:47:49 -05:00
PThorpe92	b3b95f9e0f	Add comments and cleanups	2025-11-20 12:47:49 -05:00
PThorpe92	2688f8b9c3	Add tcl tests for expr indexes and collation	2025-11-20 12:47:49 -05:00
PThorpe92	e593028b0c	Respect collation in index DDL statements	2025-11-20 12:47:48 -05:00
PThorpe92	563b1d8b6d	Fix optimizer make useable constraint require column position for index	2025-11-20 12:47:48 -05:00
PThorpe92	8376f9e46d	Fix rewriting idx expression to existing registers in upsert	2025-11-20 12:47:48 -05:00
PThorpe92	67f3ed92f4	Fix expr index handling in upsert	2025-11-20 12:47:48 -05:00
PThorpe92	6c163f5202	Fix expr_to_column_order to return None when mask extraction fails or spans >1 table	2025-11-20 12:47:47 -05:00
PThorpe92	fae566b3e0	Cleanup expr indexes, use raw pointer for ColumnTarget::Expr	2025-11-20 12:47:47 -05:00
PThorpe92	b0ac042092	Add tests for using expr indexes in query plans for ORDER BY and WHERE predicates	2025-11-20 12:47:47 -05:00
PThorpe92	9f735feb10	add some comments in optimizer	2025-11-20 12:47:46 -05:00
PThorpe92	04629d4d7b	Add some expression indexes in our fuzz tests	2025-11-20 12:47:46 -05:00
PThorpe92	cd440e2684	Make ORDER BY expression aware and wire index exprs into optimizer	2025-11-20 12:47:46 -05:00
PThorpe92	420615c4de	Wire expression indexes plumbing into UPDATE emission	2025-11-20 12:47:46 -05:00
PThorpe92	1a417ef4dc	Begin to implement index creation on arbitrary expressions	2025-11-20 12:47:45 -05:00
Pere Diaz Bou	86119b0dba	Merge 'core/mvcc/cursor: implement prev and last ' from Pere Diaz Bou ... Backward scan of a table wasn't implemented yet in MVCC so this achieves that. I added simple test for mixed btree and mvcc backward scan but I should add more intense testing for this. <!-- CURSOR_SUMMARY --> --- > [!NOTE] > Implements backward scanning and last() in MVCC lazy cursor and adds directional rowid iteration in the MVCC store, with new tests for mixed MVCC+B-Tree backward scans. > > - **MVCC Cursor (`core/mvcc/cursor.rs`)**: > - Implement `prev()` and `last()` with mixed MVCC/B-Tree coordination using `IterationDirection`. > - Add `PrevState` and extend state machine to handle backward iteration. > - Update `get_new_position_from_mvcc_and_btree(...)` to choose rowids based on direction. > - Integrate B-Tree cursor calls (`last`, `prev`) and adjust `rewind`/rowid selection; tweak next-rowid when at `End`. > - **MVCC Store (`core/mvcc/database/mod.rs`)**: > - Add `get_prev_row_id_for_table(...)` and generalized `get_row_id_for_table_in_direction(...)` supporting forward/backward scans. > - Add tracing and minor refactors around next/prev rowid retrieval. > - **Tests (`core/mvcc/database/tests.rs`)**: > - Add test for backward scan combining B-Tree and MVCC and an ignored test covering delete during backward scan. > > <sup>Written by [Cursor Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit 430bd457e6. This will update automatically on new commits. Configure [here](https://cursor.com/dashboard?tab=bugbot).</sup> <!-- /CURSOR_SUMMARY --> Closes #3980	2025-11-20 18:41:41 +01:00
Nikita Sivukhin	0c1e2a8a10	fix test	2025-11-20 18:03:32 +04:00