Commit Graph

333 Commits

Author SHA1 Message Date
pedrocarlo
2954e2e7bf shrinking: remove table assertions for non-dependent tables 2025-08-18 12:44:49 -03:00
pedrocarlo
59da828362 do not shadow FaultyQuery's immediately. Only shadow them later 2025-08-18 12:40:02 -03:00
pedrocarlo
7fb14cfc76 add File path to SimulatorFile 2025-08-18 11:52:10 -03:00
pedrocarlo
6388ed2017 FaultyQuery enabled by default 2025-08-18 11:52:10 -03:00
Jussi Saurio
6c17fa2a5e fix/sim: prevent sim from trying to create an existing table or index
We recently merged a change that panics the sim on parse errors, because
not doing so has masked many scenarios where the sim unintentionally
creates incorrect sql and we just ignore it.

We already have Property::DoubleCreateFailure to assert that the same table
cannot be created twice, so this should not hide any bugs.
2025-08-17 18:13:05 +03:00
Jussi Saurio
0a3ebf8914 Fail simulator on parse errors 2025-08-16 10:27:10 +03:00
PThorpe92
2810b5d8a2 Add --keep-files flag and force Drop of database in simulator to prevent issues with the Registry 2025-08-15 17:54:40 -04:00
pedrocarlo
2bc6edc3d4 introduce TableContext for the simulator to properly generate predicates for Joins 2025-08-15 23:44:00 +03:00
Jussi Saurio
3760d44c13 sim: force compound selects to have the same number of result columns 2025-08-15 23:43:38 +03:00
Jussi Saurio
96072509f5 sim: add Property::TableHasExpectedContent 2025-08-15 11:16:31 +03:00
Jussi Saurio
4c76191fec fix/sim: fix incorrect implementation of compound select generation
Problem:

sim was generating compound selects like this:

- pick a random `table`
- create a random single SELECT
- create `n` random compound SELECTs (UNION ALL etc)

assign a WHERE clause that always has a condition based on `table`

This can result in e.g.

```
SELECT a FROM foo WHERE bar.x = 'baz'
```

Solution:

Don't base the WHERE clause on a table that might not be referenced
in the query.

Again, the only reason this wasn't caught before was because `FaultyQuery`
and `FsyncNoWait` are the only paths where this is actually tested with an
assertion, and those are both disabled
2025-08-15 10:48:42 +03:00
Jussi Saurio
165f2aed23 make big joins even less likely 2025-08-14 19:35:16 +03:00
Jussi Saurio
f5438d14de make big compound selects even less likely 2025-08-14 19:35:16 +03:00
Jussi Saurio
aed48bb665 sim: reduce frequency of compound selects and complex joins 2025-08-14 19:35:16 +03:00
Jussi Saurio
1a11648974 sim: add Property::ReadYourUpdatesBack 2025-08-14 19:04:27 +03:00
Jussi Saurio
0b9d07db4c sim: restrict Update::arbitrary_from() to SET each column max once 2025-08-14 18:55:34 +03:00
Jussi Saurio
87bf488bbc chore: use rusqlite 0.37 with bundled sqlite everywhere 2025-08-11 15:13:57 +03:00
pedrocarlo
0c9ecbc768 simulator should delete files after a successful run 2025-08-09 16:25:51 -03:00
PThorpe92
f6a68cffc2 Remove RefCell from IO and Page apis 2025-08-05 16:24:49 -04:00
pedrocarlo
543cdb3e2c underscoring completions and IOResult to avoid warning messages 2025-07-31 11:51:17 -03:00
PThorpe92
daec8aeb22 impl pwritev for simulator file 2025-07-30 19:42:49 -04:00
PThorpe92
9c75872827 Make writing wal header able to be sync 2025-07-30 14:05:18 +03:00
PThorpe92
dcf346b9b0 Update apparently very old branch to new api 2025-07-30 13:48:05 +03:00
PThorpe92
52f63b2af0 Impl truncate for simulator 2025-07-30 13:48:05 +03:00
pedrocarlo
3831e0db39 convert must_use compile warnings to unused_variables to track locations where we need to refactor in the future 2025-07-28 16:09:26 -03:00
pedrocarlo
d30c7d54c8 change all Arc<Completion> to Completion 2025-07-28 15:32:45 -03:00
meteorgan
b5a18d7dc9 fix get_column_name() when column name doesn't exist 2025-07-25 23:49:31 +08:00
bit-aloo
9a54ef214e parser: Distinguish quoted identifiers and unify Id into Name enum
This commit replaces the `Name(pub String)` struct with a `Name` enum that
explicitly models how the name appeared in the source either as an
unquoted identifier (`Ident`) or a quoted string (`Quoted`).

In the process, the separate `Id` wrapper type has been coalesced into the
`Name` enum, simplifying the AST and reducing duplication in identifier
handling logic.

While this increases the size of some AST nodes (notably `yyStackEntry`),
it improves correctness and makes source structure more explicit for
later phases.
2025-07-24 14:40:19 +05:30
Jussi Saurio
e604a80e93 fix/sim: actually enable indexes by default 2025-07-21 13:35:11 +03:00
Jussi Saurio
dce355d9b7 test/stress&sim: enable indexes by default 2025-07-21 11:49:30 +03:00
Jussi Saurio
b1ebc1c82f Merge 'sim: show seed in 'execute_interaction()' trace' from Jussi Saurio
This helps e.g. when you hit some infinite loop

Closes #2161
2025-07-18 12:27:48 +03:00
Jussi Saurio
086a269a65 Merge 'sim: change --disable-create-index flag to --enable-create-index (default false)' from Jussi Saurio
Closes #2160
2025-07-18 12:27:35 +03:00
Jussi Saurio
687d126e25 sim: change --disable-create-index flag to --enable-create-index (default false) 2025-07-18 11:21:24 +03:00
Jussi Saurio
5b2d7315eb sim: show seed in 'execute_interaction()' trace 2025-07-18 11:21:04 +03:00
Jussi Saurio
d67a9f03fd sim: add order by to some queries 2025-07-18 10:47:36 +03:00
Jussi Saurio
746995f436 Merge 'Property FaultyQuery should fail if we encounter an error that is not expected' from Pedro Muniz
If we use `Assumption` here, the simulator just goes to the next
property instead of halting here.

Reviewed-by: Jussi Saurio <jussi.saurio@gmail.com>

Closes #2147
2025-07-17 23:44:44 +03:00
Jussi Saurio
1b52b5c764 Merge 'chore: update rust to version 1.88.0' from Nils Koch
This PR updates to version Rust 1.88.0 ([Release
notes](https://releases.rs/docs/1.88.0/)) and fixes all the clippy
errors that come with the new Rust version.
This is possible in the latest Rust version:
```rust
if let Some(foo) = bar && foo.is_cool() {
  ...
}
```
There are three complications in the migration (so far):
- A BUNCH of Clippy warnings (mostly fixed in
https://github.com/tursodatabase/limbo/pull/1827)
- Windows cross compilation failed; linking `advapi32` on windows fixes
it
  - Since Rust 1.87.0, advapi32 is not linked by default anymore
([Release notes](https://github.com/rust-
lang/rust/blob/master/RELEASES.md#compatibility-notes-1),
[PR](https://github.com/rust-lang/rust/pull/138233))
- Rust is more strict with FFIs and aligning pointers now. CI checks
failed with error below
  - Fixed in https://github.com/tursodatabase/turso/pull/2064
```
thread 'main' panicked at
core/ext/vtab_xconnect.rs:64:25:
misaligned pointer dereference: address must be
a multiple of 0x8 but is 0x7ffd9d901554
```

Closes #1807
2025-07-17 23:35:33 +03:00
pedrocarlo
4aa2c1a579 Property should fail if we encounter an error that is not expected 2025-07-17 17:01:54 -03:00
pedrocarlo
b80218324d fix merge conflicts 2025-07-17 12:25:31 -03:00
pedrocarlo
6088aa34c2 only add the sync completion after we call sync in the closure 2025-07-17 12:25:31 -03:00
pedrocarlo
0ab2f2b951 cli options for max and min tick + adjust github action to run with faster clock so no timeouts happen 2025-07-17 12:25:31 -03:00
pedrocarlo
46a7d20c12 clippy 2025-07-17 12:24:43 -03:00
pedrocarlo
4a13286d62 modify clock to use simulated time instead 2025-07-17 12:24:43 -03:00
pedrocarlo
d72a7f5d8e decrease IO latency time as Turso becomes super slow for some reason 2025-07-17 12:24:43 -03:00
pedrocarlo
7b8eec90bd edit state machine in Btree for freeing pages + Pager state machine for free_page 2025-07-17 12:24:43 -03:00
pedrocarlo
5771d1a00e disable wal sync timeout on checkpoint 2025-07-17 12:24:43 -03:00
pedrocarlo
b292e08d2b inject latency with queuing system 2025-07-17 12:24:43 -03:00
pedrocarlo
dc5f73887e refactor to require Arc<Completion> in file traits so that we can delay IO calls correctly 2025-07-17 12:24:43 -03:00
pedrocarlo
4ca5595f07 enable io latency + limit number of tests in simulator pull request CI check 2025-07-17 12:22:27 -03:00
Pekka Enberg
e8ac707190 simulator: Disable INSERT INTO .. SELECT for being slow
Refs #2129
2025-07-17 09:20:00 +03:00