Commit Graph

5376 Commits

Author SHA1 Message Date
luizgfc
5da6ed7499 core/printf: Fix null %s formatting behavior for sqlite compatibility 2025-09-29 21:26:47 -03:00
Pekka Enberg
75f088740d Merge 'core: Disallow CREATE INDEX when MVCC is enabled' from Pekka Enberg
MVCC does currently not support indexes. Therefore,
- Fail if a database with indexes is opened with MVCC
- Disallow `CREATE INDEX` when MVCC is enabled
Fixes: #3108

Closes #3426
2025-09-29 20:53:16 +03:00
Pekka Enberg
05d8cca75c core: Disallow CREATE INDEX when MVCC is enabled
MVCC does currently not support indexes. Therefore,

- Fail if a database with indexes is opened with MVCC
- Disallow `CREATE INDEX` when MVCC is enabled

Fixes: #3108
2025-09-29 19:38:08 +03:00
Preston Thorpe
cdab174350 Merge 'Fix column fetch in joins' from Glauber Costa
In comparisons for joins, we were assuming that the left column belonged
to the left join (and vice-versa). That is incorrect, because you can
write the comparison condition in any order.
Fixes #3368

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

Closes #3400
2025-09-29 12:34:45 -04:00
Preston Thorpe
da599a1fb8 Merge 'quoting fix' from Nikita Sivukhin
This PR moves part of string normalization to the parser layer.
Now, we dequote and unescape values in the parser, but we still need to
lowercase them for proper ignore-case comparison logic in the planner.
The reason to not lowercase in the parser is following:
1. SQLite (and tursodb) have ident->string conversion rule and by
lowercasing value early we will loose original representation
2. Some things like column names are preserve the case right now and we
better to not change this behaviour.

Closes #3344
2025-09-29 12:33:42 -04:00
Pekka Enberg
1f86400cec Merge 'core/vdbe: Wrap Program::n_change with AtomicI64' from Pekka Enberg
Closes #3424
2025-09-29 18:11:18 +03:00
Preston Thorpe
bbbcbf894e Merge 'translate/upsert: fix explicit conflict target of non-rowid primary key in UPSERT' from Preston Thorpe
closes #3384

Closes #3386
2025-09-29 11:09:29 -04:00
Preston Thorpe
1e974c1383 Merge 'Fix zero limit' from Nikita Sivukhin
Before, we validated that condition during program emit - which works
for fixed values of parameters but doesn't work with variables provided
externally to the prepared statement

Reviewed-by: Preston Thorpe <preston@turso.tech>

Closes #3421
2025-09-29 11:06:46 -04:00
Preston Thorpe
8d424d0ab2 Merge 'use explicit null if it set instead of column default value' from Nikita Sivukhin
Before, tursodb always used default value even if NULL was explicitly
set by the user

Reviewed-by: Preston Thorpe <preston@turso.tech>

Closes #3422
2025-09-29 11:05:14 -04:00
Preston Thorpe
8665d76c2e Merge 'Improve encryption module' from Avinash Sajjanshetty
This patch improves the encryption module:
1. Previously, we did not use the first 100 bytes in encryption. This
patch uses that portion as associated data, for protection against
tampering and corruption
2. Once the page 1 encrypted, on disk we store a special Turso header
(the first 16 bytes). During decryption we replace this with standard
SQLite's header (`"SQLite format 3\000"`). So that the upper layers (B
Tree or in Sync APIs) operate on the existing SQLite page expectations.
The format is:
```
///                    Turso Header (16 bytes)
///        ┌─────────┬───────┬────────┬──────────────────┐
///        │         │       │        │                  │
///        │  Turso  │Version│ Cipher │     Unused       │
///        │  (5)    │ (1)   │  (1)   │    (9 bytes)     │
///        │         │       │        │                  │
///        └─────────┴───────┴────────┴──────────────────┘
///         0-4      5       6        7-15
///
///        Standard SQLite Header: "SQLite format 3\0" (16 bytes)
///                            ↓
///        Turso Encrypted Header: "Turso" + Version + Cipher ID + Unused
```

Reviewed-by: Nikita Sivukhin (@sivukhin)
Reviewed-by: bit-aloo (@Shourya742)

Closes #3358
2025-09-29 11:04:31 -04:00
Pekka Enberg
5f9287304b core/vdbe: Wrap Program::n_change with AtomicI64 2025-09-29 17:09:33 +03:00
Nikita Sivukhin
0910483522 fix clippy 2025-09-29 16:30:07 +04:00
Nikita Sivukhin
a142c59de4 use explicit null if it set instead of column default value 2025-09-29 16:28:09 +04:00
Nikita Sivukhin
12863b35c4 fix compound select 2025-09-29 16:21:35 +04:00
Nikita Sivukhin
49a5617a95 fix limit for compount select 2025-09-29 15:51:39 +04:00
Nikita Sivukhin
70e18ce3f7 validate zero limit at the beginning in the VDBE program
- before, we validated that condition during program emit - which works for fixed values of parameters but doesn't work with variables provided externally to the prepared statement
2025-09-29 15:32:36 +04:00
Pekka Enberg
f247b1a2bb core/storage: Wrap Pager::commit_info with RwLock
Also remove RefCells from CommitInfo because they're not only redundant,
but cause CommitInfo not to be Send.
2025-09-29 13:54:28 +03:00
Nikita Sivukhin
af9d5ab52e fix bug 2025-09-29 12:28:27 +04:00
Nikita Sivukhin
7a1da051d0 remove usage of expr.to_string() and fix normalize util test 2025-09-29 11:36:12 +04:00
Nikita Sivukhin
00114d9674 remove quotes handling from normalization util 2025-09-29 11:00:19 +04:00
Nikita Sivukhin
86a95e813d Merge branch 'main' into quoting-fix-attempt-2 2025-09-29 10:58:51 +04:00
Avinash Sajjanshetty
ec1bf8888c refactor to adress review comments 2025-09-28 22:03:47 +05:30
Pekka Enberg
d3abeb6281 core/storage: Wrap WalFile::{max,min}_frame with AtomicU64 2025-09-28 16:47:54 +03:00
Pekka Enberg
aba596441c core/storage: Wrap WalFile::max_frame_read_lock_index with AtomicUsize 2025-09-28 13:42:32 +03:00
Jussi Saurio
959165eed1 Merge 'core/storage: Mark Page as Send and Sync' from Pekka Enberg
Closes #3399
2025-09-28 08:08:46 +03:00
Jussi Saurio
618cf2a4e7 Merge 'core/translate: rewrite default column value from identifier to string literal' from Preston Thorpe
closes #3390
closes #3389

Closes #3403
2025-09-28 08:07:11 +03:00
PThorpe92
03046faccb Translate default value Expr::Id to Literal to allow for identifier in col def 2025-09-27 13:56:04 -04:00
PThorpe92
b7fc9fef40 Persist NOT NULL column constraint to schema table 2025-09-27 13:12:19 -04:00
Glauber Costa
78ee8b8627 Fix column fetch in joins
In comparisons for joins, we were assuming that the left column belonged
to the left join (and vice-versa). That is incorrect, because you can
write the comparison condition in any order.

Fixes #3368
2025-09-27 12:08:47 -03:00
Pekka Enberg
3d3e39a958 Merge 'Make Sorter Send and Sync' from Pekka Enberg
Closes #3398
2025-09-27 16:51:27 +03:00
Pekka Enberg
d7a0a3db56 Merge 'core/translate: allow creating column called 'rowid'' from Preston Thorpe
closes #3282
includes minor refactor, removing `column_is_rowid_alias`, which is only
checking the public field of the argument Column.

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

Closes #3385
2025-09-27 16:51:09 +03:00
Pekka Enberg
5ff0044961 Merge 'length shall not count when it sees nullc' from Pavan Nambi
fixes #3317

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

Closes #3356
2025-09-27 16:50:50 +03:00
Pekka Enberg
e34a935e0a Merge 'resolve column alias after rewritting column access in the expression in returning insert clause' from Nikita Sivukhin
Fixes https://github.com/tursodatabase/turso/issues/3295

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

Closes #3355
2025-09-27 16:50:42 +03:00
Avinash Sajjanshetty
c2453046fa clippy fixes 2025-09-27 18:16:51 +05:30
Avinash Sajjanshetty
a2df313ad5 Add documentation for the encryption module 2025-09-27 18:11:27 +05:30
Pekka Enberg
2a8a0729c7 Merge 'core/storage: Wrap WalFile::syncing with AtomicBool' from Pekka Enberg
Closes #3397
2025-09-27 15:18:32 +03:00
Pekka Enberg
ce76aa11b2 core/storage: Mark Page as Send and Sync 2025-09-27 15:16:38 +03:00
Avinash Sajjanshetty
dc3d1fa36d Use the SQLite header as associated data for protection
against tampering and corruption.

Previously, we did not use the first 100 bytes in encryption
machinery. This patch changes that and uses that data as
associated data. So in case the header is corrupted or
tampered with, the decryption will fail.
2025-09-27 17:34:51 +05:30
PThorpe92
16d1e7e6a9 Rename function and update comment to match behavior 2025-09-27 07:52:21 -04:00
PThorpe92
8517355c0c make clippy happy 2025-09-27 07:52:21 -04:00
PThorpe92
5ad3e5244b Fix explicit ON CONFLICT target of non-rowid alias primary keys in UPSERT 2025-09-27 07:52:20 -04:00
Pekka Enberg
7b6fc0f3b6 core/vdbe: Wrap SortedChunk::total_bytes_read with AtomicUsize 2025-09-27 14:35:31 +03:00
Pekka Enberg
61b3f56997 core/vdbe: Wrap SortedChunk::io_state with RwLock 2025-09-27 14:28:55 +03:00
Pekka Enberg
5f39987ec0 core/vdbe: Wrap SortedChunk::buffer_len with AtomicUsize 2025-09-27 14:23:02 +03:00
Pekka Enberg
b31818f77c core/vdbe: Wrap SortedChunk::buffer with RwLock 2025-09-27 14:23:02 +03:00
Pekka Enberg
8d9d2dad1d core/storage: Wrap WalFile::syncing with AtomicBool 2025-09-27 14:07:26 +03:00
Glauber Costa
3ee97ddf36 Make sure complex expressions in filters go through Project
We had code for this, but the code had a fatal flaw: it tried to detect
a complex operation (an operation that needs projection), and return
false (no need for projection), for the others.

This is the exact opposite of what we should do: we should identify the
*simple* operations, and then return true (needs projection) for the
rest.

CAST is a special beast, since it is not a function, but rather, a
special opcode. Everything else above is the true just the same. But for
CAST, we have to do the extra work to capture it in the logical plan and
pass it down.

Fixes #3372
Fixes #3370
Fixes #3369
2025-09-27 07:21:03 -03:00
Jussi Saurio
283fba2e0d use normalized table name 2025-09-27 09:53:11 +03:00
Jussi Saurio
67d320960d ALTER TABLE: prevent dropping/renaming column referenced in VIEW 2025-09-27 09:45:15 +03:00
Jussi Saurio
3137357092 ALTER TABLE: prevent dropping indexed column in VDBE layer 2025-09-27 09:45:15 +03:00