aljaz/turso
1
0
Fork 0
mirror of https://github.com/aljazceru/turso.git synced 2025-12-25 12:04:21 +01:00
Code Issues Packages Projects Releases Wiki Activity
9,079 Commits 1 Branch 0 Tags
6b9fb2f181e428bb521640d9459ee2adcfe1dc2e
Commit Graph

306 Commits

Author SHA1 Message Date
Pere Diaz Bou
0eb2aa123d cli/app: add mvcc mode 2025-09-18 18:21:04 +02:00
Pere Diaz Bou
de8a975a0b core/mvcc: introduce MvccMode Logical Log 2025-09-18 18:21:04 +02:00
rajajisai
e605aff31b Merge branch 'main' into enc-page-1 2025-09-16 10:06:00 -04:00
rajajisai
89caa868f9 Encryption support for database header page 2025-09-16 10:04:30 -04:00
Pekka Enberg
ae25a0f088 Merge 'Implement Min/Max aggregators' from Glauber Costa 
We have not implemented them before because they require the raw
elements to be kept. It is easy to see why in the following example:
```
current_min = 3;
insert(2) => current_min = 2 // can be done without state
delete(2) => needs to look at the state to determine new min!
```
The aggregator state was a very simple key-value structure. To
accomodate for min/max, we will make it into a more complex table, where
we can encode a more complex structure.
The key insight is that we can use a primary key composed of:
```
1) storage_id
2) zset_id,
3) element
```
The storage_id and zset_id are our previous key, except they are now
exploded to support a larger range of storage_id. With more bits
available in the storage_id, we can encode information about which
column we are storing. For aggregations in multiple columns, we will
need to keep a different list of values for min/max!
The element is just the values of the columns.
Because this is a primary key, the data will be sorted in the btree. We
can then just do a prefix search in the first two components of the key
and easily find the min/max when needed.
This new format is also adequate for joins. Joins will just have a new
storage_id which encodes two "columns" (left side, right side).

Closes #3143
 2025-09-16 16:19:59 +03:00
Glauber Costa
3e9a5d93b5 hide internal tables from .schema 2025-09-15 22:30:48 -05:00
pedrocarlo
3c91ae206b move as many dependencies as possible to workspace to avoid multiple versions of the same dependency 2025-09-15 17:19:36 -03:00
TcMits
9dac467b40 support EXPLAIN QUERY PLAN 2025-09-12 19:58:45 +07:00
TcMits
830e10da8f resolve merge conflict 2025-09-11 18:13:29 +07:00
TcMits
b574b4bcea finish EXPLAIN 2025-09-11 18:04:59 +07:00
Jussi Saurio
7ac18a6952 Merge 'Remove some traces in super hot paths in btree' from Preston Thorpe 
Particularly we were tracing `ImmutableRecord` / `BTreeKey` which would
then trace the bytes of records. These are super super hot paths and I
think we can probably remove even more to under debug assertions so we
dont eat those atomics/branches all the time.
This PR also introduces the `tracing_release` feature, which turns all
`trace!` and `debug!` macro invocations to noops at compile time, and
makes that feature available for all bindings.
it also removes the unused `lru` dependency, and cleans up the makefile
a bit

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

Closes #2995
 2025-09-11 13:33:25 +03:00
TcMits
5caf9a2640 make it more safe + clippy 2025-09-11 00:14:38 +07:00
PThorpe92
ba1ed72ed8 Add tracing_release feature for benchmarks to compile tracing macros to noops 2025-09-10 09:56:12 -04:00
TcMits
eeef8b85fa always use consume instead of run_query, handle_dot_command 2025-09-10 16:54:51 +07:00
TcMits
688dc6dde3 minor 2025-09-10 16:31:57 +07:00
TcMits
65f5fbd1f6 no errors in consume 2025-09-10 16:31:12 +07:00
TcMits
ddf0df0aea merge main 2025-09-10 15:59:56 +07:00
TcMits
dbcd01bf8b make consume safer 2025-09-10 15:56:20 +07:00
Jussi Saurio
b6d99cd188 Merge 'clean print_query_result' from Lâm Hoàng Phúc 
different between output modes is how we handle `Ok(StepResult::Row)` so
let introduce `row_step_result_query` macro to reduce duplicated code.

Closes #2979
 2025-09-10 11:08:40 +03:00
TcMits
e8b853ed25 pretty mode's table need a line break 2025-09-10 12:03:59 +07:00
TcMits
28f23973a4 fix logic 2025-09-09 18:38:33 +07:00
TcMits
463fd75f0b no need QueryStatistics 2025-09-09 18:35:27 +07:00
TcMits
8a72b24c0e add writeln_fmt 2025-09-09 17:57:58 +07:00
TcMits
036235a518 clean 'print_query_result' 2025-09-09 17:41:08 +07:00
TcMits
048e72abf5 consume remaining 2025-09-09 16:27:31 +07:00
TcMits
8a6667a829 refactor cli: will write to 2025-09-09 16:23:08 +07:00
Pekka Enberg
80a4358272 cli: Fix dump compatibility in "PRAGMA foreign_keys" 
SQLite emits a semicolon after "PRAGMA foreign_keys=OFF" so let's do th
same.
 2025-09-08 12:55:29 +03:00
TcMits
f518291522 infinite loop when query starts with '--' 2025-09-05 16:36:14 +07:00
Pekka Enberg
12cf4d2e72 core: Make strict schema support experimental 
It's not tested properly so let's mark it as experimental for now.

Fixes #2775
 2025-09-02 16:40:02 +03:00
Arkoniak
2c0e3cf593 feat: consistent interrupt event processing (#2801) 2025-08-30 06:57:14 +03:00
Arkoniak
cb602d960d feat: records output (#2801) 2025-08-30 06:57:14 +03:00
PThorpe92
959bc6ba74 Remove unused argument from handle_row fn 2025-08-29 10:18:49 -04:00
PThorpe92
5b73fad5d4 Remove some code duplication in the CLI 2025-08-29 10:15:07 -04:00
Alex Miller
327936fb51 Highlight the EXPLAIN keyword 2025-08-22 22:49:58 -07:00
Alex Miller
f1250a6c86 Add syntax highlighting for ANALYZE 2025-08-22 22:47:46 -07:00
Preston Thorpe
82fe508609 Merge 'add metrics and implement the .stats command' from Glauber Costa 
This adds basic statement and connection metrics like SQLite (and
libSQL) have.
This is particularly useful to show that materialized views are working:
turso> create table t(a);
turso> insert into t(a) values (1) , (2), (3), (4), (5), (6), (7), (8),
(9), (10); turso> create materialized view v as select count(*) from t;
turso> .stats on
Stats display enabled.
turso> select count(*) from t;
┌───────────┐
│ count (*) │
├───────────┤
│        10 │
└───────────┘
Statement Metrics:
  Row Operations:
    Rows read:        10
    Rows written:     0
    [ ... other metrics ... ]
turso> select * from v;
┌───────────┐
│ count (*) │
├───────────┤
│        10 │
└───────────┘
Statement Metrics:
  Row Operations:
    Rows read:        1
    Rows written:     0
    [ ... other metrics ... ]

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

Closes #2651
 2025-08-18 20:26:48 -04:00
Glauber Costa
36fc8e8fdb add metrics and implement the .stats command 
This adds basic statement and connection metrics like SQLite (and
libSQL) have.

This is particularly useful to show that materialized views are working:

turso> create table t(a);
turso> insert into t(a) values (1) , (2), (3), (4), (5), (6), (7), (8), (9), (10);
turso> create materialized view v as select count(*) from t;
turso> .stats on
Stats display enabled.
turso> select count(*) from t;
┌───────────┐
│ count (*) │
├───────────┤
│        10 │
└───────────┘

Statement Metrics:
  Row Operations:
    Rows read:        10
    Rows written:     0
    [ ... other metrics ... ]

turso> select * from v;
┌───────────┐
│ count (*) │
├───────────┤
│        10 │
└───────────┘

Statement Metrics:
  Row Operations:
    Rows read:        1
    Rows written:     0
    [ ... other metrics ... ]
 2025-08-18 09:11:06 -05:00
Glauber Costa
40efaab382 make sure our responses are compliant with MCP 
There is an inconsistency now on how we are handling IDs and
notifications. Claude Code doesn't complain (good boi), but Claude
Desktop does.
 2025-08-18 08:02:30 -05:00
Glauber Costa
841296a39d add new commands to the MCP server 
When playing with our MCP server with Claude Code, I realized that
having the agent pre-configured to a specific path is just not very
useful.

I am now adding two new commands, to open a database and to list
the current database. Those make our server more useful.
 2025-08-16 11:00:28 -05:00
Nikita Sivukhin
6280d33b02 init tracing in CLI early 
- now we miss logs from DB initialization phase
 2025-08-15 13:11:38 +04:00
PThorpe92
9ccf79111a Support non-utf8 blobs in .clone command 2025-08-14 21:31:14 -04:00
PThorpe92
2b289157d0 Properly quote sequence value 2025-08-14 21:31:14 -04:00
PThorpe92
7abf071128 Conditionally emit pragma foreign_keys=off when not .cloning 2025-08-14 21:31:14 -04:00
PThorpe92
4a612c1586 Properly implement .clone method safely 2025-08-14 21:31:13 -04:00
Jussi Saurio
69d8a73028 Merge 'use virtual root page for sqlite_schema' from Mikaël Francoeur 
This PR fixes a problem where `sqlite_schema` could be read before page
1 was allocated.
The fix is similar to that in SQLite. In SQLite, if `btreeCursor()` sees
that the root page is 1 and that the b-tree is empty, it sets the page
to 0 ([here](https://github.com/sqlite/sqlite/blob/master/src/btree.c#L4
691-L4696)). SQLite's `moveToRoot()` then uses this special value to
return `CURSOR_INVALID` with no rows ([here](https://github.com/sqlite/s
qlite/blob/master/src/btree.c#L5538-L5540)).
Fixes https://github.com/tursodatabase/turso/issues/2449

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

Closes #2551
 2025-08-14 11:08:11 +03:00
Mikaël Francoeur
07ef47924c use virtual root page for sqlite_schema 2025-08-13 16:31:21 -04:00
PThorpe92
614a0a45a6 Relax and fix memory ordering 2025-08-13 10:09:37 -04:00
Pekka Enberg
b82d850f7a Merge 'Reset interrupt count each time a new line is entered in CLI' from 
Resolves issue #1833.
**Before:**
<img width="581" height="92" alt="Screenshot 2025-08-11 at 9 07 21 PM"
src="https://github.com/user-
attachments/assets/eff4c919-ad54-4379-bbee-28eb3e4a375f" />
**After:**
<img width="563" height="118" alt="Screenshot 2025-08-11 at 9 07 27 PM"
src="https://github.com/user-
attachments/assets/f99f84a0-f5d3-4d70-9e08-1c3871fddda3" />

Closes #2554
 2025-08-12 11:50:03 +03:00
rajajisai
5cc5b0bca1 Reset interrupt count each time a new line is entered in CLI 2025-08-11 21:16:43 -07:00
Pekka Enberg
e829201fdd cli: Fix .schema SQL statement some more 
We need to fetch name and type too, but also sort in both places
consistently.
 2025-08-11 11:52:57 +03:00