- Python package build and example usage (Pekka Enberg)
- Minor improvements to JavaScript API (Pekka Enberg)
CAST
support (Jussi Saurio)
- Fix issues found in-btree code with the DST (Pere Diaz Bou)
- Fix database truncation caused by
limbo-wasm
opening file in wrong mode (Pere Diaz Bou)
-
CREATE TABLE
support (Pere Diaz Bou) -
Add Add Database.prepare() and Statement.all() to Wasm bindings (Pekka Enberg)
-
WAL improvements (Pere Diaz Bou)
-
Primary key index scans and single-column secondary index scans (Jussi Saurio)
-
GROUP BY
support (Jussi Saurio) -
Overflow page support (Pere Diaz Bou)
-
Improvements to Python bindings (Jean Arhancet and Lauri Virtanen)
-
Improve scalar function support (Lauri Virtanen)
-
Panic in codegen with
COUNT(*)
(Jussi Saurio) -
Fix
LIKE
to be case insensitive (RJ Barman)
-
Query planner rewrite (Jussi Saurio)
-
Initial pass on Python bindings (Jean Arhancet)
-
Improve scalar function support (Kim Seon Woo and Jean Arhancet)
- Partial support for
json()
function (Jean Arhancet)
-
Initial pass on the write path. Note that the write path is not transactional yet. (Pere Diaz Bou)
-
More scalar functions:
unicode()
(Ethan Niser) -
Optimize point queries with integer keys (Jussi Saurio)
-
ORDER BY
support for nullable sorting columns and qualified identifiers (Jussi Saurio) -
Fix
.schema
command crash in the CLI (#212 (Jussi Saurio)
-
Partial
LEFT JOIN
support. -
Partial
ORDER BY
support. -
Partial scalar function support.
- Lock database file with POSIX filesystem advisory lock when database is opened to prevent concurrent processes from corrupting a file. Please note that the locking scheme differs from SQLite, which uses POSIX advisory locks for every transaction. We're defaulting to locking on open because it's faster. (Issue #94)
- Install to
~/.limbo/
instead ofCARGO_HOME
.
-
Partial
SELECT
statement support, includingWHERE
,LIKE
,LIMIT
,CROSS JOIN
, andINNER JOIN
. -
Aggregate function support.
-
EXPLAIN
statement support. -
Partial
PRAGMA
statement support, includingcache_size
. -
Asynchronous I/O support with Linux io_uring using direct I/O and Darwin kqueue.
-
Initial pass on command line shell with following commands:
.schema
command that describes the database schema..opcodes <opcode>
that describes what a VDBE opcode does.