This implements a subset of SQLite TCL tests. Let's start fixing errors
and then import more tests!
The test cases are imported from SQLite 3.50.2.
Refs #62Closes#1980
The `.headers` command takes `on` and `off` as parameter, supported by
SQLite, which controls whether result set header is printed in list
mode.
Closes#1983
Previously all constraints were disabled with an `if false {}` hack.
In this PR:
- enable PRIMARY KEYs and UNIQUE constraints if `experimental_indexes`
compile-time feature is enabled.
- Also enable PRIMARY KEY for exactly INTEGER columns, as INTEGER
PRIMARY KEY does not require an index
- NOT NULL constraint is also now enabled in both cases, as it doesn't
require an index
Closes#1981
We're now mixing different error messages, which makes compatibility
testing pretty hard. Unify on a single, SQLite compatible error message
"no such table".
Closes#1978
We're now mixing different error messages, which makes compatibility
testing pretty hard. Unify on a single, SQLite compatible error message
"no such table".
Previously `ScalarFunc::Glob` only handled Text based values. After this
fix it will be able to handle all types of values.
Now:
```SQL
turso> CREATE TABLE IF NOT EXISTS t0 (c0 REAL );
turso> UPDATE t0 SET c0='C2IS*24', c0=0Xffffffffbfc4330f, c0=0.6463854797956918 WHERE ((((((((t0.c0)AND(t0.c0)))AND(0.23913649834358142)))OR(CASE t0.c0 WHEN t0.c0 THEN 'j2' WHEN t0.c0 THEN t0.c0 WHEN t0.c0 THEN t0.c0 END)))OR(((((((((t0.c0)AND(t0.c0)))AND(t0.c0)))OR(t0.c0)))AND(t0.c0))));
turso> INSERT INTO t0 VALUES (NULL);
INSERT INTO t0 VALUES ('0&');
UPDATE t0 SET c0=2352448 WHERE ((((t0.c0)GLOB(t0.c0))) NOT NULL);
turso> SELECT * from t0;
┌───────────┐
│ c0 │
├───────────┤
│ │
├───────────┤
│ 2352448.0 │
└───────────┘
```
Fixes: https://github.com/tursodatabase/turso/issues/1953Closes#1955
I modified the `better-sqlite3.spec.mjs` test suite to use the dual test
runner from https://github.com/tursodatabase/turso/pull/1941 and I
deleted the `limbo.spec.mjs` test suite, because it's now redundant.
I looked carefully at the diff between the two test suites to make sure
that we didn't lose any coverage with this change.
Closes#1959
In reponse to [discord discussion](https://discord.com/channels/12586588
26257961020/1385754749634084885/1390535068401012858), I thought it would
be better to add a test to show how JDBC4Statement can be used.
Closes#1962
