Test against multiple SQLite versions

[asg017]

---

📚 Documentation preview 📚: https://datasette--2352.org.readthedocs.build/en/2352/

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.63%. Comparing base (64a125b) to head (eb6b27d).

❗ Current head eb6b27d differs from pull request most recent head 653c46d

Please upload reports for the commit 653c46d to get more accurate results.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2352      +/-   ##
==========================================
- Coverage   92.63%   92.63%   -0.01%     
==========================================
  Files          41       41              
  Lines        6399     6395       -4     
==========================================
- Hits         5928     5924       -4     
  Misses        471      471              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[simonw]

Looks like that test failure is caused by:

• Canned queries with named parameters fail with error against SQLite 3.46.0 #2353

    async def test_derive_named_parameters(sql, expected):
        ds = Datasette([], memory=True)
        db = ds.get_database("_memory")
        params = await utils.derive_named_parameters(db, sql)
>       assert params == expected
E       AssertionError: assert ['00', 'cat'] == ['cat']
E         
E         At index 0 diff: '00' != 'cat'
E         Left contains one more item: 'cat'
E         
E         Full diff:
E           [
E         +     '00',
E               'cat',
E           ]

[simonw]

That test is still failing because the whole point of that test is to confirm that the clever trick I came up with (and that gets skipped on SQLite 3.46.0) correctly avoids the "0:00" trick string.

datasette/tests/test_utils.py

Lines 604 to 619 in b39b01a

	@pytest.mark.asyncio
	@pytest.mark.parametrize(
	"sql,expected",
	(
	("select 1", []),
	("select 1 + :one", ["one"]),
	("select 1 + :one + :two", ["one", "two"]),
	("select 'bob' || '0:00' || :cat", ["cat"]),
	("select this is invalid :one, :two, :three", ["one", "two", "three"]),
	),
	)
	async def test_derive_named_parameters(sql, expected):
	ds = Datasette([], memory=True)
	db = ds.get_database("_memory")
	params = await utils.derive_named_parameters(db, sql)
	assert params == expected

[simonw]

Let's drop macOS for the moment, and just go with Linux.

[simonw]

Got this intermittent failure: https://github.com/simonw/datasette/actions/runs/9494472159/job/26164885108

Really frustrating, I don't think it's anything to do with SQLite versions, I think it's a flaky test.

    def test_max_csv_mb(app_client_csv_max_mb_one):
        # This query deliberately generates a really long string
        # should be 100*100*100*2 = roughly 2MB
        response = app_client_csv_max_mb_one.get(
            "/fixtures.csv?"
            + urllib.parse.urlencode(
                {
                    "sql": """
                select group_concat('ab', '')
                from json_each(json_array({lots})),
                    json_each(json_array({lots})),
                    json_each(json_array({lots}))
                """.format(
                        lots=", ".join(str(i) for i in range(100))
                    ),
                    "_stream": 1,
                    "_size": "max",
                }
            ),
        )
        # It's a 200 because we started streaming before we knew the error
>       assert response.status == 200
E       assert 400 == 200
E        +  where 400 = <datasette.utils.testing.TestResponse object at 0x7fd35cdc04d0>.status

[simonw]

Let's comment out that test so it doesn't block us on merging this PR, then fix that test in this issue:

• Flaky test_max_csv_mb test #2355

[simonw]

We could drop this, it's covered elsewhere

[simonw]

We should change this to Test SQLite versions otherwise the GitHub Actions UI has a bunch of things all called Test:

[simonw]

Flaky test failed again in a different way: https://github.com/asg017/datasette/actions/runs/9503497141/job/26194049093

    def test_max_csv_mb(app_client_csv_max_mb_one):
        # This query deliberately generates a really long string
        # should be 100*100*100*2 = roughly 2MB
        response = app_client_csv_max_mb_one.get(
            "/fixtures.csv?"
            + urllib.parse.urlencode(
                {
                    "sql": """
                select group_concat('ab', '')
                from json_each(json_array({lots})),
                    json_each(json_array({lots})),
                    json_each(json_array({lots}))
                """.format(
                        lots=", ".join(str(i) for i in range(100))
                    ),
                    "_stream": 1,
                    "_size": "max",
                }
            ),
        )
        # It's a 200 because we started streaming before we knew the error
>       assert response.status == 200
E       assert 500 == 200
E        +  where 500 = <datasette.utils.testing.TestResponse object at 0x7fe7fb6d0790>.status

/home/runner/work/datasette/datasette/tests/test_csv.py:213: AssertionError
----------------------------- Captured stderr call -----------------------------
Traceback (most recent call last):
  File "/home/runner/work/datasette/datasette/datasette/database.py", line 306, in sql_operation_in_thread
    cursor.execute(sql, params if params is not None else {})
sqlite3.OperationalError: interrupted

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/runner/work/datasette/datasette/datasette/app.py", line 1732, in route_path
    response = await view(request, send)
  File "/home/runner/work/datasette/datasette/datasette/app.py", line 1899, in async_view_for_class
    return await async_call_with_supported_arguments(
  File "/home/runner/work/datasette/datasette/datasette/utils/__init__.py", line 1022, in async_call_with_supported_arguments
    return await fn(*call_with)
  File "/home/runner/work/datasette/datasette/datasette/views/base.py", line 89, in __call__
    return await handler(request, datasette)
  File "/home/runner/work/datasette/datasette/datasette/views/database.py", line 61, in get
    return await QueryView()(request, datasette)
  File "/home/runner/work/datasette/datasette/datasette/views/base.py", line 89, in __call__
    return await handler(request, datasette)
  File "/home/runner/work/datasette/datasette/datasette/views/database.py", line 564, in get
    return await stream_csv(datasette, fetch_data_for_csv, request, db.name)
  File "/home/runner/work/datasette/datasette/datasette/views/base.py", line 441, in stream_csv
    response_or_template_contexts = await fetch_data(request)
  File "/home/runner/work/datasette/datasette/datasette/views/database.py", line 560, in fetch_data_for_csv
    results = await db.execute(sql, params, truncate=True)
  File "/home/runner/work/datasette/datasette/datasette/database.py", line 336, in execute
    results = await self.execute_fn(sql_operation_in_thread)
  File "/home/runner/work/datasette/datasette/datasette/database.py", line 282, in execute_fn
    return await asyncio.get_event_loop().run_in_executor(
  File "/opt/hostedtoolcache/Python/3.8.18/x64/lib/python3.8/concurrent/futures/thread.py", line 57, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/home/runner/work/datasette/datasette/datasette/database.py", line 280, in in_thread
    return fn(conn)
  File "/home/runner/work/datasette/datasette/datasette/database.py", line 319, in sql_operation_in_thread
    raise QueryInterrupted(e, sql, params)
datasette.database.QueryInterrupted: (OperationalError('interrupted'), "\n            select group_concat('ab', '')\n            from json_each(json_array(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99)),\n                json_each(json_array(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99)),\n                json_each(json_array(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99))\n            ", {'_stream': '1', '_size': 'max'})

[simonw]

Split out another flaky test issue:

• Very weird flaky test_create_table_ignore_replace and test_upsert tests #2356