Timeseries SQL Tests

View source code

These tests cover SQL-first timeseries behavior from the Python bindings.

Covered Behavior

1) insert, range query, and bucket aggregation

Creates a TIMESERIES TYPE, inserts records, validates BETWEEN queries (3 rows; timestamps compared up to a constant offset), and checks ts.timeBucket(...) aggregation results.

2) tag filtering and empty ranges

Verifies tag-based filtering and the no-row case for non-overlapping time windows.

3) COMPACT TIMESERIES TYPE seals the tail

Appends 5,000 samples with append_samples(), runs COMPACT TIMESERIES TYPE TempData, and asserts the result names the type, reports mutableSamples 0, and a mutableSamplesBefore between 0 and 5,000 (the background pass may have sealed part of it first), and that all 5,000 samples still count. The statement is new in 26.10.1 (ArcadeData/arcadedb#8574); 26.9.1 rejects it as a syntax error.

Runtime Guard

The wheel always ships time series, so a failing CREATE TIMESERIES TYPE fails these tests; none skips.