service.compare¶

The comparison of runs: provenance and the summaries of their tasks side by side.

CompareTable ¶

Bases: BaseModel

A table of runs.

Attributes:

Name	Type	Description
`columns`	`list[str]`	The columns in order: `run`, `group`, `agent`, `state`, then per source `<source>.changed_files` and `<source>.diff_sha256`, then every key of `out/summary.json` of every task as `<task>.<key>`.
`rows`	`list[dict[str, Any]]`	One row per run, with every column (None where a run has no value).

compare_runs(store, run_ids)

Compare runs by their provenance and the summaries their merges wrote.

Parameters:

Name	Type	Description	Default
`store`	`RunStore`	The run store.	required
`run_ids`	`list[str]`	The runs, one row each in this order.	required

Returns:

Type	Description
`CompareTable`	The table.

Raises:

Type	Description
`KeyError`	A run does not exist.