Files
Patrick Buckley 1f3b89610a test(832): replay-parity harness + pre-fold baselines
Thirteen scenario scripts drawn from the chunk-field-to-UI grid, each driven
through the streaming seam against a scripted provider fake that arms
cancel_ref eagerly (the classifier the fold introduces distinguishes
creation-vs-midstream failures by that arming, so the fake must mirror the
real adapters' eager append). The captured records — ordered UI events,
committed-message projection, mid-stream usage, raised class — are the
OLD-WORLD baselines: this commit's session.py is byte-identical to main,
which is what makes them the record. The assert path applies only the
behavior deltas the design table rules, each transform citing its row; a
difference outside a ruled transform is a fold regression.
2026-08-06 01:04:32 -07:00

41 lines
676 B
JSON

{
"cancelled_partial": null,
"last_usage": {
"cache_creation_tokens": 0,
"cache_read_tokens": 0,
"completion_tokens": 7,
"prompt_tokens": 11,
"total_tokens": 18
},
"raised": null,
"result": {
"content": "Calling.",
"provider_content": null,
"tool_calls": [
{
"function": {
"arguments": "{\"city\": \"Paris\"}",
"name": "get_weather"
},
"id": "call_1",
"type": "function"
}
]
},
"scenario": "tools_simple",
"ui_events": [
[
"thinking_stop",
""
],
[
"content",
"Calling."
],
[
"stream_end",
""
]
]
}