Context
From PR #7672 review feedback by @wbreza:
The functional test recordings (YAML files in test/functional/testdata/recordings/) can be very large. For example, the AI quota preflight tests produce recordings with ~607KB per request for the CognitiveServices model catalog response, which includes costs, SKU variants, capabilities, deprecation timelines, and rate limits — none of which the test logic actually inspects.
The 3 RG-scoped recordings in #7672 are nearly identical (~1.56MB each / 5,206 lines) because they hit the same API responses with only minor parameter differences.
Proposal
Investigate reducing recording size across all functional tests:
- Strip responses to only include fields the test logic reads
- Share common base recordings and parameterize differences
- Compress or deduplicate large API responses in the recorder
This would reduce PR footprint and CI artifact sizes for all functional tests, not just the quota check tests.
Related
Context
From PR #7672 review feedback by @wbreza:
The functional test recordings (YAML files in
test/functional/testdata/recordings/) can be very large. For example, the AI quota preflight tests produce recordings with ~607KB per request for the CognitiveServices model catalog response, which includes costs, SKU variants, capabilities, deprecation timelines, and rate limits — none of which the test logic actually inspects.The 3 RG-scoped recordings in #7672 are nearly identical (~1.56MB each / 5,206 lines) because they hit the same API responses with only minor parameter differences.
Proposal
Investigate reducing recording size across all functional tests:
This would reduce PR footprint and CI artifact sizes for all functional tests, not just the quota check tests.
Related