zhengruifeng commented on PR #56026: URL: https://github.com/apache/spark/pull/56026#issuecomment-4518499897
### `prompt_for_components` coverage on real-world titles Among the parseable titles from the previous sample (983 commits, 172 open PRs), how many would lack a primary tag and trigger the prompt at merge time: | Source | Parseable | Need prompt | Rate | |---|---|---|---| | Commits | 983 | 78 | ~8% | | Open PRs | 172 | 14 | ~8% | Breakdown of *why* the prompt would fire: | Category | Commits | PRs | |---|---|---| | No tags at all (just `[SPARK-NNNNN] ...`) | 26 | 8 | | Only `[PS]` (alone or with `[TEST]` / `[FOLLOWUP]`) | 38 | 3 | | Only `[TEST]` (alone or with `[FOLLOWUP]`) | 9 | 0 | | Only `[FOLLOWUP]` | 2 | 0 | | Only `[UDF]` | 0 | 2 | | Other (`[WEBIUI]`, `[SHS]`, `[EXAMPLE]`, `[REPL]`) | 3 | 1 | | **Total** | **78** | **14** | Observations: - `[PS]` (Pandas API on Spark) accounts for ~half of the prompts. Promoting it to primary would cut the prompt rate roughly in half. - `[WEBIUI]` looks like a typo of `[WEBUI]`, which is already an alias of the primary `UI` tag. - The "no tags at all" group (26 commits, 8 PRs) is the prompt working as intended — titles like `[SPARK-NNNNN] ...` with no component tag get caught. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
