11 Commits
Author SHA1 Message Date
Classic298 2d18727ab8 perf: build info log messages lazily so raising the log level actually saves work (#27837)
Raising GLOBAL_LOG_LEVEL to WARNING buys quieter output but not less work: 241 INFO call sites interpolate their payload into an f-string before the logging call gets to drop it. The heaviest is get_doc, which logs every chunk id and metadata dict in a collection, so on the full-context retrieval path that is the entire knowledge base, once per chat request.

That one line at WARNING, CPython 3.12:

| knowledge base | payload | before   | after   |
| -------------- | ------- | -------- | ------- |
| top-k of 3     | 1.2 kB  | 3.8 us   | 0.07 us |
| 500 chunks     | 201 kB  | 583.6 us | 0.08 us |
| 5000 chunks    | 2.0 MB  | 5.8 ms   | 0.15 us |

The lazy form log.info('query_doc:result %s %s', result.ids, result.metadatas) hands the payload to record.getMessage(), which the InterceptHandler only reaches once a record has passed the level check. Output at INFO is byte-identical. Two sites that already built their message eagerly, one str concat and one % operator, move to the same lazy form.
2026-08-02 15:39:10 -05:00
Classic298 52cfb02c72 perf: build debug log messages lazily so disabled debug logs cost nothing (#27834)
GLOBAL_LOG_LEVEL defaults to INFO, so every log.debug(...) in the backend is discarded, but the message is built first: 187 call sites interpolate their payload into an f-string before the logging call runs, so the work happens on every request and the result is thrown away. The worst one sits in process_chat_payload and stringifies the whole request body, full conversation history included, once per chat completion.

That one line with DEBUG disabled, CPython 3.12:

| conversation | payload | before   | after   |
| ------------ | ------- | -------- | ------- |
| 4 messages   | 1.2 kB  | 3.4 us   | 0.07 us |
| 20 messages  | 17 kB   | 24.8 us  | 0.07 us |
| 60 messages  | 123 kB  | 216.6 us | 0.07 us |

The lazy form log.debug('form_data: %s', form_data) hands the payload to record.getMessage(), which the InterceptHandler only reaches once a record has passed the level check. With DEBUG enabled the emitted lines are byte-identical, f'{x=}' sites included: those map to %r. MistralLoader._debug_log callers get the same treatment, since that wrapper already forwards *args.
2026-07-31 19:09:01 -05:00
Timothy Jaeryang Baek 23d03d6aae refac 2026-06-28 23:10:14 -05:00
Timothy Jaeryang Baek 6d0295588e refac: modernize type annotations (PEP 604 / PEP 585) 2026-05-12 17:10:15 +09:00
Timothy Jaeryang Baek de3317e26b refac 2026-03-17 17:58:01 -05:00
Ethan T. a229f9ea42 fix: replace bare except with except Exception (#22473)
Replace bare except clauses with except Exception to follow Python best practices and avoid catching unexpected system exceptions like KeyboardInterrupt and SystemExit.
2026-03-15 17:48:23 -05:00
Timothy Jaeryang Baek fe653a1336 refac 2025-12-20 18:12:03 +04:00
Timothy Jaeryang Baek afaa404fe4 enh: mineru api timeout 2025-12-20 17:39:33 +04:00
Timothy Jaeryang Baek 4673e120c4 refac/fix: mineru params
breaking change
2025-11-11 00:30:11 -05:00
palazski 288b323df8 feat: use MINERU_PARAMS json field for mineru settings 2025-10-15 22:59:59 +03:00
palazski 40e9d9c330 feat: add mineru as document parser support with both local and managed api 2025-10-13 21:09:52 +03:00