Dymas/whisper-api-server
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Simplify architecture: remove overengineering, flatten structure

Browse Source 
- Remove dead code: cache.py, context_managers.py, decorators.py, flask-limiter
- Replace AudioSource ABC + FakeFile (5 classes) with 4 plain functions
- Replace TempFileManager class with create_temp_file/cleanup_temp_files functions
- Simplify RequestLogger from 233 to 65 lines, remove file reading side-effect
- Convert HistoryLogger and AudioUtils classes to module-level functions
- Remove unused speed_up_audio and audio_speed_factor config
- Flatten single-file directories: shared/, api/, storage/, validation/, async_tasks/, logging/
- Merge logging config + request_logger into single infrastructure/log.py
- Fix request_logging config key (was request_logger)
- Trim CLAUDE.md to high-level only

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Serge Zaigraeff
2026-03-22 05:28:34 +03:00
co-authored by Claude Opus 4.6
parent 38785398b3
commit 2e2bad8255
30 changed files with 556 additions and 1407 deletions
Download Patch File Download Diff File Expand all files Collapse all files
RULES.md
+5 -5
View File
@@ -46,13 +46,13 @@ Non-negotiable. Violation = stop and fix before continuing.
## Audio Processing Rules
* FFmpeg and SoX are external dependencies. Always check subprocess return codes.
* Temp files must be created via `TempFileManager.temp_file()` context manager -- never raw `tempfile.mktemp`.
* Audio pipeline order matters: convert to WAV 16kHz -> normalize -> compress/expand -> speedup -> add silence.
* Temp files must be created via `create_temp_file()` from `app/infrastructure/storage.py` -- never raw `tempfile.mktemp`.
* Audio pipeline order matters: convert to WAV 16kHz -> normalize -> compress/expand -> add silence.
## Flask / API Rules
* New endpoints go in `app/api/routes.py` inside `_register_routes()`.
* All transcription endpoints delegate to `TranscriptionService.transcribe_from_source()`.
* New audio input methods: subclass `AudioSource`, implement `get_audio_file()`.
* New endpoints go in `app/routes.py` inside `_register_routes()`.
* All transcription endpoints delegate to `TranscriptionService.transcribe()`.
* New audio input methods: add a `get_*_file()` function in `sources.py` returning `(temp_path, filename, error)`.
* File validation runs through `FileValidator` -- never validate inline in routes.
* Configuration values accessed via `self.config.get()` with sensible defaults, except critical params which must crash if missing.