adsist-cms/python-detector-worker
3
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions

Refactor streaming readers: Split into modular files and implement base class
Some checks failed
Build Worker Base and Application Images / check-base-changes (push) Successful in 7s
Details
Build Worker Base and Application Images / build-base (push) Successful in 10m17s
Details
Build Worker Base and Application Images / build-docker (push) Successful in 3m54s
Details
Build Worker Base and Application Images / deploy-stack (push) Has been cancelled
Details

Browse source 
- Removed the existing `readers.py` file and created separate modules for `FFmpegRTSPReader`, `HTTPSnapshotReader`, and utility functions.
- Introduced an abstract base class `VideoReader` to standardize the interface for video stream readers.
- Updated `FFmpegRTSPReader` and `HTTPSnapshotReader` to inherit from `VideoReader` and implement required methods.
- Enhanced logging utilities for better readability and maintainability.
- Removed `pycuda` from requirements as it is no longer needed.
This commit is contained in:
Siwat Sirichai 2025-09-26 20:03:09 +07:00
parent d8d1b33cd8
commit 2b382210eb
7 changed files with 672 additions and 558 deletions
Download patch file Download diff file Expand all files Collapse all files

1
requirements.base.txt
View file

@ -9,5 +9,4 @@ lap>=0.5.12
pynvml
PyTurboJPEG
PyNvVideoCodec
pycuda
cupy-cuda12x