Fixing a bug with byte order marks
a day ago
- The author was converting SRT subtitle files to WebVTT format for use with HTML5 video on static websites.
- A bug occurred because byte order marks (BOM) in UTF-8 SRT files prevented proper removal of sequence numbers, leaving the BOM and first number in the output.
- The fix involved opening files with encoding='utf-8-sig' in Python to automatically strip the BOM.
- To repair already converted files, the author used ripgrep to find BOMs and a Python script to remove them from all .vtt files.
- The experience highlighted the importance of understanding low-level text encoding details when managing media files.