Add filter for files with bad extensions in language handler

This commit is contained in:
mrT23
2025-03-11 17:03:05 +02:00
parent 0f33750035
commit 9a574e0caa

View File

@ -45,6 +45,8 @@ def sort_files_by_main_languages(languages: Dict, files: list):
else:
main_extensions.append([])
# filter out files bad extensions
files_filtered = filter_bad_extensions(files)
# sort files by their extension, put the files that are in the main extension first
# and the rest files after, map languages_sorted to their respective files
@ -53,7 +55,7 @@ def sort_files_by_main_languages(languages: Dict, files: list):
# if no languages detected, put all files in the "Other" category
if not languages:
files_sorted = [({"language": "Other", "files": list(files)})]
files_sorted = [({"language": "Other", "files": list(files_filtered)})]
return files_sorted
main_extensions_flat = []
@ -62,7 +64,7 @@ def sort_files_by_main_languages(languages: Dict, files: list):
for extensions, lang in zip(main_extensions, languages_sorted_list): # noqa: B905
tmp = []
for file in files:
for file in files_filtered:
extension_str = f".{file.filename.split('.')[-1]}"
if extension_str in extensions:
tmp.append(file)