explicit allowance for ć, ś, ń, ł etc

This commit is contained in:
erew123 authored and GitHub committed 2024-02-07 12:44:42 +00:00
1 parent c0d01ee931
commit 00b93cfe1b
1 file changed
+2 -10
+2 -10
View File
@@ -663,11 +663,7 @@ def output_modifier(string, state):
# Replace multiple exclamation marks, question marks, or other punctuation with a single instance
cleaned_part = re.sub(r"([!?.])\1+", r"\1", part)
# Further clean to remove any other unwanted characters
cleaned_part = re.sub(
r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\$]',
"",
cleaned_part,
)
cleaned_part = re.sub(r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0100-\u017F\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\$]', '', cleaned_part)
# Remove all newline characters (single or multiple)
cleaned_part = re.sub(r"\n+", " ", cleaned_part)
@@ -694,11 +690,7 @@ def output_modifier(string, state):
# Replace multiple instances of certain punctuation marks with a single instance
cleaned_part = re.sub(r"([!?.])\1+", r"\1", cleaned_part)
# Further clean to remove any other unwanted characters
cleaned_part = re.sub(
r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\$]',
"",
cleaned_part,
)
cleaned_part = re.sub(r'[^a-zA-Z0-9\s\.,;:!?\-\'"\u0100-\u017F\u0400-\u04FFÀ-ÿ\u0150\u0151\u0170\u0171\$]', '', cleaned_part)
# Remove all newline characters (single or multiple)
cleaned_part = re.sub(r"\n+", " ", cleaned_part)
# Process the part and give it a non-character name if being used vai API or standalone.