o vZh@s<ddlmZmZddZddZddZdd Zd d Zd S) ) RegexBuildersymbolscCttjdddjS)zKeep tone-modifying punctuation by matching following character. Assumes the `tone_marks` pre-processor was run for cases where there might not be any space after a tone-modifying punctuation mark. cS d|S)Nz(?<={}).formatxr M/var/www/auris/lib/python3.10/site-packages/gtts/tokenizer/tokenizer_cases.py  ztone_marks..Z pattern_argsZ pattern_func)rr TONE_MARKSregexr r r r tone_markss  rcCr)aJPeriod and comma case. Match if not preceded by "." and only if followed by space. Won't cut in the middle/after dotted abbreviations; won't cut numbers. Note: Won't match if a dotted abbreviation ends a sentence. Note: Won't match the end of a sentence if not followed by a space. cSr)Nz(?.r)rr PERIOD_COMMArr r r r period_commas  rcCr)zColon case. Match a colon ":" only if not preceded by a digit. Mainly to prevent a cut in the middle of time notations e.g. 10:01 cSr)Nz (?.r)rrCOLONrr r r r colon#s  rcCs@dttjttjttjttj}t|dddjS)zMatch other punctuation. Match other punctuation to split on; punctuation that naturally inserts a break in speech. cSrNz{}rrr r r r <r z#other_punctuation..r) joinsetrALL_PUNCrrrrrZpuncr r r other_punctuation/srcCstj}t|dddjS)z[Match all punctuation. Use as only tokenizer case to mimic gTTS 1.x tokenization. cSrrrrr r r r Er z(legacy_all_punctuation..r)rrrrrr r r legacy_all_punctuation?srN)Zgtts.tokenizerrrrrrrrr r r r s