o qi @s2dZgdZgdZddZdedefddZd S) z'Rule based Sentence tokenization module)u جنہیںuجسuجنuجوuاورuاگرu اگرچہuلیکنuمگرuپرuیاuتاہمuکہuکرuتوگےگی)u کیجیےu کیجئےuگئیںuتھیںuہوںu خریداru ہونگےuگاu چاہیےu ہوئیںruتھاuتھیuتھےuہیںuہےcCs2|sgSttt|d}|||||S)z&Replace end of sentence with separator)chrordmaxreplacesplit)_str separatormax_pr M/home/ubuntu/.local/lib/python3.10/site-packages/urduhack/tokenization/eos.py_split_and_keep srtextreturnc Cs,g}t|d}|D] }|rt|dkrd|vrt|d}|D]w}|}d}d}t|D]O\}} |r:d}q1| tvrz|dt|krz||dtvrz|dt|krq||ddvrq|d| d||dd 7}d }q1|d| d 7}q1|d| 7}q1|d D]} | rt| dkr|| qq#q |}d}d}t|D]O\}} |rd}q| tvr|dt|kr||dtvr|dt|kr||ddvr|d| d||dd 7}d }q|d| d 7}q|d| 7}q|d D]} | rt| dkr|| qq |S) aGenerate a list of urdu sentences from a given string. This function automatically fixes multiple whitespaces or new lines so you just need to pass the data and get sentences in return. Args: text (str): base string Returns: list ۔u؟Fr)ru،  T)rlenr enumerate_URDU_NEWLINE_WORDS_URDU_CONJUNCTIONSappendstrip) r all_sentences sentencessentence q_sentences_sennew_sentis_contindexwordsenr r r _generate_sentencessd       r&N)__doc__rrrstrlistr&r r r r s