o
    Zh                     @   s*   d dl mZ d dlmZ G dd dZdS )    )Rule)AbbreviationReplacerc                   @   s   e Zd Zg dZeddZeddZeddZeddZed	d
Z	eddZ
G dd deZG dd deZG dd deZG dd deZG dd deZG dd deZG dd deZdS )Standard)   。   ．.   ！!?   ？u   (?<=[a-zA-z]°)\.(?=\s*\d+)   ∯z(?<=\s)\.(?=(jpe?g|png|gif|tiff?|pdf|ps|docx?|xlsx?|svg|bmp|tga|exif|odt|html?|txt|rtf|bat|sxw|xml|zip|exe|msi|blend|wmv|mp[34]|pptx?|flac|rb|cpp|cs|js)\s)z\n   ȹz\?(?=(\'|\"))   &ᓷ&z\s{3,} u   &⎋&'c                   @   s2   e Zd ZdZg dZg dZg dZeddZdS )zStandard.Abbreviationz:Defines the abbreviations for each language (if available))ZadjadmZadvalZalaZaltaaprarcZarizZarkartZassnZasstattysaugZaveZbartZbldZbldgZblvdbrigZbrosZbtwcalZcalifcaptclcmdrcocolZcoloZcomdrconconnZcorpcplZcresctzd.philZdakdecdelZdeptdetdistdrzdr.philz	dr.philosZdrsze.gZensZespZesqetcexpZexpyextfebZfedZflaftZfwyfygagengovZhonZhosphrZhwayZhwyzi.eiaidZidaZillincindingZinspisjanZjrjuljunZkanZkansZkenkyZlaltZltdmajmanmarZmassmaymdmeZmedmessrsZmexZmfgZmichminZminnZmissZmllemmZmmemoZmontmrmrsmsZmsgrmssrsmtZmtnZnebZnebrZnevnonosnovnroctokZoklaZontopordZoreppapdZpdeZpennZpennaZpfcphzph.dplZplzppprofZpvtZquerdrsrefreprepsresrevrtZsasksecsensenssepZseptZsfcsgtsrstsuptZsurgZtceZtenntexunivZusafazu.sutvavverZvizvsvtZwashZwisZwiscZwyZwyoZyukfig)!r   r   r   r   r   r   r!   r%   r'   r/   r0   r6   r<   r=   rG   rH   rI   rK   rC   rJ   rZ   rW   r^   r_   ra   rd   re   rg   ri   rj   ro   rq   rs   )r   r*   rL   rM   rT   rY   z ([a-zA-Z0-9_])(\.)([a-zA-Z0-9_])u   \1∮\3N)	__name__
__module____qualname____doc__ZABBREVIATIONSZPREPOSITIVE_ABBREVIATIONSZNUMBER_ABBREVIATIONSr   ZWithMultiplePeriodsAndEmailRule rx   rx   Q/var/www/html/lang_env/lib/python3.10/site-packages/pysbd/lang/common/standard.pyAbbreviation   s    rz   c                   @   sD   e Zd ZeddZeddZeddZeddZd	ZeeeegZ	d
S )zStandard.DoublePunctuationRulesz\?!   ☉z!\?   ☈z\?\?   ☇!!   ☄z\?!|!\?|\?\?|!!N)
rt   ru   rv   r   Z	FirstRuleZ
SecondRuleZ	ThirdRuleZ	ForthRuleZDoublePunctuationAllrx   rx   rx   ry   DoublePunctuationRules#   s    



r   c                   @   s4   e Zd ZeddZeddZeddZeeegZdS )zStandard.ExclamationPointRulesz\!(?=(\'|\"))   &ᓴ&z\!(?=\,\s[a-z])z\!(?=\s[a-z])N)rt   ru   rv   r   ZInQuotationRuleZBeforeCommaMidSentenceRuleZMidSentenceRuler   rx   rx   rx   ry   ExclamationPointRules+   s
    


r   c                   @   s   e Zd ZeddZeddZeddZeddZed	d
ZeddZ	eddZ
eddZeddZeddZeddZeddZeddZeddZeddZedd Zed!d"Zeeeeee	e
eeeeeeeeeegZd#S )$zStandard.SubSymbolsRulesr   r   u   ♬u   ،u   ♭:u   &ᓰ&r   u   &ᓱ&r   u   &ᓳ&r   r   r	   r   r
   u   &ᓸ&r   r{   z?!r}   z??r|   z!?r   r~   u   &✂&(u   &⌬&)u   ȸ r   
N)rt   ru   rv   r   ZPeriodZArabicCommaZ	SemiColonZFullWidthPeriodZSpecialPeriodZFullWidthExclamationZExclamationPointZQuestionMarkZFullWidthQuestionMarkZMixedDoubleQEZMixedDoubleQQZMixedDoubleEQZMixedDoubleEEZ
LeftParensZRightParensZTemporaryEndingPunctutationZNewliner   rx   rx   rx   ry   SubSymbolsRules7   s0    

















r   c                   @   sL   e Zd ZeddZeddZeddZeddZed	dZeeeeegZ	d
S )zStandard.EllipsisRulesz\.\.\.(?=\s+[A-Z])u   ☏☏.z(?<=\S)\.{3}(?=\.\s[A-Z])   ƪƪƪz(\s\.){3}\s   ♟♟♟♟♟♟♟z(?<=[a-z])(\.\s){3}\.($|\\n)   ♝♝♝♝♝♝♝z\.\.\.N)
rt   ru   rv   r   ZThreeConsecutiveRuleZFourConsecutiveRuleZThreeSpaceRuleZFourSpaceRuleZOtherThreePeriodRuler   rx   rx   rx   ry   EllipsisRulesO   s    




r   c                   @   sL   e Zd ZeddZeddZeddZeddZed	d
ZeeeeegZ	dS )zStandard.ReinsertEllipsisRulesr   z...r   z . . . r   z. . . .u   ☏☏z..u   ∮r   N)
rt   ru   rv   r   ZSubThreeConsecutivePeriodZSubThreeSpacePeriodZSubFourSpacePeriodZSubTwoConsecutivePeriodZSubOnePeriodr   rx   rx   rx   ry   ReinsertEllipsisRulesd   s    




r   c                   @   s   e Zd ZddZdS )zStandard.AbbreviationReplacerzgA Being Did For He How However I In It Millions More She That The There They We What When Where Who Whyr   N)rt   ru   rv   splitZSENTENCE_STARTERSrx   rx   rx   ry   r   o   s    r   N)rt   ru   rv   ZPunctuationsr   ZGeoLocationRuleZFileFormatRuleZSingleNewLineRuleZQuestionMarkInQuotationRuleZExtraWhiteSpaceRuleZSubSingleQuoteRuleobjectrz   r   r   r   r   r   r   rx   rx   rx   ry   r      s    





r   N)Zpysbd.utilsr   Zpysbd.abbreviation_replacerr   r   rx   rx   rx   ry   <module>   s   