o
    ÂifÁ
  ã                   @   s@   d dl mZ G dd„ deƒZG dd„ deƒZG dd„ deƒZdS )	é    )ÚRulec                   @   sÔ   e Zd ZeddƒZeddƒZeddƒZeddƒZeddƒZedd	ƒZ	ed
dƒZ
edd	ƒZeddƒZeddƒZeddƒZeddƒZeddƒZdZeedƒZdZeedƒZg d¢ZdZeddƒZeddƒZeddƒZdS )Ú
CleanRulesz\n(?=[a-zA-Z]{1,2}\n)Ú z\n \núz\n\nz\n(?=\.(\s|\n))z\nz\\nÚ
z\\rz\\\ nz\\\ rz!{b\^&gt;\d*&lt;b\^}|{b\^>\d*<b\^}z\.{4,}\s*\d+-*\d*z\.{5,}ú z\/{3}z(?<=[a-z])\.(?=[A-Z])z. z(?<=\d)\.(?=[A-Z]))ú@Úhttpz.comÚnetÚwwwz//z(?<=\s)\n(?=([a-z]|\())u
   \n(?=â€¢')z''ú"z``N)Ú__name__Ú
__module__Ú__qualname__r   ÚNewLineInMiddleOfWordRuleÚDoubleNewLineWithSpaceRuleÚDoubleNewLineRuleÚNewLineFollowedByPeriodRuleÚ$ReplaceNewlineWithCarriageReturnRuleÚEscapedNewLineRuleÚEscapedCarriageReturnRuleÚTypoEscapedNewLineRuleÚTypoEscapedCarriageReturnRuleÚInlineFormattingRuleÚTableOfContentsRuleÚConsecutivePeriodsRuleÚConsecutiveForwardSlashRuleÚ NO_SPACE_BETWEEN_SENTENCES_REGEXÚNoSpaceBetweenSentencesRuleÚ&NO_SPACE_BETWEEN_SENTENCES_DIGIT_REGEXÚ NoSpaceBetweenSentencesDigitRuleÚURL_EMAIL_KEYWORDSÚ#NEWLINE_IN_MIDDLE_OF_SENTENCE_REGEXÚNewLineFollowedByBulletRuleÚQuotationsFirstRuleÚQuotationsSecondRule© r&   r&   úJ/var/www/html/corbot_env/lib/python3.10/site-packages/pysbd/clean/rules.pyr      s.    
















r   c                   @   s(   e Zd ZeddƒZeddƒZeegZdS )ÚHTMLzD<\/?\w+((\s+\w+(\s*=\s*(?:\".*?\"|'.*?'|[\^'\">\s]+))?)+\s*|\s*)\/?>r   z&lt;\/?[^gt;]*gt;N)r   r   r   r   ÚHTMLTagRuleÚEscapedHTMLTagRuleÚAllr&   r&   r&   r'   r(   A   s    

r(   c                   @   s    e Zd ZeddƒZeddƒZdS )ÚPDFz(?<=[^\n]\s)\n(?=\S)r   z\n(?=[a-z])r   N)r   r   r   r   ÚNewLineInMiddleOfSentenceRuleÚ%NewLineInMiddleOfSentenceNoSpacesRuler&   r&   r&   r'   r,   K   s    
r,   N)Úpysbd.utilsr   Úobjectr   r(   r,   r&   r&   r&   r'   Ú<module>   s   <
