o
     ¢iÐ  ã                
   @   s´   d dl Z d dlmZ e j ddg d¢fg¡dd„ ƒZe j ddg d	¢fd
g d¢fdg d¢fdg d¢fdg d¢fg¡dd„ ƒZe j dg d¢¡dd„ ƒZe j dg d¢¡dd„ ƒZdS )é    N©Úlike_numztext,expected_tokensu2   ×¤×™×™×ª×•×Ÿ ×”×™× ×©×¤×ª ×ª×›× ×•×ª ×“×™× ×ž×™×ª)u   ×¤×™×™×ª×•×Ÿu   ×”×™×u   ×©×¤×ªu
   ×ª×›× ×•×ªu   ×“×™× ×ž×™×ªc                 C   s&   | |ƒ}dd„ |D ƒ}||ksJ ‚d S )Nc                 S   s   g | ]}|j s|j‘qS © )Úis_spaceÚtext©Ú.0Útokenr   r   úV/home/ubuntu/.local/lib/python3.10/site-packages/spacy/tests/lang/he/test_tokenizer.pyÚ
<listcomp>   s    z:test_he_tokenizer_handles_abbreviation.<locals>.<listcomp>r   )Úhe_tokenizerr   Úexpected_tokensÚtokensÚ
token_listr   r   r
   Ú&test_he_tokenizer_handles_abbreviation   s   r   u1   ×¢×§×‘×ª ××—×¨×™×• ×‘×›×œ ×¨×—×‘×™ ×”×ž×“×™× ×”.)õ   ×¢×§×‘×ªõ
   ××—×¨×™×•õ   ×‘×›×œõ   ×¨×—×‘×™õ   ×”×ž×“×™× ×”Ú.u1   ×¢×§×‘×ª ××—×¨×™×• ×‘×›×œ ×¨×—×‘×™ ×”×ž×“×™× ×”?)r   r   r   r   r   ú?u1   ×¢×§×‘×ª ××—×¨×™×• ×‘×›×œ ×¨×—×‘×™ ×”×ž×“×™× ×”!)r   r   r   r   r   ú!u2   ×¢×§×‘×ª ××—×¨×™×• ×‘×›×œ ×¨×—×‘×™ ×”×ž×“×™× ×”..)r   r   r   r   r   z..u3   ×¢×§×‘×ª ××—×¨×™×• ×‘×›×œ ×¨×—×‘×™ ×”×ž×“×™× ×”...)r   r   r   r   r   z...c                 C   s"   | |ƒ}|dd„ |D ƒksJ ‚d S )Nc                 S   s   g | ]}|j ‘qS r   )r   r   r   r   r
   r   +   s    z3test_he_tokenizer_handles_punct.<locals>.<listcomp>r   )r   r   r   r   r   r   r
   Útest_he_tokenizer_handles_punct   s   r   z
text,match))Ú10T)Ú1T)z10,000T)z10,00T)z999.0T)u   ××—×“T)u
   ×©×ª×™×™×T)õ
   ×ž×œ×™×•×ŸT)u   ×›×œ×‘F)ú,F)z1/2Tc                 C   s.   | |ƒ}t |ƒdksJ ‚|d j|ksJ ‚d S )Né   r   )Úlenr   )r   r   Úmatchr   r   r   r
   Útest_lex_attrs_like_number.   s   r!   Úword)u
   ×©×œ×™×©×™r   u
   ×¢×©×™×¨×™u   ×ž××”u   ×¢×©×¨u   ××—×“ ×¢×©×¨c                 C   s   t | ƒsJ ‚d S )Nr   )r"   r   r   r
   Ú)test_he_lex_attrs_like_number_for_ordinalD   s   r#   )	ÚpytestÚspacy.lang.he.lex_attrsr   ÚmarkÚparametrizer   r   r!   r#   r   r   r   r
   Ú<module>   sD    þ
þþþþþïþ
þ
