o
     ¢i"  ã                   @   s¤   d dl Z d dlmZmZ dd„ Ze j dg d¢¡dd„ ƒZe j d	g d
¢¡dd„ ƒZe j dg d¢¡dd„ ƒZ	e j ddg¡dd„ ƒZ
e j dg d¢¡dd„ ƒZdS )é    N)Úlike_numÚnorm_customc                 C   s    d}| |ƒ}t |ƒdksJ ‚d S )NuŽ  OnÃ¨ ap fÃ¨t pou ansyen lidÃ¨ Pati TravayÃ¨ Britanik

Moun atravÃ¨ lemond ap voye onÃ¨ pou ansyen lidÃ¨
Pati TravayÃ¨ a, John Smith, ki mouri pi bonÃ¨ jodi a apre li te fÃ¨ yon gwo kriz kadyak a laj 55 an.

Nan Washington, Depatman Deta Etazini pibliye yon deklarasyon ki eksprime "regre lanmÃ² twÃ² bonÃ¨" avoka ak palmantÃ¨ eskoze a.

"Misye Smith, pandan tout karyÃ¨ li ki te make ak distenksyonéT   ©Úlen)Úht_tokenizerÚtextÚtokens© r
   úQ/home/ubuntu/.local/lib/python3.10/site-packages/spacy/tests/lang/ht/test_text.pyÚ#test_ht_tokenizer_handles_long_text   s   r   ztext,length))z;Map manje gato a pandan map gade televizyon lem lakay mwen.é   )z@M'ap vini, eske wap la avek lajan'm? Si oui, di'l non pou fre'w.é   )zM ap teste sa (pou kounye a).é
   c                 C   s   | |ƒ}t |ƒ|ksJ ‚d S ©Nr   )r   r   Úlengthr	   r
   r
   r   Útest_ht_tokenizer_handles_cnts   s   	r   z
text,match))Ú10T)Ú1T)z10,000T)z10,00T)z999.0T)ÚenT)ÚdeT)ÚmilyaT)ÚdogF)ú,F)z1/2Tc                 C   s.   | |ƒ}t |ƒdksJ ‚|d j|ksJ ‚d S )Né   r   )r   r   )r   r   Úmatchr	   r
   r
   r   Útest_lex_attrs_like_number    s   r   Úword)u   ventyÃ¨mu   MilyonnyÃ¨mu   3yÃ¨mu   SantyÃ¨mu   25yÃ¨mu   52yÃ¨mc                 C   s   t | ƒsJ ‚d S r   )r   ©r   r
   r
   r   Ú)test_ht_lex_attrs_like_number_for_ordinal6   s   r   Úonzc                 C   s    t | ƒsJ ‚t |  ¡ ƒsJ ‚d S r   )r   Úupperr   r
   r
   r   Útest_ht_lex_attrs_capitals=   s   r"   zword, expected))z'mÚmwen)z'nÚnou)z'lÚli)z'yÚyo)z'wÚouc                 C   s   t | ƒ|ksJ ‚d S r   )r   )r   Úexpectedr
   r
   r   Útest_ht_lex_attrs_norm_customC   s   r)   )ÚpytestÚspacy.lang.ht.lex_attrsr   r   r   ÚmarkÚparametrizer   r   r   r"   r)   r
   r
   r
   r   Ú<module>   s0    þ
þ
ÿ

þ
