U
    >vf,&                     @   s*  d Z ddlmZmZ ddlZddlZddlZddlZddl	Z	ddl
mZmZ ddlmZ e	jdZe	jdZdd Zed	d
 Zee	jdddge	jddddgdd Zdd Zdd Ze	jddi eddgifdddieddgifdd dgiedd!d"gifd#dgd$d%eddgifd#dgd&d%edejdgifgd'd( Zd)d* Ze	jd+d,d-d.gd/d0 Zee	jd1d$d&gd2d3 Zed4d5 Ze	jdd6dd7d8d9gd:d; Z ee	jdd6ddd<d=gd>d? Z!ed@dA Z"edBdC Z#e	jde	jdDdEdFgdGdH Z$dS )IzZ
Tests encoding functionality during parsing
for all of the parsers defined in parsers.py
    )BytesIOTextIOWrapperN)	DataFrameread_csvZpyarrow_skipZpyarrow_xfailc                 C   sL   d}| }t d|}|j|d|d}tddggddgd	}t|| d S )
Ncp1255u   שלום:1234
562:123:)sepencodingi2  {   u   שלוםZ1234columnsr   encoder   r   tmassert_frame_equal)all_parsersr	   parserdataresultexpected r   H/tmp/pip-unpacked-wheel-vdrwu74i/pandas/tests/io/parser/test_encoding.pytest_bytes_io_input   s    r   c                 C   s@   | }t d }|j|ddd d}tddgg}t|| d S )Nu   Łaski, Jan;1;utf-8)r   r	   headeru   Łaski, Jan   r   )r   r   r   r   r   r   r   r   test_read_csv_unicode%   s
    r   r   ,	r	   utf-16zutf-16lezutf-16bec              
   C   s   | }d d|}dt  d}|dd}d}t|}||}t|d}	|	| W 5 Q R X tt	|||d	2}
|j
|fd
|i|}|j
|
fd
|i|}W 5 Q R X t|| W 5 Q R X d S )Nz)skip this
skip this too
A,B,C
1,2,3
4,5,6r   __z__.csv   )r   Zskiprowsr   wbr	   r	   )replaceuuidZuuid4r   ensure_cleanr   openwriter   r   r   r   )r   r   r	   r   r   pathkwargsutf8
bytes_datafZbytes_bufferr   r   r   r   r   test_utf16_bom_skiprows/   s      

 r/   c                 C   s6   t j|d}| }|j|ddd}t|dks2td S )Nzutf16_ex.txtr    r   )r	   r   2   )osr*   joinr   lenAssertionError)r   csv_dir_pathr*   r   r   r   r   r   test_utf16_exampleL   s    r6   c                 C   sL   t j|d}| }|j|d dd}|d}|d d }d}||ksHtd S )Nunicode_series.csvlatin-1)r   r	   r   r   i`  u$   Á köldum klaka (Cold Fever) (1994))r1   r*   r2   r   Z	set_indexr4   )r   r5   r*   r   r   gotr   r   r   r   test_unicode_encodingS   s    
r:   zdata,kwargs,expectedza
1ar   z"a"
1	quotechar"zb
1namesb1
1T)r>   skip_blank_linesFc                    sx   | }d d fdd}|j dkrN|dkrN|ddrN|jtjjd	d
 |j||fdi|}t	|| d S )Nu   ﻿r   c                    s    |   }t|S )N)r   r   )_dataZbom_databomr,   r   r   _encode_data_with_bomx   s    z,test_utf8_bom.<locals>._encode_data_with_bomZpyarrowrA   rB   TzPyarrow can't read blank lines)reasonr	   )
enginegetnodeZ
add_markerpytestmarkZxfailr   r   r   )r   r   r+   r   requestr   rF   r   r   rD   r   test_utf8_bom_   s    
rN   c                 C   sL   t dgdgd}| }||}d|}|jt||d}t|| d S )Ng333333@test)Zmb_numZ	multibytezmb_num,multibyte
4.8,testr$   )r   formatr   r   r   r   r   )r   	utf_valueencoding_fmtr   r   r	   r   r   r   r   r   test_read_csv_utf_aliases   s    

rS   zfile_path,encoding))ior   csvz	test1.csvr   ))rT   r   r   r7   r8   ))rT   r   r   zsauron.SHIFT_JIS.csvshiftjisc           
   	   C   s   | }|| }|j ||d}t||d}| |}|jr<tW 5 Q R X t|| t|dd}	|j |	|d}|	jrxtW 5 Q R X t|| t|ddd}	|j |	|d}|	jrtW 5 Q R X t|| d S )Nr$   rbmoder   )rY   	buffering)r   r(   closedr4   r   r   )
r   	file_pathr	   datapathr   Zfpathr   far   Zfbr   r   r   test_binary_mode_file_buffers   s    
r_   pass_encodingc           	   	   C   sr   | }| |}tddgi}tjd|dd<}|d |d |j||rP|nd d}t|| W 5 Q R X d S )	NZfoobarzw+T)rY   r	   Zreturn_filelikezfoo
barr   r$   )rP   r   r   r'   r)   seekr   r   )	r   rQ   rR   r`   r   r	   r   r.   r   r   r   r   test_encoding_temp_file   s    


rc   c              	   C   s~   | }d}d}d}t ||gi}t N}|| d| | |d |j||d}t|| |j	rpt
W 5 Q R X d S )Nz	shift-jisu	   てすとu   こむ
r   r$   )r   tempfileNamedTemporaryFiler)   r   rb   r   r   r   r[   r4   )r   r   r	   titler   r   r.   r   r   r   r   test_encoding_named_temp_file   s    

rh   r   z	utf-16-bez	utf-16-lezutf-32c                 C   sR   d}t || }t|d| d}tddgddgdd	ggd
dgd}t|| d S )Nu   a	b
：foo	0
bar	1
baz	2r   )	delimiterr	   u   ：foor   ra   r   Zbazr"   r;   r?   )r   r   r   )r	   r   Zencoded_datar   r   r   r   r   %test_parse_encoded_special_characters   s
    "rj   r   r8   c              	   C   sp   | }t ddddgddddgd	d
ddgd}t &}|j|d|d |j||dd}W 5 Q R X t|| d S )NZRaphaelZ	DonatellozMiguel AngelZLeonardoZredZpurpleZorangeZblueZsaizbo staffZnunchunkZkatana)namemaskZweaponF)indexr	   T)r	   
memory_map)r   r   r'   to_csvr   r   )r   r	   r   r   filedfr   r   r   test_encoding_memory_map   s    



rr   c              	   C   sh   | }t dgd d}d|jd< td*}|j|dddd	 |j|d
ddd}W 5 Q R X t|| d
S )zO
    Chunk splits a multibyte character with memory_map=True

    GH 43540
    Zaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaai   )r   u   aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaąi  zbug-gh43540.csvFr   rm   r   r	   NTc)r   rn   rH   )r   Zilocr   r'   ro   r   r   )r   r   rq   fnamedfrr   r   r    test_chunk_splits_multibyte_char   s    
rw   c              	   C   s   g }d}d}d}t t|t||D ]X}ddd t ||d D d }z|d W n tk
rp   Y q$Y nX || q$| }t|}td	,}	|j	|	d
d
dd |j
|	ddddd}
W 5 Q R X t||
 dS )zg
    GH 43787

    Test correct handling of UTF-8 chars when memory_map=True and encoding is UTF-8
        u   𐂀 c                 S   s   g | ]}t |qS r   )chr).0rt   r   r   r   
<listcomp>  s     z,test_readcsv_memmap_utf8.<locals>.<listcomp>rd   r   zutf8test.csvFrs   NTrt   )r   rn   rH   r	   )rangeordr2   r   UnicodeEncodeErrorappendr   r   r'   ro   r   r   )r   linesZline_lengthZ
start_charZend_charlnumliner   rq   ru   rv   r   r   r   test_readcsv_memmap_utf8  s.    "
    r   rY   zw+bzw+tc              	   C   sh   | }d}d|krd}t j|d$}|| |d ||}W 5 Q R X tg dgd}t|| d S )Ns   abcdtZabcdrX   r   r   )re   SpooledTemporaryFiler)   rb   r   r   r   r   )r   rY   r   contenthandlerq   r   r   r   r   test_not_readable.  s    

r   )%__doc__rT   r   r   r1   re   r&   ZnumpynprK   Zpandasr   r   Zpandas._testingZ_testingr   rL   ZusefixturesZskip_pyarrowZxfail_pyarrowr   r   Zparametrizer/   r6   r:   nanrN   rS   r_   rc   rh   rj   rr   rw   r   r   r   r   r   r   <module>   sz   
	



 



