o
    dW                    @   sz  d dl Z d dlZd dl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Zd dl	Z	d dl
Zd dlZd dlmZ d dlZd dlmZ d dlmZmZ d dlmZ d dlmZ d dlmZmZmZmZm Z m!Z!m"Z"m#Z#m$Z$ ej%dd	 Z&ej%d
d Z'G dd dZ(ej)*dg ddd Z+dd Z,dd Z-ej)*dg dej)*dddgej)*dddgdd Z.ej)*dddgej)*dg d d!d" Z/ej)*dg dd#d$ Z0d%d& Z1ej)*d'd(d)d* Z2d+d, Z3d-d. Z4d/d0 Z5ej)*d1ddgej)*d2ddgd3d4 Z6d5d6 Z7d7d8 Z8d9d: Z9ej)*dg dej)*d;ej:ej;ej<ej=ej>ej?ej@ejAejBg	d<d= ZCdS )>    N)datetime)is_categorical_dtype)	DataFrameSeries)_compression_to_extensionread_csv)	CategoricalConversionWarningInvalidColumnNamePossiblePrecisionLossStataMissingValueStataReaderStataWriterStataWriterUTF8ValueLabelTypeMismatch
read_statac                   C   s   t g dg dg ddS )N            )      ?      @g      ;@g     @T@)ZAtlantaZ
BirminghamZ
CincinnatiZDetroitabc)r    r   r   N/app/.heroku/python/lib/python3.10/site-packages/pandas/tests/io/test_stata.pymixed_frame'   s   r   c                 C   s&   | dddd}t |dd}d|j_|S )Niodatastatazstata5_114.dtaTconvert_datesindex)r   r%   name)datapathZ	dta14_114
parsed_114r   r   r   r(   2   s   r(   c                   @   s  e Zd Zdd Zdd Zejdg ddd Zejdg dd	d
 Z	ejdddgdd Z
dd Zejdg ddd Zejdg ddd Zdd Zdd Zdd Zdd Zejdg dd d! Zd"d# Zd$d% Zejdg dd&d' Zd(d) Zejdg dd*d+ Zd,d- Zejdg dejdg d.d/d0 Zejdg d1d2d3 Zejdg dd4d5 Zejdg dd6d7 Zd8d9 Zejdg dd:d; Zd<d= Zd>d? Z d@dA Z!dBdC Z"dDdE Z#dFdG Z$dHdI Z%dJdK Z&ejdg dejdLdMdNgdOdP Z'dQdR Z(dSdT Z)dUdV Z*dWdX Z+ejdg dYdZd[ Z,d\d] Z-d^d_ Z.d`da Z/ejdg dej0dbdcdd Z1dedf Z2ejdg ddgdh Z3ejddidjgdkdl Z4ejddmdngdodp Z5ejddidjgdqdr Z6ejdg dsejdtdudvgejdwdxdygejdzdxdygd{d| Z7e8d}e9d~e9fddZ:dd Z;ejdg dejdtdudvgejdwdxdygejdzdxdygdd Z<dd Z=ejdg ddd Z>ejdg ddd Z?ejdddgdd Z@dd ZAdd ZBdd ZCdd ZDdd ZEdd ZFdd ZGdd ZHejdeIjJeIjJ gdd ZKdd ZLdd ZMejddydxgdd ZNdd ZOejdg ddd ZPdd ZQdd ZRdd ZSejdg ddd ZTdd ZUdd ZVdd ZWejdg ddd ZXejdg ddd ZYddÄ ZZddń Z[ejj\ddǄ Z]ejdg dȢddʄ Z^dd̄ Z_dS )	TestStatac                 C      t |ddS )NTr#   )r   selffiler   r   r   read_dta;   s   zTestStata.read_dtac                 C   r*   )NT)Zparse_datesr   r+   r   r   r   r   ?   s   zTestStata.read_csvversionr   u   v   w   Nc                 C   s^   t dgd}t }|j|d|d t|}t|| W d    d S 1 s(w   Y  d S )NunitcolumnsFwrite_indexr/   )r   tmensure_cleanto_statar   assert_frame_equal)r,   r/   Zempty_dspathZ	empty_ds2r   r   r   test_read_empty_dtaB   s   
"zTestStata.test_read_empty_dtac                 C   s   t tdg dd}t }|j|d|d t|}W d    n1 s&w   Y  t|jtj	s4J |
 }|d tj|d< tj||dd d S )	N   )b1b2Zb3Zb4Zb5)r   r   Fr8   r   TZcheck_index_type)r   ranger:   r;   r<   r   
isinstancer%   pd
RangeIndexcopyastypenpint32r=   )r,   r/   dfr>   Zread_dfexpectedr   r   r   test_read_index_col_noneK   s   

z"TestStata.test_read_index_col_noner-   Z
stata1_114
stata1_117c                 C   sh   |ddd| d}|  |}ttjtjtjtjtjfgg dd}|d tj|d< t|| d S )Nr    r!   r"   .dta
float_missZdouble_missZ	byte_missZint_missZ	long_missr6   rR   )r.   r   rJ   nanrI   float32r:   r=   r,   r-   r'   parsedrM   r   r   r   test_read_dta1X   s   
zTestStata.test_read_dta1c                 C   s  t jtdddddddtdd	dtdd	d
tdd	d	tddd	tdd	d	tdd	d	ftdddddddtdddtdddtdd	d	tddd	tdd	d	tdd	d	ftjtjtjtjtjtjtjtjfgg dd}|d d|d< tjdd:}td | 	|dd d!d"}| 	|dd d!d#}| 	|dd d!d$}d%d& |D }t
|dksJ W d    n1 sw   Y  tj||dd' tj||dd' tj||dd' d S )(N                 l   *}ai  r      i     i        r   iZli  
   r   i     i  )Z
datetime_cZdatetime_big_cdateZweekly_dateZmonthly_dateZquarterly_dateZhalf_yearly_dateyearly_dater6   re   OTrecordalwaysr    r!   r"   zstata2_114.dtazstata2_115.dtazstata2_117.dtac                 S   s   g | ]	}|j tu r|qS r   )categoryUserWarning).0xr   r   r   
<listcomp>       z,TestStata.test_read_dta2.<locals>.<listcomp>check_datetimelike_compat)r   from_recordsr   rF   NaTrI   warningscatch_warningssimplefilterr.   lenr:   r=   )r,   r'   rM   wr(   
parsed_115
parsed_117r   r   r   test_read_dta2j   sR   











"
#
	zTestStata.test_read_dta2)Z
stata3_113Z
stata3_114
stata3_115
stata3_117c                 C   sv   |ddd| d}|  |}| |dddd}|tj}|d tj|d< |d tj|d< t|| d S )Nr    r!   r"   rP   
stata3.csvyearquarter)	r.   r   rI   rJ   rT   int16int8r:   r=   rU   r   r   r   test_read_dta3   s   
zTestStata.test_read_dta3)Z
stata4_113Z
stata4_114
stata4_115
stata4_117c           	      C   s  |ddd| d}|  |}tjg dg dg dg dd	d
dtjd	gd
d	dtjd
gdddtjdgdddtjdgdddtjdgdddtjdgg
g dd}|D ]2}||  }t|d |  }|dkri|}|dj	}|j
|dd}|jjd dd |||< qPt|| d S ) Nr    r!   r"   rP   )onetenr   r   r   )twoniner   r   r   )threeeightr   r   r   )foursevenr   r   r   fivesixr@   rc   r   r   r_   r   r   r^   r   r   	   r   r   )fully_labeledfully_labeled2incompletely_labeledlabeled_with_missingsfloat_labelledr6   r   r   rj   TorderedZinplace)r.   r   rr   rJ   rS   rH   ZasarrayZnotnarI   _valuesset_categories
categoriesrenamer:   r=   )	r,   r-   r'   rV   rM   colorigr   catr   r   r   test_read_dta4   s4   

zTestStata.test_read_dta4c                 C   sL   |  |dddd}tjg dg dg dgg dd	}tj||d
d d S )Nr    r!   r"   zstata12_117.dta)r   abcZ	abcdefghi)r   ZcbaZqwertywertyqwerty)]    Zstrl)rm   yzr6   Fcheck_dtype)r.   r   rr   r:   r=   )r,   r'   rz   rM   r   r   r   test_read_dta12   s   	zTestStata.test_read_dta12c              	   C   s
  |  |dddd}|d d|d< tjg ddd	d
tjtjtjtjgg dg dg dgg dd}|d tj|d< |jD ]}t	|| ||  qAt
|dddd$}| }dddddddd}t|| |jdkssJ W d    d S 1 s~w   Y  d S )Nr    r!   r"   zstata14_118.dtaBytesrf   )ZCatZBogotau   Bogotár   r   u   option b Ünicoder   ZDogZBostonu   Uzunköprü)ZPlaneZRomeu   Tromsør           option ar   )ZPotatoZTokyou	   Elâzığ      @r   r   )r   r   r   r   gT?r   gUUUUUU?)ThingsCitiesUnicode_Cities_StrlIntsFloatsr   Longsr6   r   u'   Here are some strls with Ünicode charsz	long datazHere are some thingsz	byte datazint datazHere are some citiesz
float data)r   r   r   r   r   r   r   u   This is a  Ünicode data label)r.   rI   r   rr   rJ   rS   rT   r7   r:   Zassert_almost_equalr   variable_labelsZassert_dict_equal
data_label)r,   r'   Z
parsed_118rM   r   rdrZvlZvl_expectedr   r   r   test_read_dta18   s6   
	"zTestStata.test_read_dta18c                 C   s   t tjtjtjtjtjfgg dd}d|j_t }|j|d d | |}W d    n1 s2w   Y  |	 }|j
tj|_t|d| d S )NrQ   r6   r%   r#   )r   rJ   rS   r%   r&   r:   r;   r<   r.   rH   rI   rK   r=   	set_index)r,   originalr>   written_and_read_againrM   r   r   r   test_read_write_dta5   s   
zTestStata.test_read_write_dta5c                 C   s   |  |dddd}d|j_|jtj|_|d tj|d< |d tj|d< t  }|j|d d | 	|}tj
|d|d	d
 W d    d S 1 sQw   Y  d S )Nr    r!   r"   r~   r%   r   r   r#   FrC   )r   r%   r&   rI   rJ   rK   r:   r;   r<   r.   r=   r   )r,   r'   r   r>   r   r   r   r   test_write_dta6/  s   

"zTestStata.test_write_dta6c                 C   s   t ddddtdggg dd}t|d td|d< d	|j_|jtj|_|d
 tj|d
< t	
 #}|j|ddi|d | |}t	j|d	|dd W d    d S 1 s[w   Y  d S )Nstringobjectr   皙?
2003-12-25)r   r   integerfloatingr   r!   r7   dtyper%   r   r   tcr$   r/   FrC   )r   rJ   
datetime64r   r   r%   r&   rI   rK   r:   r;   r<   r.   r=   r   )r,   r/   r   r>   r   r   r   r   test_read_write_dta10?  s"   

"zTestStata.test_read_write_dta10c                 C   sR   t  }ttjddtdd}|| W d    d S 1 s"w   Y  d S )Nrb   r   ABr6   )r:   r;   r   rJ   randomrandnlistr<   )r,   r>   rL   r   r   r   test_stata_doc_examplesT  s   
"z!TestStata.test_stata_doc_examplesc                 C   s   t jd tt jddtdd}t j|jdddf< | }t	
 }|j|d	d
 W d    n1 s7w   Y  t	|| d S )Ni  r@   r   abcdr6   r   r   r   Fr9   )rJ   r   seedr   r   r   rS   locrH   r:   r;   r<   r=   )r,   rL   Zdf_copyr>   r   r   r   test_write_preserves_originalY  s   
z'TestStata.test_write_preserves_originalc           	      C   s   t |dddd}t |dddd}|jd }|jd }||ks"J t|ts)J t }|j|d|d t |}t|| W d    d S 1 sKw   Y  d S )Nr    r!   r"   zstata1_encoding.dtar   Fr8   )r   	kreis1849rE   strr:   r;   r<   r=   )	r,   r/   r'   rawencodedresultrM   r>   reread_encodedr   r   r   test_encodingc  s   


"zTestStata.test_encodingc              	   C   s   t dgg dd}t dgg dd}d|j_|tj}t )}tt	 |j
|d d W d    n1 s8w   Y  | |}W d    n1 sLw   Y  | }|jtj|_t|d| d S )Nr   )goodu   bädZ8numberZ%astringwithmorethan32characters______r6   )r   Zb_dZ_8number astringwithmorethan32characters_r%   r#   )r   r%   r&   rI   rJ   rK   r:   r;   assert_produces_warningr
   r<   r.   rH   r=   r   )r,   r   	formattedr>   r   rM   r   r   r   test_read_write_dta11s  s&   	
zTestStata.test_read_write_dta11c              	   C   s   t dgg dd}t dgg dd}d|j_|tj}t 9}tj	dd}t
dt |j|d |d	 t|d
ks>J W d    n1 sHw   Y  | |}W d    n1 s\w   Y  | }|jtj|_t|d| d S )Nr   r   r   r   r@   rc   )Z!astringwithmorethan32characters_1Z!astringwithmorethan32characters_2+-Zshortdeleter6   )r   Z _0astringwithmorethan32character_Z_1_Z_short_deleter%   Trg   ri   r   r   )r   r%   r&   rI   rJ   rK   r:   r;   rt   ru   rv   r
   r<   rw   r.   rH   r=   r   )r,   r/   r   r   r>   rx   r   rM   r   r   r   test_read_write_dta12  s*   
	zTestStata.test_read_write_dta12c           	      C   s   t dtjd}t dtjd}t dtjd}t|||d}d|j_|}|d tj	|d< t
 }|| | |}W d    n1 sFw   Y  | }|jtj|_t
|d| d S )Ni   r   i   l        )r   rK   int64r%   r   )r   rJ   r   rK   r   r   r%   r&   rI   float64r:   r;   r<   r.   rH   r=   r   )	r,   s1s2s3r   r   r>   r   rM   r   r   r   test_read_write_dta13  s   

zTestStata.test_read_write_dta13)Z
stata5_113Z
stata5_114
stata5_115
stata5_117c           	      C   s   |ddd| d}|  |}d|j_t|| t }|j|ddi|d |  |}W d    n1 s7w   Y  | }|jt	j
|_t|d| d S )	Nr    r!   r"   rP   r%   date_tdtdr   )r.   r%   r&   r:   r=   r;   r<   rH   rI   rJ   rK   r   )	r,   r-   r(   r/   r'   rV   r>   r   rM   r   r   r   test_read_write_reread_dta14  s   

z&TestStata.test_read_write_reread_dta14)Z
stata6_113Z
stata6_114
stata6_115
stata6_117c                 C   s   |  |dddd}|d tj|d< |d tj|d< |d tj|d< |d tj|d< |d	 tj|d	< |d
 jt	j
dd|d
< |ddd| d}| |}t|| d S )Nr    r!   r"   
stata6.csvbyte_int_long_float_double_r   z%Y-%m-%dargsrP   )r   rI   rJ   r   r   rK   rT   r   applyr   strptimer.   r:   r=   )r,   r-   r'   rM   rV   r   r   r   test_read_write_reread_dta15  s   

z&TestStata.test_read_write_reread_dta15c              	   C   s   t dgdgd}tddddd}d	}t <}|j||||d
 t|}|jdks,J |j|ks3J W d    n1 s=w   Y  W d    d S W d    d S 1 sUw   Y  d S )Nr   variabler6     r            zThis is a data file.)
time_stampr   r/   z29 Feb 2000 14:21)r   r   r:   r;   r<   r   r  r   )r,   r/   r   r  r   r>   readerr   r   r   test_timestamp_and_label  s   

"z"TestStata.test_timestamp_and_labelc              	   C   s   t dgdgd}d}t 2}d}tjt|d |j|||d W d    n1 s+w   Y  tj	|r8J W d    d S 1 sCw   Y  d S )Nr   r  r6   z01 Jan 2000, 00:00:00z"time_stamp should be datetime typematch)r  r/   )
r   r:   r;   pytestraises
ValueErrorr<   osr>   isfile)r,   r/   r   r  r>   msgr   r   r   test_invalid_timestamp  s   
"z TestStata.test_invalid_timestampc              	   C   s   t ttdd}d|j_t '}tt	 |
| W d    n1 s)w   Y  | |}W d    n1 s=w   Y  |d}t|j}dd }t|||_| }|jtj|_t|| d S )Ng      9@)r@   r@   r%   c                 S   s   t | d S )Nr   )intrm   r   r   r   <lambda>  s    z5TestStata.test_numeric_column_names.<locals>.<lambda>)r   rJ   reshapearanger%   r&   r:   r;   r   r
   r<   r.   r   r   r7   maprH   rI   rK   r=   )r,   r   r>   r   r7   Zconvert_col_namerM   r   r   r   test_numeric_column_names  s   


z#TestStata.test_numeric_column_namesc                 C   s   t tdtjd}t tdtjd}tj|d d d< tj|dd d< t||d}d|j_t	
 }|j||d | |}W d    n1 sJw   Y  |d}| }|jtj|_t	|| d S )Nr   r   r   r   r   r   r%   r/   )r   rJ   r  rT   r   rS   r   r%   r&   r:   r;   r<   r.   r   rH   rI   rK   r=   )r,   r/   r   r   r   r>   r   rM   r   r   r   test_nan_to_missing_value  s   

z#TestStata.test_nan_to_missing_valuec              	   C   s   ddg}t ttdd|d}d|j_t :}|j|dd | 	|}t
jt|jjd	 |d  W d    n1 s>w   Y  W d    d S W d    d S 1 sVw   Y  d S )
Nrm   r         $@)r@   r   r6   Zindex_not_writtenFr   r	  )r   rJ   r  r  r%   r&   r:   r;   r<   r.   r  r  KeyError)r,   r7   r   r>   r   r   r   r   test_no_index+  s   


"zTestStata.test_no_indexc                 C   s   t ddg}t ddgtjd}t||d}d|j_t }|| | 	|}W d    n1 s3w   Y  |
 }|jtj|_t|d| d S )Nr   zA longer stringr          @r   r  r%   )r   rJ   r   r   r%   r&   r:   r;   r<   r.   rH   rI   rK   r=   r   )r,   r   r   r   r>   r   rM   r   r   r   test_string_no_dates5  s   

zTestStata.test_string_no_datesc           	   	   C   s8  t ddgtjd}t ddgtjd}t ddgtjd}t ddgtjd}t||||d}d|j_t	 '}t
t || W d    n1 sKw   Y  | |}W d    n1 s_w   Y  | }t |d	 tjd|d	< t |d
 tjd|d
< t |d tjd|d< |jtj|_t|d| d S )Nr   c   r      i  l    )s0r   r   r   r%   r   r   r   )r   rJ   r   r   r   r   r%   r&   r:   r;   r   r   r<   r.   rH   rK   r   rI   r=   r   )	r,   r#  r   r   r   r   r>   r   modifiedr   r   r   test_large_value_conversionB  s$   
z%TestStata.test_large_value_conversionc              	   C   s   t tddddddg}d|j_t +}tt |j|dd	id
 W d    n1 s.w   Y  | 	|}W d    n1 sBw   Y  |
 }dg|_|jtj|_t|d| d S )NrX   rY   rZ   r[   r\   r]   r%   r   r   r#   Z_0)r   r   r%   r&   r:   r;   r   r
   r<   r.   rH   r7   rI   rJ   rK   r=   r   )r,   r   r>   r   r$  r   r   r   test_dates_invalid_columnV  s   
z#TestStata.test_dates_invalid_columnc                 C   s   |dddd}t |}g dg dg dg}t|}g d|_|d	 tj|d	< |d
 tj|d
< |d tj|d< |d tj|d< t	|
d| d S )Nr    r!   r"   S4_EDUC1.dta)r   r   r   )r   r   r   r(  )r   r   r   r(  )clustnumpri_schlpsch_numpsch_disr)  r*  r+  r,  r   )r   r   r7   rI   rJ   r   r   rT   r:   r=   head)r,   r'   dpathrL   Zdf0r   r   r   test_105d  s   
zTestStata.test_105c                 C   sL   |dddd}t |}| i ksJ W d    d S 1 sw   Y  d S )Nr    r!   r"   r'  )r   value_labels)r,   r'   r.  r  r   r   r   test_value_labels_old_formatr  s   
"z&TestStata.test_value_labels_old_formatc           	   
   C   s  g d}dd |D }t ddddddgt| }t|g|d	}d
|j_t ddddddt dddt dddt dddt dddt dddt dddg}t|gtjdgtjd
d|d}t	
 }|j||d | |}W d    n1 sww   Y  t	|d
| d S )N)r   r   Ztwr:   Ztqthtyc                 S   s   i | ]}||qS r   r   rl   r   r   r   r   
<dictcomp>}  s    z6TestStata.test_date_export_formats.<locals>.<dictcomp>rX   rY   r]   r[   r\   r6   r%   rZ   r   rb   r_   r   r   r&   r%   r7   r#   )r   rw   r   r%   r&   rF   IndexrJ   rK   r:   r;   r<   r.   r=   r   )	r,   r7   Zconversionsr!   r   expected_valuesrM   r>   r   r   r   r   test_date_export_formats{  s.   







z"TestStata.test_date_export_formatsc                 C   s   t dgd ggdgd}t dgdggtjddgtjdddgd	}t }|| | |}W d    n1 s9w   Y  t	|
d| d S )
N1Zfoor6   r   r   r   r%   r6  r7  )r   rF   r8  rJ   rK   r:   r;   r<   r.   r=   r   )r,   r   rM   r>   r   r   r   r   test_write_missing_strings  s   


z$TestStata.test_write_missing_strings	byteorder><c              	   C   sZ  t g dtjd}t g dtjd}t g dtjd}t g dtjd}t g dtjd}t g dtjd}t g dtjd}	t|||||||	d	}
d
|
j_|
	 }|
j
tj|_tjtjtjtjtjtjtjf}t|j|D ]\}}|| 
|||< qqt }|
j|||d | |}W d    n1 sw   Y  |d
}t|| d S )N)r   r   Tr   )r   r   d   )r   r      )r   r   i  )r   r   i  )r   r   i)r   r   l    )r#  r   r   r   s4s5s6r%   )r=  r/   )r   rJ   Zbool_Zuint8Zuint16Zuint32r   r%   r&   rH   rI   rK   r   r   r   zipr7   r:   r;   r<   r.   r   r=   )r,   r=  r/   r#  r   r   r   rB  rC  rD  r   rM   Zexpected_typesr   tr>   r   r   r   r   test_bool_uint  s:   	

zTestStata.test_bool_uintc           	      C   s   t |dddd}| }W d    n1 sw   Y  t |dddd}| }W d    n1 s5w   Y  d}d}| D ]\}}||v sLJ ||| ksTJ ||v sZJ ||v s`J qBd S )Nr    r!   r"   zstata7_115.dtazstata7_117.dta)Zvar1Zvar2Zvar3)label1label2Zlabel3)r   r   items)	r,   r'   r   Zsr_115Zsr_117keyslabelskvr   r   r   test_variable_labels  s   

zTestStata.test_variable_labelsc           
   	   C   s  d}i }|D ]}t d| d| d| g|dt| < qt|}t ]}|j|dd t|9}|  t|j	|j
|jD ]#\}}}	t|dd  t|dd	 ksUJ t|dd  |	ksaJ q>W d    n1 slw   Y  W d    d S W d    d S 1 sw   Y  d S )
N)r   r@     r   r   r   sFr   r   )r   r   r   r:   r;   r<   r   Z_ensure_openrE  Z_varlistZ_fmtlistZ_typlistr  )
r,   str_lensrQ  str_lenr   r>   srr  fmttypr   r   r   test_minimal_size_col  s&   

$"zTestStata.test_minimal_size_colc              	   C   s   d}i }|D ]}t d| d| d| g|dt| < qt|}d}tjt|d* t }|| W d    n1 s>w   Y  W d    d S W d    d S 1 sVw   Y  d S )N)r   rP    r   r   r   rQ  zFixed width strings in Stata \.dta files are limited to 244 \(or fewer\)\ncharacters\.  Column 's500' does not satisfy this restriction\. Use the\n'version=117' parameter to write the newer \(Stata 13 and later\) format\.r	  )	r   r   r   r  r  r  r:   r;   r<   )r,   rS  rQ  rT  r   r  r>   r   r   r   test_excessively_long_string  s    
"z&TestStata.test_excessively_long_stringc              	   C   sh  d}t dggdgd}t $}|| t|}|j}W d    n1 s'w   Y  W d    n1 s6w   Y  dd tdD }|dd	 |D ]!}|| d
 }tddD ]}	t|d
 |	 }
|
j	||	 kslJ qYqLtt
ddd }
|
j	d	ksJ tt
ddd }
|
j	dksJ tt
ddd }
|
j	d	ksJ tt
ddd }
|
j	dksJ d S )N)r   hlr   r   r6   c                 S   s   g | ]
}d t d|  qS ).a   )chrrl   ir   r   r   rn     s    z:TestStata.test_missing_value_generator.<locals>.<listcomp>   r   r]  r      z<fs      s     z.zz<ds         s        )r   r:   r;   r<   r   ZVALID_RANGErD   insertr   r   structunpack)r,   typesrL   r>   r   Zvalid_ranger9  rF  offsetra  valr   r   r   test_missing_value_generator  s:   


z&TestStata.test_missing_value_generator)Z
stata8_113
stata8_115
stata8_117c           	         s   g d}t d}t|j g }tdD ]  fddtdD }|| qt||d}t|dd	d
| ddd}t	|| d S )N)Zint8_Zint16_Zint32_Zfloat32_Zfloat64_e   rc  c                    s    g | ]}t  |d    qS )rc  )r   )rl   jra  rK  r   r   rn      s     z;TestStata.test_missing_value_conversion.<locals>.<listcomp>r@   r6   r    r!   r"   rP   TZconvert_missing)
r   sortedZMISSING_VALUESrK  rD   appendr   r   r:   r=   )	r,   r-   r'   r7   smvr!   rowrM   rV   r   ro  r   test_missing_value_conversion  s   z'TestStata.test_missing_value_conversionc                 C   s  g d}g d}g d}g d}g d}g d}g }t ||||||D ]>\}	}
}}}}g }tdD ]*}|dkrD|t|	|
|||| q1|dkrR|t|	d	d	 q1|t|	|
| q1|| q#|tjgd  g d
}tddd|d d< tddd	|d d< tddd	|d d< tddd	|d d< tddd|d d< tddd	 |d d< |d d< tdd	d	 |d d< |d d< tddd	 |d d<  |d d< |d d< tdd	d	 |d d< |d d< t||td}t|dddd}t|dddd}t	j
||dd t	j
||dd dd  |D }t	 }d!|j_|j||d" | |}W d    n	1 s7w   Y  t	j
|d!||jtjdd d S )#N)i  r  '  r@      )r   r   r`   r   r   r   )r   r   ra   r      r[   )r   r   r[   r   r   r   )r   r   ;   r   r   r   r_   r   rc   r   )Zdate_tcr   Zdate_twZdate_tmZdate_tqZdate_thZdate_tyrv  r`      r   r   rb   r   r@   rw     rx  i  )r7   r   r    r!   r"   zstata9_115.dtazstata9_117.dtaTrp   c                 S   s   i | ]	}||d d qS )r(  Nr   r4  r   r   r   r5  W  ro   z,TestStata.test_big_dates.<locals>.<dictcomp>r%   r#   )rE  rD   rr  r   rF   rs   r   r   r   r:   r=   r;   r%   r&   r<   r.   r   rI   rJ   rK   )r,   r'   yrmoddhrmmssrM   r   monthdayhourminutesecondrt  rn  r7   ry   rz   Zdate_conversionr>   r   r   r   r   test_big_dates)  sV   "  , 

zTestStata.test_big_datesc                 C   s  |  |dddd}|d tj|d< |d tj|d< |d tj|d< |d tj|d< |d	 tj|d	< |d
 jt	j
dd|d
< t|dddddd}t|| t|ddddddd}|  |dddd}|d
 jt	j
dd|d
< t|| d S )Nr    r!   r"   r   r   r   r   r   r   r   r   r   stata6_117.dtaTr#   F)r$   Zpreserve_dtypes)r   rI   rJ   r   r   rK   rT   r   r   r   r   r   r:   r=   )r,   r'   rM   Zno_conversion
conversionr   r   r   test_dtype_conversiond  s.   

zTestStata.test_dtype_conversionc                 C   s  |  |dddd}|d tj|d< |d tj|d< |d tj|d< |d tj|d< |d	 tj|d	< |d
 jt	j
dd|d
< g d}|| }t|ddddd|d}t|| g d}|| }t|ddddd|d}t|| d}tjt|d ddg}t|ddddd|d W d    n1 sw   Y  d}tjt|d g d}t|ddddd|d W d    d S 1 sw   Y  d S )Nr    r!   r"   r   r   r   r   r   r   r   r   r   )r   r   r   r  T)r$   r7   )r   r   r   z"columns contains duplicate entriesr	  zEThe following columns were not found in the Stata data set: not_found)r   r   r   	not_found)r   rI   rJ   r   r   rK   rT   r   r   r   r   r   r:   r=   r  r  r  )r,   r'   rM   r7   Zdropped	reorderedr  r   r   r   test_drop_column  sV   
"zTestStata.test_drop_columnz;ignore:\nStata value:pandas.io.stata.ValueLabelTypeMismatchc           	         s  t jg dg dg dg ddddtjddgdddtjddgd	d
dtjd	dgdddtjddgdddtjddgdddtjddgg
g dd   }tj fdd D dd |jd	tj
|_|d t|d< |d t|d< |D ])}||  }|	dj}| }|dkr|j|dd }|jjd dd! |||< qvt } j||d" | |}W d    n1 sw   Y  |d}t|| d S )#N)r   r   r   r   r   r   )r   r   r   r   r   r   )r   r   r   r   r   r   )r   r   r   r   r   r   r   r   r@   rc   r   r   r_   r   r   r^   r   r   r   r   r   rb   )r   r   r   r   r   	unlabeledr6   c                       g | ]	} |  d qS rj   rI   rl   r   r   r   r   rn     ro   z6TestStata.test_categorical_writing.<locals>.<listcomp>r   Zaxisr%   r   r  rj   Tr   r   r  )r   rr   rJ   rS   rH   rF   concatr%   Z	set_namesrI   rK   r   r   r   
as_orderedr   r   r   r:   r;   r<   r.   r   r=   )	r,   r/   rM   r   r   r   r>   r   resr   r  r   test_categorical_writing  sF   


z"TestStata.test_categorical_writingc              	      s*  t jdd gdd gdd gdd ggdgd tj fdd	 D d
d t &}d}tjt|d  	| W d    n1 sDw   Y  W d    n1 sSw   Y  t jdgdgdgdgd
ggdgd tj fdd	 D d
d t
t  	| W d    d S 1 sw   Y  d S )Nr   i'  r   r   dZToo_longr6   c                    r  r  r  r  r  r   r   rn     ro   zBTestStata.test_categorical_warnings_and_errors.<locals>.<listcomp>r   r  zbStata value labels for a single variable must have a combined length less than 32,000 characters\.r	  c                    r  r  r  r  r  r   r   rn     ro   )r   rr   rF   r  r:   r;   r  r  r  r<   r   r   )r,   r>   r  r   r  r   $test_categorical_warnings_and_errors  s0   "
"z.TestStata.test_categorical_warnings_and_errorsc           
         s   dd t dD }|tjg tj|dgd tj fdd D dd d	 j_	t
 } j||d
 | |}W d    n1 sEw   Y  |d	}  }|D ]}|| j}| j}	|j|	dd}|||< qU|jtj|_t
|| d S )Nc                 S   s   g | ]	}d t | gqS )r   )r   r`  r   r   r   rn     ro   zHTestStata.test_categorical_with_stata_missing_values.<locals>.<listcomp>x   Zmany_labelsr6   c                    r  r  r  r  r  r   r   rn     ro   r   r  r%   r  Tr   )rD   rr  rJ   rS   r   rr   rF   r  r%   r&   r:   r;   r<   r.   r   rH   r   remove_unused_categoriesr   r   rI   rK   r=   )
r,   r/   valuesr>   r   r  rM   r   r   Znew_catsr   r  r   *test_categorical_with_stata_missing_values	  s(   




z4TestStata.test_categorical_with_stata_missing_valuesstata10_115stata10_117c           
      C   s~  ddg dt dfddg dt dd d d fddg dt g dfdd	g dt d
dfddg dt g dfddg dt dfddg dt dfg}g }|D ]#\}}}}|rm||tjj||ddf qV||t|t jdf qVt	
t|}|ddd| d}t|}	t||	 |D ]%}t|| jrt|| jj|	| jj t|| jj|	| jj qd S )NTr   )r   r   r   r  er@   reverserR  Znoorder)r   r   r   r   r   r   r   Zfloat_missing)r   r  r  )r   r   r   rR  rR  FZnolabel)r   r  r   r   g      @Zint32_mixed)r  r   r  r   r   r   r   r    r!   r"   rP   )rJ   r  arrayrr  rF   Categorical
from_codesr   rT   r   	from_dictdictr   r:   r=   r   r   assert_series_equalr   codesassert_index_equalr   )
r,   r-   r'   rM   colsZis_catr   rL  r  rV   r   r   r   test_categorical_order!  s6   	z TestStata.test_categorical_orderstata11_115stata11_117c                 C   sx   t |ddd| d}|jddd}tt||_g d}g d	}tjj||d
d}t|dd}t	
||d  d S )Nr    r!   r"   rP   Zsrhfirst)Zna_position)
rR  rR  r   r   r   r   r   r   r   r   )ZPoorZFairZGoodz	Very goodZ	ExcellentT)r  r   r   r&   )r   Zsort_valuesrF   rG   rw   r%   r  r  r   r:   r  )r,   r-   r'   rV   r  r   r   rM   r   r   r   test_categorical_sortingE  s   z"TestStata.test_categorical_sortingc                 C   sf   |ddd| d}t |}t |dd}|D ]}t|| js q|| jjs(J || jjr0J qd S )Nr    r!   r"   rP   F)order_categoricals)r   r   r   r   r   )r,   r-   r'   rV   Zparsed_unorderedr   r   r   r   test_categorical_orderingV  s   z#TestStata.test_categorical_ordering)rO   Z
stata2_117r}   r   r   r   Z
stata7_117rl  Z
stata9_117r  r  	chunksizer   r   convert_categoricalsFTr$   c                 C   sX  |ddd| d}t jdd t d t|||d}W d    n1 s'w   Y  t|d||d	n}d
}	tdD ]V}
t jdd% t d z||}W n ty`   Y W d     n:w W d    n1 skw   Y  |j|	|	| d d f  }| 	|}t
j||ddd |	|7 }	q;W d    d S W d    d S 1 sw   Y  d S )Nr    r!   r"   rP   Trg   ri   r  r$   )iteratorr  r$   r   r@   Fr   rq   rt   ru   rv   r   rD   readStopIterationilocrH   _convert_categoricalr:   r=   r,   r-   r  r  r$   r'   fnamerV   itrposrn  chunk
from_framer   r   r   test_read_chunks_117b  sL   



"zTestStata.test_read_chunks_117r  returnc                 C   sX   | D ]'}| | }t |jr)|j }|jjtkr%tj|jj}|	|}|| |< q| S )zX
        Emulate the categorical casting behavior we expect from roundtripping.
        )
r   r   r   r  r   r   rF   r8  Z_with_inferr   )r  r   Zserr   r   r   r   r   r    s   


zTestStata._convert_categoricalc                 C   s  |dddd}t |}t |dd}|d}t|jddd d f | W d    n1 s0w   Y  t |dd	}t|}t|jddd d f |d  W d    n1 s[w   Y  t |dd}|d}t|jddd d f | W d    n1 sw   Y  t |dd	}| }t|jddd d f | W d    n1 sw   Y  t |d
d	}t|}W d    n1 sw   Y  t|| d S )Nr    r!   r"   stata3_117.dtaTr  r@   r   r  r   )	r   r  r:   r=   r  r   Z	get_chunkrF   r  )r,   r'   r  rV   r  r  Zfrom_chunksr   r   r   test_iterator  s,   
 $
  zTestStata.test_iterator)
Z
stata2_115r|   r   r   r   Z
stata7_115rk  Z
stata9_115r  r  c                 C   sX  |ddd| d}t jdd t d t|||d}W d    n1 s'w   Y  t|d||d	n}d
}	tdD ]V}
t jdd% t d z||}W n ty`   Y W d     n:w W d    n1 skw   Y  |j|	|	| d d f  }| 	|}t
j||ddd |	|7 }	q;W d    d S W d    d S 1 sw   Y  d S )Nr    r!   r"   rP   Trg   ri   r  )r  r$   r  r   r@   Fr  r  r  r   r   r   test_read_chunks_115  sL   
	


"zTestStata.test_read_chunks_115c                 C   s   |dddd}g d}d}t ||d}t |dd	@}d
}tdD ](}|j||d}	|	d u r/ n"|j||| d d f }
tj|
|	dd ||7 }q W d    d S W d    d S 1 s\w   Y  d S )Nr    r!   r"   r  )r   Zcpim1r   r6   Tr  r   r@   Fr   )r   rD   r  r  r:   r=   )r,   r'   r  r7   r  rV   r  r  rn  r  r  r   r   r   test_read_chunks_columns  s"   
"z"TestStata.test_read_chunks_columnsc              	   C   s  d|j _dddd}t 5}|j|||d t|}| }W d    n1 s*w   Y  ddddd}||ks<J W d    n1 sFw   Y  d	|d< t /}|j|||d t|}| }W d    n1 sow   Y  ||kszJ W d    d S 1 sw   Y  d S )
Nr%   	City RankCity ExponentCityr   r   r/   r   )r%   r   r   r   z	The Index)r%   r&   r:   r;   r<   r   r   )r,   r/   r   r   r>   rU  Zread_labelsZexpected_labelsr   r   r   test_write_variable_labels  s,   





"z$TestStata.test_write_variable_labelsc              	   C   s   d|j _dddd}t 2}d}tjt|d |j|||d W d    n1 s+w   Y  W d    d S W d    d S 1 sCw   Y  d S )	Nr%   Zvery longvery longvery longvery longvery longvery longvery longvery longvery longvery longr  r  r   .Variable labels must be 80 characters or fewerr	  r  r%   r&   r:   r;   r  r  r  r<   )r,   r/   r   r   r>   r  r   r   r   test_invalid_variable_labels  s   
"z&TestStata.test_invalid_variable_labelsr1   r2   c              	   C   s   d|j _dddd}d|d< t 0}tjtdd	 |j|||d
 W d    n1 s-w   Y  W d    d S W d    d S 1 sEw   Y  d S )Nr%   r  r  r  r   u   invalid character Œr   z,Variable labels must contain only charactersr	  r  r  )r,   r/   r   r   r>   r   r   r   $test_invalid_variable_label_encoding(  s   
"z.TestStata.test_invalid_variable_label_encodingc              	   C   s  g d}ddd |d}d}tjt|d# t }|j||d W d    n1 s-w   Y  W d    n1 s<w   Y  ddd	d}d
}tjt|d, t }|j||d W d    n1 sgw   Y  W d    d S W d    d S 1 sw   Y  d S )N)u   Ρ   Αu   Νu   Δr  u   Σr  r  r   r   zKVariable labels must contain only characters that can be encoded in Latin-1r	  )r   zqA very, very, very long variable label that is too long for Stata which means that it has more than 80 charactersr  )joinr  r  r  r:   r;   r<   )r,   r   r  Zvariable_labels_utf8r  r>   Zvariable_labels_longr   r   r    test_write_variable_label_errors5  s2   

"z*TestStata.test_write_variable_label_errorsc                 C   s  t dddddddt dddddddt dd	d
d	d
d	dg}tg dg d|d}t P}|j|dd t|dd}t|| |j|dddid t|dd}t|| |j	 
d}|j|d|did t|dd}t|| W d    d S 1 sw   Y  d S )N  r`   ra   .    r  R    r_   r     r   r  r   applebananaZcherrynumsstrsdatesFr   Tr#   r  r   )r9   r$   )dtr   r   r:   r;   r<   r   r=   r7   tolistr%   )r,   r  r   r>   rereaddirectZ	dates_idxr   r   r   test_default_date_conversionS  s,   
"z&TestStata.test_default_date_conversionc              	   C   s   t dddgi}d}tjt|d* t }|| W d    n1 s&w   Y  W d    d S W d    d S 1 s>w   Y  d S )Nr   y      ?       @y       @      @z"Data type complex128 not supportedr	  )r   r  r  NotImplementedErrorr:   r;   r<   )r,   r   r  r>   r   r   r   test_unsupported_typep  s   
"zTestStata.test_unsupported_typec                 C   sb  t dddddddt dddddddt dd	d
d	d
d	dg}tg dg d|d}d}tjt|d% t }|j|ddid W d    n1 sNw   Y  W d    n1 s]w   Y  t	j
dddd}tg dg d|d}tjtdd* t }|| W d    n1 sw   Y  W d    d S W d    d S 1 sw   Y  d S )Nr  r`   ra   r  r  r  r  r  r_   r   r  r  r  r  zFormat %tC not implementedr	  r  ZtCr#   z1-1-1990r   zAsia/Hong_Kong)ZperiodstzzData type datetime64)r  r   r   r  r  r  r:   r;   r<   rF   Z
date_range)r,   r  r   r  r>   r   r   r   test_unsupported_datetypex  s>   

"z#TestStata.test_unsupported_datetypec                 C   sN   d}t jt|d t|dddddd W d    d S 1 s w   Y  d S )	Na*  
Value labels for column ethnicsn are not unique. These cannot be converted to
pandas categoricals.

Either read the file with `convert_categoricals` set to False or use the
low level interface in `StataReader` to separately read the values and the
value_labels.

The repeated labels are:
-+
wolof
r	  r    r!   r"   zstata15.dtaTr  )r  r  r  r   )r,   r'   r  r   r   r   test_repeated_column_labels  s   
"z%TestStata.test_repeated_column_labelsc                 C   s   t |dddd}tdddddddtjddg
ddddtjd	dd
ddg
dtjd
dd	d	ddddg
g dd}|g d }t|| d S )Nr    r!   r"   zstata7_111.dtar   r   r   r   r   r@   rc   )
r   r   r   r  r  r   gr[  ra  rn  )r   rm   rx   r   )r   r   rJ   NaNr:   r=   )r,   r'   rL   r   r   r   r   test_stata_111  s   zTestStata.test_stata_111c              	   C   s   t dttjjdgdttjjttjjgd}d}tjt|d* t	
 }|| W d    n1 s:w   Y  W d    d S W d    d S 1 sRw   Y  d S )Nr   g      ZColumnOkColumnTooBigzZColumn ColumnTooBig has a maximum value \(.+\) outside the range supported by Stata \(.+\)r	  )r   rJ   finfodoubleepsmaxr  r  r  r:   r;   r<   )r,   rL   r  r>   r   r   r   test_out_of_range_double  s   
"z"TestStata.test_out_of_range_doublec                 C   s   t dttjjttjjd gdttjjttjjgd}d|j_|D ]}|| tj||< q)t	
 }|| t|}W d    n1 sNw   Y  |d tj|d< | }|jtj|_t	|d| d S )Nr   r  r  r%   r  )r   rJ   r  rT   r  r  r%   r&   rI   r:   r;   r<   r   r   rH   rK   r=   r   )r,   r   r   r>   r  rM   r   r   r   test_out_of_range_float  s,   


z!TestStata.test_out_of_range_floatinfvalc              	   C   s   t ddgd|gd}d}tjt|d* t }|| W d    n1 s)w   Y  W d    d S W d    d S 1 sAw   Y  d S )Nr   r   r  )Z
WithoutInfZWithInfz[Column WithInf contains infinity or -infinitywhich is outside the range supported by Stata.r	  )r   r  r  r  r:   r;   r<   )r,   r  rL   r  r>   r   r   r   test_inf  s   
"zTestStata.test_infc                 C   6   t  }d|j_dd }t |j|}t || d S )Nr%   c                 S      t | dS Nr%   r   r   r  r   r   r   r        z-TestStata.test_path_pathlib.<locals>.<lambda>)r:   makeDataFramer%   r&   Zround_trip_pathlibr<   r=   r,   rL   r  r   r   r   r   test_path_pathlib  
   zTestStata.test_path_pathlibc                 C   r  )Nr%   c                 S   r  r  r  r  r   r   r   r    r  z6TestStata.test_pickle_path_localpath.<locals>.<lambda>)r:   r  r%   r&   Zround_trip_localpathr<   r=   r  r   r   r   test_pickle_path_localpath  r	  z$TestStata.test_pickle_path_localpathr9   c              	   C   s   dg di}t |d}|d d|d< t )}|j||d t|dd}| }W d    n1 s5w   Y  W d    n1 sDw   Y  |dddd	d
diksVJ d S )NA)BECr  r  )r!   rj   r   Tr  r  r  r  )r   r   r   r   )r   rI   r:   r;   r<   r   r0  )r,   r9   r  rL   r>   Zdta_iterr0  r   r   r   test_value_labels_iterator  s   


z$TestStata.test_value_labels_iteratorc                 C   s^   t  }d|j_t  }|| t|dd}W d    n1 s"w   Y  t || d S )Nr%   	index_col)r:   r  r%   r&   r;   r<   r   r=   )r,   rL   r>   r  r   r   r   test_set_index  s   

zTestStata.test_set_indexcolumn)msr  weekr  ZqtrZhalfr}  c                 C   sB   t |dddd}|jd|f }|jd|d f }||ksJ d S )Nr    r!   r"   zstata13_dates.dtar   _fmt)r   r   )r,   r  r'   rL   Zunformattedr   r   r   r   (test_date_parsing_ignores_format_details  s   z2TestStata.test_date_parsing_ignores_format_detailsc                 C   sP  t dddddddtddddd dgd	d
dddddtdddddggg dd}t|d td|d< t|d tjd|d< t|d tjd|d< |d tj|d< t|d tj	d|d< d|j
_|j
tj|_
| }t +}|j|ddidgdd | |}tj|d|dd t|| W d    d S 1 sw   Y  d S )Nr   r   r   r   r   r   A  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaai  zstring-1zobject-1z
2003-12-26r   A  bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbr   )r   r   r   r   rK   rT   r   r   r   Zs2045Zsrtlforced_strlr   r   r   r   rK   rT   r%   r   r   r  r2   )r$   convert_strlr/   FrC   )r   rJ   r   r   r   r   r   rI   rK   rT   r%   r&   rH   r:   r;   r<   r.   r=   r   )r,   r   rH   r>   r   r   r   r   test_writer_117)  sj   .

"zTestStata.test_writer_117c              	   C   s   t g dg dgg dd}d|j_ttE t (}|j|ddgdd	 | |}|	d}|j
|_
tj||d
d W d    n1 sGw   Y  W d    d S W d    d S 1 s_w   Y  d S )N)A  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaar  r  )A  bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbr  r  )Z2long1long1long1long1long1long1long1long1long1long1longr   r6   r%   r  r   r2   )r  r/   FrC   )r   r%   r&   r:   r   r
   r;   r<   r.   r   r7   r=   )r,   r   r>   r  r   r   r   test_convert_strl_name_swapp  s    


"z%TestStata.test_convert_strl_name_swapc                 C   s   t dddddddt dddddddt dd	d
d	d
d	dg}tg dg d|d}t 3}d}tjt|d |j|ddid W d    n1 sNw   Y  W d    d S W d    d S 1 sfw   Y  d S )Nr  r`   ra   r  r  r  r  r  r_   r   r  r  r  r  z0convert_dates key must be a column or an integerr	  Z
wrong_namer   r#   )	r  r   r   r:   r;   r  r  r  r<   )r,   r  r   r>   r  r   r   r   test_invalid_date_conversion  s$   
"z&TestStata.test_invalid_date_conversionc              	   C   s   t  }t }d|j_t 6}|j||d |d t	|d}|
|  W d    n1 s4w   Y  t|dd}W d    n1 sIw   Y  t|| d S )Nr%   r  r   wbr  )r    BytesIOr:   r  r%   r&   r;   r<   seekopenwriter  r   r=   )r,   r/   biorL   r>   dtar  r   r   r   test_nonfile_writing  s   

zTestStata.test_nonfile_writingc              	   C   s   t  }d|j_t  A}t|d}|j|dd W d    n1 s%w   Y  t|d}t|dd}W d    n1 sAw   Y  W d    n1 sPw   Y  t 	|| d S )Nr%   r  r1   r  rbr  )
r:   r  r%   r&   r;   gzipGzipFiler<   r   r=   )r,   rL   r>   gzr  r   r   r   test_gzip_writing  s   
zTestStata.test_gzip_writingc                 C   sf   |  |dddd}g d}g dg dg dg d	g d
g dg dg}t||d}t|| d S )Nr    r!   r"   zstata16_118.dta)utf8latin1asciiZ	utf8_strlZ
ascii_strl)   ραηδαςu   PÄNDÄSpr.  r/  )   ƤĀńĐąŜ   Ör   r0  r   )   ᴘᴀᴎᴅᴀS   Ünr2  r4  )      r5  r  r5  r  ) r   r   r6  r   )r   r   rQ  r   rQ  )r   r   r6  r   r6  r6   )r.   r   r:   r=   )r,   r'   Z
unicode_dfr7   r  rM   r   r   r   test_unicode_dta_118  s   	zTestStata.test_unicode_dta_118c                 C   s   dddd ddg}t |}|jd|_t =}|j|ddd t|}|d	}t|| d |d
< |j|dd
gdd t|}|d	}t|| W d    d S 1 sYw   Y  d S )NA  stringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringstringr   )mixednumberr   rK   Fr2   r8   r   r8  )r9   r  r/   )	r   r9  rI   r:   r;   r<   r   fillnar=   )r,   outputr>   r  rM   r   r   r   test_mixed_string_strl  s    



"z TestStata.test_mixed_string_strlc              	   C   s   dddd ddg}t |}d |d< t /}tjtdd |j||d W d    n1 s0w   Y  W d    d S W d    d S 1 sHw   Y  d S )Nnoner   )r=  r9  r   z Column `none` cannot be exportedr	  r  )r   r:   r;   r  r  r  r<   )r,   r/   r;  r>   r   r   r   test_all_none_exception  s   
"z!TestStata.test_all_none_exceptionc              	   C   s   d}t |gdgd}t 6}d}d}tjt| d| d || W d    n1 s/w   Y  W d    d S W d    d S 1 sGw   Y  d S )Nu9   Here is one __�__ Another one __·__ Another one __½__invalidr6   z\'latin-1' codec can't encode character '\\ufffd' in position 14: ordinal not in range\(256\)zP'ascii' codec can't decode byte 0xef in position 14: ordinal not in range\(128\)|r	  )r   r:   r;   r  r  UnicodeEncodeErrorr<   )r,   r/   contentrL   r>   Zmsg1Zmsg2r   r   r   test_invalid_file_not_written  s   
	"z'TestStata.test_invalid_file_not_writtenc           
   	   C   s,  t dgd dgd gddgd}t x}|j|ddgd t|d	Q}| }d}|d
|v s3J |d|v s<J |dd dd }|ddd  D ]}|dd }||dd  }	t	||	d kslJ qPW d    n1 sww   Y  W d    d S W d    d S 1 sw   Y  d S )Npandasr   u   þâÑÐÅ§Zvar_strZvar_strlr6   r2   )r/   r  r&  zlatin-1zutf-8s   strlsr   r(  s   GSO       )
r   r:   r;   r<   r!  r  encodesplitfindrw   )
r,   r;  r>   r  rB  rM   ZgsosZgsori  sizer   r   r   test_strl_latin1  s(   
"zTestStata.test_strl_latin1c                 C   s   d}|dddd}t t }t|}t|dksJ |d jjd |ks'J W d    n1 s1w   Y  tdggd d	gd
}t || d S )Na  
One or more strings in the dta file could not be decoded using utf-8, and
so the fallback encoding of latin-1 is being used.  This can happen when a file
has been incorrectly encoded by Stata or some other software. You should verify
the string values returned are correct.r    r!   r"   zstata1_encoding_118.dta   r   u   Düsseldorfr   r6   )	r:   r   UnicodeWarningr   rw   messager   r   r=   )r,   r'   r  r>   rx   r   rM   r   r   r   test_encoding_latin1_118  s   z"TestStata.test_encoding_latin1_118c              
   C   s   t |ddddd}t|}W d    n1 sw   Y  |jdks&J |jd dks/J |jd	 d
ks8J |jd dksAJ |jd ttddddddksTJ d S )Nr    r!   r"   zstata1_119.dta.gzr&  )r   i  )r   rc   A  AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA)r   r_   gQ	@)r   rR  r   )r   r   r  r`   r  )r'  r!  r   shaper  rF   	Timestampr   )r,   r'   r)  rL   r   r   r   test_stata_119  s   
*zTestStata.test_stata_119)r3   r4   Nc                 C   sv  t jg ddd}tg dg dg dgg dd}||d	< d
ddddd}d}dddddi}|d tj|d< t s}t|||dg|d||d}|	  t
|}	|d d|d< |d |d dj |d< t||	 t|}
|
j|ksJ |
 |ksJ W d    n1 sw   Y  |j||dd t
|}t|| W d    d S 1 sw   Y  d S )N)r      β   ĉTr   )r   r   u   ᴬu   ᴀ relatively long ŝtring)r  r   u   ᴮr   )r   r   u   ᴰN)   ÅrS  rT  strlsr6   	   ᴐᴬᵀr  u	   ᵈᵉᵊu   ᴎტჄႲႳႴႶႺzLong Stringsr   )rU  rS  rT  rV  rW  u   ᴅaᵀa-labelrS  labelu   æøåu   ŋot valid latin-1r   r   r   rV  F)r   r  r   r9   r/   r0  rj   )r/   r9   )rF   r  r   rI   rJ   rK   r:   r;   r   
write_filer   r:  replacer   r  r=   r   r   r   r<   )r,   r/   r   r!   r   r   r0  r>   writerr   r  Zreread_to_statar   r   r   test_utf8_writer  sV   


"zTestStata.test_utf8_writerc              	   C   s   t tjdtjd}t &}tjtdd t	||dd W d    n1 s(w   Y  W d    n1 s7w   Y  t /}tjtdd t	||dd W d    n1 sZw   Y  W d    d S W d    d S 1 srw   Y  d S )	N)r   i  r   z"version must be either 118 or 119.r	  r2   r  zYou must use version 119r3   )
r   rJ   Zzerosr   r:   r;   r  r  r  r   )r,   rL   r>   r   r   r   test_writer_118_exceptionsP  s   

"z$TestStata.test_writer_118_exceptionsN)`__name__
__module____qualname__r.   r   r  markparametrizer?   rN   rW   r{   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r  r  r  r  r  r   r%  r&  r/  r1  r:  r<  rG  rO  rX  rZ  rj  ru  r  r  r  filterwarningsr  r  r  r  r  r  r  staticmethodr   r  r  r  r  r  r  r  r  r  r  r  r  r  r  r  rJ   infr   r  r
  r  r  r  r  r  r  r%  r*  r7  r<  r>  rC  rK  rO  ZslowrR  r]  r^  r   r   r   r   r)   :   s$   


C

,(



'


	

	"
;14

#

!$





	
G




1r)   r/   )i   l   o   q   r1   c                 C   sT   |ddd}t j|d}t j|d|  d}t|}t|}tj||dd d S )	Nr    r!   r"   stata-compat-118.dtazstata-compat-rP   Fr   )r  r>   r  r   r:   r=   )r/   r'   Z	data_baserefoldrM   Zold_dtar   r   r   test_backward_compatZ  s   rn  c              
   C   sv  | dddd}t |}| jrJ t|jtjrJ W d    n1 s&w   Y  t|d*}t |}| jr=J |j|u sDJ W d    n1 sNw   Y  W d    n1 s]w   Y  t|dJ}t| *}t |}| jr|J |j|u sJ W d    n1 sw   Y  W d    n1 sw   Y  W d    d S W d    d S 1 sw   Y  d S )Nr    r!   r"   rk  r&  )r   r  emptyrE   Z_path_or_bufr    r  r!  )r'   Zmonkeypatch	file_pathr  fpr#  r   r   r   test_direct_readd  s.   


"rr  c                 C   s   | dddd}t jtdd t|}|  W d    n1 s!w   Y  t jtdd |  W d    d S 1 s=w   Y  d S )Nr    r!   r"   rk  zwithout using a context managerr	  zis not part of the public API)r:   r   ResourceWarningr   r  FutureWarningclose)r'   rp  rU  r   r   r   0test_statareader_warns_when_used_without_context{  s   

"rv  r0   use_dictTFinferc              	   C   s  d}| r|r	| }nt |  }|d| 7 }| }|rd}|r d| i}ttjddtdd}d	|j_t	|}|j
|||d
 | dkrat|d}	t|	 }
W d    n1 s[w   Y  n| dkrt|d}	t|	|	jd }
W d    n1 sw   Y  n| dkrt|}t|| d  }
W d    n1 sw   Y  ny| dkrt|d}	t|	 }
W d    n1 sw   Y  nW| dkrtd}||d}	t|	 }
W d    n1 sw   Y  n0| dkr!td}||d}	t|	 }
W d    n	1 sw   Y  n| d u r(|}
t|
d	d}W d    n	1 s9w   Y  | }|jtj|_t|| d S )Nzdta_inferred_compression.dtar]  rx  methodrb   r   r   r6   r%   )r/   compressionr'  r&  rE  rr   tarbz2zstdZ	zstandardxzlzmar  )r   r   rJ   r   r   r   r%   r&   r:   r;   r<   r'  r!  r    r  r  zipfileZipFilefilelisttarfileextractfilegetnamesr}  r  Zimportorskipr   rH   rI   rK   r=   )rz  r/   rw  rx  	file_namefile_extZcompression_argrL   r>   comprq  r|  r~  r  r  rM   r   r   r   test_compression  sl   



r  ry  rE  r  )Nr$  rE  c              	   C   s$  d| }d}t tjddtdd}d|j_t|Y}| |d}|j	||d	 | d
ks2|d
krgt
|d&}t|jdksBJ |jd j|ksLJ t||jd }W d    n1 saw   Y  n|}t|dd}	W d    n1 syw   Y  | }
|
jtj|
_t|	|
 d S )Ntest.ztest.dtarb   r   r   r6   r%   )ry  archive_namerz  rE  r{  r   r   r  )r   rJ   r   r   r   r%   r&   r:   r;   r<   r  r  rw   r  filenamer    r  r  r   rH   rI   rK   r=   )ry  r  r  r  rL   r>   rz  Zzprq  r  rM   r   r   r   test_compression_dict  s(   

r  c                 C   s   t dtg dddi}d|j_| }|jtj|_t	 V}|j
|| d t|ddd	/}t|D ]"\}}|d}d|v sCJ t|j|jjd| d|d
    q4W d    n1 saw   Y  W d    d S W d    d S 1 syw   Y  d S )Ncats)r   r   r   r   r   rj   r   r%   r  r   F)r  r  r   )r   r   r%   r&   rH   rI   rJ   rK   r:   r;   r<   r   	enumerater   r  r  r  )r/   rL   rM   r>   r  ra  blockr   r   r   test_chunked_categorical  s$   

"r  c           	   	   C   sT  | dddd}g d}t |ddV}ttA t|D ]4\}}t|j|d| d|d   ks3J |dk r?td	d
g}ntjdgdd}t	|jj
j| qW d    n1 s[w   Y  W d    n1 sjw   Y  tt" t |dd}| }W d    n1 sw   Y  W d    n1 sw   Y  t|}t|| d S )Nr    r!   r"   stata-dta-partially-labeled.dta)r   r   r   r   r   r   r  r   r   r   r   r   r   r@   )r   r:   r   r	   r  r   r  rF   r8  r  r   r   __next__r   r=   )	r'   dta_filer  r  ra  r  idxZlarge_chunkr  r   r   r    test_chunked_categorical_partial  s,   &	
r  r  )rR  r   r  c              	   C   s~   | dddd}t jtdd' t||d W d    n1 s w   Y  W d    d S W d    d S 1 s8w   Y  d S )Nr    r!   r"   r  zchunksize must be a positiver	  r  )r  r  r  r   )r'   r  r  r   r   r   test_iterator_errors  s   "r  c               
      s  ddgdgd   t  fddtdD } t b}| j|dd	 tjg d
dd}t|dd3}t|D ]&\}}tdD ]}t	|j
| j| q>t|| j|d |d d   q6W d    n1 sgw   Y  W d    d S W d    d S 1 sw   Y  d S )Nc_labelb_labela_labelrY  c                    s"   i | ]}d | t j ddqS )r   Tr   )rF   r  )rl   rM  r  r   r   r5    s   " z.test_iterator_value_labels.<locals>.<dictcomp>r   Fr   )r  r  r  r   r   r@  r  r   )r   rD   r:   r;   r<   rF   r8  r   r  r  dtypesr   r=   r  )rL   r>   rM   r  rn  r  ra  r   r  r   test_iterator_value_labels  s   
$"r  c               	   C   s  t tdd tdD tdd tdD ggddgd} t X}tjtdd	 | j|d
d W d    n1 s;w   Y  t|}t	t
jt
jgddgd}t|j| |jd | jd kscJ |jd t| jd ksqJ W d    d S 1 s|w   Y  d S )Nc                 s   s    | ]}d | V  qdS )r   Nr   r`  r   r   r   	<genexpr>  s    z&test_precision_loss.<locals>.<genexpr><   4   biglittler6   z&Column converted from int64 to float64r	  Fr   )r%   )r   r  )r   r  )r   sumrD   r:   r;   r   r   r<   r   r   rJ   r   r  r  r   float)rL   r>   r  Zexpected_dtr   r   r   test_precision_loss  s    ,
"r  c              	   C   s   t g dg dgddgg dd}d|j_t F}|j|| d t|| dd	}t|| t|| }t	
| }W d    n1 sFw   Y  t|dd
}t|| W d    d S 1 sbw   Y  d S )Ng~Ϛ?g9DܜJ?g>D)?g.4i(@g33333@g̤Ar  r  XYZr7  r%   r  rz  r  r  )r   r%   r&   r:   r;   r<   r   r=   Zdecompress_filer    r  r  )rz  rL   r>   r  fhcontentsr   r   r   test_compression_roundtrip!  s    
"r  to_infer
read_inferc                 C   s   | }t | }d| }tg dg dgddgg dd}d|j_|r$d	n|}|r*d	n|}t|}	|j|	|d
 t|	|dd}
t|
| W d    d S 1 sQw   Y  d S )Nr  r  r  r  r  r  r7  r%   rx  r  r  )	r   r   r%   r&   r:   r;   r<   r   r=   )Zcompression_onlyr  r  rz  extr  rL   Zto_compressionZread_compressionr>   r   r   r   r   test_stata_compression5  s    
"r  c               	   C   s  t g dddtjdtjgg dtg dd} t }dd	d
ddd	dd}i |dddddi}t|| |d}|  t	|}|
 }||ksPJ W d    n1 sZw   Y  d}tjt|d ddddi}t|| |d W d    n1 sw   Y  d}tjt|d ddddddi}t|| |d W d    n1 sw   Y  W d    d S W d    d S 1 sw   Y  d S )N)r   r   r   r   r   r   r  g      "@)r_   r_   r   r^   rb   )rn  rM  r\  rM  rn  )fully_labelledpartially_labelledr  r  r   r   r   rY  )r   r  )r  r  r  rn  rM  r\  )r   r   r   r0  zCCan't create value labels for notY, it wasn't found in the dataset.r	  ZnotYrH  rI  )r_   r^   zUCan't create value labels for Z, value labels can only be applied to numeric columns.r   ra  r   )r   rJ   rS   rF   r  r:   r;   r   rZ  r   r0  r  r  r  r  )r!   r>   r0  rM   r\  r  reader_value_labelsr  r   r   r   !test_non_categorical_value_labelsM  s@   
	

"r  c               	   C   s.  t g dg dg dg dg dd} ddddddd	d
dddiddid}ddddddd	d
dddiddid}t K}tt | j||d W d    n1 sZw   Y  t|}| }||ksnJ W d    n1 sxw   Y  W d    d S W d    d S 1 sw   Y  d S )N)r   r   r   r   r@   r^   )r^   r^   r   r   r^   r^   )r   r@   r@   rc   rc   r   r   )z	invalid~!Z	6_invalidZ&invalid_name_longer_than_32_charactersZ	aggregater   r   rH  rI  r  r   r   )r^   r   r@   r   r   r   )Z	invalid__Z
_6_invalidZ invalid_name_longer_than_32_charZ
_aggregateZ_1__2_r  )r   r:   r;   r   r
   r<   r   r0  )r!   r0  rM   r>   r  r  r   r   r   0test_non_categorical_value_label_name_conversions  s>   	

"r  c               	   C   s  dddddi} t dg di}t h}|j|| d t|dd}| }W d    n1 s1w   Y  || ks<J d}d	d
dg }d| d| d
}tjt	|d t
|dd W d    n1 sgw   Y  W d    d S W d    d S 1 sw   Y  d S )NZrepeated_labelsZTenzMore than ten)rb   r]   (   )rb   rb   r]   r]   r  r  r  Fr  zQ--------------------------------------------------------------------------------

z
Value labels for column a    are not unique. These cannot be converted to
pandas categoricals.

Either read the file with `convert_categoricals` set to False or use the
low level interface in `StataReader` to separately read the values and the
value_labels.

The repeated labels are:
r	  T)r   r:   r;   r<   r   r0  r  r  r  r  r   )r0  r!   r>   r  r  r   Zrepeatsr  r   r   r   ;test_non_categorical_value_label_convert_categoricals_error  s2   

	"r  r   c           
      C   s  t tg dtdtjtjg| jdtg dd}|jjjj}|dd}|dkr-d	}n|d
kr3d}t	j
| }t	|}td||gtdd}tg ddd}t -}|j|d|d t|dd}	t|j|	j t|	j| t|	j| W d    d S 1 sw   Y  d S )Nr  r   r   )r   r   Nr   ur   r   rK   boolr   r   r6  )r   r   r   r   r  Fr8   Trp  )r   r   rF   ZNAr&   r   r   Znumpy_dtyper[  r   ZBASE_MISSING_VALUESr   r:   r;   r<   r   r  r   r   )
r   r/   rL   Z
dtype_namevaluers  Z
expected_bZ
expected_cr>   r  r   r   r   test_nullable_support  s.   



"r  )Dr}  r   r  r'  r    r  re  r  rt   r  numpyrJ   r  Zpandas.core.dtypes.commonr   rD  rF   Zpandas._testingZ_testingr:   Zpandas.core.framer   r   Z pandas.tests.io.test_compressionr   Zpandas.io.parsersr   Zpandas.io.statar	   r
   r   r   r   r   r   r   r   Zfixturer   r(   r)   rb  rc  rn  rr  rv  r  r  r  r  r  r  r  r  r  r  r  r  ZBooleanDtypeZ	Int8DtypeZ
Int16DtypeZ
Int32DtypeZ
Int64DtypeZ
UInt8DtypeZUInt16DtypeZUInt32DtypeZUInt64Dtyper  r   r   r   r   <module>   s    ,


              .
	/

&%&