File: //lib/python3/dist-packages/bs4/__pycache__/diagnose.cpython-38.pyc
U
t�^) � @ s d Z dZddlZddlmZ ddlmZ ddlZddlmZm Z ddl
m Z ddl Z ddl
Z
ddlZddlZddlZddlZddlZddlZdd � Zd#d d �ZG d
d� de�Zdd� ZdZdZd$dd�Zd%dd�Zd&dd�Zd'd d �Z d(d d!�Z e d"k�reej � � � dS ))z=Diagnostic functions, mainly for use when doing tech support.ZMIT� N)�StringIO)�
HTMLParser)�
BeautifulSoup� __version__)�builder_registryc C sJ t dt � t dtj � dddg}|D ]4}tjD ]}||jkr2 q(q2|�|� t d| � q(d|kr�|�d� z*dd l m
} t d
d � t t
|j�� � W n* tk
r� } z t d � W 5 d
}~X Y nX d|k�rzdd
l}t d|j � W n, tk
�r } z t d� W 5 d
}~X Y nX t| d��r.| �� } n�| �d��sF| �d��r^t d| � t d� d
S z:tj�| ��r�t d| � t| ��}|�� } W 5 Q R X W n tk
�r� Y nX t � |D ]�}t d| � d} zt| |d�}
d} W n8 tk
�r } zt d| � t� � W 5 d
}~X Y nX | �r:t d| � t |
� � � t d � �q�d
S ) z�Diagnostic suite for isolating common problems.
:param data: A string containing markup that needs to be explained.
:return: None; diagnostics are printed to standard output.
z'Diagnostic running on Beautiful Soup %szPython version %s� html.parser�html5lib�lxmlz;I noticed that %s is not installed. Installing it may help.zlxml-xmlr ��etreezFound lxml version %s�.z.lxml is not installed or couldn't be imported.NzFound html5lib version %sz2html5lib is not installed or couldn't be imported.�readzhttp:zhttps:z<"%s" looks like a URL. Beautiful Soup is not an HTTP client.zpYou need to use some other library to get the document behind the URL, and feed that document to Beautiful Soup.z7"%s" looks like a filename. Reading data from the file.z#Trying to parse your markup with %sF)�featuresT� %s could not parse the markup.z#Here's what %s did with the markup:zP--------------------------------------------------------------------------------) �printr �sys�versionr Zbuildersr �remove�appendr r �join�map�strZ LXML_VERSION� ImportErrorr �hasattrr
�
startswith�os�path�exists�open�
ValueErrorr � Exception� traceback� print_excZprettify) �dataZ
basic_parsers�nameZbuilderr �er �fp�parser�success�soup� r* �./usr/lib/python3/dist-packages/bs4/diagnose.py�diagnose sr
��
�
�
r, Tc K sJ ddl m} |jt| �fd|i|��D ] \}}td||j|jf � q&dS )a� Print out the lxml events that occur during parsing.
This lets you see how lxml parses a document when no Beautiful
Soup code is running. You can use this to determine whether
an lxml-specific problem is in Beautiful Soup's lxml tree builders
or in lxml itself.
:param data: Some markup.
:param html: If True, markup will be parsed with lxml's HTML parser.
if False, lxml's XML parser will be used.
r r
�htmlz %s, %4s, %sN)r r Z iterparser r �tag�text)r# r- �kwargsr Zevent�elementr* r* r+ �
lxml_trace] s "r2 c @ s` e Zd ZdZdd� Zdd� Zdd� Zdd � Zd
d � Zd d
� Z dd� Z
dd� Z dd� Z dd� Z
dS )�AnnouncingParserz�Subclass of HTMLParser that announces parse events, without doing
anything else.
You can use this to get a picture of how html.parser sees a given
document. The easiest way to do this is to call `htmlparser_trace`.
c C s t |� d S )N)r )�self�sr* r* r+ �_pu s zAnnouncingParser._pc C s | � d| � d S )Nz%s START�r6 )r4 r$ Zattrsr* r* r+ �handle_starttagx s z AnnouncingParser.handle_starttagc C s | � d| � d S )Nz%s ENDr7 �r4 r$ r* r* r+ �
handle_endtag{ s z AnnouncingParser.handle_endtagc C s | � d| � d S )Nz%s DATAr7 �r4 r# r* r* r+ � handle_data~ s z AnnouncingParser.handle_datac C s | � d| � d S )Nz
%s CHARREFr7 r9 r* r* r+ �handle_charref� s z AnnouncingParser.handle_charrefc C s | � d| � d S )Nz %s ENTITYREFr7 r9 r* r* r+ �handle_entityref� s z!AnnouncingParser.handle_entityrefc C s | � d| � d S )Nz
%s COMMENTr7 r; r* r* r+ �handle_comment� s z AnnouncingParser.handle_commentc C s | � d| � d S )Nz%s DECLr7 r; r* r* r+ � handle_decl� s z AnnouncingParser.handle_declc C s | � d| � d S )Nz%s UNKNOWN-DECLr7 r; r* r* r+ � unknown_decl� s z AnnouncingParser.unknown_declc C s | � d| � d S )Nz%s PIr7 r; r* r* r+ � handle_pi� s zAnnouncingParser.handle_piN)�__name__�
__module__� __qualname__�__doc__r6 r8 r: r< r= r> r? r@ rA rB r* r* r* r+ r3 m s r3 c C s t � }|�| � dS )z�Print out the HTMLParser events that occur during parsing.
This lets you see how HTMLParser parses a document when no
Beautiful Soup code is running.
:param data: Some markup.
N)r3 Zfeed)r# r' r* r* r+ �htmlparser_trace� s rG ZaeiouZbcdfghjklmnpqrstvwxyz� c C s: d}t | �D ](}|d dkr"t}nt}|t�|�7 }q |S )z#Generate a random word-like string.� � r )�range� _consonants�_vowels�random�choice)�lengthr5 �i�tr* r* r+ �rword� s rS � c C s d� dd� t| �D ��S )z'Generate a random sentence-like string.� c s s | ]}t t�d d��V qdS )rT � N)rS rN �randint)�.0rQ r* r* r+ � <genexpr>� s z rsentence.<locals>.<genexpr>)r rK )rP r* r* r+ � rsentence� s rZ �� c C s� dddddddg}g }t | �D ]r}t�dd �}|dkrPt�|�}|�d
| � q |d krp|�tt�d d ��� q |d
kr t�|�}|�d| � q dd�|� d S )z+Randomly generate an invalid HTML document.�pZdiv�spanrQ �bZscript�tabler � z<%s>� rT rJ z</%s>z<html>�
z</html>)rK rN rW rO r rZ r )� num_elementsZ tag_names�elementsrQ rO Ztag_namer* r* r+ �rdoc� s
re 順 c
C s$ t dt � t| �}t dt|� � dddgddfD ]z}d}z"t�� }t||�}t�� }d}W n6 tk
r� } zt d | � t�� W 5 d
}~X Y nX |r4t d ||| f � q4d d
l m
} t�� }|� |� t�� }t d|| � d d
l } | �
� }t�� }|�|� t�� }t d|| � d
S )z.Very basic head-to-head performance benchmark.z1Comparative parser benchmark on Beautiful Soup %sz3Generated a large invalid HTML document (%d bytes).r r- r r FTr Nz"BS4+%s parsed the markup in %.2fs.r r
z$Raw lxml parsed the markup in %.2fs.z(Raw html5lib parsed the markup in %.2fs.)r r re �len�timer r r! r" r r ZHTMLr r �parse)
rc r# r' r( �ar) r^ r% r r r* r* r+ �benchmark_parsers� s4
rk r c C sX t �� }|j}t| �}tt||d�}t�d|||� t� |�}|�
d� |� dd� dS )z7Use Python's profiler on a randomly generated document.)�bs4r# r' z bs4.BeautifulSoup(data, parser)Z
cumulativez
_html5lib|bs4�2 N) �tempfileZNamedTemporaryFiler$ re �dictrl �cProfileZrunctx�pstatsZStatsZ
sort_statsZ print_stats)rc r' Z
filehandle�filenamer# �varsZstatsr* r* r+ �profile� s
rt �__main__)T)rH )rT )r[ )rf )rf r )!rF Z __license__rp �ior Z html.parserr rl r r Z bs4.builderr r rq rN rn rh r! r r, r2 r3 rG rM rL rS rZ re rk rt rC �stdinr
r* r* r* r+ �<module> s8 G
&