U
    \ShV                     @  s  U d 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	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mZmZmZmZmZ ddlmZ ddlmZm Z  dd	l!m"Z"m#Z# dd
l$m%Z% ddl&m'Z' G dd dZ(ddl)m*Z* dddddZ+dde*ddddddddddd d!d"d#d$Z,d%e-d&< G d'd( d(Z.e. Z/d)d*d+d,d-d.Z0ej1G d/d0 d0Z2G d1d2 d2ej3Z4G d3d4 d4Z5dS )5z3Create a full-text search index for offline search.    )annotationsN)import_module)path)IOAnyCallableDict	GeneratorIterableIteratorListOptionalSequenceSetTupleTypeUnion)nodes)ElementNode)addnodespackage_dir)BuildEnvironment)split_index_msgc                   @  s   e Zd ZU dZdZded< dZded< e Zded< dZ	d	ed
< dZ
ded< dZedZdddddZdddddZd	ddddZd	d	dddZd	ddddZdS )SearchLanguagea<  
    This class is the base class for search natural language preprocessors.  If
    you want to add support for a new language, you should override the methods
    of this class.

    You should override `lang` class property too (e.g. 'en', 'fr' and so on).

    .. attribute:: stopwords

       This is a set of stop words of the target language.  Default `stopwords`
       is empty.  This word is used for building index and embedded in JS.

    .. attribute:: js_splitter_code

       Return splitter function of JavaScript version.  The function should be
       named as ``splitQuery``.  And it should take a string and return list of
       strings.

       .. versionadded:: 3.0

    .. attribute:: js_stemmer_code

       Return stemmer class of JavaScript version.  This class' name should be
       ``Stemmer`` and this class must have ``stemWord`` method.  This string is
       embedded as-is in searchtools.js.

       This class is used to preprocess search word which Sphinx HTML readers
       type, before searching index. Default implementation does nothing.
    N
str | Nonelanglanguage_nameset[str]	stopwords strjs_splitter_codejs_stemmer_rawcodez
/**
 * Dummy stemmer for languages without stemming rules.
 */
var Stemmer = function() {
  this.stemWord = function(w) {
    return w;
  }
}
z\w+dictNone)optionsreturnc                 C  s   || _ | | d S N)r&   initselfr&    r,   I/root/rtd-docs/venv/lib/python3.8/site-packages/sphinx/search/__init__.py__init__U   s    zSearchLanguage.__init__c                 C  s   dS )zK
        Initialize the class with the options the user has given.
        Nr,   r*   r,   r,   r-   r)   Y   s    zSearchLanguage.init	list[str])inputr'   c                 C  s   | j |S )z
        This method splits a sentence into words.  Default splitter splits input
        at white spaces, which should be enough for most languages except CJK
        languages.
        )_word_refindall)r+   r0   r,   r,   r-   split^   s    zSearchLanguage.split)wordr'   c                 C  s   |S )a  
        This method implements stemming algorithm of the Python version.

        Default implementation does nothing.  You should implement this if the
        language has any stemming rules.

        This class is used to preprocess search words before registering them in
        the search index.  The stemming of the Python version and the JS version
        (given in the js_stemmer_code attribute) must be compatible.
        r,   r+   r4   r,   r,   r-   stemf   s    zSearchLanguage.stemboolc                 C  sT   t |dkpRt |dk r8dt|d   k o2dk n  pPt|d dk oP|| jk S )z
        Return true if the target word should be registered in the search index.
        This method is called after stemming.
        r      iA0  i0     )lenordr   r5   r,   r,   r-   word_filters   s    ,zSearchLanguage.word_filter)__name__
__module____qualname____doc__r   __annotations__r   setr   r"   r#   js_stemmer_coderecompiler1   r.   r)   r3   r6   r<   r,   r,   r,   r-   r   %   s   

r   )SearchEnglishr!   r   )sourcer'   c                 C  s4   t  }|  D ] }|dd }||  q|S )zs
    Parse snowball style word list like this:

    * http://snowball.tartarus.org/algorithms/finnish/stop.txt
    |r   )rB   
splitlinesr3   update)rG   resultliner,   r,   r-   parse_stop_word   s
    rM   zsphinx.search.da.SearchDanishzsphinx.search.de.SearchGermanzsphinx.search.es.SearchSpanishzsphinx.search.fi.SearchFinnishzsphinx.search.fr.SearchFrenchz sphinx.search.hu.SearchHungarianzsphinx.search.it.SearchItalianzsphinx.search.ja.SearchJapanesezsphinx.search.nl.SearchDutchz sphinx.search.no.SearchNorwegianz!sphinx.search.pt.SearchPortuguesezsphinx.search.ro.SearchRomanianzsphinx.search.ru.SearchRussianzsphinx.search.sv.SearchSwedishzsphinx.search.tr.SearchTurkishzsphinx.search.zh.SearchChinese)dadeenesfifrhuitjanlnoptrorusvtrzhz%dict[str, str | type[SearchLanguage]]	languagesc                   @  sZ   e Zd ZdZdZdZdddddZddd	d
dZddddddZdddddZ	dS )_JavaScriptIndexz
    The search index as JavaScript file that calls a function
    on the documentation search object to register the index.
    zSearch.setIndex()r   r!   )datar'   c                 C  s   | j t| | j S r(   )PREFIXjsondumpsSUFFIX)r+   rb   r,   r,   r-   re      s    z_JavaScriptIndex.dumps)sr'   c                 C  sH   |t | jt | j  }|r6|| jr6|| js>tdt|S )Nzinvalid data)r:   rc   rf   
startswithendswith
ValueErrorrd   loads)r+   rg   rb   r,   r,   r-   rk      s    
z_JavaScriptIndex.loadsr   r%   )rb   fr'   c                 C  s   | | | d S r(   )writere   )r+   rb   rl   r,   r,   r-   dump   s    z_JavaScriptIndex.dump)rl   r'   c                 C  s   |  | S r(   )rk   read)r+   rl   r,   r,   r-   load   s    z_JavaScriptIndex.loadN)
r=   r>   r?   r@   rc   rf   re   rk   rn   rp   r,   r,   r,   r-   r`      s   r`   z
nodes.metar   r7   )noder   r'   c                 C  s4   |  ddkr0|  d}|d kr$dS ||kr0dS dS )Nnamekeywordsr   TF)get)rq   r   Z	meta_langr,   r,   r-   _is_meta_keywords   s    
ru   c                   @  sJ   e Zd ZU ejedZded< ejedZded< ejedZ	ded< dS )	WordStore)default_factoryr/   wordszlist[tuple[str, str]]titlestitle_wordsN)
r=   r>   r?   dataclassesfieldlistrx   rA   ry   rz   r,   r,   r,   r-   rv      s   
rv   c                      s:   e Zd ZdZdddd fddZddd	d
dZ  ZS )WordCollectorzG
    A special visitor that collects words for the `IndexBuilder`.
    nodes.documentr   r%   )documentr   r'   c                   s(   t  | g | _g | _g | _|| _d S r(   )superr.   found_wordsfound_titlesfound_title_wordsr   )r+   r   r   	__class__r,   r-   r.      s
    zWordCollector.__init__r   )rq   r'   c                 C  sX  t |tjrtjn>t |tjrd|dd krtjdd|	 tj
tjB d}tjdd|tj
tjB d}tdd|}| j| j| tjnt |tjr| j| j|	  nt |tjr|	 }|jd }| j||r|d	 nd f | j| j| nDt |trTt|| jjrT|d
 }dd |dD }| j| d S )Nhtmlformatr    <style.*?</style>flags<script.*?</script><[^<]+?>idsr   contentc                 S  s   g | ]}|  qS r,   strip.0keywordr,   r,   r-   
<listcomp>   s     z0WordCollector.dispatch_visit.<locals>.<listcomp>,)
isinstancer   commentZSkipNoderawrt   r3   rD   subastext
IGNORECASEDOTALLr   extendr   Texttitleparentr   appendr   r   ru   )r+   rq   nodetextr   r   rs   r,   r,   r-   dispatch_visit   s(    

zWordCollector.dispatch_visit)r=   r>   r?   r@   r.   r   __classcell__r,   r,   r   r-   r~      s   r~   c                   @  s   e Zd ZdZeedZdddddddd	Zd
dddddZd
dddddZ	dddddZ
dddddZddddZddddZddd d!d"Zdddd#dd$d%d&Zd#d'd(d)d*Zddd+d,Zd-dd.d/Zd0dd1d2Zddd3d4Zd5S )6IndexBuilderzi
    Helper class that creates a search index based on the doctrees
    passed to the `feed` method.
    )rd   pickler   r!   r$   r%   )envr   r&   scoringr'   c           	   	   C  s   || _ |j| _|j| _|j| _|j| _|j	| _
|j| _|j| _|j| _t|}|d krtd|krtt|dd }|d krt|| _n>t|tr|dd\}}tt||}||| _n
||| _|rt|d}|  | _W 5 Q R X nd| _d| _d S )N_r   .   rbr    ) r   Z_search_index_titles_titlesZ_search_index_filenames
_filenamesZ_search_index_mapping_mappingZ_search_index_title_mapping_title_mappingZ_search_index_all_titles_all_titlesZ_search_index_index_entries_index_entriesZ_search_index_objtypes	_objtypesZ_search_index_objnames	_objnamesr_   rt   r3   rF   r   r   r!   rsplitgetattrr   openro   decodejs_scorer_coder"   )	r+   r   r   r&   r   Z
lang_classmodule	classnamefpr,   r,   r-   r.   	  s0    


zIndexBuilder.__init__r   r   )streamr   r'   c           
        s  t |tr| j| }||}t |tr:|d| jjkrBtd|d  tt	 |d | _
tt	 |d | _i | _| j D ]}g | j|< q|d  D ].\}}|D ] \}}| j |  ||f qqddd	 fd
d}	|	|d | _|	|d | _dS )zReconstruct from frozen data.
envversionz
old formatdocnames	filenamesry   	alltitlesdict[str, Any]zdict[str, set[str]])mappingr'   c                   sJ   i }|   D ]8\}}t|tr. | h||< q fdd|D ||< q|S )Nc                   s   h | ]} | qS r,   r,   )r   iZindex2fnr,   r-   	<setcomp>M  s     z8IndexBuilder.load.<locals>.load_terms.<locals>.<setcomp>)itemsr   int)r   rvkvr   r,   r-   
load_termsG  s    
z%IndexBuilder.load.<locals>.load_termsterms
titletermsN)r   r!   formatsrp   r$   rt   r   versionrj   zipr   r   r   keysr   r   r   r   )
r+   r   r   frozendocnamer   Z
doc_tuplesdoctitleidr   r,   r   r-   rp   3  s&    



	zIndexBuilder.loadc                 C  s(   t |tr| j| }||  | dS )z"Dump the frozen index to a stream.N)r   r!   r   rn   freeze)r+   r   r   r,   r,   r-   rn   T  s    

zIndexBuilder.dumpzdict[str, int]z/dict[str, list[tuple[int, int, int, str, str]]])fn2indexr'   c                 C  sJ  i }| j }| j}t| jj D ]"\}}t| D ]
\}}}	}
}}|
|krRq6|dk r\q6t|}t|}|	d\}}}|
|g }z|||	f }W n` tk
r   t|}||||	f< |j|	}|r||	t||f||< n||	|	f||< Y nX ||krd}n||	d | kr&d}n|}|||
 ||||f q6q |S )Nr   r   r    -)r   r   sortedr   domainsr   get_objectsr   escape
rpartition
setdefaultKeyErrorr:   Zobject_typesrt   r!   Zget_type_namer   )r+   r   r   ZotypesZonamesZ
domainnamedomainfullnameZdispnametyper   anchorprioprefixr   rr   plistZ	typeindexotypeZshortanchorr,   r,   r-   r   Z  s@    




zIndexBuilder.get_objectsz1tuple[dict[str, list[str]], dict[str, list[str]]]c                   s|   i i f}t || j| jfD ]\\}}| D ]J\}}t|dkrZ|\}| krt | ||< q*t fdd|D ||< q*q|S )Nr   c                   s   g | ]}| kr | qS r,   r,   )r   fnr   r,   r-   r     s      z*IndexBuilder.get_terms.<locals>.<listcomp>)r   r   r   r   r:   r   )r+   r   Zrvsr   r   r   r   r   r,   r   r-   	get_terms  s    zIndexBuilder.get_termsr   )r'   c                   s  t t j  \}} fdd|D }dd t|D } |\}} |}dd  j D } j}	i }
 j	 D ]0\}}|D ]"\}}|

|g || |f qqzi } j D ]6\}}|D ](\}}}|
| g || |f qqt|||||||	| jj|
|dS )z/Create a usable data structure for serializing.c                   s   g | ]} j |qS r,   )r   rt   )r   r   r+   r,   r-   r     s     z'IndexBuilder.freeze.<locals>.<listcomp>c                 S  s   i | ]\}}||qS r,   r,   )r   r   rl   r,   r,   r-   
<dictcomp>  s      z'IndexBuilder.freeze.<locals>.<dictcomp>c                 S  s&   i | ]\}}||d  d |d  qS )r   :r   r,   )r   r   r   r,   r,   r-   r     s      )r   r   ry   r   objectsobjtypesobjnamesr   r   r   Zindexentries)r   r   r   r   	enumerater   r   r   r   r   r   r   r   lowerr$   r   r   )r+   r   ry   r   r   r   Ztitle_termsr   r   r   r   r   Z	titlelistr   r   Zindex_entriesentriesentryZentry_idZ
main_entryr,   r   r-   r     s0    
"
    zIndexBuilder.freezec                 C  s   | j j d| j j  dS )Nz (code: ra   )r   r   r   r,   r,   r-   label  s    zIndexBuilder.labelzIterable[str])r   r'   c                 C  s   i }i }i }|D ]8}|| j kr| j | ||< | j| ||< | j| ||< q|| _ || _|| _| j D ]}|| qf| j D ]}|| qdS )z-Remove data for all docnames not in the list.N)r   r   r   r   valuesintersection_updater   )r+   r   Z
new_titlesZnew_alltitlesZnew_filenamesr   Z	wordnamesr,   r,   r-   prune  s    
zIndexBuilder.pruner   )r   filenamer   doctreer'   c              
     s  || j |< || j|< | |}| jj}| jj tjddddd fdd}|j| j	|< |j
D ]J}||}	||	r| j|	t | q`||r`| j|t | q`|jD ]T}||}	||	s||r|}	|| j|	dk}
||	r|
s| j|	t | qt }|tjD ]}|d D ]v\}}}zt||}W n tk
rX   Y nBX pbd	|d
kr||d f |fdd|D O }q&qt|| j|< dS )zFeed a doctree to the index.N)maxsizer!   )word_to_stemr'   c                   s    |   S r(   )r   )r   )_stemr,   r-   r6     s    zIndexBuilder.feed.<locals>.stemr,   r   r    >   seeseealsor   c                   s   h | ]}| fqS r,   r,   )r   x)main	target_idr,   r-   r     s     z$IndexBuilder.feed.<locals>.<setcomp>)r   r   _word_collectorr   r<   r6   	functools	lru_cachery   r   rz   r   r   rB   addrx   rt   r   r2   r   indexr   rj   r   r   )r+   r   r   r   r   
word_storeZ_filterr6   r4   Zstemmed_wordZalready_indexedr   rq   Z
entry_typevalueZ_category_keyrK   r,   )r   r  r  r-   feed  sB    







 zIndexBuilder.feedrv   )r   r'   c                   s4    fdd t  | jj| jj | S )Nc                   sP  t | tjrd S t | tjrd| dd krtjdd|  tj	tj
B d}tjdd|tj	tj
B d}tdd|}j| d S t | tjrt| rdd	 | d
 dD }j| npt | tjrj|   nNt | tjr6|  }| jd }j||r|d nd f j| | jD ]} | q<d S )Nr   r   r    r   r   r   r   c                 S  s   g | ]}|  qS r,   r   r   r,   r,   r-   r      s     zFIndexBuilder._word_collector.<locals>._visit_nodes.<locals>.<listcomp>r   r   r   r   )r   r   r   r   rt   r3   rD   r   r   r   r   rx   r   metaru   r   r   r   ry   r   rz   children)rq   r   rs   r   r   child_visit_nodeslanguager3   r
  r,   r-   r    s8    




z2IndexBuilder._word_collector.<locals>._visit_nodes)rv   r   r3   )r+   r   r,   r  r-   r    s    zIndexBuilder._word_collectorc                 C  s:   | j jr| j j}n| j}|  tt| j j| j|dS )N)Zsearch_language_stemming_codeZsearch_language_stop_wordsZsearch_scorer_toolZsearch_word_splitter_code)r   r"   get_js_stemmer_coderd   re   r   r   r   )r+   r"   r,   r,   r-   context_for_searchtool  s    
z#IndexBuilder.context_for_searchtoolr/   c                 C  s&   | j jrdd d| j jfD S g S dS )z8Returns a list of non-minified stemmer JS files to copy.c                 S  s   g | ]}t td d|qS )searchznon-minified-js)r   joinr   )r   fnamer,   r,   r-   r   #  s   z8IndexBuilder.get_js_stemmer_rawcodes.<locals>.<listcomp>base-stemmer.jsN)r   r#   r   r,   r,   r-   get_js_stemmer_rawcodes   s
    
z$IndexBuilder.get_js_stemmer_rawcodesr   c                 C  s   d S r(   r,   r   r,   r,   r-   get_js_stemmer_rawcode*  s    z#IndexBuilder.get_js_stemmer_rawcodec              	   C  s   | j jr|ttdd}tt|ddd}| }W 5 Q R X tt|| j jdd}| }W 5 Q R X d||| j jf S | j jS dS )z<Returns JS code that will be inserted into language_data.js.r  zminified-jsr  zutf-8)encodingz%s
%s
Stemmer = %sStemmer;N)	r   r#   r   r  r   r   ro   r   rC   )r+   Zjs_dirZjs_fileZbase_jsZlanguage_jsr,   r,   r-   r  -  s    z IndexBuilder.get_js_stemmer_codeN)r=   r>   r?   r@   rd   r   r   r.   rp   rn   r   r   r   r   r   r  r  r  r  r  r  r,   r,   r,   r-   r      s$   *!%4%
r   )6r@   
__future__r   r{   r  r   rd   r   rD   	importlibr   osr   typingr   r   r   r   r	   r
   r   r   r   r   r   r   r   r   Zdocutilsr   Zdocutils.nodesr   r   Zsphinxr   r   Zsphinx.environmentr   Zsphinx.utilr   r   Zsphinx.search.enrF   rM   r_   rA   r`   Zjs_indexru   	dataclassrv   NodeVisitorr~   r   r,   r,   r,   r-   <module>   sV   @\&