U
    {h=:                     @   s  d Z ddlZddlmZ dZdZdZdZed	Z	d
d Z
dd Zdd Zdd Zdd ZG dd dZdd ZG dd deZG dd deZe ZG dd deZG dd deZG d d! d!eZG d"d# d#eZG d$d% d%eZe Zd&e_ d'e_d(d) Zd*d+ Zd,d- Zd.d/ Zed0Zd1e_ d2e_dCd3d4Z d5d6 Z!d7d8 Z"d9d: Z#d;d< Z$eeZ%d=e%_ d>e%_eeZ&d?e&_ d@e&_eeZ'dAe'_ dBe'_dS )Dz.
Python Lexical Analyser

Regular Expressions
    N   )ErrorsiZboleoleof
c                 C   s   t | }|  d}t|}g }||k rt|| }|d }|d7 }||k rn|t|| krn|d7 }|d7 }qD|| || q |S )z
    Return a list of character codes consisting of pairs
    [code1a, code1b, code2a, code2b,...] which cover all
    the characters in |s|.
    r   r   )listsortlenordappend)sZ	char_listinresultcode1code2 r   7/tmp/pip-unpacked-wheel-fhl22ezh/Cython/Plex/Regexps.pychars_to_ranges   s    

r   c                 C   sP   t | td}t|tdd }||k rHtdtd }|| || fS dS dS )z
    If the range of characters from code1 to code2-1 includes any
    lower case letters, return the corresponding upper case range.
    azr   ANmaxr
   minr   r   Zcode3Zcode4dr   r   r   uppercase_range3   s    r   c                 C   sP   t | td}t|tdd }||k rHtdtd }|| || fS dS dS )z
    If the range of characters from code1 to code2-1 includes any
    upper case letters, return the corresponding lower case range.
    r   Zr   r   Nr   r   r   r   r   lowercase_rangeA   s    r   c                    s&    fddt dt dD }t| S )z
    Given a list of codes as returned by chars_to_ranges, return
    an RE which will match a character in any of the ranges.
    c                    s"   g | ]}t  |  |d   qS )r   )	CodeRange).0r   	code_listr   r   
<listcomp>T   s     zCodeRanges.<locals>.<listcomp>r      )ranger	   Alt)r#   re_listr   r"   r   
CodeRangesO   s    r)   c                 C   sB   | t   kr|k r4n ntt| t ttt d |S t| |S dS )z
    CodeRange(code1, code2) is an RE which matches any character
    with a code |c| in the range |code1| <= |c| < |code2|.
    r   N)nl_coder'   RawCodeRange
RawNewliner   r   r   r   r   r    X   s    
r    c                   @   sd   e Zd ZdZdZdZdZdd Zdd Zdd	 Z	d
d Z
dd Zdd Zdd Zdd Zdd ZdS )REa  RE is the base class for regular expression constructors.
    The following operators are defined on REs:

         re1 + re2         is an RE which matches |re1| followed by |re2|
         re1 | re2         is an RE which matches either |re1| or |re2|
    r   Nc                 C   s   t d| jj dS )aM  
        This method should add states to |machine| to implement this
        RE, starting at |initial_state| and ending at |final_state|.
        If |match_bol| is true, the RE must be able to match at the
        beginning of a line. If nocase is true, upper and lower case
        letters should be treated as equivalent.
        z %s.build_machine not implementedN)NotImplementedError	__class____name__)selfmachineinitial_statefinal_state	match_bolnocaser   r   r   build_machineu   s    	zRE.build_machinec                 C   s"   |  }|| ||| |S )z~
        Given a state |s| of machine |m|, return a new state
        reachable from |s| on character |c| or epsilon.
        )	new_statelink_toadd_transition)r2   mr4   cr   r   r   r   	build_opt   s    
zRE.build_optc                 C   s
   t | |S N)Seqr2   otherr   r   r   __add__   s    z
RE.__add__c                 C   s
   t | |S r?   )r'   rA   r   r   r   __or__   s    z	RE.__or__c                 C   s   | j r| j S |  S d S r?   )strcalc_strr2   r   r   r   __str__   s    z
RE.__str__c                 C   s   t |ts| ||d d S )NzPlex.RE instance)
isinstancer.   
wrong_typer2   numvaluer   r   r   check_re   s    
zRE.check_rec                 C   s   t |tk	r| ||d d S )Nstring)typerE   rJ   rK   r   r   r   check_string   s    zRE.check_stringc                 C   s8   |  || t|dkr4td|| jjt|f d S )Nr   zOInvalid value for argument %d of Plex.%s.Expected a string of length 1, got: %s)rQ   r	   r   ZPlexValueErrorr0   r1   reprrK   r   r   r   
check_char   s      zRE.check_charc                 C   sL   t |tjkr$d|jj|jjf }n
t |j}td|| jj||f d S )Nz%s.%s instancez<Invalid type for argument %d of Plex.%s (expected %s, got %s)rP   typesZInstanceTyper0   
__module__r1   r   ZPlexTypeError)r2   rL   rM   expectedgotr   r   r   rJ      s     
   zRE.wrong_type)r1   rU   __qualname____doc__nullablematch_nlrE   r8   r>   rC   rD   rH   rN   rQ   rS   rJ   r   r   r   r   r.   i   s   
r.   c                 C   s>   t | dkr$tt| t| d }nt| }dt|  |_|S )z;
    Char(c) is an RE which matches the character |c|.
    r   zChar(%s))r	   r    r
   SpecialSymbolrR   rE   )r=   r   r   r   r   Char   s
    r]   c                   @   s<   e Zd ZdZdZdZdZdZdZdd Z	dd Z
dd	 ZdS )
r+   z
    RawCodeRange(code1, code2) is a low-level RE which matches any character
    with a code |c| in the range |code1| <= |c| < |code2|, where the range
    does not include newline. For internal use only.
    r   Nc                 C   s&   ||f| _ t||| _t||| _d S r?   )r&   r   r   )r2   r   r   r   r   r   __init__   s    
zRawCodeRange.__init__c                 C   sP   |r|  ||t}|| j| |rL| jr8|| j| | jrL|| j| d S r?   )r>   BOLr;   r&   r   r   r2   r<   r4   r5   r6   r7   r   r   r   r8      s    zRawCodeRange.build_machinec                 C   s   d| j | jf S )NzCodeRange(%d,%d)r-   rG   r   r   r   rF      s    zRawCodeRange.calc_str)r1   rU   rX   rY   rZ   r[   r&   r   r   r^   r8   rF   r   r   r   r   r+      s   
r+   c                   @   s    e Zd ZdZdZdZdd ZdS )_RawNewlinezd
    RawNewline is a low-level RE which matches a newline character.
    For internal use only.
    r   r   c                 C   s8   |r|  ||t}|  ||t}|ttd f| d S Nr   )r>   r_   EOLr;   r*   )r2   r<   r4   r5   r6   r7   r   r   r   r   r8      s    z_RawNewline.build_machineN)r1   rU   rX   rY   rZ   r[   r8   r   r   r   r   ra      s   ra   c                   @   s,   e Zd ZdZdZdZdZdd Zdd ZdS )r\   zx
    SpecialSymbol(sym) is an RE which matches the special input
    symbol |sym|, which is one of BOL, EOL or EOF.
    r   Nc                 C   s
   || _ d S r?   )sym)r2   rd   r   r   r   r^      s    zSpecialSymbol.__init__c                 C   s.   |r| j tkr| ||t}|| j | d S r?   )rd   rc   r>   r_   r;   r`   r   r   r   r8     s    zSpecialSymbol.build_machine)	r1   rU   rX   rY   rZ   r[   rd   r^   r8   r   r   r   r   r\      s   r\   c                   @   s(   e Zd ZdZdd Zdd Zdd ZdS )	r@   z]Seq(re1, re2, re3...) is an RE which matches |re1| followed by
    |re2| followed by |re3|...c                 G   sx   d}t |D ]\}}| || |o(|j}q|| _|| _t|}d}|rn|d8 }|| }|jrdd}qn|jsDqnqD|| _d S )Nr   r   )	enumeraterN   rZ   r(   r	   r[   )r2   r(   rZ   r   rer[   r   r   r   r^     s"    zSeq.__init__c                 C   s   | j }t|dkr|| n^|}t|}t|D ]H\}	}
|	|d k rP| }n|}|
||||| |}|
jpx|ox|
j}q2d S Nr   r   )r(   r	   r:   re   r9   r8   r[   rZ   )r2   r<   r4   r5   r6   r7   r(   s1r   r   rf   s2r   r   r   r8      s    
zSeq.build_machinec                 C   s   dd tt| j S )NzSeq(%s),joinmaprE   r(   rG   r   r   r   rF   0  s    zSeq.calc_strNr1   rU   rX   rY   r^   r8   rF   r   r   r   r   r@   	  s   r@   c                   @   s(   e Zd ZdZdd Zdd Zdd ZdS )	r'   zRAlt(re1, re2, re3...) is an RE which matches either |re1| or
    |re2| or |re3|...c                 G   s~   || _ d}d}g }g }d}|D ]B}| || |jrD|| d}n
|| |jrXd}|d7 }q|| _|| _|| _|| _d S rg   )r(   rN   rZ   r   r[   nullable_resnon_nullable_res)r2   r(   rZ   r[   ro   rp   r   rf   r   r   r   r^   8  s&    


zAlt.__init__c                 C   sX   | j D ]}|||||| q| jrT|r6| ||t}| jD ]}||||d| q<d S )Nr   )ro   r8   rp   r>   r_   )r2   r<   r4   r5   r6   r7   rf   r   r   r   r8   N  s    

zAlt.build_machinec                 C   s   dd tt| j S )NzAlt(%s)rj   rk   rG   r   r   r   rF   W  s    zAlt.calc_strNrn   r   r   r   r   r'   4  s   	r'   c                   @   s(   e Zd ZdZdd Zdd Zdd ZdS )	Rep1z@Rep1(re) is an RE which matches one or more repetitions of |re|.c                 C   s&   |  d| || _|j| _|j| _d S rb   )rN   rf   rZ   r[   )r2   rf   r   r   r   r^   ^  s    zRep1.__init__c                 C   sN   |  }|  }|| | j||||p0| jj| || || d S r?   )r9   r:   rf   r8   r[   )r2   r<   r4   r5   r6   r7   rh   ri   r   r   r   r8   d  s    

zRep1.build_machinec                 C   s
   d| j  S )NzRep1(%s)rf   rG   r   r   r   rF   l  s    zRep1.calc_strNrn   r   r   r   r   rq   [  s   rq   c                   @   s0   e Zd ZdZdZdZdd Zdd Zdd ZdS )	
SwitchCasez
    SwitchCase(re, nocase) is an RE which matches the same strings as RE,
    but treating upper and lower case letters according to |nocase|. If
    |nocase| is true, case is ignored, otherwise it is not.
    Nc                 C   s    || _ || _|j| _|j| _d S r?   )rf   r7   rZ   r[   )r2   rf   r7   r   r   r   r^   y  s    zSwitchCase.__init__c                 C   s   | j ||||| j d S r?   )rf   r8   r7   r`   r   r   r   r8     s    zSwitchCase.build_machinec                 C   s   | j rd}nd}d|| jf S )NNoCaseCasez%s(%s))r7   rf   )r2   namer   r   r   rF     s    zSwitchCase.calc_str)	r1   rU   rX   rY   rf   r7   r^   r8   rF   r   r   r   r   rs   p  s   rs   z8
    Empty is an RE which matches the empty string.
    Emptyc                 C   s$   t ttt|  }dt|  |_|S )z@
    Str1(s) is an RE which matches the literal string |s|.
    Str(%s))r@   tuplerm   r]   rR   rE   r   r   r   r   r   Str1  s    r{   c                  G   sH   t | dkrt| d S tttt|  }ddtt|  |_|S dS )z
    Str(s) is an RE which matches the literal string |s|.
    Str(s1, s2, s3, ...) is an RE which matches any of |s1| or |s2| or |s3|...
    r   r   rx   rj   N)r	   r{   r'   ry   rm   rl   rR   rE   )strsr   r   r   r   Str  s
    r}   c                 C   s   t t| }dt|  |_|S )zH
    Any(s) is an RE which matches any character in the string |s|.
    zAny(%s))r)   r   rR   rE   rz   r   r   r   Any  s    r~   c                 C   s:   t | }|dt  |t t|}dt|  |_|S )zp
    AnyBut(s) is an RE which matches any character (including
    newline) which is not in the string |s|.
    r   z
AnyBut(%s))r   insertmaxintr   r)   rR   rE   )r   rangesr   r   r   r   AnyBut  s    
r    zT
    AnyChar is an RE which matches any single character (including a newline).
    AnyCharc              	   C   s   |r*t t| t|d }d| |f |_nXg }tdt| dD ],}|t t| | t| |d  d  q>t| }dt|  |_|S )a  
    Range(c1, c2) is an RE which matches any single character in the range
    |c1| to |c2| inclusive.
    Range(s) where |s| is a string of even length is an RE which matches
    any single character in the ranges |s[0]| to |s[1]|, |s[2]| to |s[3]|,...
    r   zRange(%s,%s)r   r%   z	Range(%s))r    r
   rE   r&   r	   r   r'   rR   )rh   ri   r   r   r   r   r   r   Range  s    *r   c                 C   s   t | t}d|  |_|S )zI
    Opt(re) is an RE which matches either |re| or the empty string.
    zOpt(%s))r'   rw   rE   rf   r   r   r   r   Opt  s    

r   c                 C   s   t t| }d|  |_|S )zJ
    Rep(re) is an RE which matches zero or more repetitions of |re|.
    zRep(%s))r   rq   rE   r   r   r   r   Rep  s    
r   c                 C   s   t | ddS )z
    NoCase(re) is an RE which matches the same strings as RE, but treating
    upper and lower case letters as equivalent.
    r   r7   rs   rr   r   r   r   rt     s    rt   c                 C   s   t | ddS )z
    Case(re) is an RE which matches the same strings as RE, but treating
    upper and lower case letters as distinct, i.e. it cancels the effect
    of any enclosing NoCase().
    r   r   r   rr   r   r   r   ru     s    ru   z=
    Bol is an RE which matches the beginning of a line.
    Bolz7
    Eol is an RE which matches the end of a line.
    Eolz9
    Eof is an RE which matches the end of the file.
    Eof)N)(rY   rT   r   r   r   r_   rc   EOFr
   r*   r   r   r   r)   r    r.   r]   r+   ra   r,   r\   r@   r'   rq   rs   rw   rE   r{   r}   r~   r   r   r   r   r   rt   ru   r   r   r   r   r   r   r   <module>   sf   	O+'"		
		