3
 UP?                 @   s  d dl mZ d dlZyd dlmZ W n  ek
rD   d dlmZ Y nX ddlm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e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/d0 Z$e$d1Z%d2e%_d3e%_ dDd4d5Z&d6d7 Z'd8d9 Z(d:d; Z)d<d= Z*ee	Z+d>e+_d?e+_ ee
Z,d@e,_dAe,_ eeZ-dBe-_dCe-_ dS )E    )absolute_importN)maxsize)maxint   )Errorsboleoleof
c             C   s   t | }|j  d}t|}g }xj||k rt|| }|d }|d7 }x,||k rr|t|| krr|d7 }|d7 }qHW |j| |j| q"W |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)s	char_listinresultcode1code2 r   c/home/psgendb/BIRCHDEV/lib-linux-x86_64/python/lib64/python3.6/site-packages/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   code3code4dr   r   r   uppercase_range9   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   N)r   r   r   )r   r   r   r    r!   r   r   r   lowercase_rangeG   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>Z   s    zCodeRanges.<locals>.<listcomp>r      )ranger   Alt)r'   re_listr   )r'   r   
CodeRangesU   s    r-   c             C   sB   | t   ko|k n  r4t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%   ^   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_machine{   s    	zRE.build_machinec             C   s"   |j  }|j| |j|| |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)r5   mr7   cr   r   r   r   	build_opt   s    
zRE.build_optc             C   s
   t | |S )N)Seq)r5   otherr   r   r   __add__   s    z
RE.__add__c             C   s
   t | |S )N)r+   )r5   rC   r   r   r   __or__   s    z	RE.__or__c             C   s   | j r| j S | j S d S )N)strcalc_str)r5   r   r   r   __str__   s    z
RE.__str__c             C   s   t |ts| j||d d S )NzPlex.RE instance)
isinstancer1   
wrong_type)r5   numvaluer   r   r   check_re   s    
zRE.check_rec             C   s"   t |t dkr| j||d d S )N string)typerJ   )r5   rK   rL   r   r   r   check_string   s    zRE.check_stringc             C   s8   | j || t|dkr4tj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   PlexValueErrorr3   r4   repr)r5   rK   rL   r   r   r   
check_char   s    zRE.check_charc             C   sL   t |tjkr$d|jj|jjf }n
t |j}tjd|| jj||f d S )Nz%s.%s instancez<Invalid type for argument %d of Plex.%s (expected %s, got %s)rP   typesInstanceTyper3   
__module__r4   r   PlexTypeError)r5   rK   rL   expectedgotr   r   r   rJ      s    
zRE.wrong_type)r4   rW   __qualname____doc__nullablematch_nlrF   r;   rA   rD   rE   rH   rM   rQ   rT   rJ   r   r   r   r   r1   o   s   
r1   c             C   s>   t | dkr$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   SpecialSymbolrS   rF   )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 )N)r*   r"   r$   )r5   r   r   r   r   r   __init__   s    
zRawCodeRange.__init__c             C   sP   |r| j ||t}|j| j| |rL| jr8|j| j| | jrL|j| j| d S )N)rA   BOLr>   r*   r"   r$   )r5   r?   r7   r8   r9   r:   r   r   r   r;      s    zRawCodeRange.build_machinec             C   s   d| j | jf S )NzCodeRange(%d,%d))r   r   )r5   r   r   r   rG     s    zRawCodeRange.calc_str)r4   rW   r[   r\   r]   r^   r*   r"   r$   ra   r;   rG   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| j ||t}| j ||t}|jttd f| d S )Nr   )rA   rb   EOLr>   r.   )r5   r?   r7   r8   r9   r:   r   r   r   r   r;     s    z_RawNewline.build_machineN)r4   rW   r[   r\   r]   r^   r;   r   r   r   r   rc     s   rc   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 )N)sym)r5   re   r   r   r   ra   #  s    zSpecialSymbol.__init__c             C   s.   |r| j tkr| j||t}|j| j | d S )N)re   rd   rA   rb   r>   )r5   r?   r7   r8   r9   r:   r   r   r   r;   &  s    zSpecialSymbol.build_machine)	r4   rW   r[   r\   r]   r^   re   ra   r;   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 )	rB   z]Seq(re1, re2, re3...) is an RE which matches |re1| followed by
    |re2| followed by |re3|...c             G   s   d}x*t |D ]\}}| j|| |o*|j}qW || _|| _t|}d}x,|rt|d8 }|| }|jrjd}P |jsJP qJW || _d S )Nr   r   )	enumeraterM   r]   r,   r   r^   )r5   r,   r]   r   rer^   r   r   r   ra   2  s"    zSeq.__init__c             C   s   | j }t|dkr|j| nb|}t|}xTt|D ]H\}	}
|	|d k rR|j }n|}|
j||||| |}|
jpz|oz|
j}q4W d S )Nr   r   )r,   r   r=   rf   r<   r;   r^   r]   )r5   r?   r7   r8   r9   r:   r,   s1r   r   rg   s2r   r   r   r;   E  s    
zSeq.build_machinec             C   s   ddj tt| j S )NzSeq(%s),)joinmaprF   r,   )r5   r   r   r   rG   U  s    zSeq.calc_strN)r4   rW   r[   r\   ra   r;   rG   r   r   r   r   rB   .  s   rB   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}xJ|D ]B}| j|| |jrF|j| d}n
|j| |jrZd}|d7 }q W || _|| _|| _|| _d S )Nr   r   )r,   rM   r]   r   r^   nullable_resnon_nullable_res)r5   r,   r]   r^   rm   rn   r   rg   r   r   r   ra   ]  s&    


zAlt.__init__c             C   s`   x | j D ]}|j||||| qW | jr\|r:| j||t}x | jD ]}|j|||d| qBW d S )Nr   )rm   r;   rn   rA   rb   )r5   r?   r7   r8   r9   r:   rg   r   r   r   r;   s  s    zAlt.build_machinec             C   s   ddj tt| j S )NzAlt(%s)rj   )rk   rl   rF   r,   )r5   r   r   r   rG   |  s    zAlt.calc_strN)r4   rW   r[   r\   ra   r;   rG   r   r   r   r   r+   Y  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&   | j d| || _|j| _|j| _d S )Nr   )rM   rg   r]   r^   )r5   rg   r   r   r   ra     s    zRep1.__init__c             C   sN   |j  }|j  }|j| | jj||||p0| jj| |j| |j| d S )N)r<   r=   rg   r;   r^   )r5   r?   r7   r8   r9   r:   rh   ri   r   r   r   r;     s    

zRep1.build_machinec             C   s
   d| j  S )NzRep1(%s))rg   )r5   r   r   r   rG     s    zRep1.calc_strN)r4   rW   r[   r\   ra   r;   rG   r   r   r   r   ro     s   ro   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 )N)rg   r:   r]   r^   )r5   rg   r:   r   r   r   ra     s    zSwitchCase.__init__c             C   s   | j j||||| j d S )N)rg   r;   r:   )r5   r?   r7   r8   r9   r:   r   r   r   r;     s    zSwitchCase.build_machinec             C   s   | j rd}nd}d|| jf S )NNoCaseCasez%s(%s))r:   rg   )r5   namer   r   r   rG     s    zSwitchCase.calc_str)	r4   rW   r[   r\   rg   r:   ra   r;   rG   r   r   r   r   rp     s   rp   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|.
    zStr(%s))rB   tuplerl   r`   rS   rF   )r   r   r   r   r   Str1  s    rv   c              G   sH   t | dkrt| d S tttt|  }ddj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   zStr(%s)rj   N)r   rv   r+   ru   rl   rk   rS   rF   )strsr   r   r   r   Str  s
    rx   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   rS   rF   )r   r   r   r   r   Any  s    ry   c             C   s:   t | }|jdt  |j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   insertr   r   r-   rS   rF   )r   rangesr   r   r   r   AnyBut  s    
r|   rN   zT
    AnyChar is an RE which matches any single character (including a newline).
    AnyCharc             C   s   |r*t t| t|d }d| |f |_n\g }x@tdt| dD ],}|jt t| | t| |d  d  q@W 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   rF   r*   r   r   r+   rS   )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+   rt   rF   )rg   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   ro   rF   )rg   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   )r:   )rp   )rg   r   r   r   rq     s    rq   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:   )rp   )rg   r   r   r   rr      s    rr   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).
__future__r   rU   sysr   r   ImportErrorrN   r   rb   rd   EOFr   r.   r   r"   r$   r-   r%   objectr1   r`   r/   rc   r0   r_   rB   r+   ro   rp   rt   r\   rF   rv   rx   ry   r|   r}   r~   r   r   rq   rr   r   r   r   r   r   r   r   <module>	   sb   	n+'!	

		