o
    NiM9                     @   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boleoleof
c                 C   s   t | }|  d}t|}g }||k rPt|| }|d }|d7 }||k rB|t|| krB|d7 }|d7 }||k rB|t|| ks.|| || ||k s|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   G/home/ubuntu/.local/lib/python3.10/site-packages/Cython/Plex/Regexps.pychars_to_ranges   s"   

	r   c                 C   L   t | td}t|tdd }||k r$tdtd }|| || f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_range3      r"   c                 C   r   )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   r#   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   s>   | t   kr
|k rn ntt| t ttt d |S t| |S )z
    CodeRange(code1, code2) is an RE which matches any character
    with a code |c| in the range |code1| <= |c| < |code2|.
    r   )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 )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 implemented)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)r8   mr:   cr   r   r   r   	build_opt   s   
zRE.build_optc                 C   
   t | |S N)Seqr8   otherr   r   r   __add__      
z
RE.__add__c                 C   rE   rF   )r-   rH   r   r   r   __or__   rK   z	RE.__or__c                 C   s   | j r| j S |  S rF   )strcalc_strr8   r   r   r   __str__   s   z
RE.__str__c                 C   s    t |ts| ||d d S d S )NzPlex.RE instance)
isinstancer4   
wrong_typer8   numvaluer   r   r   check_re   s   
zRE.check_rec                 C   s"   t |tur| ||d d S d S )Nstring)typerM   rR   rS   r   r   r   check_string   s   zRE.check_stringc                 C   s8   |  || t|dkrtd|| jjt|f d S )Nr   zOInvalid value for argument %d of Plex.%s.Expected a string of length 1, got: %s)rY   r
   r   PlexValueErrorr6   r7   reprrS   r   r   r   
check_char   s   zRE.check_charc                 C   s.   t d|dd| jj d| dt|j )NzInvalid type for argument r!   z of z (expected z, got )r   PlexTypeErrorr6   __qualname__rX   r7   )r8   rT   rU   expectedr   r   r   rR      s   zRE.wrong_type)r7   
__module__r^   __doc__nullablematch_nlrM   r>   rD   rJ   rL   rP   rV   rY   r\   rR   r   r   r   r   r4   i   s    
r4   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   SpecialSymbolr[   rM   )rC   r   r   r   r   Char   s
   re   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 )
r1   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 rF   )r,   r"   r%   )r8   r   r   r   r   r   __init__   s   
zRawCodeRange.__init__c                 C   sX   |r	|  ||t}|| j| |r(| jr|| j| | jr*|| j| d S d S d S rF   )rD   BOLrA   r,   r"   r%   r8   rB   r:   r;   r<   r=   r   r   r   r>      s   zRawCodeRange.build_machinec                 C   s   d| j | jf S )NzCodeRange(%d,%d)r3   rO   r   r   r   rN      s   zRawCodeRange.calc_str)r7   r`   r^   ra   rb   rc   r,   r"   r%   rf   r>   rN   r   r   r   r   r1      s    
r1   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   )rD   rg   EOLrA   r0   )r8   rB   r:   r;   r<   r=   r   r   r   r   r>      s   z_RawNewline.build_machineN)r7   r`   r^   ra   rb   rc   r>   r   r   r   r   ri      s
    ri   c                   @   s,   e Zd ZdZdZdZdZdd Zdd ZdS )rd   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 rF   )sym)r8   rl   r   r   r   rf      rK   zSpecialSymbol.__init__c                 C   s.   |r| j tkr| ||t}|| j | d S rF   )rl   rk   rD   rg   rA   rh   r   r   r   r>      s   zSpecialSymbol.build_machine)	r7   r`   r^   ra   rb   rc   rl   rf   r>   r   r   r   r   rd      s    rd   c                   @   (   e Zd ZdZdd Zdd Zdd ZdS )	rG   z]Seq(re1, re2, re3...) is an RE which matches |re1| followed by
    |re2| followed by |re3|...c                 G   sz   d}t |D ]\}}| || |o|j}q|| _|| _t|}d}|r8|d8 }|| }|jr2d}n|js6n|s$|| _d S )Nr   r   )	enumeraterV   rb   r.   r
   rc   )r8   r.   rb   r   rerc   r   r   r   rf   	  s$   
zSeq.__init__c                 C   s   | j }t|dkr|| d S |}t|}t|D ]$\}	}
|	|d k r)| }n|}|
||||| |}|
jp=|o=|
j}qd S Nr   r   )r.   r
   r@   rn   r?   r>   rc   rb   )r8   rB   r:   r;   r<   r=   r.   s1r   r   ro   s2r   r   r   r>     s   
zSeq.build_machinec                 C      dd tt| j S )NzSeq(%s),joinmaprM   r.   rO   r   r   r   rN   ,     zSeq.calc_strNr7   r`   r^   ra   rf   r>   rN   r   r   r   r   rG     s
    rG   c                   @   rm   )	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 ]!}| || |jr"|| d}n|| |jr,d}|d7 }q|| _|| _|| _|| _d S rp   )r.   rV   rb   r   rc   nullable_resnon_nullable_res)r8   r.   rb   rc   rz   r{   r   ro   r   r   r   rf   4  s&   



zAlt.__init__c                 C   s\   | j D ]}|||||| q| jr*|r| ||t}| jD ]}||||d| qd S d S )Nr   )rz   r>   r{   rD   rg   )r8   rB   r:   r;   r<   r=   ro   r   r   r   r>   J  s   

zAlt.build_machinec                 C   rs   )NzAlt(%s)rt   ru   rO   r   r   r   rN   S  rx   zAlt.calc_strNry   r   r   r   r   r-   0  s
    	r-   c                   @   rm   )	Rep1z@Rep1(re) is an RE which matches one or more repetitions of |re|.c                 C   s&   |  d| || _|j| _|j| _d S rj   )rV   ro   rb   rc   )r8   ro   r   r   r   rf   Z  s   zRep1.__init__c                 C   sN   |  }|  }|| | j||||p| jj| || || d S rF   )r?   r@   ro   r>   rc   )r8   rB   r:   r;   r<   r=   rq   rr   r   r   r   r>   `  s   

zRep1.build_machinec                 C   s
   d| j  S )NzRep1(%s)ro   rO   r   r   r   rN   h  rK   zRep1.calc_strNry   r   r   r   r   r|   W  s
    r|   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 rF   )ro   r=   rb   rc   )r8   ro   r=   r   r   r   rf   u  s   zSwitchCase.__init__c                 C   s   | j ||||| j d S rF   )ro   r>   r=   rh   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=   ro   )r8   namer   r   r   rN     s   zSwitchCase.calc_str)	r7   r`   r^   ra   ro   r=   rf   r>   rN   r   r   r   r   r~   l  s    r~   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))rG   tuplerw   re   r[   rM   r   r   r   r   r   Str1  s   r   c                  G   sD   t | dkrt| d S tttt|  }ddtt|  |_|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   r   rt   )r
   r   r-   r   rw   rv   r[   rM   )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   r[   rM   r   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/   r[   rM   )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 |_|S g }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   rM   r,   r
   r   r-   r[   )rq   rr   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-   r   rM   ro   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   r|   rM   r   r   r   r   Rep  s   
r   c                 C      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=   r~   r}   r   r   r   r     s   r   c                 C   r   )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   r}   r   r   r   r     s   r   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.
    EofrF   )(ra   typesr   r   r   rg   rk   EOFr   r0   r   r"   r%   r/   r&   r4   re   r1   ri   r2   rd   rG   r-   r|   r~   r   rM   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   <module>   sh    	K+'"		
		
