
    FglN              	       r   S SK r S SKrS SKrS SKJrJrJrJrJrJ	r	J
r
JrJrJrJr  S SKJrJr  S SKJr  \R(                  " \5      r\R.                  r\R0                  r\R2                  r\R4                  r\R6                  r " S S5      r " S S\5      r " S	 S
\5      r\" S\\5      r " S S\\   5      r \ " \5      r!\ " \5      r"\!RF                  r$\"RF                  r%\%" S5      r&\%" S5      r'\%" S5      r(\%" S5      r)\%" S5      r*\%" S5      r+S\S\,4S jr-S\S\4S jr.\R^                  " S5      r0\R^                  " S5      r1\R^                  " S5      r2\R^                  " S5      r3\R^                  " S5      r4\R^                  " S5      r5\R^                  " S5      r6\R^                  " S5      r7\R^                  " S5      r8\R^                  " S 5      r9\R^                  " S!5      r:S"S#S$S%S&S'S(S)S*.r;\\<\=\\\>4   r? " S+ S,5      r@\" S-5      rA\\,\<\=\\>\	\\A4   rB\\C\B\A   4   rD " S. S/\@\\A   5      rEg)0    N)AnyBinaryIODictGenericIteratorListOptionalTupleTypeTypeVarUnion)psexceptionssettings)choplistc                       \ rS rSrSrSrg)PSObject!   z0Base class for all PS or PDF-related data types. N)__name__
__module____qualname____firstlineno____doc____static_attributes__r       `/Users/admin/workspace/ai/PDFMathTranslate/myenv/lib/python3.13/site-packages/pdf2zh/psparser.pyr   r   !   s    :r   r   c                   F    \ rS rSrSr\\\4   rS\SS4S jr	S\4S jr
Srg)		PSLiteral%   aE  A class that represents a PostScript literal.

Postscript literals are used as identifiers, such as
variable names, property names and dictionary keys.
Literals are case sensitive and denoted by a preceding
slash sign (e.g. "/Name")

Note: Do not create an instance of PSLiteral directly.
Always use PSLiteralTable.intern().
namereturnNc                     Xl         g Nr    selfr    s     r   __init__PSLiteral.__init__3       	r   c                 $    U R                   nSU-  $ Nz/%rr$   r%   s     r   __repr__PSLiteral.__repr__6       yyt|r   r$   )r   r   r   r   r   r   strbytesNameTyper'   r,   r   r   r   r   r   r   %   s6    	 S%Z HX $ # r   r   c                   8    \ rS rSrSrS\SS4S jrS\4S jrSr	g)		PSKeyword;   a6  A class that represents a PostScript keyword.

PostScript keywords are a dozen of predefined words.
Commands and directives in PostScript are expressed by keywords.
They are also used to denote the content boundaries.

Note: Do not create an instance of PSKeyword directly.
Always use PSKeywordTable.intern().
r    r!   Nc                     Xl         g r#   r$   r%   s     r   r'   PSKeyword.__init__F   r)   r   c                 $    U R                   nSU-  $ r+   r$   r%   s     r   r,   PSKeyword.__repr__I   r.   r   r$   )
r   r   r   r   r   r0   r'   r/   r,   r   r   r   r   r3   r3   ;   s&    U t # r   r3   _SymbolTc                   V    \ rS rSrSrS\\   SS4S jrS\R                  S\4S jr
S	rg)
PSSymbolTableQ   z{A utility class for storing PSLiteral/PSKeyword objects.

Interned objects can be checked its identity with "is" operator.
klassr!   Nc                     0 U l         Xl        g r#   dictr=   )r&   r=   s     r   r'   PSSymbolTable.__init__W   s    8:	%*
r   r    c                     XR                   ;   a  U R                   U   nU$ U R                  U5      nX R                   U'   U$ r#   r?   )r&   r    lits      r   internPSSymbolTable.intern[   s@    99))D/C 
 **T"C!IIdO
r   r?   )r   r   r   r   r   r   r9   r'   r   r1   rD   r   r   r   r   r;   r;   Q   s7    
+d8n + +9-- ( r   r;      {   }   [   ]s   <<s   >>xr!   c                 X   [        U [        5      (       aB  [        U R                  [        5      (       a  U R                  $  [        U R                  S5      $ [
        R                  (       a  [        SU < 35      e[        U 5      $ ! [         a    [        U R                  5      s $ f = f)Nutf-8zLiteral required: )
isinstancer   r    r/   UnicodeDecodeErrorr   STRICTPSTypeError)rJ   s    r   literal_namerQ   r   s    !Yaffc""66M	qvvw'' ?? 21%8991v " 	qvv;	s   B B)(B)c                     [        U [        5      (       d(  [        R                  (       a  [	        SU -  5      eU n U$ [        U R                  SS5      nU$ )NzKeyword required: %rrL   ignore)rM   r3   r   rO   rP   r/   r    )rJ   r    s     r   keyword_namerT      sL    a##??4q899D K 1667H-Kr   s   [\r\n]s   \ss   \Ss   [0-9a-fA-F]s   [#/%\[\]()<>{}\s]s   [^\s0-9a-fA-F]s   [0-9a-fA-F]{2}|.s   [^0-9]s   [()\134]s   [0-7]   	   
         (   )   \   )   b   t   n   f   r   (   )   \c                      \ rS rSrSrSrS\SS4S jrS\4S jr	S%S	 jr
S%S
 jrS\4S jrS&S\\   S\SS4S jjrS\SS4S jrS%S jrS\\\4   4S jrS\\   4S jrS\S\S\4S jrS\SS4S jrS\S\S\4S jrS\S\S\4S jrS\S\S\4S jrS\S\S\4S jrS\S\S\4S jrS\S\S\4S jrS\S\S\4S jr S\S\S\4S jr!S\S\S\4S  jr"S\S\S\4S! jr#S\S\S\4S" jr$S\\\4   4S# jr%S$r&g)'PSBaseParser   z=Most basic PostScript parser that performs only tokenization.i   fpr!   Nc                 2    Xl         U R                  S5        g Nr   )rh   seekr&   rh   s     r   r'   PSBaseParser.__init__   s    		!r   c                 b    SU R                   R                  U R                  U R                  4-  $ )Nz<%s: %r, bufpos=%d>)	__class__r   rh   bufposr&   s    r   r,   PSBaseParser.__repr__   s&    $(?(?$++'VVVr   c                     g r#   r   rq   s    r   flushPSBaseParser.flush       r   c                 $    U R                  5         g r#   )rt   rq   s    r   closePSBaseParser.close   s    

r   c                 4    U R                   U R                  -   $ r#   )rp   charposrq   s    r   tellPSBaseParser.tell   s    {{T\\))r   posnc                     U R                   R                  5       nU(       d  U R                  U R                  -   nU R                   R	                  U5        U R                   R	                  U5        g r#   )rh   r|   rp   r{   rk   )r&   r~   r   pos0s       r   pollPSBaseParser.poll   sF    ww||~++,CSTr   c                     U R                   R                  U5        Xl        SU l        SU l        U R
                  U l        SU l        SU l        / U l	        g)z'Seeks the parser to the given position.r   r   N)
rh   rk   rp   bufr{   _parse_main_parse1	_curtoken_curtokenpos_tokensr&   r~   s     r   rk   PSBaseParser.seek   sJ     	S''<>r   c                 $   U R                   [        U R                  5      :  a  g U R                  R	                  5       U l        U R                  R                  U R                  5      U l        U R                  (       d  [        S5      eSU l         g )NzUnexpected EOFr   )	r{   lenr   rh   r|   rp   readBUFSIZPSEOFrq   s    r   fillbufPSBaseParser.fillbuf   s\    <<#dhh-'gglln77<<,xx())r   c                 p   SnU R                   U R                  -   nSn U R                  5         U(       aI  U R                  U R                  U R                  S-    nUS:X  a  X-  nU =R                  S-  sl         X!4$ [        R                  U R                  U R                  5      nU(       aP  XR                  U R                  UR                  S5       -  nUR                  S5      U l        USS S:X  a  S	nO9 X!4$ XR                  U R                  S -  n[        U R                  5      U l        GM  )
z3Fetches a next line that ends either with \r or \n.r   F      
r   N   T)rp   r{   r   r   EOLsearchendr   )r&   linebuflineposeolcms         r   nextlinePSBaseParser.nextline   s   ++,LLNHHT\\DLL1,<=:LGLLA%L !! 

488T\\2A88DLL1558<< uuQx23<5(C !!	 88DLLN33"488}' r   c              #     #    U R                   R                  S[        R                  5        U R                   R	                  5       nSnUS:  a  Un[        SXR                  -
  5      nU R                   R                  U5        U R                   R                  X1-
  5      nU(       d  g [        UR                  S5      UR                  S5      5      nUS:X  a  XB-   nOXES U-   v   USU nSnMG  US:  a  M  gg7f)zYFetches a next line backword.

This is used to locate the trailers at the end of a file.
r   r   r   r   r   N)	rh   rk   ioSEEK_ENDr|   maxr   r   rfind)r&   r~   r   prevpossr   s         r   revreadlinesPSBaseParser.revreadlines   s     
 	Q$ggllnAgGa{{*+CGGLLW]+A77'Ceck!bqE  Ags   C5C;9C;r   ic                 |   [         R                  X5      nU(       d  [        U5      $ UR                  S5      nXUS-    nU R                  U-   U l        US:X  a  SU l        U R                  U l        US-   $ US:X  a  SU l        U R                  U l        US-   $ US;   d  UR                  5       (       a  XPl        U R                  U l        US-   $ US:X  a  XPl        U R                  U l        US-   $ UR                  5       (       a  XPl        U R                  U l        US-   $ US:X  a$  SU l        SU l        U R                   U l        US-   $ US	:X  a  SU l        U R"                  U l        US-   $ US
:X  a  SU l        U R$                  U l        US-   $ US:X  a  US-   $ U R'                  [)        U5      5        US-   $ )Nr   r      %   /r   s   -+   .rb      <   >    )NONSPCr   r   startrp   r   r   _parse_commentr   _parse_literalisdigit_parse_number_parse_floatisalpha_parse_keywordparen_parse_string_parse_wopen_parse_wclose
_add_tokenKWDr&   r   r   r   jr   s         r   r   PSBaseParser._parse_main  s   MM!q6MGGAJ!a%L KK!O9!DN..DLq5L$Y DN..DLq5L%Z199;;N--DLq5L$YN,,DLq5LYY[[N..DLq5L$Y DNDJ--DLq5L$Y DN,,DLq5L$Y DN--DLq5L'\q5LOOCF#q5Lr   objc                 R    U R                   R                  U R                  U45        g r#   )r   appendr   )r&   r   s     r   r   PSBaseParser._add_token=  s    T..45r   c                     [         R                  X5      nU(       d"  U =R                  XS  -  sl        [        U5      $ UR	                  S5      nU =R                  XU -  sl        U R
                  U l        U$ rj   )r   r   r   r   r   r   r   r&   r   r   r   r   s        r   r   PSBaseParser._parse_comment@  sa    JJqNNae#Nq6MGGAJ!a& '' r   c                    [         R                  X5      nU(       d"  U =R                  XS  -  sl        [        U5      $ UR	                  S5      nU =R                  XU -  sl        XUS-    nUS:X  a  SU l        U R                  U l        US-   $  [        U R                  S5      nU R                  [        U5      5        U R                  U l        U$ ! [         a    U R                  n NEf = f)Nr   r      #r   rL   )END_LITERALr   r   r   r   hex_parse_literal_hexr   r/   	Exceptionr   LITr   )r&   r   r   r   r   r   r    s          r   r   PSBaseParser._parse_literalL  s    q$NNae#Nq6MGGAJ!a& !a%L9DH22DLq5L	"&)$..'&BD 	D	"''	  	">>D	"s   C C.-C.c                 X   XUS-    n[         R                  U5      (       a3  [        U R                  5      S:  a  U =R                  U-  sl        US-   $ U R                  (       a3  U =R                  [        [        U R                  S5      45      -  sl        U R                  U l        U$ )Nr         )	HEXmatchr   r   r   r0   intr   r   r&   r   r   r   s       r   r   PSBaseParser._parse_literal_hex`  s{    !a%L99Q<<CMA-HHMHq5L88NNeS2%6$899N**r   c                    [         R                  X5      nU(       d"  U =R                  XS  -  sl        [        U5      $ UR	                  S5      nU =R                  XU -  sl        XUS-    nUS:X  a+  U =R                  U-  sl        U R
                  U l        US-   $  U R                  [        U R                  5      5        U R                  U l        U$ ! [         a     Nf = f)Nr   r   r   )
END_NUMBERr   r   r   r   r   r   r   r   
ValueErrorr   r   s         r   r   PSBaseParser._parse_numberj  s    a#NNae#Nq6MGGAJ!a& !a%L9NNaN,,DLq5L	OOC/0 ''  		s    $C 
C$#C$c                 ^   [         R                  X5      nU(       d"  U =R                  XS  -  sl        [        U5      $ UR	                  S5      nU =R                  XU -  sl         U R                  [        U R                  5      5        U R                  U l	        U$ ! [         a     Nf = frj   )
r   r   r   r   r   r   floatr   r   r   r   s        r   r   PSBaseParser._parse_float}  s    a#NNae#Nq6MGGAJ!a& 	OOE$..12 ''  		s   ($B 
B,+B,c                    [         R                  X5      nU(       a)  UR                  S5      nU =R                  XU -  sl        O"[	        U5      nU =R                  XS  -  sl        U R                  S:X  a  SnO(U R                  S:X  a  SnO[        U R                  5      nU R                  U5        U R                  U l        U$ )Nr   s   trueTs   falseF)	END_KEYWORDr   r   r   r   r   r   r   r   r&   r   r   r   r   tokens         r   r   PSBaseParser._parse_keyword  s    q$
ANNa!f$N
 AANNae#N>>W$,0E^^x'E'E''r   c                 z   [         R                  X5      nU(       d"  U =R                  XS  -  sl        [        U5      $ UR	                  S5      nU =R                  XU -  sl        XUS-    nUS:X  a  SU l        U R                  U l        US-   $ US:X  a/  U =R                  S-  sl        U =R                  U-  sl        US-   $ US:X  a@  U =R                  S-  sl        U R                  (       a  U =R                  U-  sl        US-   $ U R                  U R                  5        U R                  U l        US-   $ )Nr   r   rd   r   rb   rc   )
END_STRINGr   r   r   r   oct_parse_string_1r   r   r   r   r   s         r   r   PSBaseParser._parse_string  s   a#NNae#Nq6MGGAJ!a& !a%L:DH//DLq5L9JJ!OJNNaNq5L9JJ!OJzz!#1u'''1ur   c                    XUS-    n[         R                  U5      (       a3  [        U R                  5      S:  a  U =R                  U-  sl        US-   $ U R                  (       am  [	        U R                  S5      nUS:  d   S[        U R                  5      U4-  5       eU =R                  [        U45      -  sl        U R                  U l	        U$ U[        ;   a'  U =R                  [        [        U   45      -  sl        O+US:X  a%  [        U5      US-   :  a  XS-   US-    S:X  a  US-  nU R                  U l	        US-   $ )	z+Parse literal strings

PDF Reference 3.2.3
r      rU      zInvalid octal %s (%d)r   r   r   )
OCT_STRINGr   r   r   r   reprr   r0   r   r   
ESC_STRING)r&   r   r   r   chrcodes        r   r   PSBaseParser._parse_string_1  s   
 !a%LA3txx=1#4HHMHq5LXX$((A&GS=U"9T$((^W<U"UU=NNeWJ//N--DLH*_NNeZ]$455N%ZCFQUNqQQ/?5/H FA ))1ur   c                     XUS-    nUS:X  a-  U R                  [        5        U R                  U l        US-  nU$ U R                  U l        U$ )Nr   r   )r   KEYWORD_DICT_BEGINr   r   _parse_hexstringr   s       r   r   PSBaseParser._parse_wopen  sT    !a%L9OO./++DLFA   00DLr   c                 v    XUS-    nUS:X  a  U R                  [        5        US-  nU R                  U l        U$ )Nr   r   )r   KEYWORD_DICT_ENDr   r   r   s       r   r   PSBaseParser._parse_wclose  s>    !a%L9OO,-FA''r   c                    [         R                  X5      nU(       d"  U =R                  XS  -  sl        [        U5      $ UR	                  S5      nU =R                  XU -  sl        [
        R                  S [        R                  SU R                  5      5      nU R                  U5        U R                  U l
        U$ )Nr   c                 L    [        [        U R                  S5      S5      45      $ )Nr   r   )r0   r   group)r   s    r   <lambda>/PSBaseParser._parse_hexstring.<locals>.<lambda>  s    eSR023r   r   )END_HEX_STRINGr   r   r   r   HEX_PAIRsubSPCr   r   r   r   s         r   r   PSBaseParser._parse_hexstring  s    !!!'NNae#Nq6MGGAJ!a& 3GGC(
 	''r   c                     U R                   (       dN  U R                  5         U R                  U R                  U R                  5      U l        U R                   (       d  MN  U R                   R                  S5      nU$ rj   )r   r   r   r   r{   pop)r&   r   s     r   	nexttokenPSBaseParser.nexttoken  sR    ,,LLN<<$,,?DL ,,,   #r   )r   r   r   r   r   rp   r{   rh   r   r   r   r!   N)NP   )'r   r   r   r   r   r   r   r'   r/   r,   rt   rx   r   r|   r	   r   rk   r   r
   r0   r   r   r   r   PSBaseParserTokenr   r   r   r   r   r   r   r   r   r   r   r   r	  r   r   r   r   rf   rf      s    GF8  W# W*c *  d ? ? ?"%U
+ ":huo 0,U ,s ,s ,\6/ 6D 6
 
# 
# 
 # # (E c c u   &e    # # *u   4 3 3 :e   u   % C C 5&7!78 r   rf   ExtraTc                      \ rS rSrS\SS4S jrSS jrS\SS4S jrS	\	\
   SS4S
 jrS\S\\	\
      4S jrS\\	\
      4S jrS	\	\
   SS4S jrS\S\SS4S jrS\S\\\\\
      4   4S jrS\S\SS4S jrS\	\
   4S jrSrg)PSStackParseri  rh   r!   Nc                 N    [         R                  X5        U R                  5         g r#   )rf   r'   resetrl   s     r   r'   PSStackParser.__init__  s    d'

r   c                 <    / U l         S U l        / U l        / U l        g r#   )contextcurtypecurstackresultsrq   s    r   r  PSStackParser.reset  s    TV&*4635r   r~   c                 N    [         R                  X5        U R                  5         g r#   )rf   rk   r  r   s     r   rk   PSStackParser.seek  s    $$

r   objsc                 :    U R                   R                  U5        g r#   )r  extendr&   r  s     r   pushPSStackParser.push  s    T"r   r   c                 F    U R                   U* S  n/ U R                   U* S & U$ r#   r  )r&   r   r  s      r   r  PSStackParser.pop  s+    }}aRS!qbcr   c                 ,    U R                   n/ U l         U$ r#   r#  r  s     r   popallPSStackParser.popall#  s    }}r   c                 :    U R                   R                  U5        g r#   )r  r  r  s     r   add_resultsPSStackParser.add_results(  s    
 	D!r   typec                     U R                   R                  XR                  U R                  45        U/ sU l        U l        g r#   )r  r   r  r  )r&   r~   r+  s      r   
start_typePSStackParser.start_type/  s1    S,,>?)-r%t}r   c                     U R                   U:w  a  [        SU R                   < SU< 35      eU R                   VVs/ s H  u  p#UPM	     nnnU R                  R	                  5       u  oPl         U l        XT4$ s  snnf )NzType mismatch: z != )r  rP   r  r  r  )r&   r+  _r   r  r~   s         r   end_typePSStackParser.end_type4  sm    <<4/?tD8LMM$(MM2MM2-1\\-=-=-?*lDM{ 3s   A8r   c                     g r#   r   )r&   r~   r   s      r   
do_keywordPSStackParser.do_keyword<  rv   r   c           	         SnU R                   (       Gd6  U R                  5       u  p#[        U[        [        [
        [        [        [        45      (       a  U R                  X#45        GOU[        :X  a  U R                  US5        GOU[        :X  a#   U R                  U R                  S5      5        GO^U["        :X  a  U R                  US5        GO@U[$        :X  ax   U R                  S5      u  p$['        U5      S-  S:w  a  SU-  n[)        U5      e[+        SU5       VVs0 s H  u  pgUc  M
  [-        U5      U_M     nnnU R                  X(45        OU[.        :X  a  U R                  US5        OU[0        :X  a"   U R                  U R                  S5      5        Ou[        U[2        5      (       a'  UR4                  S:X  a  US	-   nU R7                  X#5        O9[8        R;                  S
UUU R<                  5        U R7                  X#5        [>        eU R@                  (       a  GM$  U RC                  5         U R                   (       d  GM6  U R                   RE                  S5      n	X4$ ! [         a    [        R                   (       a  e  Nxf = fs  snnf ! [         a    [        R                   (       a  e  Nf = f! [         a    [        R                   (       a  e  Nf = f)zYields a list of objects.

Arrays and dictionaries are represented as Python lists and
dictionaries.

:return: keywords, literals, strings, numbers, arrays and dictionaries.
Nadr   r   z Invalid dictionary construct: %rps   endobj   z)unknown token: pos=%r, token=%r, stack=%r)#r  r	  rM   r   r   boolr/   r0   r   r   KEYWORD_ARRAY_BEGINr-  KEYWORD_ARRAY_ENDr1  rP   r   rO   r   r   r   PSSyntaxErrorr   rQ   KEYWORD_PROC_BEGINKEYWORD_PROC_ENDr3   r    r4  logerrorr  PSExceptionr  rt   r  )
r&   r   r~   r   r  	error_msgkvr8  r   s
             r   
nextobjectPSStackParser.nextobject?  sh    ,,,>>+LS%#udC	!JKK		3,'--S)++IIdmmC01 ,,S)**"&--"4KS4y1})$F$M	+I66 '/q$&7&7FQ +Q*&7  
 IIsh' ,,S)**IIdmmC01 E9-- ::y(AC+		?MM	 +!!||

E ,,,F llq!
 wy #  ' #  ' #  'sO    I) AJ "	J/J?J = J; ) JJJ  J87J8; KK)r  r  r  r  r  )r   r   r   r   r   r'   r  r   rk   PSStackEntryr  r   r   r  r&  r)  r/   r-  r
   PSStackTyper1  r3   r4  rG  r   r   r   r   r  r    s    8  6  #,v. #4 #S T,v"67 
\&12 
"f!5 "$ "3c 3 3 3
S U3[5H0I+I%J c )  QL0 Qr   r  )Fr   loggingretypingr   r   r   r   r   r   r	   r
   r   r   r   pdf2zhr   r   pdf2zh.utilsr   	getLoggerr   rA  rC  r   r>  rP   PSValueErrorr   r   r3   r9   r;   PSLiteralTablePSKeywordTablerD   r   r   r?  r@  r<  r=  r   r   r/   rQ   rT   compiler   r  r   r   r   r  r  r   r   r   r   r   r   r;  r0   r  rf   r  rJ  r   rI  r  r   r   r   <module>rU     sw   	  	    * !! &&**&&((; ; , & :y)4GH% * y)y)Y t9 $i I Z u: C C C C  	jjjj	F	jj!jj././::*+ZZ
#
jj./ZZ%
ZZ	"







	
 %y)UBC Y YD 
	CidFJKS+f--.DL'&/ Dr   