File indexing completed on 2024-04-21 04:44:10

 # -*- coding: utf-8 -*-
 #     Copyright 2007-8 Jim Bublitz <jbublitz@nwinternet.com>
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation; either version 2 of the License, or
 # (at your option) any later version.
 #
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
 #
 # You should have received a copy of the GNU General Public License
 # along with this program; if not, write to the
 # Free Software Foundation, Inc.,
 # 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
 
 import re
 import ply.lex as lex
 # handles the evaluation of conditionals
 from .exprparser import ExpressionParser
 
 newtext   = []
 macros    = []
 bitBucket = False
 sentinel  = False
 
 preprocessor_tokens =  ['cond', 'else', 'endif', 'include', 'define', 'undef', 'line', 'error', 'pragma', 'warning']
 tokens = preprocessor_tokens + ['anyline']
 
 values = {}
 evaluate = ExpressionParser ().parse
 
 # Completely ignored characters
 t_ANY_ignore           = ' \t\x0c'
 
 def stripComment (s):
     pos1 = s.find ('\/\*')
     pos2 = s.find ('\/\/')
     if pos1 > 0 and pos2 > 0:
         pos = min (pos1, pos2)
     elif pos1 < 0 and pos2 < 0:
         pos = -1
     else:
         pos = max (pos1, pos2)
     
     if pos > 0:
         return s [:pos].strip (), s[pos:].strip ()
     else:
         return s, ''
 
 def t_cond (t):
     r'\#\s*(?P<ifType>ifdef\s|ifndef\s|if\s|elif\s)\s*(?P<cond>.*?)\n'
 
     # All conditionals that perform a test are handled here
     global newtext
     ifType = t.lexer.lexmatch.group ('ifType').strip ()
     condition, comment  = stripComment (t.lexer.lexmatch.group ('cond'))
     
     # #if/#elif look for True/False, others for defintion only
     # #if defined - 'defined' is handled as an operator by the
     # expression parser which evaluates the conditional
     if ifType in ['if', 'elif']:
         mode = 'calc'
     else:
         mode = 'def'
 
     ifCondition = evaluate (condition, mode, values)
         
     global bitBucket, sentinel
     bitBucket = ((not ifCondition) and (ifType != 'ifndef')) or (ifCondition and (ifType == 'ifndef'))
     
     # remove #define <sentinel>?
     sentinel = not bitBucket and ('_h' in condition or '_H' in condition)
 
     # A multiline comment could begin on a preprocessor line
     # that's being eliminated here
     if bitBucket and comment:
         newtext.append (comment + '\n')
     else:
         newtext.append ('\n')
 
     t.lexer.lineno += 1
     
 def t_else (t):
     r'\#\s*else(.*?)\n'  # comments?
     global bitBucket, newtext
     bitBucket = not bitBucket
     t.lexer.lineno += 1
     newtext.append ('\n')
     
 def t_endif (t):
     r'\#\s*endif(.*?)\n'
     global bitBucket, newtext
     bitBucket = False
     t.lexer.lineno += 1
     newtext.append ('\n')
 
 def t_include (t):
     r'\#\s*include.*?\n'
     global newtext
     t.lexer.lineno += 1
     newtext.append ('\n')
     
 def t_line (t):
     r'\#\s*line.*?\n'
     global newtext
     t.lexer.lineno += 1
     newtext.append ('\n')
 
 def t_error (t):    
     r'\#\s*error.*?\n'
     global newtext
     t.lexer.lineno += 1
     newtext.append ('\n')
     
 def t_pragma (t):    
     r'\#\s*pragma.*?\n'
     global newtext
     t.lexer.lineno += 1
     newtext.append ('\n')
     
 def t_warning (t):    
     r'\#\s*warning.*?\n'
     global newtext
     t.lexer.lineno += 1
     newtext.append ('\n')
 
 def t_undef (t):
     r'\#\s*undef\s*(?P<item>.*?)\n'
     global macros, values, newtext
     item = t.lexer.lexmatch.group ('item').strip ()
     if item in values:
         macros = [macro for macro in macros if len(macro)==2 or macro[2] != item]
         del values [item]
     t.lexer.lineno += 1
     newtext.append ('\n')
     
 def t_define (t):
     r'\#\s*define\s*(?P<first>[\S]+)\s*?(?P<second>[^\n]*?)\n'    
     global sentinel, values, macros, newtext
     a = t.lexer.lexmatch.group ('first')
     b = t.lexer.lexmatch.group ('second')
     
     # Append any continuation lines
     newlines = 1
     start = t.lexer.lexpos
     if b and b.endswith ('\\'):              
         data = t.lexer.lexdata
         for i in range (start, len (data)):
             if data [i] == '\n':
                 t.lexer.lineno += 1
                 newlines += 1
             if data [i] == '\n' and data [i - 1] != '\\':                
                 break
         t.lexer.lexpos = i + 1
         b += data [start:t.lexer.lexpos].replace ('\\\n', ' ')
 
     if '(' in a and not ')' in a:
         pos = b.find (')')
         if pos < 0:
             return
         a += b [:pos + 1]
         b = b [pos + 1:]
     
     # remove #define <sentinel>
     sentinel = sentinel and not b and ('_h' in a or '_H' in a)
     if not sentinel:
         if not b or '(' in a:
             values [a] = ''
             macros.insert (0, (re.compile (a), '', a))
         else:
             values [a] = b
             macros.insert (0, (re.compile (a), b.strip (), a))
     
     sentinel = False
             
     newtext.append (newlines *'\n')
     t.lexer.lineno += 1
     
 def t_anyline (t):
     r'[^\n]*?\n(([^#\n][^\n]*\n)|\n)*'
     """
     Process anything that's not a preprocesor directive.
 
     Apply all #define macros to each line. Code that has
     been #if'd out (bitBucket == True) is replaced by
     a single newline for each line removed.
     """
     global sentinel, newtext
     sentinel = False
     if not bitBucket:
         line = t.value
         for m in macros:
             line = m[0].sub(m[1], line)
         newtext.append (line)
         t.lexer.lineno += line.count('\n')
     else:
         c = t.value.count('\n')
         for x in range(c):
             newtext.append('\n')
         t.lexer.lineno += c
 
 # this needs to be HERE - not above token definitions
 ppLexer = lex.lex (debug=0)
 
     
 def preprocess (text, global_values={}, global_macros=[]):
     """
     Preprocess a C/C++ header file text
     
     Preprocesses h files - does #define substitutions and
     evaluates conditionals to include/exclude code. No
     substitutions are performed on preprocessor lines (any
     line beginning with '#'). Global #defines are applied
     LAST, so they override any local #defines.
 
     All C preprocessor code is stripped, and along with any
     lines eliminated conditionally, is replaced with newlines
     so that error messages still refer to the correct line in
     the original file.
     
     Arguments:
     text -- The text to process.
     global_values -- Dict mapping string variable names to values.
     global_macros -- List of tuples. The first value in a tuple is a
                      regular expression object. The second is that
                      replacement string which may contain re module
                      back references.
     
     Returns the processed string.
     """
     global newtext, bitBucket, macros, values
     newtext   = []
     bitBucket = False
     macros    = [] + global_macros
     values    = {}
         
     values.update (global_values)
     if text[-1]!='\n':
         text = text + '\n'
     ppLexer.input (text)
     token = ppLexer.token()
     #print(newtext)
     #return "".join (fixDoc (newtext))
     return "".join(newtext)
 
 def fixDoc (textList):
     doReplace = False
     doBackReplace = False
     nLines    = len (textList)
     for i in range (nLines):
         if i >= nLines - 1:
             break
             
         if textList [i].startswith ('/////'):
             textList [i] = '\n'
             continue
             
         haveBackCmt = textList [i].find ('///<') >= 0
         haveCmt = textList [i].find ('///') >= 0 and not haveBackCmt
         if haveBackCmt:
             if not doBackReplace:
                 doBackReplace = textList [i + 1].strip ().startswith ('///<')
                 if doBackReplace:
                     textList [i] = textList [i].replace ('///<', '/**<')
             else:
                 textList [i] = textList [i].replace ('///<', '*')            
         elif doBackReplace:
             textList.insert (i, '*/\n')
             doBackReplace = False
             
         if not haveBackCmt and haveCmt:
             if not doReplace:
                 doReplace = textList [i + 1].strip ().startswith ('///')
                 if doReplace:
                     textList [i] = textList [i].replace ('///', '/**')
             else:
                 textList [i] = textList [i].replace ('///', '*')
         elif doReplace:
             textList.insert (i, '*/\n')
             doReplace = False
             
     return textList
     
 if __name__ == '__main__':
     text = """#define foo bar"""