mkheader.py

convert C programs for use in C++ compiler environment

#!/usr/bin/env python

"""
   Make Header
   C to C++ conversion tools (c) 2001-2003 by D.G. Sureau

   This script builds or completes the .h header file to include,
   from a C source file.
   Uses a list of files, or processes recursively all files from the
   main file of a project.
   C++ files are no handled.
   This script should be used before converting a C program to C++,
   or to update automatically header files when changes are performed
   in a C source.

   Syntax is: python mkheader.py filename.c

   Requires:  - the python interpreter (look at www.python.org)
              - wstring.py, pattern.py, lexer.py (included)

   Processing:
   - A prototype is created for functions which don't has one.
   - Prototypes are corrected to match definitions of functions.
   - Multi-lines declarations are changed to one-lines ones.
   - Variables defined as static are declared static in header.
     They are not added to header if not already present since they
     are local to the file.
   - Other variables (no static) are added to the header if
     not already declared, to become attributes at class making.
   - Prototypes and declaration of variables are corrected in
     the header file to match the definitions in the source file.

   License:
    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., 675 Mass Ave, Cambridge, MA 02139, USA.

    [e-mail]
    http://www.scriptol.org

"""


import sys
import os
import string
import wstring
import pattern
import lexer

DEBUG = 0

TRUE  = 1
FALSE = 0


# --- Processing a c source file and its header file

def processfile(cfile):
 global extension

 node, ext = os.path.splitext(cfile)
 headername = node + ".h"

 print "mkheader - processing", cfile, "and" , headername

 infile  = open(cfile, "rb")
 if infile is None:
   print "Enable to open", cfile
   sys.exit(0)



 varlist = []          # list of declarations
 funclist = []         # list of prototypes
 defdict = {}          # define statements may be needed
 defflag = {}          # flags for things to be inside header
 defline = {}          # the original line
 linecount = 0
 defcount = 0


 # Pass 1 - Processing the C file
 # ------------------------------
 # make a list of
 # - each function definition
 # - each global variable declaration (simple or array)
 # - make also a dictionary of define statements

 COMMENT = FALSE      # This flag is true inside a comment
 INSBLK = FALSE      # This one inside a function's body


 while(1):
   line = infile.readline()
   if not line: break

   linecount = linecount + 1
   line = wstring.chop(line)    # Removing line separators
   line = wstring.strip(line)   # Removing spaces (with better strip)

   if len(line) == 0: continue     # Empty line ignored

   # If we enter a comment block, remove it
   if not COMMENT: COMMENT = lexer.opencomment(line)
   if COMMENT:
     i = string.find(line, "/*")
     left = line[:i]
     while COMMENT:
       COMMENT = lexer.closecomment(line)
       if COMMENT:
         line = infile.readline()
         if not line: COMMENT = FALSE
         linecount = linecount + 1
       if not COMMENT:
         i = string.find(line, "*/")
         right = line[i + 2:]
     line = left + right

   # If we are inside a compound block, skip the content

   # print linecount, ') [' + str(lexer.blocklevel) + ']', line

   if not INSBLK: INSBLK = lexer.openblock(line)
   if INSBLK:
     i = string.find(line, '{')
     left = line[:i]                   # Getting the code at left of {
     while(INSBLK):
       INSBLK = lexer.closeblock(line)
       if INSBLK:
         line = infile.readline()      # Skipping the content
         linecount = linecount + 1
         # print "skiping", lexer.blocklevel,  wstring.chop(line)
         if not line: INSBLK = FALSE
       if not INSBLK:
         i = string.find(line, '}')
         right = line[i + 1:]          # Getting the code at right of }
     line = left + right
   # "shortened" may change inside the call of the test functions

   if line is None: continue
   if line == "": continue

   # A define statement
   if line[0] == '#':      # Define or pragma statement
     if line[0:7] == "#define":
       key, value = lexer.splitdefine(line)
       if key != None:
         defdict[key] = value         # Add to dictionary
         defflag[key] = FALSE         # By default not to be included in header
         defline[key] = line
         #print "define added", line, "value=", value
     continue

   if not lexer.isdefinition(line):  continue

   #print "is typed", line

   # Simple keywords of type on the line
   if lexer.typeonly(line):
     while(1):
      next = infile.readline()
      if not next: break
      next = wstring.clean(next)
      if next == "": continue
      linecount = linecount + 1
      line = line + ' ' + next
      break
     

   # A function prototype: ignore it
   if lexer.isprototype(line): continue

   # A function definition
   if lexer.isfunction(line) == TRUE:
     line = lexer.shortened
     #print "is function ", line

     # The heading may use several lines
     while string.find(line, ')') == -1:
      next = infile.readline()
      linecount = linecount + 1
      if next == None:
         print "Error, function uncomplete"
         exit()
      next = wstring.chop(next)
      next = wstring.strip(next)
      line = line + next


     # Handling the old format with declarations on next lines
     if lexer.oldformat(line):
       #print "old format", line
       tlist = []
       while(1):
        l = infile.readline()
        if not l: break
        linecount = linecount + 1
        l = wstring.chop(l)
        l = wstring.strip(l)
        l = lexer.getdecl(l)
        if lexer.isvardef(l):
         tlist.append(l)
        else: break
        if '{' in l: break

       # Rebuilding the heading according to the new format
       line = lexer.argreplace(line, tlist)
       #print "args replace", line, "with list", tlist

     i = string.find(line, '{')
     if i != -1: line = line[:i]
     line = lexer.removecomment(line)
     line = lexer.addsemicolon(line)
     #print "append to funclist", line

     # There is no prototype for the main function
     funcname = lexer.getident(line)
     if funcname is None:
       print "Error in (" + linecount + ')', line
       sys.exit(0)
     funcname = string.lower(funcname)
     if funcname == "main": continue

     funclist.append(line)
     defcount = defcount + 1
     #print line
     continue

   # An array declaration
   # is static, ignored
   # Removing from = to end of line
   # Adding each size parameter to defdict (done in processarray)
   # if initializer, become static for C++
   line = lexer.getdecl(line)       # removing extra codes

   if lexer.isarray(line):
      #print "is array", line
      # Udate defflag
      vlist = []
      if lexer.multivar(line):
        vlist = lexer.splitvardef(line)
        #print "multivar", vlist
      else:
        vlist.append(line)
      for vardef in vlist:
        j = string.find(vardef, '=')
        k = string.find(vardef, ';')
        if k == -1: k = len(vardef)
        if j < k:
          vardef = vardef[:j] + ';'

        lexer.processarray(vardef, defdict, defflag)
        var = lexer.shortened
        #print "in main shortened", var
        var = lexer.addextern(var)
        varlist.append(var)
        #print "added array", var
        defcount = defcount + 1
        #print line
      continue


   # A variable declaration
   if lexer.isvardef(line):
      #print "255 var", line
      vlist = []
      if lexer.multivar(line):
        vlist = lexer.splitvardef(line)
        #print "multivar", vlist