""" Unicode Mapping Parser and Codec Generator.

This script parses Unicode mapping files as available from the Unicode
site (ftp://ftp.unicode.org/Public/MAPPINGS/) and creates Python codec
modules from them. The codecs use the standard character mapping codec
to actually apply the mapping.

Synopsis: gencodec.py dir codec_prefix

All files in dir are scanned and those producing non-empty mappings
will be written to <codec_prefix><mapname>.py with <mapname> being the
first part of the map's filename ('a' in a.b.c.txt) converted to
lowercase with hyphens replaced by underscores.

The tool also writes marshalled versions of the mapping tables to the
same location (with .mapping extension).

Written by Marc-Andre Lemburg ([email protected]).

(c) Copyright CNRI, All Rights Reserved. NO WARRANTY.
(c) Copyright Guido van Rossum, 2000.

Table generation:
(c) Copyright Marc-Andre Lemburg, 2005.
   Licensed to PSF under a Contributor Agreement.

"""#"

import re, os, time, marshal, codecs

# Maximum allowed size of charmap tables
MAX_TABLE_SIZE = 8192

# Standard undefined Unicode code point
UNI_UNDEFINED = unichr(0xFFFE)

mapRE = re.compile('((?:0x[0-9a-fA-F]+\+?)+)'
                  '\s+'
                  '((?:(?:0x[0-9a-fA-Z]+|<[A-Za-z]+>)\+?)*)'
                  '\s*'
                  '(#.+)?')

def parsecodes(codes,
              len=len, filter=filter,range=range):

   """ Converts code combinations to either a single code integer
       or a tuple of integers.

       meta-codes (in angular brackets, e.g. <LR> and <RL>) are
       ignored.

       Empty codes or illegal ones are returned as None.

   """
   if not codes:
       return None
   l = codes.split('+')
   if len(l) == 1:
       return int(l[0],16)
   for i in range(len(l)):
       try:
           l[i] = int(l[i],16)
       except ValueError:
           l[i] = None
   l = filter(lambda x: x is not None, l)
   if len(l) == 1:
       return l[0]
   else:
       return tuple(l)

def readmap(filename):

   f = open(filename,'r')
   lines = f.readlines()
   f.close()
   enc2uni = {}
   identity = []
   unmapped = range(256)

   # UTC mapping tables per convention don't include the identity
   # mappings for code points 0x00 - 0x1F and 0x7F, unless these are
   # explicitly mapped to different characters or undefined
   for i in range(32) + [127]:
       identity.append(i)
       unmapped.remove(i)
       enc2uni[i] = (i, 'CONTROL CHARACTER')

   for line in lines:
       line = line.strip()
       if not line or line[0] == '#':
           continue
       m = mapRE.match(line)
       if not m:
           #print '* not matched: %s' % repr(line)
           continue
       enc,uni,comment = m.groups()
       enc = parsecodes(enc)
       uni = parsecodes(uni)
       if comment is None:
           comment = ''
       else:
           comment = comment[1:].strip()
       if enc < 256:
           if enc in unmapped:
               unmapped.remove(enc)
           if enc == uni:
               identity.append(enc)
           enc2uni[enc] = (uni,comment)
       else:
           enc2uni[enc] = (uni,comment)

   # If there are more identity-mapped entries than unmapped entries,
   # it pays to generate an identity dictionary first, and add explicit
   # mappings to None for the rest
   if len(identity) >= len(unmapped):
       for enc in unmapped:
           enc2uni[enc] = (None, "")
       enc2uni['IDENTITY'] = 256

   return enc2uni

def hexrepr(t, precision=4):

   if t is None:
       return 'None'
   try:
       len(t)
   except:
       return '0x%0*X' % (precision, t)
   try:
       return '(' + ', '.join(['0x%0*X' % (precision, item)
                               for item in t]) + ')'
   except TypeError, why:
       print '* failed to convert %r: %s' % (t, why)
       raise

def python_mapdef_code(varname, map, comments=1, precisions=(2, 4)):

   l = []
   append = l.append
   if map.has_key("IDENTITY"):
       append("%s = codecs.make_identity_dict(range(%d))" %
              (varname, map["IDENTITY"]))
       append("%s.update({" % varname)
       splits = 1
       del map["IDENTITY"]
       identity = 1
   else:
       append("%s = {" % varname)
       splits = 0
       identity = 0

   mappings = map.items()
   mappings.sort()
   i = 0
   key_precision, value_precision = precisions
   for mapkey, mapvalue in mappings:
       mapcomment = ''
       if isinstance(mapkey, tuple):
           (mapkey, mapcomment) = mapkey
       if isinstance(mapvalue, tuple):
           (mapvalue, mapcomment) = mapvalue
       if mapkey is None:
           continue
       if (identity and
           mapkey == mapvalue and
           mapkey < 256):
           # No need to include identity mappings, since these
           # are already set for the first 256 code points.
           continue
       key = hexrepr(mapkey, key_precision)
       value = hexrepr(mapvalue, value_precision)
       if mapcomment and comments:
           append('    %s: %s,\t#  %s' % (key, value, mapcomment))
       else:
           append('    %s: %s,' % (key, value))
       i += 1
       if i == 4096:
           # Split the definition into parts to that the Python
           # parser doesn't dump core
           if splits == 0:
               append('}')
           else:
               append('})')
           append('%s.update({' % varname)
           i = 0
           splits = splits + 1
   if splits == 0:
       append('}')
   else:
       append('})')

   return l

def python_tabledef_code(varname, map, comments=1, key_precision=2):

   l = []
   append = l.append
   append('%s = (' % varname)

   # Analyze map and create table dict
   mappings = map.items()
   mappings.sort()
   table = {}
   maxkey = 0
   if map.has_key('IDENTITY'):
       for key in range(256):
           table[key] = (key, '')
       maxkey = 255
       del map['IDENTITY']
   for mapkey, mapvalue in mappings:
       mapcomment = ''
       if isinstance(mapkey, tuple):
           (mapkey, mapcomment) = mapkey
       if isinstance(mapvalue, tuple):
           (mapvalue, mapcomment) = mapvalue
       if mapkey is None:
           continue
       table[mapkey] = (mapvalue, mapcomment)
       if mapkey > maxkey:
           maxkey = mapkey
   if maxkey > MAX_TABLE_SIZE:
       # Table too large
       return None

   # Create table code
   for key in range(maxkey + 1):
       if key not in table:
           mapvalue = None
           mapcomment = 'UNDEFINED'
       else:
           mapvalue, mapcomment = table[key]
       if mapvalue is None:
           mapchar = UNI_UNDEFINED
       else:
           if isinstance(mapvalue, tuple):
               # 1-n mappings not supported
               return None
           else:
               mapchar = unichr(mapvalue)
       if mapcomment and comments:
           append('    %r\t#  %s -> %s' % (mapchar,
                                           hexrepr(key, key_precision),
                                           mapcomment))
       else:
           append('    %r' % mapchar)

   append(')')
   return l

def codegen(name, map, encodingname, comments=1):

   """ Returns Python source for the given map.

       Comments are included in the source, if comments is true (default).

   """
   # Generate code
   decoding_map_code = python_mapdef_code(
       'decoding_map',
       map,
       comments=comments)
   decoding_table_code = python_tabledef_code(
       'decoding_table',
       map,
       comments=comments)
   encoding_map_code = python_mapdef_code(
       'encoding_map',
       codecs.make_encoding_map(map),
       comments=comments,
       precisions=(4, 2))

   if decoding_table_code:
       suffix = 'table'
   else:
       suffix = 'map'

   l = [
       '''\
""" Python Character Mapping Codec %s generated from '%s' with gencodec.py.

"""#"

import codecs

### Codec APIs

class Codec(codecs.Codec):

   def encode(self,input,errors='strict'):
       return codecs.charmap_encode(input,errors,encoding_%s)

   def decode(self,input,errors='strict'):
       return codecs.charmap_decode(input,errors,decoding_%s)
''' % (encodingname, name, suffix, suffix)]
   l.append('''\
class IncrementalEncoder(codecs.IncrementalEncoder):
   def encode(self, input, final=False):
       return codecs.charmap_encode(input,self.errors,encoding_%s)[0]

class IncrementalDecoder(codecs.IncrementalDecoder):
   def decode(self, input, final=False):
       return codecs.charmap_decode(input,self.errors,decoding_%s)[0]''' %
       (suffix, suffix))

   l.append('''
class StreamWriter(Codec,codecs.StreamWriter):
   pass

class StreamReader(Codec,codecs.StreamReader):
   pass

### encodings module API

def getregentry():
   return codecs.CodecInfo(
       name=%r,
       encode=Codec().encode,
       decode=Codec().decode,
       incrementalencoder=IncrementalEncoder,
       incrementaldecoder=IncrementalDecoder,
       streamreader=StreamReader,
       streamwriter=StreamWriter,
   )
''' % encodingname.replace('_', '-'))

   # Add decoding table or map (with preference to the table)
   if not decoding_table_code:
       l.append('''
### Decoding Map
''')
       l.extend(decoding_map_code)
   else:
       l.append('''
### Decoding Table
''')
       l.extend(decoding_table_code)

   # Add encoding map
   if decoding_table_code:
       l.append('''
### Encoding table
encoding_table=codecs.charmap_build(decoding_table)
''')
   else:
       l.append('''
### Encoding Map
''')
       l.extend(encoding_map_code)

   # Final new-line
   l.append('')

   return '\n'.join(l).expandtabs()

def pymap(name,map,pyfile,encodingname,comments=1):

   code = codegen(name,map,encodingname,comments)
   f = open(pyfile,'w')
   f.write(code)
   f.close()

def marshalmap(name,map,marshalfile):

   d = {}
   for e,(u,c) in map.items():
       d[e] = (u,c)
   f = open(marshalfile,'wb')
   marshal.dump(d,f)
   f.close()

def convertdir(dir, dirprefix='', nameprefix='', comments=1):

   mapnames = os.listdir(dir)
   for mapname in mapnames:
       mappathname = os.path.join(dir, mapname)
       if not os.path.isfile(mappathname):
           continue
       name = os.path.split(mapname)[1]
       name = name.replace('-','_')
       name = name.split('.')[0]
       name = name.lower()
       name = nameprefix + name
       codefile = name + '.py'
       marshalfile = name + '.mapping'
       print 'converting %s to %s and %s' % (mapname,
                                             dirprefix + codefile,
                                             dirprefix + marshalfile)
       try:
           map = readmap(os.path.join(dir,mapname))
           if not map:
               print '* map is empty; skipping'
           else:
               pymap(mappathname, map, dirprefix + codefile,name,comments)
               marshalmap(mappathname, map, dirprefix + marshalfile)
       except ValueError, why:
           print '* conversion failed: %s' % why
           raise

def rewritepythondir(dir, dirprefix='', comments=1):

   mapnames = os.listdir(dir)
   for mapname in mapnames:
       if not mapname.endswith('.mapping'):
           continue
       name = mapname[:-len('.mapping')]
       codefile = name + '.py'
       print 'converting %s to %s' % (mapname,
                                      dirprefix + codefile)
       try:
           map = marshal.load(open(os.path.join(dir,mapname),
                              'rb'))
           if not map:
               print '* map is empty; skipping'
           else:
               pymap(mapname, map, dirprefix + codefile,name,comments)
       except ValueError, why:
           print '* conversion failed: %s' % why

if __name__ == '__main__':

   import sys
   if 1:
       apply(convertdir,tuple(sys.argv[1:]))
   else:
       apply(rewritepythondir,tuple(sys.argv[1:]))

You are viewing proxied material from sdf.org. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.