#!/usr/bin/env python
'''transcode.py

Script to convert a textfile in one encoding into another encoding.

Copyright (c) 2011 CLiPS. All rights reserved.

# License: GNU General Public License, see http://www.clips.ua.ac.be/~vincent/scripts/LICENSE.txt
'''
__author__="Vincent Van Asch"
__date__="April 2011"
__version__ = '1.1'
import codecs, os, sys, getopt, re


def transcode(fname, source, target, verbose=False, rewhite=False):
    '''
    fname: file to transcode
    source: encoding of the file
    target: encoding of the output file
   
    Creates a file: fname.target
    '''
    name = os.path.abspath(os.path.expanduser(fname))
    
    try:
        f=codecs.open(name, 'rU', source)
    except LookupError:
        print >>sys.stderr, 'ERROR: encoding "%s" is not known.' %source
        return
        
    txt=None
    try:
        txt = f.read()
    except UnicodeDecodeError:
        print >>sys.stderr, 'ERROR: Could not open "%s" with source encoding "%s". File has probably another encoding.' %(os.path.basename(name), source)
    finally:
        f.close()
    
    if txt is None: return
    
    
    name = name+'.'+target
    try:
        f=codecs.open(name, 'w', target)
    except LookupError:
        print >>sys.stderr, 'ERROR: encoding "%s" is not known.' %target
        os.remove(name)
        return
        
    try:
        if rewhite:
            if verbose: print >>sys.stderr, '''Replacing all (multiple) whitespaces with %s and newlines with '\\n'.''' %repr(rewhite) 
            # Split in lines
            lines = txt.split('\r')
            
            lines2=[]
            for line in lines:
                lines2.extend(line.split('\n'))
            lines=lines2[:]
            del lines2
            
            # Split in part, join and write
            for line in lines:
                line = re.sub('\s+', ' ', line)
                f.write(rewhite.join(line.split()))
                f.write('\n')
        else:
            f.write(txt)
    except UnicodeEncodeError:
        print >>sys.stderr, 'ERROR: Could not write "%s" with target encoding "%s" because of unavailable characters. Try another target encoding.' %(os.path.basename(name), target)
        os.remove(name)
        return
    finally:
        f.close()
       
    if verbose: print >>sys.stderr, 'Transcoded file from "%s" to "%s": %s' %(source, target, name)
    return name
    
    
def _usage():
    print >>sys.stderr, '''Convert a file into another encoding (version %s)
    
USAGE
    ./transcode.py [-s enc] [-t enc] [-v] [-r] fname1 [fname2 ...]
    
OPTIONS
    -s enc: The encoding of the source file. (default: latin1)
    -t enc: The encoding of the target file. (default: utf8)
    -S    : Normalize whitespace to a single space and newlines to "\n"
    -T    : Normalize whitespace to a tab and newlines to "\n" 
    -v    : Print some info
    
    A list of available encodings can be found at:
    http://docs.python.org/library/codecs.html#standard-encodings
    
    
%s''' %(__version__, __date__)
       
       
       
if __name__ == '__main__':
    try:
        opts,args=getopt.getopt(sys.argv[1:],'hs:t:vST', ['help', 'source=', 'target=', 'verbose', 'rewhite', 'retab'])
    except getopt.GetoptError:
        # print help information and exit:
        _usage()
        sys.exit(2)
    
    
    source='latin1'
    target='utf8'
    verbose=False
    rewhite = False
    
    for o, a in opts:
        if o in ('-h', '--help'):
            _usage()
            sys.exit()
        if o in ('-s', '--source'):
            source = a
        if o in ('-t', '--target'):
            target = a
        if o in ('-v', '--verbose'):
            verbose=True
        if o in ('-S', '--rewhite'):
            rewhite = ' '
        if o in ('-T', '--retab'):
            rewhite = '\t'
        
            
    if len(args) < 1:
        _usage()
        sys.exit(1)
            
            
    for fname in args:
        transcode(fname, source, target, verbose=verbose, rewhite=rewhite)