#!/usr/bin/python
'''Newline converter

 - Converts \\r and \\r\\n line endings into \\n line endings*
 - Removes byte order marks (BOM)

* When run on UNIX or Mac OS X (not tested on Windows)

# License: GNU General Public License, see http://www.clips.ua.ac.be/~vincent/scripts/LICENSE.txt
'''

__author__ = "Kim Luyckx, Vincent Van Asch"
__date__ = "July 2011"
__version__="1.0"
import sys, os
import getopt
from codecs import BOM_BE, BOM_LE, BOM_UTF8

def _usage():
    print >>sys.stderr, '''
Convert line endings (version %s)
USAGE:
    python ./reline.py [-b] [-n] inputfile outputfile
    
    Converts \\r and \\r\\n line endings into \\n line endings and 
    removes the byte order mark (BOM) at the start of the file.
    
OPTIONS
    -b : If present: Preserve byte order marks for UTF8, UTF16 (LE and BE)
         on the first line.
    -n: Leave the newlines as they are
    
NOTE
    - Setting -b and -n simultaneously means doing nothing.
    - Tested UNIX ans Mac OS X; not tested on Windows.
    
%s, %s
''' %(__version__, __author__, __date__)
    
if __name__ == "__main__":
    try:
        opts,args=getopt.getopt(sys.argv[1:],'hbn', ['help'])
    except getopt.GetoptError:
        # print help information and exit:
        _usage()
        sys.exit(2)

    remove_byteorder=True
    mode = "rU"

    for o, a in opts:
        if o in ('-h', '--help'):
            _usage()
            sys.exit()
        if o in ('-b',):
            remove_byteorder=False
        if o in ('-n',):
            mode="r"

    if len(args) == 2:
        infile = os.path.expanduser(args[0])
        outfile = os.path.expanduser(args[1])
    
        f=open(infile, mode)
        input=f.readlines()
        f.close()
        
        if remove_byteorder:
            input[0] = input[0].lstrip(BOM_BE)
            input[0] = input[0].lstrip(BOM_LE)
            input[0] = input[0].lstrip(BOM_UTF8)

        out=open(outfile,"w")
        out.write("".join(input))
        out.close()
    else:
        _usage()
        sys.exit(1)