#!/usr/bin/env python
"""Usage: xhtmlforward [FILE]
Reformats XHTML, as generated by XML tools like XSLT processors, to make it 
work in browsers that do not handle XML properly. Reads from FILE or standard 
input and writes to standard output. More specifically:

1. Inserts a space before the closing slash in any empty-element shorthand tag 
declared EMPTY in the specs, so that:

  <img src="popotan.jpg" alt="Puni puni ha petan!"/>
  <hr/>

is rewritten to:
     
  <img src="popotan.jpg" alt="Puni puni ha petan!" />
  <hr />

2. Rewrites any other empty-element shorthand tag to a start-tag immediately 
followed by an end-tag, so that:

  <script type="text/javascript" src="sidebar.js"/>

is rewritten to:

  <script type="text/javascript" src="sidebar.js"></script> 
"""

__version__ = "0.91"
__author__ = "Thijs van der Vossen <http://www.vandervossen.net/>"
__copyright__ = "Copyright 2003, Fingertips <http://www.fngtps.com/>"
__license__ = "Python"
__history__ = """
0.91 - 19/10/2003 - Thijs - fixed buffer bug
0.9 - 8/8/2003 - Thijs - first beta release
"""

# non-deprecated empty html elements
empty = ["area", "base", "br", "col", "frame", "hr", "img", "input", "link", 
         "meta", "param"]

def reformat(data):
    """Quick convenience function for reformatting xhtml in one go."""
    xhtmlfilter = filter()
    xhtmlfilter.write(data)
    return xhtmlfilter.read()

class filter:
    """Filter for reformatting XHTML to make it more backwards-compatible."""
    def __init__(self):
        self.buffer = []
        self.tag = 0
        self.tagname = ""

    def write(self, data):
        """Writes xhtml input to the filter."""
        for char in data:
            if char == "<":
                self.tag = 1
            elif self.tag and not self.tagname and (char.isspace() or char == "/"):
                self.tagname = "".join(self.buffer[-self.tag+2:]) 
            elif self.tag and char == ">":
                if self.buffer[-1] == "/":
                    if self.tagname not in empty:
                        self.buffer[-1] = "></"
                        self.buffer.append(self.tagname)
                    else:
                        if not self.buffer[-2].isspace():
                            self.buffer[-1] = " "
                            self.buffer.append("/")
                self.tag = 0  
                self.tagname = ""
            if self.tag:
                self.tag += 1   

            self.buffer.append(char)

    def read(self):
        """Reads reformatted xhtml output from the filter."""
        if self.tag:
            r = "".join(self.buffer[:-self.tag])
            self.buffer = self.buffer[-self.tag:]
        else:
            r = "".join(self.buffer)
            self.buffer = []
        return r

if __name__ == "__main__":
    import sys
    import getopt

    try:
        opts, args = getopt.getopt(sys.argv[1:], "h", ["help"])
    except getopt.error, msg:
        print msg
        print "for help use --help"
        sys.exit(2)
    for o, a in opts:
        if o in ("-h", "--help"):
            print __doc__
            sys.exit(0)

    try:
        infile = open(args[0])
    except IndexError:
        infile = sys.stdin

    xhtmlfilter = filter()
    while 1:
        input = infile.read(4096)
        xhtmlfilter.write(input)
        sys.stdout.write(xhtmlfilter.read())
        if not input:
            break