#! /usr/bin/env python
import sys
import getopt
import os, os.path
import re
from zipfile import ZipFile, ZIP_STORED, ZIP_DEFLATED

def sanitize_html( html ):
    '''
    Probably lots of opporunities for others here, too
    '''
    return re.sub(r'<meta name="Adept.resource" [^>]*>','',html)

def main(args):
    (opts, optargs) = getopt.getopt( args[1:], 'f' )
    overwrite_ok = False
    for o, oa in opts:
        if '-f' == o:
            overwrite_ok = True
    epubfn = optargs[0]

    zf = ZipFile(epubfn)

    newdn = os.path.dirname( epubfn )
    newfn =  '_clean_%s' % os.path.basename(epubfn)
    epubfn2 = os.path.join( newdn, newfn )
    del newdn, newfn

    if os.path.exists( epubfn2 ) and not overwrite_ok:
        print "I will not overwrite %s" % epubfn2
        return 1
    zf2 = ZipFile( epubfn2 ,'w')

    html_re = re.compile(r'\.x?html?$')
    interesting_ext = ['.ncx','.opf','.xml']
    for entry in zf.infolist():
        name = entry.filename
        name_lc = name.lower()
        for ext in interesting_ext:
            if name_lc.endswith( ext ):
                print "Confirm(%s):" % name, zf.read(name)
                print ''
        if not html_re.search(name.lower()):
            if 'mimetype' == name:
                entry.compress_type = ZIP_STORED
            else:
                entry.compress_type = ZIP_DEFLATED
            zf2.writestr( entry, zf.read(name) )
            continue
        print "Sanitizing entry",name
        html = zf.read(name)
        html = sanitize_html( html )
        zf2.writestr( entry, html )
    zf2.close()
    zf.close()
    return 0

if __name__ == '__main__': sys.exit(main(sys.argv))