demaff.py

## For use with https://addons.mozilla.org/en-US/firefox/addon/212/
## If named to match title of page, this extracts zipped contents for source control usage
## inspired by {{{ http://code.activestate.com/recipes/465649/ (r3)
## 
import os, zipfile, sys, subprocess, platform
from cStringIO import StringIO

def extract( filename, dir ):
    zfn = filename + '.maff'
    print 'Processing: ' + zfn
    zf = zipfile.ZipFile( zfn )
    namelist = zf.namelist()
    # all entries in the maff are in a numbered directory 122123123223_222/ which we don't want.
    numbered_directory_name_len = len(namelist[0].split('/')[0])+1 
    filelist = filter( lambda x: not x.endswith( '/' ), namelist )
    # make base
    pushd = os.getcwd()
    if not os.path.isdir( dir ):
        os.mkdir( dir )
    os.chdir( dir )
    for fn in filelist:
        to_fn = fn[numbered_directory_name_len:]
        if to_fn.endswith('index.html'): to_fn = filename
        elif to_fn.endswith('index.rdf'): to_fn = filename.replace('.html','.rdf')
        if(to_fn.find('/') >= 0 and not os.path.isdir(os.path.dirname(to_fn))):
            os.makedirs(os.path.dirname(to_fn))
        out = open( to_fn, 'wb' )
        buffer = StringIO( zf.read( fn ))
        buflen = 2 ** 20
        datum = buffer.read( buflen )
        while datum:
            out.write( datum )
            datum = buffer.read( buflen )
        out.close()
        if to_fn.endswith('.html'):
           # sed is different on Mac and Linux :-(
           if platform.system() == 'Linux':
               subprocess.call(['sed', '-i', '-E', '-s', 's/^\s*//g', to_fn])
           if platform.system() == 'Darwin':
               subprocess.call(['sed', '-i', '.bkp', '-E', 's/^[[:space:]]*//g', to_fn])
    print '    with ' + str(len(filelist)-2) + ' references.'
    os.chdir( pushd )

def main():       
    extract(sys.argv[1].replace('.maff', ''), '.')

if __name__ == '__main__': 
    main()