root/confluence/feedfilters/onemanga.py

#!/usr/bin/env python

# This is a generic filter for selecting items in an rss feed on the basis of subelement blacklists and whitelists
#
# Written by Adrianna Pinska
# Licence: GPLv3

import sys
import re
import xml.dom.minidom

include = {
    "title": ["Ressentiment", "Pluto", "Vinland Saga", "Twin Spica", "Historie", "Full Metal Alchemist", "Billy Bat", "Cherry", "Cesare"],
}

exclude = {
}

whitelist = {}
blacklist = {}

for element, lst in include.items():
    whitelist[element] = re.compile("(%s)" % "|".join(lst))
for element, lst in exclude.items():
    blacklist[element] = re.compile("(%s)" % "|".join(lst))

elements_of_interest = set(whitelist.keys()+blacklist.keys())

xml = xml.dom.minidom.parseString(sys.stdin.read())
channel = xml.getElementsByTagName("rss")[0].getElementsByTagName("channel")[0]
for item in channel.getElementsByTagName("item"):
    elements = {}
    for element in elements_of_interest:
        elements[element] = item.getElementsByTagName(element)[0].firstChild.data

    if whitelist:
        allowed = False
        for element, regex in whitelist.items():
            if regex.search(elements[element]):
                allowed = True
                break
        if not allowed:
            channel.removeChild(item)
            item.unlink()
            continue

    for element, regex in blacklist.items():
        if regex.search(elements[element]):
            channel.removeChild(item)
            item.unlink()
            continue

text = xml.toxml(encoding="UTF-8")
text = re.sub(" *\n","\n",text)
text = re.sub("\n+","\n",text)
print text