#!/usr/bin/env python
"""Parse active NOAA NWS CAP advisories and output an ATOM feed"""

__author__ = 'Daniel Washburn'
__copyright__ = 'Copyright (c) 2013, 2014, All rights reserved.'
__license__ = 'http://opensource.org/licenses/BSD-3-Clause'

import cgi
import getopt
import sys
from lxml import etree
from lxml.etree import SubElement

ATOM_URI = 'http://www.w3.org/2005/Atom'
ATOM_NAMESPACE = {'atom': ATOM_URI}
CAP_URI = 'urn:oasis:names:tc:emergency:cap:1.1'
CAP_NAMESPACE = {'cap': CAP_URI}
CAP_SCHEMA_URL = 'http://docs.oasis-open.org/emergency/cap/v1.1/errata/approved/cap.xsd'
CAP_XSLT_URL = 'capatomcustom.xsl'

CAP_SCHEMA = etree.XMLSchema(etree.parse(CAP_SCHEMA_URL))
CAP_XSLT = xslt_transform = etree.XSLT(etree.parse(CAP_XSLT_URL))

ATOM_PARSER = etree.XMLParser(ns_clean=True, remove_comments=True)
CAP_PARSER = etree.XMLParser(ns_clean=True, remove_comments=True,
        remove_blank_text=True, schema=CAP_SCHEMA)

def usage():
    print """nwscap.py [-h] -t (region|zone) -a area

-h|--help		Print this message and exit
-t|--type		Area type. must be either region or zone
-a|--area		The area to report on
"""

try:
    opts, args = getopt.gnu_getopt(sys.argv[1:],
            "ht:a:",
            ["help", "type=", "area="])
except getopt.GetoptError as err:
    print str(err)
    usage()
    sys.exit(2)
area_type=False
area=False
url=False

for o, a in opts:
    if o in ("-h", "--help"):
        usage()
        sys.exit()
    elif o in ("-t", "--type"):
        if a in ("region", "zone"):
            area_type = a
        else:
            print """type must be one of: region, zone"""
            usage()
            sys.exit()
    elif o in ("-a", "--area"):
        area = a
    else:
        assert False, "unhandled option"

if area_type and area:
    if area_type == "region":
        url='https://alerts.weather.gov/cap/' + area + '.php?x=0'
    elif area_type == "zone":
        url='https://alerts.weather.gov/cap/wwaatmget.php?x=' + area + '&y=0'
else:
    print """Both type and area must be specified"""
    usage()
    sys.exit()

in_feed = etree.parse(url, ATOM_PARSER)

out_feed = etree.XML('''<?xml version='1.0' encoding='UTF-8' standalone='yes'?>
        <feed xmlns='http://www.w3.org/2005/Atom'/>''')
SubElement(out_feed, 'id').text = "%s" % \
        in_feed.xpath('/atom:feed/atom:id',
        namespaces=ATOM_NAMESPACE)[0].text
SubElement(out_feed, 'updated').text = "%s" % \
        in_feed.xpath('/atom:feed/atom:updated',
        namespaces=ATOM_NAMESPACE)[0].text
out_feed_author = SubElement(out_feed, 'author')
SubElement(out_feed_author, 'name').text = "%s" % \
        in_feed.xpath('/atom:feed/atom:author/atom:name',
        namespaces=ATOM_NAMESPACE)[0].text
SubElement(out_feed, 'title').text = "%s" % \
        in_feed.xpath('/atom:feed/atom:title',
        namespaces=ATOM_NAMESPACE)[0].text
SubElement(out_feed, 'link', href="%s" % \
        in_feed.xpath('/atom:feed/atom:link/@href',
        namespaces=ATOM_NAMESPACE)[0])

# get URLs of active alerts
entry_titles = in_feed.xpath('/atom:feed/atom:entry/atom:title',
        namespaces=ATOM_NAMESPACE)
alert_links=in_feed.xpath('/atom:feed/atom:entry/atom:link/@href',
        namespaces=ATOM_NAMESPACE)

no_alerts_title = 'There are no active watches, warnings or advisories'
if no_alerts_title in entry_titles[0].text:
    alert_links = []

for link in alert_links:
    cap=etree.parse(link, CAP_PARSER)
    cap_entry = SubElement(out_feed, 'entry')
    SubElement(cap_entry, 'id').text = "%s" % \
            cap.xpath('/cap:alert/cap:identifier',
                namespaces=CAP_NAMESPACE)[0].text

    cap_entry_author = SubElement(cap_entry, 'author')

    SubElement(cap_entry_author, 'name').text = "%s" % \
            cap.xpath('/cap:alert/cap:sender',
                namespaces=CAP_NAMESPACE)[0].text

    SubElement(cap_entry, 'title').text = "%s" % \
            cap.xpath('/cap:alert/cap:info/cap:headline',
                namespaces=CAP_NAMESPACE)[0].text

    SubElement(cap_entry, 'published').text = "%s" % \
            cap.xpath('/cap:alert/cap:sent',
                namespaces=CAP_NAMESPACE)[0].text

    SubElement(cap_entry, 'link', href=link)

    SubElement(cap_entry, "content", type="text/html").text = "%s" % \
            etree.tostring(xslt_transform(cap), pretty_print=True)

out_tree = etree.ElementTree(out_feed)
print(etree.tostring(out_tree, pretty_print=True, method='xml'))