#!/usr/bin/env python

from feedhelpers import newChild, rssDate, atomDate
from filehelpers import all_files
import email.Utils, os, re, string, sys, time
import cElementTree as ElementTree

if __name__ == '__main__':

    usage = """
    USAGE: %s

    """ % os.path.basename(sys.argv[0])

    # put feed .xml files in ./
    feed_dir = ''

    # tarballs are in ./copyright/bulk
    tgz_dir = 'copyright/bulk'

    # number of tarballs to list in feed
    items_to_show = 50

    path_list = []
    for path in all_files(tgz_dir, 'hid*tgz'):
        path_list.append(path)

    if len(path_list) == 0:
        sys.exit()

    # Reverse order of files, so newest are first
    path_list.reverse() 

    # Common feed elements
    feed_title = "copyright.gov bulk feed by public.resource.org"
    feed_link = "http://public.resource.org/"
    feed_copyrights = "public domain"
    feed_imageurl = "http://public.resource.org/seal.head.jpg"
    feed_description = string.join([
        "Latest registrations in the United States Copyright Office ",
        "Catalog of Registrations (1978-present). For bulk data to 1978, see ",
        "http://bulk.resource.org/copyright/hids/ . There is also an ",
        "up-to-the-minute feed of new records as they are added, go to ",
        "http://rss.resource.org/ for this and our other feeds."], '')
    now_tstamp = os.path.getmtime(path_list[0])


    # RSS channel elements
    now_rss = rssDate(now_tstamp) # RFC 822
    rss = ElementTree.Element("rss")
    rss.set("version", "2.0")
    rss_c = ElementTree.SubElement(rss, "channel")
    newChild(rss_c, "title", feed_title)
    newChild(rss_c, "link", feed_link)
    newChild(rss_c, "description", feed_description) 
    newChild(rss_c, "copyright", feed_copyrights)
    newChild(rss_c, "pubDate", now_rss)
    newChild(rss_c, "lastBuildDate", now_rss)
    newChild(rss_c, "docs", "http://www.rssboard.org/rss-specification")
    newChild(rss_c, "ttl", "5")
    rss_im = newChild(rss_c, "image")
    newChild(rss_im, "url", feed_imageurl)
    newChild(rss_im, "title", "The Seal of Approval")
    newChild(rss_im, "link", "http://public.resource.org/")

    # Atom channel elements
    now_atom = atomDate(now_tstamp) # RFC 3339
    url_atom = 'http://rss.resource.org/copyright.bulk.atom.xml'
    atom = ElementTree.Element("feed")
    atom.set("xmlns", "http://www.w3.org/2005/Atom")
    newChild(atom, "title", feed_title)
    newChild(atom, "link", None, {'title':'public.resource.org homepage','rel':'related','href':feed_link})
    newChild(atom, "link", None, {'title':'this feed','rel':'self','href':url_atom})
    atom_a = newChild(atom, "author")
    newChild(atom_a, "name", "public.resource.org")
    newChild(atom_a, "uri", feed_link)
    newChild(atom, "id", url_atom)
    newChild(atom, "updated", now_atom)
    newChild(atom, "logo", feed_imageurl)
    newChild(atom, "icon", feed_imageurl)
    newChild(atom, "rights", feed_copyrights)
    newChild(atom, "subtitle", feed_description)
    
    # Add items
    items_shown = 0
    for path in path_list:
        items_shown = items_shown + 1
        if items_shown > items_to_show:
            break
        # Common item vars
        item_filename = os.path.basename(path)
        item_link = "http://rss.resource.org/" + path
        item_desc = "Archive of copyright records from copyright.gov: " + item_filename
        item_tstamp = os.path.getmtime(path)
        item_size = str(os.path.getsize(path))

        # RSS item
        rss_it = newChild(rss_c, "item")
        newChild(rss_it, "title", item_filename)
        newChild(rss_it, "link", item_link)
        newChild(rss_it, "enclosure", None, {'url':item_link, 'length':item_size, 'type':'application/x-tar-gz'})
        rss_it_desc = newChild(rss_it, "description", item_desc)
        newChild(rss_it, "guid", item_link)
        newChild(rss_it, "pubDate", rssDate(item_tstamp))

        # Atom entry
        atom_ent = newChild(atom, "entry")
        newChild(atom_ent, "title", item_filename)
        newChild(atom_ent, "id", item_link)
        newChild(atom_ent, "updated", atomDate(item_tstamp))
        newChild(atom_ent, "content", item_desc, {'type':'text'})
        newChild(atom_ent, "link", None, {'title':'Tarball of copyright records in Latin-1 MARC format', 'type':'application/x-tar-gz', 'rel':'enclosure', 'length':item_size, 'href':item_link})

    rss_tree = ElementTree.ElementTree(rss)
    rss_tree.write("copyright.bulk.rss.xml", encoding="utf-8")

    atom_tree = ElementTree.ElementTree(atom)
    atom_tree.write("copyright.bulk.atom.xml", encoding="utf-8")