#!/home2/depalati/python/bin/python
"""genfeed.py version 0.1.0 (27 July 2009)

Generates an rss or atom feed from specified directories. The feed
filename will be <directory>.xml. This requires the rfc3339 module
which can be downloaded from
http://henry.precheur.org/2008/9/3/RFC_3339_formatting_in_Python.html.

At present this is a lie. Only atom feeds can be generated.

genfeed.py is free software and distributed under the GNU GPL v. 2.
"""

import os, os.path
import re
import shelve
from datetime import datetime
from rfc3339 import rfc3339

## Settings
#############

# Path to check for updates as well as the URL of the path
path = "/home2/depalati/mike/docs"
path_url_base = "http://mike.depalatis.net/docs"

# Maximum number of entries per feed.
max_feed_items = 20

# Set to True to use RSS 2.0 instead of Atom 1.0.
use_rss = False

# Feed title, author, etc. settings
title = "Unblog"
author = "Michael V. DePalatis"
email = "mvd@fake.email.com"
url = "http://mike.depalatis.net/docs.html"
feed_url = "http://mike.depalatis.net/docs.xml"

## Functions
##############

def sanitize(text):
    """Changes < and > characters to &lt; and &gt; respectively in
    order to allow for valid XML."""
    gt = re.compile(">")
    lt = re.compile("<")
    return gt.sub("&gt;", lt.sub("&lt;", text))

def genFeed(path, use_rss=False):
    """Generates the feed for the directory path and returns it as an
    xml string. If use_rss is True, the returned string will be in RSS
    2.0 instead of Atom 1.0."""
    xml = ''
    re_update = re.compile(r"<!--UPDATE-->", re.IGNORECASE)

    # Get a list of files in the path and sort based on their
    # modification time.
    contents = os.listdir(path)
    entries = []
    for entry in contents:
        if entry == "header.txt" or entry == "footer.txt":
            continue
        if entry[-3:].lower() == "txt":
            entries.append(entry)
    entries = sorted(entries, key=lambda f: os.path.getmtime(path + "/" + f))
    entries.reverse()

    # Open entry database for entry times
    entry_db = shelve.open(path + "/entries.db")

    # Generate XML data.
    if use_rss:
        # Not yet implemented.
        pass
    else:
        # Feed-wide stuff
        xml += """<?xml version="1.0" encoding="utf-8"?>
        <feed xmlns="http://www.w3.org/2005/Atom">
        <title>%s</title>
        <link href="%s" />
        <link href="%s" rel="self" />
        <author><name>%s</name></author>
        <updated>%s</updated>
        <id>%s</id>
        """ % (title, url, feed_url, author, rfc3339(datetime.today()), url)
        limit = max_feed_items

        # Per entry stuff
        if len(entries) < max_feed_items:
            limit = len(entries)
        for j in range(limit):
            entry = entries[j]
            entry_file = open(path + "/" + entry, 'r')
            entry_title = entry_file.readline()[2:]
            entry_link = path_url_base + "/" + entry[:-3] + "html"
            entry_text = entry_file.read(800) + "..."
            do_update = not re.search(re_update, entry_text)
            if entry_db.has_key(entry) and do_update:
                updated = entry_db[entry]
            else:
                updated = rfc3339(os.path.getmtime(path + "/" + entry))
                if do_update:
                    entry_db[entry] = updated
            entry_file.close()
            xml += """<entry>
            <title>%s</title>
            <link href="%s"/>
            <id>%s</id>
            <updated>%s</updated>
            <summary>%s</summary>
            </entry>
            """ % (entry_title, entry_link, entry_link,
                   updated, sanitize(entry_text))
        xml += "</feed>\n"
    entry_db.close()
    return xml

## Main
#########

if __name__ == "__main__":
    print "Updating feed..."
    xml = genFeed(path)
    xml_file_name = os.path.basename(os.path.normpath(path)) + ".xml"
    xml_file = open(xml_file_name, 'w')
    xml_file.write(xml)
    xml_file.close()