#!/usr/bin/env python

# vim: set fileencoding=utf-8 tabstop=4 shiftwidth=4 expandtab smartindent:

u"""

Files to Feed

-------------

Create an ATOM feed from the files in a directory

.. :Authors:

      Aurélien Bompard <aurelien@bompard.org> <http://aurelien.bompard.org>

.. :License:

      GNU GPL v3 or later

"""

import os

import sys

import socket

import datetime

import hashlib

import cgi

import urllib

import optparse

def main(directory, options):

    feed = open(options.output, "w")

    feed.write("""<?xml version="1.0" encoding="utf-8" ?>

<feed xmlns="http://www.w3.org/2005/Atom">

    <title>%(title)s</title>

    <link rel="self" href="%(address)s/%(feed_name)s" />

    <updated>%(date)s</updated>

    <id>%(address)s</id>

""" % { "feed_name": os.path.basename(options.output),

        "title": options.title,

        "address": options.url,

        "date": datetime.datetime.now().strftime("%Y-%m-%dT%H:%M:%SZ")})

    files_raw = []

    for root, dirs, cur_files in os.walk(directory):

        for subdir in dirs:

            # Exclude dirs

            if subdir in options.exclude or subdir.startswith("."):

                del dirs[dirs.index(subdir)]

                if options.debug:

                    print "excluding", subdir

                continue

            # Recurse into symlinks to dirs (included in python >= 2.6)

            if os.path.islink(os.path.join(root,subdir)):

                for subroot, subsubdirs, subcur_files in \

                        os.walk(os.path.join(root,subdir)):

                    for subsubdir in subsubdirs:

                        # Exclude dirs

                        if subsubdir in options.exclude \

                                or subsubdir.startswith("."):

                            del subsubdirs[subsubdirs.index(subsubdir)]

                            if options.debug:

                                print "excluding", subsubdir

                            continue

                    for file in subcur_files:

                        files_raw.append(os.path.join(subroot,file))

        for file in cur_files:

            files_raw.append(os.path.join(root,file))

    files = []

    for file in files_raw:

        if os.path.basename(file) in options.exclude \

                or os.path.basename(file).startswith("."):

            continue

        if os.path.basename(file) == os.path.basename(sys.argv[0]):

            continue

        if os.path.basename(file) == os.path.basename(options.output):

            continue

        try:

            date = os.path.getmtime(file)

            size = os.path.getsize(file)

            files.append( (file.replace(directory,""), date, size) )

        except OSError:

            print "WARNING: can't access file %s" % file

    files.sort(cmp=lambda x,y: cmp(x[1],y[1]), reverse=True)

    for file in files[:options.max]:

        id = hashlib.md5()

        id.update(file[0])

        id = id.hexdigest()

        feed.write("""    <entry>

        <title>%(basename)s</title>

        <updated>%(mtime)s</updated>

        <link href=\"%(address)s%(filename)s\" />

        <summary type=\"html\">&lt;p&gt;%(basename)s (%(size)sKo)&lt;br/&gt;In: %(path)s&lt;/p&gt;</summary>

        <id>tag:%(host)s:%(md5)s</id>

    </entry>

""" % { "basename": cgi.escape(os.path.basename(file[0])), 

        "mtime": datetime.datetime.fromtimestamp(file[1]).strftime("%Y-%m-%dT%H:%M:%SZ"),

        "host": options.hostname,

        "filename": urllib.quote(file[0]),

        "path": cgi.escape(os.path.dirname(file[0])),

        "size": int(file[2]) / 1024,

        "year": datetime.date.today().year,

        "md5": id,

        "address": options.url,

      })

    feed.write("</feed>\n")

    feed.close()

def parse_opts():

    usage = "usage: %prog [options] [directory]"

    parser = optparse.OptionParser(usage)

    parser.add_option("-o", "--output", dest="output", metavar="FILE",

                      default="files.xml",

                      help="write to this file (default: %default)")

    parser.add_option("-m", "--max", dest="max", default=50, type="int",

                      help="the maximum number of files to include in the "

                      "feed (default: %default)")

    parser.add_option("-t", "--title", dest="title", default="Files on %s",

          help="title for the feed. You can use %s to include the hostname.")

    parser.add_option("-x", "--exclude", dest="exclude", default="",

                      metavar="DIR1,DIR2,...", help="a comma-sparated list "

                      "of directories that should not be crawled")

    parser.add_option("-r", "--root", dest="root", help="the document root "

                      "(where the directory starts to be network-accessible")

    parser.add_option("--hostname", dest="hostname",

              default=socket.gethostname(),

              help="the hostname to make links point to (default: %default)")

    parser.add_option("--url-scheme", dest="url_scheme", default="http",

                      help="the URL method (default: %default)")

    parser.add_option("--url-port", dest="url_port",

                      help="the URL port, if not default for the chosen scheme")

    parser.add_option("-d", "--debug", dest="debug", action="store_true",

                      help="Debug mode")

    opts, args = parser.parse_args()

    if len(args) == 0:

        directory = os.getcwd()

    elif len(args) == 1:

        directory = args[0]

    else:

        parser.error("Only one feed can be given as argument")

    if not os.path.isdir(directory):

        parser.error("The directory '%s' does not exist" % directory)

    directory = os.path.abspath(directory)

    if opts.title.count("%"):

        opts.title = opts.title % opts.hostname

    opts.exclude = [ d.strip() for d in opts.exclude.split(",") ]

    if not opts.root:

        parser.error("You must provide a document root")

    url = [opts.url_scheme, "://", opts.hostname]

    if opts.url_port:

        url.append(":", opts.url_port)

    path = directory.replace(opts.root, "")

    if not path.startswith("/"):

        url.append("/")

    url.append(path)

    url = "".join(url)

    opts.url = url

    return opts, directory

if __name__ == "__main__":

    opts, directory = parse_opts()

    main(directory, opts)