svn.andrew.net.au/scripts/movies.py

#!/usr/bin/python

import copy
import urllib2
import time
import datetime
import lxml.html
import xml.sax.saxutils

crufty_classes = [
        "rating rating-list",
        "add_to_watchlist",
        "see-more",
        "rating_txt",
        "header",
]

def main():
        imdb = urllib2.urlopen("http://www.imdb.com/movies-in-theaters/")
        doc = lxml.html.fromstring("".join(imdb.readlines()))
        for element in doc.iter(tag=lxml.etree.Element):
            if element.tag.endswith("div"):
                if element.attrib.get("id", "") == "main":
                    break
        new_releases = copy.deepcopy(element)
        # Go deeper to ditch the Box Office Top Ten
        for element in new_releases.iter(tag=lxml.etree.Element):
            if element.attrib.get("class", "") == "list detail sub-list":
                    break
        new_releases = copy.deepcopy(element)
        for crufty_class in crufty_classes:
            for cruft in new_releases.find_class(crufty_class):
                cruft.drop_tree()
        movies = xml.sax.saxutils.escape(lxml.html.tostring(new_releases))
        print """<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">

<link href="http://home.andrew.net.au/~apollock/movies.xml" rel="self"/>

<title>This week's movies from IMDb</title>
<updated>%(updated)sZ</updated>
<author>
        <name>Andrew Pollock</name>
</author>
<id>http://www.andrew.net.au/</id>

<entry>
<id>http://home.andrew.net.au/movies/%(timestamp)s</id>

<updated>%(updated)sZ</updated>
<title>This week's movies</title>

<content type="html">
%(movies)s
</content>
</entry>
</feed>
""" % { 'updated': datetime.datetime.utcnow().isoformat()[0:19], 'movies': movies, 'timestamp': int(time.time()) }

if __name__ == "__main__":
        main()
1	apollock	6	#!/usr/bin/python
2
3	apollock	50	import copy
4	apollock	49	import urllib2
5	apollock	6	import time
6			import datetime
7	apollock	50	import lxml.html
8	apollock	6	import xml.sax.saxutils
9
10	apollock	51	crufty_classes = [
11			"rating rating-list",
12			"add_to_watchlist",
13			"see-more",
14			"rating_txt",
15	apollock	52	"header",
16	apollock	51	]
17
18	apollock	6	def main():
19	apollock	50	imdb = urllib2.urlopen("http://www.imdb.com/movies-in-theaters/")
20			doc = lxml.html.fromstring("".join(imdb.readlines()))
21			for element in doc.iter(tag=lxml.etree.Element):
22			if element.tag.endswith("div"):
23			if element.attrib.get("id", "") == "main":
24			break
25			new_releases = copy.deepcopy(element)
26	apollock	53	# Go deeper to ditch the Box Office Top Ten
27			for element in new_releases.iter(tag=lxml.etree.Element):
28			if element.attrib.get("class", "") == "list detail sub-list":
29			break
30			new_releases = copy.deepcopy(element)
31	apollock	51	for crufty_class in crufty_classes:
32			for cruft in new_releases.find_class(crufty_class):
33			cruft.drop_tree()
34	apollock	50	movies = xml.sax.saxutils.escape(lxml.html.tostring(new_releases))
35	apollock	6	print """<?xml version="1.0" encoding="utf-8"?>
36			<feed xmlns="http://www.w3.org/2005/Atom">
37
38			<link href="http://home.andrew.net.au/~apollock/movies.xml" rel="self"/>
39
40			<title>This week's movies from IMDb</title>
41			<updated>%(updated)sZ</updated>
42			<author>
43			<name>Andrew Pollock</name>
44			</author>
45			<id>http://www.andrew.net.au/</id>
46
47			<entry>
48	apollock	49	<id>http://home.andrew.net.au/movies/%(timestamp)s</id>
49	apollock	6
50			<updated>%(updated)sZ</updated>
51			<title>This week's movies</title>
52
53			<content type="html">
54			%(movies)s
55			</content>
56			</entry>
57			</feed>
58	apollock	16	""" % { 'updated': datetime.datetime.utcnow().isoformat()[0:19], 'movies': movies, 'timestamp': int(time.time()) }
59	apollock	6
60			if __name__ == "__main__":
61			main()