def get_rss(client, hours, cat):
"""
Generate and return the aggregate rss feed.
"""
from PyRSS2Gen import RSS2, RSSItem
from StringIO import StringIO
t = time.time() - int(hours or 24)*60*60
start_row = build_key(cat or '__all__', t)
stop_row = build_key(cat or '__all__', time.time())
items = []
scanner = db.Scanner(client, 'UrlsIndex', ['Url:'], start_row, stop_row)
for row in scanner:
url = client.getRow('Urls', row.columns['Url:'].value)[0]
items.append(RSSItem(
title = url.columns['Content:title'].value.decode('utf-8', 'replace'),
link = url.row,
description = url.columns['Content:raw'].value.decode('utf-8', 'replace'),
pubDate = datetime.fromtimestamp(float(url.columns['Meta:updated'].value))
))
items.reverse()
rss = RSS2(
title = 'Aggregated feed',
link = 'http://example.com/rss',
description = 'Hbase aggregated feed',
lastBuildDate = datetime.now(),
items = items
)
out = StringIO()
rss.write_xml(out)
return out.getvalue()
def dump_urls(client, hours, cat):
"""
Dump on stdout an aggregated list of urls
"""
t = time.time() - int(hours or 24)*60*60
start_row = build_key(cat or '__all__', t)
stop_row = build_key(cat or '__all__', time.time())
scanner = db.Scanner(client, 'UrlsIndex', ['Url:'], start_row, stop_row)
urls = [row.columns['Url:'].value for row in scanner]
urls.reverse()
for url in urls: print url