def test_unicode(self):
from planet.spider import filename
index = idindex.create()
iri = 'http://www.\xe8\xa9\xb9\xe5\xa7\x86\xe6\x96\xaf.com/'
index[filename('', iri)] = 'data'
index[filename('', iri.decode('utf-8'))] = 'data'
index[filename('', u'1234')] = 'data'
index.close()
def test_unicode(self):
from planet.spider import filename
index = idindex.create()
iri = 'http://www.\xe8\xa9\xb9\xe5\xa7\x86\xe6\x96\xaf.com/'
index[filename('', iri)] = 'data'
index[filename('', iri.decode('utf-8'))] = 'data'
index[filename('', u'1234')] = 'data'
index.close()
def test_expunge(self):
config.load(configfile)
# create test entries in cache with correct timestamp
for entry in glob.glob(testentries):
e=minidom.parse(entry)
e.normalize()
eid = e.getElementsByTagName('id')
efile = filename(workdir, eid[0].childNodes[0].nodeValue)
eupdated = e.getElementsByTagName('updated')[0].childNodes[0].nodeValue
emtime = time.mktime(feedparser._parse_date_w3dtf(eupdated))
if not eid or not eupdated: continue
shutil.copyfile(entry, efile)
os.utime(efile, (emtime, emtime))
# create test feeds in cache
sources = config.cache_sources_directory()
for feed in glob.glob(testfeeds):
f=minidom.parse(feed)
f.normalize()
fid = f.getElementsByTagName('id')
if not fid: continue
ffile = filename(sources, fid[0].childNodes[0].nodeValue)
shutil.copyfile(feed, ffile)
# verify that exactly nine entries + one source dir were produced
files = glob.glob(workdir+"/*")
self.assertEqual(10, len(files))
# verify that exactly four feeds were produced in source dir
files = glob.glob(sources+"/*")
self.assertEqual(4, len(files))
# expunge...
expungeCache()
# verify that five entries and one source dir are left
files = glob.glob(workdir+"/*")
self.assertEqual(6, len(files))
# verify that the right five entries are left
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test1,1') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test2,1') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test3,3') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test4,2') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test4,3') in files)
def test_expunge(self):
config.load(configfile)
# create test entries in cache with correct timestamp
for entry in glob.glob(testentries):
e=minidom.parse(entry)
e.normalize()
eid = e.getElementsByTagName('id')
efile = filename(workdir, eid[0].childNodes[0].nodeValue)
eupdated = e.getElementsByTagName('updated')[0].childNodes[0].nodeValue
emtime = time.mktime(feedparser._parse_date_w3dtf(eupdated))
if not eid or not eupdated: continue
shutil.copyfile(entry, efile)
os.utime(efile, (emtime, emtime))
# create test feeds in cache
sources = config.cache_sources_directory()
for feed in glob.glob(testfeeds):
f=minidom.parse(feed)
f.normalize()
fid = f.getElementsByTagName('id')
if not fid: continue
ffile = filename(sources, fid[0].childNodes[0].nodeValue)
shutil.copyfile(feed, ffile)
# verify that exactly nine entries + one source dir were produced
files = glob.glob(workdir+"/*")
self.assertEqual(10, len(files))
# verify that exactly four feeds were produced in source dir
files = glob.glob(sources+"/*")
self.assertEqual(4, len(files))
# expunge...
expungeCache()
# verify that five entries and one source dir are left
files = glob.glob(workdir+"/*")
self.assertEqual(6, len(files))
# verify that the right five entries are left
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test1,1') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test2,1') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test3,3') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test4,2') in files)
self.assertTrue(os.path.join(workdir,
'bzr.mfd-consult.dk,2007,venus-expunge-test4,3') in files)
def create():
from planet import logger as log
cache = config.cache_directory()
index = os.path.join(cache, 'index')
if not os.path.exists(index): os.makedirs(index)
import anydbm
index = anydbm.open(filename(index, 'id'), 'c')
try:
import libxml2
except:
libxml2 = False
from xml.dom import minidom
for file in glob(cache + "/*"):
if os.path.isdir(file):
continue
elif libxml2:
try:
doc = libxml2.parseFile(file)
ctxt = doc.xpathNewContext()
ctxt.xpathRegisterNs('atom', 'http://www.w3.org/2005/Atom')
entry = ctxt.xpathEval('/atom:entry/atom:id')
source = ctxt.xpathEval('/atom:entry/atom:source/atom:id')
if entry and source:
index[filename('', entry[0].content)] = source[0].content
doc.freeDoc()
except:
log.error(file)
else:
try:
doc = minidom.parse(file)
doc.normalize()
ids = doc.getElementsByTagName('id')
entry = [e for e in ids if e.parentNode.nodeName == 'entry']
source = [e for e in ids if e.parentNode.nodeName == 'source']
if entry and source:
index[filename('',entry[0].childNodes[0].nodeValue)] = \
source[0].childNodes[0].nodeValue
doc.freeDoc()
except:
log.error(file)
log.info(str(len(index.keys())) + " entries indexed")
index.close()
return open()
def create():
from planet import logger as log
cache = config.cache_directory()
index=os.path.join(cache,'index')
if not os.path.exists(index): os.makedirs(index)
import dbhash
index = dbhash.open(filename(index, 'id'),'c')
try:
import libxml2
except:
libxml2 = False
from xml.dom import minidom
for file in glob(cache+"/*"):
if os.path.isdir(file):
continue
elif libxml2:
try:
doc = libxml2.parseFile(file)
ctxt = doc.xpathNewContext()
ctxt.xpathRegisterNs('atom','http://www.w3.org/2005/Atom')
entry = ctxt.xpathEval('/atom:entry/atom:id')
source = ctxt.xpathEval('/atom:entry/atom:source/atom:id')
if entry and source:
index[filename('',entry[0].content)] = source[0].content
doc.freeDoc()
except:
log.error(file)
else:
try:
doc = minidom.parse(file)
doc.normalize()
ids = doc.getElementsByTagName('id')
entry = [e for e in ids if e.parentNode.nodeName == 'entry']
source = [e for e in ids if e.parentNode.nodeName == 'source']
if entry and source:
index[filename('',entry[0].childNodes[0].nodeValue)] = \
source[0].childNodes[0].nodeValue
doc.freeDoc()
except:
log.error(file)
log.info(str(len(index.keys())) + " entries indexed")
index.close()
return open()
def test_filename(self):
self.assertEqual(os.path.join('.', 'example.com,index.html'),
filename('.', 'http://example.com/index.html'))
self.assertEqual(os.path.join('.',
'planet.intertwingly.net,2006,testfeed1,1'),
filename('.', u'tag:planet.intertwingly.net,2006:testfeed1,1'))
self.assertEqual(os.path.join('.',
'00000000-0000-0000-0000-000000000000'),
filename('.', u'urn:uuid:00000000-0000-0000-0000-000000000000'))
# Requires Python 2.3
try:
import encodings.idna
except:
return
self.assertEqual(os.path.join('.', 'xn--8ws00zhy3a.com'),
filename('.', u'http://www.\u8a79\u59c6\u65af.com/'))
def destroy():
from planet import logger as log
cache = config.cache_directory()
index = os.path.join(cache, 'index')
if not os.path.exists(index): return None
idindex = filename(index, 'id')
if os.path.exists(idindex): os.unlink(idindex)
os.removedirs(index)
log.info(idindex + " deleted")
def destroy():
from planet import logger as log
cache = config.cache_directory()
index=os.path.join(cache,'index')
if not os.path.exists(index): return None
idindex = filename(index, 'id')
if os.path.exists(idindex): os.unlink(idindex)
os.removedirs(index)
log.info(idindex + " deleted")
def open():
try:
cache = config.cache_directory()
index = os.path.join(cache, 'index')
if not os.path.exists(index): return None
import anydbm
return anydbm.open(filename(index, 'id'), 'w')
except Exception, e:
if e.__class__.__name__ == 'DBError': e = e.args[-1]
from planet import logger as log
log.error(str(e))
def open():
try:
cache = config.cache_directory()
index=os.path.join(cache,'index')
if not os.path.exists(index): return None
import dbhash
return dbhash.open(filename(index, 'id'),'w')
except Exception, e:
if e.__class__.__name__ == 'DBError': e = e.args[-1]
from planet import logger as log
log.error(str(e))
def open():
try:
cache = config.cache_directory()
index = os.path.join(cache, "index")
if not os.path.exists(index):
return None
return dbhash.open(filename(index, "id"), "w")
except Exception, e:
if e.__class__.__name__ == "DBError":
e = e.args[-1]
from planet import logger as log
log.error(str(e))
def run(script, doc, output_file=None, options={}):
""" process an Genshi template """
context = Context(**options)
tmpl_fileobj = open(script)
tmpl = MarkupTemplate(tmpl_fileobj, script)
tmpl_fileobj.close()
if not output_file:
# filter
context.push({'input':XMLParser(StringIO(doc))})
else:
# template
import time
from planet import config,feedparser
from planet.spider import filename
# gather a list of subscriptions, feeds
global subscriptions
feeds = []
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data=feedparser.parse(filename(sources,sub))
data.feed.config = norm(dict(config.parser.items(sub)))
if data.feed.has_key('link'):
feeds.append((data.feed.config.get('name',''),data.feed))
subscriptions.append(norm(sub))
feeds.sort()
# annotate each entry
new_date_format = config.new_date_format()
vars = feedparser.parse(StringIO(doc))
vars.feeds = [value for name,value in feeds]
last_feed = None
last_date = None
for entry in vars.entries:
entry.source.config = find_config(config, entry.source)
# add new_feed and new_date fields
if 'id' in entry.source:
entry.new_feed = entry.source.id
else:
entry.new_feed = None
entry.new_date = date = None
if entry.has_key('published_parsed'): date=entry.published_parsed
if entry.has_key('updated_parsed'): date=entry.updated_parsed
if date: entry.new_date = time.strftime(new_date_format, date)
# remove new_feed and new_date fields if not "new"
if entry.new_date == last_date:
entry.new_date = None
if entry.new_feed == last_feed:
entry.new_feed = None
else:
last_feed = entry.new_feed
elif entry.new_date:
last_date = entry.new_date
last_feed = None
# add streams for all text constructs
for key in entry.keys():
if key.endswith("_detail") and entry[key].has_key('type') and \
entry[key].has_key('value'):
streamify(entry[key],entry.source.planet_bozo)
if entry.has_key('content'):
for content in entry.content:
streamify(content,entry.source.planet_bozo)
# add cumulative feed information to the Genshi context
vars.feed.config = dict(config.parser.items('Planet',True))
context.push(vars)
# apply template
output=tmpl.generate(context).render('xml')
if output_file:
out_file = open(output_file,'w')
out_file.write(output)
out_file.close()
else:
return output
def run(script, doc, output_file=None, options={}):
""" process an Genshi template """
context = Context(**options)
tmpl_fileobj = open(script)
tmpl = MarkupTemplate(tmpl_fileobj, script)
tmpl_fileobj.close()
if not output_file:
# filter
context.push({'input':XMLParser(StringIO(doc))})
else:
# template
import time
from planet import config,feedparser
from planet.spider import filename
# gather a list of subscriptions, feeds
global subscriptions
feeds = []
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data=feedparser.parse(filename(sources,sub))
data.feed.config = norm(dict(config.parser.items(sub)))
if data.feed.has_key('link'):
feeds.append((data.feed.config.get('name',''),data.feed))
subscriptions.append(norm(sub))
feeds.sort()
# annotate each entry
new_date_format = config.new_date_format()
vars = feedparser.parse(StringIO(doc))
vars.feeds = [value for name,value in feeds]
last_feed = None
last_date = None
for entry in vars.entries:
entry.source.config = find_config(config, entry.source)
# add new_feed and new_date fields
entry.new_feed = entry.source.id
entry.new_date = date = None
if entry.has_key('published_parsed'): date=entry.published_parsed
if entry.has_key('updated_parsed'): date=entry.updated_parsed
if date: entry.new_date = time.strftime(new_date_format, date)
# remove new_feed and new_date fields if not "new"
if entry.new_date == last_date:
entry.new_date = None
if entry.new_feed == last_feed:
entry.new_feed = None
else:
last_feed = entry.new_feed
elif entry.new_date:
last_date = entry.new_date
last_feed = None
# add streams for all text constructs
for key in entry.keys():
if key.endswith("_detail") and entry[key].has_key('type') and \
entry[key].has_key('value'):
streamify(entry[key],entry.source.planet_bozo)
if entry.has_key('content'):
for content in entry.content:
streamify(content,entry.source.planet_bozo)
# add cumulative feed information to the Genshi context
vars.feed.config = dict(config.parser.items('Planet',True))
context.push(vars)
# apply template
output=tmpl.generate(context).render('xml')
if output_file:
out_file = open(output_file,'w')
out_file.write(output)
out_file.close()
else:
return output
# Create the blacklist dir if it does not exist
if not os.path.exists(blacklist):
os.mkdir(blacklist)
print "<p>Created directory %s</p>" % blacklist
# find list of urls, in the form bl[n]=url
for key in form.keys():
if not key.startswith("bl"): continue
url = unquote(form[key].value)
# find corresponding files
cache_file = filename(cache, url)
blacklist_file = filename(blacklist, url)
# move to blacklist if found
if os.path.exists(cache_file):
os.rename(cache_file, blacklist_file)
print "<p>Blacklisted <a href='%s'>%s</a></p>" % (url, url)
else:
print "<p>Unknown file: %s</p>" % cache_file
print """
<p>Note that blacklisting does not automatically
def downloadReadingList(list, orig_config, callback, use_cache=True, re_read=True):
from planet import logger
import config
try:
import urllib2, StringIO
from planet.spider import filename
# list cache file name
cache_filename = filename(config.cache_lists_directory(), list)
# retrieve list options (e.g., etag, last-modified) from cache
options = {}
# add original options
for key in orig_config.options(list):
options[key] = orig_config.get(list, key)
try:
if use_cache:
cached_config = ConfigParser()
cached_config.read(cache_filename)
for option in cached_config.options(list):
options[option] = cached_config.get(list,option)
except:
pass
cached_config = ConfigParser()
cached_config.add_section(list)
for key, value in options.items():
cached_config.set(list, key, value)
# read list
curdir=getattr(os.path, 'curdir', '.')
if sys.platform.find('win') < 0:
base = urljoin('file:', os.path.abspath(curdir))
else:
path = os.path.abspath(os.path.curdir)
base = urljoin('file:///', path.replace(':','|').replace('\\','/'))
request = urllib2.Request(urljoin(base + '/', list))
if options.has_key("etag"):
request.add_header('If-None-Match', options['etag'])
if options.has_key("last-modified"):
request.add_header('If-Modified-Since',
options['last-modified'])
response = urllib2.urlopen(request)
if response.headers.has_key('etag'):
cached_config.set(list, 'etag', response.headers['etag'])
if response.headers.has_key('last-modified'):
cached_config.set(list, 'last-modified',
response.headers['last-modified'])
# convert to config.ini
data = StringIO.StringIO(response.read())
if callback: callback(data, cached_config)
# write to cache
if use_cache:
cache = open(cache_filename, 'w')
cached_config.write(cache)
cache.close()
# re-parse and proceed
logger.debug("Using %s readinglist", list)
if re_read:
if use_cache:
orig_config.read(cache_filename)
else:
cdata = StringIO.StringIO()
cached_config.write(cdata)
cdata.seek(0)
orig_config.readfp(cdata)
except:
try:
if re_read:
if use_cache:
if not orig_config.read(cache_filename): raise Exception()
else:
cdata = StringIO.StringIO()
cached_config.write(cdata)
cdata.seek(0)
orig_config.readfp(cdata)
logger.info("Using cached %s readinglist", list)
except:
logger.exception("Unable to read %s readinglist", list)
# load config files (default: config.ini)
for arg in sys.argv[1:]:
config.load(arg)
if len(sys.argv) == 1:
config.load('config.ini')
from Queue import Queue
from threading import Thread
# determine which subscriptions have no icon but do have a html page
fetch_queue = Queue()
html = ['text/html', 'application/xhtml+xml']
sources = config.cache_sources_directory()
for sub in config.subscriptions():
data = feedparser.parse(filename(sources, sub))
if data.feed.get('icon'): continue
if not data.feed.get('links'): continue
for link in data.feed.links:
if link.rel == 'alternate' and link.type in html:
fetch_queue.put((sub, link.href))
break
# find the favicon for a given webpage
def favicon(page):
parser = html5parser.HTMLParser(tree=treebuilders.getTreeBuilder('dom'))
doc = parser.parse(urlopen(page))
favicon = urljoin(page, '/favicon.ico')
for link in doc.getElementsByTagName('link'):
if link.hasAttribute('rel') and link.hasAttribute('href'):
def downloadReadingList(list, orig_config, callback, use_cache=True, re_read=True):
from planet import logger
import config
try:
import urllib2, StringIO
from planet.spider import filename
# list cache file name
cache_filename = filename(config.cache_lists_directory(), list)
# retrieve list options (e.g., etag, last-modified) from cache
options = {}
# add original options
for key in orig_config.options(list):
options[key] = orig_config.get(list, key)
try:
if use_cache:
cached_config = ConfigParser()
cached_config.read(cache_filename)
for option in cached_config.options(list):
options[option] = cached_config.get(list,option)
except:
pass
cached_config = ConfigParser()
cached_config.add_section(list)
for key, value in options.items():
cached_config.set(list, key, value)
# read list
curdir=getattr(os.path, 'curdir', '.')
if sys.platform.find('win') < 0:
base = urljoin('file:', os.path.abspath(curdir))
else:
path = os.path.abspath(os.path.curdir)
base = urljoin('file:///', path.replace(':','|').replace('\\','/'))
request = urllib2.Request(urljoin(base + '/', list))
if options.has_key("etag"):
request.add_header('If-None-Match', options['etag'])
if options.has_key("last-modified"):
request.add_header('If-Modified-Since',
options['last-modified'])
response = urllib2.urlopen(request)
if response.headers.has_key('etag'):
cached_config.set(list, 'etag', response.headers['etag'])
if response.headers.has_key('last-modified'):
cached_config.set(list, 'last-modified',
response.headers['last-modified'])
# convert to config.ini
data = StringIO.StringIO(response.read())
if callback: callback(data, cached_config)
# write to cache
if use_cache:
cache = open(cache_filename, 'w')
cached_config.write(cache)
cache.close()
# re-parse and proceed
logger.debug("Using %s readinglist", list)
if re_read:
if use_cache:
orig_config.read(cache_filename)
else:
cdata = StringIO.StringIO()
cached_config.write(cdata)
cdata.seek(0)
orig_config.readfp(cdata)
except:
try:
if re_read:
if use_cache:
if not orig_config.read(cache_filename): raise Exception()
else:
cdata = StringIO.StringIO()
cached_config.write(cdata)
cdata.seek(0)
orig_config.readfp(cdata)
logger.info("Using cached %s readinglist", list)
except:
logger.exception("Unable to read %s readinglist", list)
# Create the blacklist dir if it does not exist
if not os.path.exists(blacklist):
os.mkdir(blacklist)
print "<p>Created directory %s</p>" % blacklist
# find list of urls, in the form bl[n]=url
for key in form.keys():
if not key.startswith("bl"): continue
url = unquote(form[key].value)
# find corresponding files
cache_file = filename(cache, url)
blacklist_file = filename(blacklist, url)
# move to blacklist if found
if os.path.exists(cache_file):
os.rename(cache_file, blacklist_file)
print "<p>Blacklisted <a href='%s'>%s</a></p>" % (url, url)
else:
print "<p>Unknown file: %s</p>" % cache_file
print """
<p>Note that blacklisting does not automatically
