def parse_mobile(params):
global mobiles, errors, req
url = params['url']
comp_name = params['name']
print 'Request sent for - ' + url
req += 1
try:
mobiles += 1
tree = html.fromstring(session.get(url).text)
name = tree.xpath('//h1[@class="specs-phone-name-title"]//text()')[0]
print 'Total[%i], Mobile found : %s' % (len(data.keys()), name)
data[name] = {}
d = data[name]
d['Url'] = url
d['Brand'] = comp_name
for x in tree.xpath('//div[@id="specs-list"]//table//tr'):
name = x.xpath('.//td[@class="ttl"]/a/text()')
if name != []: d[name[0]] = x.xpath('.//td[@class="nfo"]/text()')
d['Battery'] = tree.xpath(
'//th[text()="Battery"]/ancestor::tr//td[@class="nfo"]/text()')
except Exception as e:
print 'Error in %s Restarting.... \n Error message : %s' % (url,
e.message)
queue.put_nowait((parse_mobile, params))
errors += 1
def recv(self, sock, queue):
"""Receiver."""
# Note that this is not really an actor, but we want to send
# messages anyway, so we need to access the actor.
sock.setTimeout(None)
while True:
data = sock.read()
queue.put_nowait(data)
self.send(self.address, ('data',))
def recv(self, sock, queue):
"""Receiver."""
# Note that this is not really an actor, but we want to send
# messages anyway, so we need to access the actor.
sock.setTimeout(None)
while True:
data = sock.read()
queue.put_nowait(data)
self.send(self.address, ('data',))
def UpdateTweet(self, tweet_text, tweet_dic):
for regexp_watcher in self.regexp_watcher_list.values():
if 're_prog' not in regexp_watcher or 'queue' not in regexp_watcher:
continue
re_prog = regexp_watcher['re_prog']
queue = regexp_watcher['queue']
if re_prog is None or queue is None:
continue
match_result = re_prog.findall(tweet_text)
if not match_result:
continue
#print "streaming hit for %s (%s)" % (regexp_watcher['user_name'].decode('utf-8'), regexp_watcher['description'].decode('utf-8'))
queue.put_nowait((tweet_dic, match_result))
def UpdateTweet(self, tweet_text, tweet_dic):
for regexp_watcher in self.regexp_watcher_list.values():
if 're_prog' not in regexp_watcher or 'queue' not in regexp_watcher:
continue
re_prog = regexp_watcher['re_prog']
queue = regexp_watcher['queue']
if re_prog is None or queue is None:
continue
match_result = re_prog.findall(tweet_text)
if not match_result:
continue
#print "streaming hit for %s (%s)" % (regexp_watcher['user_name'].decode('utf-8'), regexp_watcher['description'].decode('utf-8'))
queue.put_nowait((tweet_dic, match_result))
def handle_post(post):
if not post.get('approxLoc'):
print 'Post %r does not have approxLoc field' % post['id']
return
print "New post", post['id']
data = {
'post': anonymize(post),
'lat': post['approxLoc']['lat'],
'lng': post['approxLoc']['lng'],
}
try:
queue.put_nowait(data)
except gevent.queue.Full:
return
def handle_post(post):
if not post.get('approxLoc'):
print 'Post %r does not have approxLoc field' % post['id']
return
print "New post", post['id']
data = {
'post': anonymize(post),
'lat': post['approxLoc']['lat'],
'lng': post['approxLoc']['lng'],
}
try:
queue.put_nowait(data)
except gevent.queue.Full:
return
def parse_comp(params):
global errors, req
url = params['url']
comp_name = params['name']
dp.append(url)
print 'Request sent for [%s] - %s' % (comp_name, url)
req += 1
try:
tree = html.fromstring(session.get(url).text)
for x in tree.xpath('//div[@class="nav-pages"]//a/@href'):
if domain + x not in dp:
queue.put_nowait((parse_comp, {
'url': domain + x,
'name': comp_name
}))
for x in tree.xpath('//div[@class="makers"]//a/@href'):
queue.put_nowait((parse_mobile, {
'url': domain + x,
'name': comp_name
}))
except Exception as e:
print 'Error in %s Restarting.... \n Error message : %s' % (url,
e.message)
queue.put_nowait((parse_comp, params))
errors += 1
def scrape_base_url():
global data
startTime = datetime.now()
tree = html.fromstring(session.get(base_url).text)
func = lambda x: queue.put_nowait((parse_comp, {
'url': domain + x.xpath('./@href')[0],
'name': x.xpath('./text()')[0]
}))
[
func(x) for x in tree.xpath('//div[@class="st-text"]//td/a')
if x.xpath('./text()') != []
]
while not queue.empty() and not pool.full():
for x in xrange(0, min(queue.qsize(), pool.free_count())):
t = queue.get_nowait()
pool.start(pool.spawn(t[0], t[1]))
pool.join()
print 'Time Taken : ', datetime.now() - startTime
with open('data.json', 'w') as fp:
json.dump(data, fp)
def SendHeartbeat(self):
for regexp_watcher in self.regexp_watcher_list.values():
queue = regexp_watcher['queue']
#print "send heartbeat to %s (%s)" % (regexp_watcher['user_name'].decode('utf-8'), regexp_watcher['description'].decode('utf-8'))
queue.put_nowait((None, None))
self.put_to_queue(link['href'])
def run_parse(self):
try:
self._queue.put(self._start_url)
self.parse_html(self.get_content())
except Exception, e:
print e, self._start_url, self.get_content()
def pack_url_md5(self, url):
m = hashlib.md5(url)
return m.hexdigest()[8:-8]
queue = queue.Queue()
queue.put_nowait("")
threads = []
threads.append(
gevent.spawn(
Consumers(queue, 'thread1', "http://beijing.lashou.com/").run_parse))
threads.append(
gevent.spawn(
Consumers(queue, 'thread2', "http://beijing.lashou.com/").run_parse))
threads.append(
gevent.spawn(
Consumers(queue, 'thread3', "http://beijing.lashou.com/").run_parse))
gevent.joinall(threads)
# if __name__ == "__main__":
# pass
# parse_html(get_content("http://beijing.lashou.com/"))
# print pack_url_md5("http://beijing.lashou.com/")
htmlobj = BeautifulSoup(response.read())
links = htmlobj.findAll(href=re.compile(r'^(http|/page)'))
print response.geturl()
for link in links:
print link['href'],link.string
self.put_to_queue(link['href'])
def run_parse(self):
try:
self._queue.put(self._start_url)
self.parse_html(self.get_content())
except Exception,e:
print e,self._start_url,self.get_content()
def pack_url_md5(self,url):
m = hashlib.md5(url)
return m.hexdigest()[8:-8]
queue = queue.Queue()
queue.put_nowait("")
threads = []
threads.append(gevent.spawn(Consumers(queue,'thread1',"http://beijing.lashou.com/").run_parse))
threads.append(gevent.spawn(Consumers(queue,'thread2',"http://beijing.lashou.com/").run_parse))
threads.append(gevent.spawn(Consumers(queue,'thread3',"http://beijing.lashou.com/").run_parse))
gevent.joinall(threads)
# if __name__ == "__main__":
# pass
# parse_html(get_content("http://beijing.lashou.com/"))
# print pack_url_md5("http://beijing.lashou.com/")
def SendHeartbeat(self):
for regexp_watcher in self.regexp_watcher_list.values():
queue = regexp_watcher['queue']
#print "send heartbeat to %s (%s)" % (regexp_watcher['user_name'].decode('utf-8'), regexp_watcher['description'].decode('utf-8'))
queue.put_nowait((None, None))
def _notify(waiters, data):
for queue in waiters:
queue.put_nowait(data)
