def handle_urls(self, gm_worker, gm_job):
self._urldata = dict()
self._urls = gm_job.data
self._num_processed = 0
self._gm_job = gm_job
logger.info(self._urls)
AsyncBatch(self._urls, self.parse_response)
return self._urldata
def handle_urls(self, gm_worker, gm_job):
self._urldata = dict()
self._urls = gm_job.data
self._num_processed = 0
self._gm_job = gm_job
logger.info(self._urls)
AsyncBatch(self._urls, self.parse_response)
return self._urldata
def parse_response(self, url, response, response_utf8):
if response.error:
logger.info('error: %s' % response.error)
if response_utf8 is not None:
logger.info('received %s' % url)
self._urldata[url] = list()
try:
d = pq(response_utf8)
except ValueError:
logger.info('Parse of %s failed, trying soup parser...' % url)
d = pq(response_utf8, parser='soup')
for link in d('a'):
href = link.get('href')
if href is None:
continue
split_url = urlsplit(href)
if split_url.hostname is None:
continue
url_match = self._url_re.match(href)
if url_match:
logger.info('Twitter User Found: %s' % url_match.group(4))
if url_match.group(4) not in self._urldata[url]:
self._urldata[url].append(url_match.group(4))
else:
logger.debug('Parse of %s failed' % url)
self._urldata[url] = None
self._num_processed += 1
def __init__(self, filename):
self._gm_client = JSONGearmanClient([options.jobserver])
try:
urls = list()
for line in open(filename, 'r'):
urls.append(line.strip())
url_chunks = [urls[i : i + options.chunk_size] for i in range(0, len(urls), options.chunk_size)]
jobs = list()
for url_chunk in url_chunks:
jobs.append({'task': 'tweet_scout', 'data': url_chunk});
submitted_job_requests = self._gm_client.submit_multiple_jobs(jobs, background=False, wait_until_complete=False)
job_count = len(submitted_job_requests)
complete_count = 0
p = ProgressBar(maxval=job_count).start()
while complete_count != job_count:
try:
self._gm_client.wait_until_jobs_completed(submitted_job_requests, poll_timeout=1)
self._gm_client.get_job_statuses(submitted_job_requests, poll_timeout=100)
except:
pass
complete_count = 0
for job in submitted_job_requests:
if job.complete is True:
complete_count += 1
p.update(complete_count)
count = 0
twitter_user_counts = defaultdict(int)
for job in submitted_job_requests:
for url, twitter_users in job.result.iteritems():
logger.debug('%s: %s' % (url, twitter_users))
if twitter_users is not None:
count += 1
for twitter_user in twitter_users:
twitter_user_counts[twitter_user] += 1
print '\nFound %d Twitter users in %d successfully parsed pages:' % (len(twitter_user_counts), count)
for user in sorted(twitter_user_counts, key=twitter_user_counts.get, reverse=True):
print '%s,%d' % (user, twitter_user_counts[user])
except KeyboardInterrupt:
logger.info('Exiting')
pass
except Exception, e:
logger.exception('Exiting - %s' % e)
def parse_response(self, url, response, response_utf8):
if response.error:
logger.info('error: %s' % response.error)
if response_utf8 is not None:
logger.info('received %s' % url)
self._urldata[url] = list()
try:
d = pq(response_utf8)
except ValueError:
logger.info('Parse of %s failed, trying soup parser...' % url)
d = pq(response_utf8, parser='soup')
for link in d('a'):
href = link.get('href')
if href is None:
continue
split_url = urlsplit(href)
if split_url.hostname is None:
continue
url_match = self._url_re.match(href)
if url_match:
logger.info('Twitter User Found: %s' % url_match.group(4))
if url_match.group(4) not in self._urldata[url]:
self._urldata[url].append(url_match.group(4))
else:
logger.debug('Parse of %s failed' % url)
self._urldata[url] = None
self._num_processed += 1
def __init__(self):
self._urldata = dict()
self._urls = list()
self._gm_job = None
self._gm_worker = JSONGearmanWorker([options.jobserver])
self._gm_worker.register_task('tweet_scout', self.handle_urls)
self._url_re = re.compile('^http(s)?://(www\.)?twitter\.com/(?!share)(?!home)(?!intent)(#!/)?([a-zA-Z0-9_]{1,15}[^/])(/\w+)*$')
try:
logger.info('TweetScout initialized and ready for work');
self._gm_worker.work()
except KeyboardInterrupt:
logger.info('Exiting')
pass
except Exception, e:
logger.error('Exiting - %s' % e)
def __init__(self):
self._urldata = dict()
self._urls = list()
self._gm_job = None
self._gm_worker = JSONGearmanWorker([options.jobserver])
self._gm_worker.register_task('tweet_scout', self.handle_urls)
self._url_re = re.compile(
'^http(s)?://(www\.)?twitter\.com/(?!share)(?!home)(?!intent)(#!/)?([a-zA-Z0-9_]{1,15}[^/])(/\w+)*$'
)
try:
logger.info('TweetScout initialized and ready for work')
self._gm_worker.work()
except KeyboardInterrupt:
logger.info('Exiting')
pass
except Exception, e:
logger.error('Exiting - %s' % e)
def __init__(self, filename):
self._gm_client = JSONGearmanClient([options.jobserver])
try:
urls = list()
for line in open(filename, 'r'):
urls.append(line.strip())
url_chunks = [
urls[i:i + options.chunk_size]
for i in range(0, len(urls), options.chunk_size)
]
jobs = list()
for url_chunk in url_chunks:
jobs.append({
'task': 'tweet_scout',
'data': url_chunk
})
submitted_job_requests = self._gm_client.submit_multiple_jobs(
jobs, background=False, wait_until_complete=False)
job_count = len(submitted_job_requests)
complete_count = 0
p = ProgressBar(maxval=job_count).start()
while complete_count != job_count:
try:
self._gm_client.wait_until_jobs_completed(
submitted_job_requests, poll_timeout=1)
self._gm_client.get_job_statuses(submitted_job_requests,
poll_timeout=100)
except:
pass
complete_count = 0
for job in submitted_job_requests:
if job.complete is True:
complete_count += 1
p.update(complete_count)
count = 0
twitter_user_counts = defaultdict(int)
for job in submitted_job_requests:
for url, twitter_users in job.result.iteritems():
logger.debug('%s: %s' % (url, twitter_users))
if twitter_users is not None:
count += 1
for twitter_user in twitter_users:
twitter_user_counts[twitter_user] += 1
print '\nFound %d Twitter users in %d successfully parsed pages:' % (
len(twitter_user_counts), count)
for user in sorted(twitter_user_counts,
key=twitter_user_counts.get,
reverse=True):
print '%s,%d' % (user, twitter_user_counts[user])
except KeyboardInterrupt:
logger.info('Exiting')
pass
except Exception, e:
logger.exception('Exiting - %s' % e)
