def putRequest(queue, payload=None):
response = {}
statusCode = {}
data = {}
while not queue.empty():
resourceURI = queue.get(timeout=DMON_TIMEOUT)
response['Node'] = resourceURI
try:
if payload is None:
r = requests.put(resourceURI, timeout=20)
else:
r = requests.put(resourceURI, data=payload, timeout=20)
if r.headers['Content-Type'] == 'application/json':
data = r.json
else:
data = r.text
response['StatusCode'] = r.status_code
response['Data'] = data
except requests.exceptions.Timeout:
response['StatusCode'] = 408
response['Data'] = data
except requests.exceptions.ConnectionError:
response['Node'] = resourceURI
statusCode['StatusCode'] = 404
response['Data'] = 'n/a'
GreenletRequests.NodeResponsesPost.append(response)
# print 'Threaded PUT with ID ' + str(GreenletRequests.npo) + ' executed for ' + resourceURI
app.logger.info(
'[%s] : [INFO] Thread PUT with ID %s executed for %s',
datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S'),
str(GreenletRequests.ng), resourceURI)
GreenletRequests.npo += 1
gevent.sleep(0)
def randomT(queue, name):
while not queue.empty():
t = queue.get(timeout=1)
gevent.sleep(5)
print "I am + " + name + " executing " + str(GreenletRequests.ng)
GreenletRequests.ng += 1
gevent.sleep(0)
def putRequest(queue, payload=None):
response = {}
statusCode = {}
data = {}
while not queue.empty():
resourceURI = queue.get(timeout=1)
response["Node"] = resourceURI
try:
if payload is None:
r = requests.put(resourceURI, timeout=20)
else:
r = requests.put(resourceURI, data=payload, timeout=20)
if r.headers["Content-Type"] == "application/json":
data = r.json
else:
data = r.text
response["StatusCode"] = r.status_code
response["Data"] = data
except requests.exceptions.Timeout:
response["StatusCode"] = 408
response["Data"] = data
except requests.exceptions.ConnectionError:
response["Node"] = resourceURI
statusCode["StatusCode"] = 404
response["Data"] = "n/a"
GreenletRequests.NodeResponsesPost.append(response)
print "Threaded PUT with ID " + str(GreenletRequests.npo) + " executed for " + resourceURI
GreenletRequests.npo += 1
gevent.sleep(0)
def deleteRequest(queue):
response = {}
while not queue.empty():
resURI = queue.get(timeout=DMON_TIMEOUT)
try:
r = requests.delete(resURI, timeout=DMON_TIMEOUT)
data = r.json()
response['Node'] = resURI
response['StatusCode'] = r.status_code
response['Data'] = data
except requests.exceptions.Timeout:
response['Node'] = resURI
response['StatusCode'] = 408
response['Data'] = 'n/a'
except requests.exceptions.ConnectionError:
response['Node'] = resURI
response['StatusCode'] = 404
response['Data'] = 'n/a'
GreenletRequests.NodeResponsesGet.append(response)
# print 'Threaded DELETE with ID ' + str(GreenletRequests.nd) + ' executed for ' + resURI
app.logger.info(
'[%s] : [INFO] Thread DELETE with ID %s executed for %s',
datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S'),
str(GreenletRequests.ng), resURI)
GreenletRequests.nd += 1
gevent.sleep(0)
def _download_helper():
while not queue.empty():
h = queue.get()
if not h:
break
r = requests.get(VT_DOWNLOAD, params={"apikey": apikey, "hash": h})
open(h, "wb").write(r.content)
def _download_helper():
while not queue.empty():
h = queue.get()
if not h:
break
r = requests.get(VT_DOWNLOAD, params={"apikey": apikey, "hash": h})
open(h, "wb").write(r.content)
def response_generator():
ii = 0
while npending or not queue.empty():
ii += 1
result = queue.get()
msg = '{} {}\n'.format(ii, result)
print(msg, end='')
yield msg
t2 = datetime.datetime.now()
print('====', t2 - t1)
def RegexpMatchWait(queue):
if queue.empty():
gevent.sleep(1)
return ''
(tweet_dic, match_result) = queue.get()
if tweet_dic is None or match_result is None:
return "\n"
result_dic = tweet_dic.copy()
result_dic['match_result'] = match_result
logging.info('waiting tweet text got: %s' % str(result_dic))
return "%s\n" % json.dumps(result_dic)
def randomT(queue, name):
while not queue.empty():
t = queue.get(timeout=1)
gevent.sleep(5)
# print 'I am + ' + name + ' executing ' + str(GreenletRequests.ng)
app.logger.info(
'[%s] : [INFO] %s executing %s',
datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S'),
name, str(GreenletRequests.ng))
GreenletRequests.ng += 1
gevent.sleep(0)
def RegexpMatchWait(queue):
if queue.empty():
gevent.sleep(1)
return ''
(tweet_dic, match_result) = queue.get()
if tweet_dic is None or match_result is None:
return "\n"
result_dic = tweet_dic.copy()
result_dic['match_result'] = match_result
logging.info('waiting tweet text got: %s' % str(result_dic))
return "%s\n" % json.dumps(result_dic)
def send_message(socket):
global queue
while True:
try:
if not queue.empty():
#print("QUEUE NOT EMPTY")
message = queue.get(block=False)
if not socket.closed:
socket.send(json.dumps(message))
#print('Sent response')
#We need a sleep call so that other greenlets can run
gevent.sleep()
except Exception as e:
print("SEND: %s" % e)
raise e
def init():
#queue init
#main.queue.put("")
#main.pool.spawn(getLink).join()
#give worker pool
print('start crwaling')
#while not pool.free_count() == 15:
while not queue.empty():
gevent.sleep(0.8)
for x in range(0, min(queue.qsize(), pool.free_count())):
pool.spawn(getData)
#wait for everything complete
pool.join()
def send_message(socket):
global queue
while True:
try:
if not queue.empty():
#print("QUEUE NOT EMPTY")
message = queue.get(block=False)
if not socket.closed:
socket.send(json.dumps(message))
#print('Sent response')
#We need a sleep call so that other greenlets can run
gevent.sleep()
except Exception as e:
print("SEND: %s" % e)
raise e
def task_thread(self, queue):
"""
Executes tasks in queue
"""
while not self.shutdown.is_set():
if queue.empty() is False:
(job, task) = queue.get_nowait()
# Don't run the task if the job is done
if job.status in [Status.ERROR, Status.ABORT]:
task.status = Status.ABORT
else:
options = {}
gpu_id = -1
try:
if isinstance(task, model_tasks.TrainTask):
### Select GPU
if len(self.gpu_list):
for gpu in self.gpu_list:
if not gpu['active']:
gpu_id = gpu['index']
gpu['active'] = True
break
assert gpu_id != -1, 'no available GPU'
else:
gpu_id = None
options['gpu_id'] = gpu_id
task.run(**options)
except Exception as e:
logger.error('%s: %s' % (type(e).__name__, e),
job_id=job.id())
task.exception = e
task.traceback = traceback.format_exc()
task.status = Status.ERROR
finally:
### Release GPU
if gpu_id != -1 and gpu_id is not None:
for gpu in self.gpu_list:
if gpu['index'] == gpu_id:
gpu['active'] = False
else:
# Wait before checking again for a task
time.sleep(utils.wait_time())
def _download_helper():
t = time.time()
while not queue.empty():
h = queue.get()
if not h:
break
if h == "wait":
time.sleep(max(0, 60 - time.time() + t))
t = time.time()
continue
if os.path.exists(h):
print "skipping..", h
continue
r = requests.get(VT_DOWNLOAD, params={"apikey": apikey, "hash": h})
open(h, "wb").write(r.content)
def task_thread(self, queue):
"""
Executes tasks in queue
"""
while not self.shutdown.is_set():
if queue.empty() is False:
(job, task) = queue.get_nowait()
# Don't run the task if the job is done
if job.status in [Status.ERROR, Status.ABORT]:
task.status = Status.ABORT
else:
options = {}
gpu_id = -1
try:
if isinstance(task, model_tasks.TrainTask):
### Select GPU
if len(self.gpu_list):
for gpu in self.gpu_list:
if not gpu['active']:
gpu_id = gpu['index']
gpu['active'] = True
break
assert gpu_id != -1, 'no available GPU'
else:
gpu_id = None
options['gpu_id'] = gpu_id
task.run(**options)
except Exception as e:
logger.error('%s: %s' % (type(e).__name__, e), job_id=job.id())
task.exception = e
task.traceback = traceback.format_exc()
task.status = Status.ERROR
finally:
### Release GPU
if gpu_id != -1 and gpu_id is not None:
for gpu in self.gpu_list:
if gpu['index'] == gpu_id:
gpu['active'] = False
else:
# Wait before checking again for a task
time.sleep(utils.wait_time())
def getrequestFile(queue, output):
response = {}
while not queue.empty():
resURI = queue.get(timeout=1)
app.logger.info(
'[%s] : [INFO] Thread File GET with ID %s starts execution for %s',
datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S'),
str(GreenletRequests.ng), resURI)
hostURL = urlparse(resURI)
hostID = hostURL.hostname
logName = 'worker-%s.tar' % hostID
logDump = os.path.join(output, logName)
try:
r = requests.get(resURI, timeout=DMON_TIMEOUT, stream=True)
if r.status_code == 200:
with open(
logDump,
'wb') as out_file: # TODO investaigate chunck writter
shutil.copyfileobj(r.raw, out_file)
response['Node'] = resURI
response['StatusCode'] = r.status_code
response['LogName'] = logDump
response['Headers'] = r.headers
del r
except requests.exceptions.Timeout:
response['Node'] = resURI
response['StatusCode'] = 408
response['LogName'] = logDump
except requests.exceptions.ConnectionError:
response['Node'] = resURI
response['StatusCode'] = 404
response['LogName'] = logDump
GreenletRequests.NodeResponsesGet.append(response)
# print 'Threaded GET with ID ' + str(GreenletRequests.ng) + ' executed for ' + resURI
app.logger.info(
'[%s] : [INFO] Thread File GET with ID %s executed for %s',
datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S'),
str(GreenletRequests.ng), resURI)
GreenletRequests.ng += 1
gevent.sleep(0)
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 deleteRequest(queue):
response = {}
while not queue.empty():
resURI = queue.get(timeout=1)
try:
r = requests.delete(resURI, timeout=2)
data = r.json()
response["Node"] = resURI
response["StatusCode"] = r.status_code
response["Data"] = data
except requests.exceptions.Timeout:
response["Node"] = resURI
response["StatusCode"] = 408
response["Data"] = "n/a"
except requests.exceptions.ConnectionError:
response["Node"] = resURI
response["StatusCode"] = 404
response["Data"] = "n/a"
GreenletRequests.NodeResponsesGet.append(response)
print "Threaded DELETE with ID " + str(GreenletRequests.nd) + " executed for " + resURI
GreenletRequests.nd += 1
gevent.sleep(0)
def getData():
#가지고 있는 url만큼만 loop
global error_count
error_log = open('./err.txt', mode='a')
while not queue.empty():
#저장되어있는 link를 queue에서 가져옴
#pool의 worker들이 link로 request 동기보다 n배 빠름
link = queue.get(timeout=0)
if link != "":
gevent.sleep(0.3)
getdata = requests.get(link)
soup = BS(getdata.text, 'lxml-xml')
#validation check
okflag = soup.find('resultCode')
try:
if okflag.text != '00':
print("okflag: ", okflag.text)
raise ValueError('okcode is not 00')
else:
#검색잘되면 엑셀 파싱
#pool map method vs pool map_async
#어떤것이 더 효율이 좋을지 결정필요
print(len(soup.find_all('item')))
pool_excel.map(makeCSV, soup.find_all('item'))
except:
error_log.write(link + '\n')
error_log.write('==================================\n')
error_count += 1
error_log.write(str(error_count) + '\n')
queue.put(link)
print('stop crwaling')
print(main_row)
error_log.close()
if result != '':
print 'Found [%s][%s] in %s' % (result, link, tag)
a += 1
if tag in json_dict:
json_dict[tag].append((result, link, imgs[i]))
else:
json_dict[tag] = list()
json_dict[tag].append((result, link, imgs[i]))
r = session.get(url)
tree = html.fromstring(r.text)
a_tags = tree.xpath('//li[@class="menu-item"]//a')
tags = [(x.xpath('.//@href'), repr(x.xpath('.//text()'))) for x in a_tags]
for t in tags:
url = t[0]
result = regex.findall(t[1])
# print url, result
# scrape(url[0], result[0])
queue.put((url[0], result[0]))
while not queue.empty() and not pool.full():
for x in xrange(0, min(queue.qsize(), pool.free_count())):
pool.spawn(worker)
pool.join()
print a
print 'Time Taken : ', datetime.now() - start_time
with open('data.json', 'w') as fp:
json.dump(json_dict, fp)
(response.status_code, url))
except gevent.queue.Empty:
print('queue empty')
break
if __name__ == '__main__':
if len(sys.argv) != 3:
print('USAGE:\n\t%s <base_url> <entry_path>' % sys.argv[0])
sys.exit(1)
if validators.url(sys.argv[1]) != True:
print('Invalid Url')
sys.exit(1)
queue.put(getUrl(sys.argv[2]))
pool.spawn(crawler)
while 1:
if queue.empty() and pool.free_count() == WORKER_COUNT:
print('No more links left and nothing running')
break
for x in range(0, min(queue.qsize(), pool.free_count())):
pool.spawn(crawler)
gevent.sleep(0.1)
# Wait for everything to complete
pool.join()
break
print "job done"
handler.log("job done")
print "so far crawled %s pages" % crawled
handler.log("so far crawled %s pages" % crawled)
queue.put(start_url_1)
queue.put(start_url_2)
pool.spawn(crawler)
handler = Handler()
print 'starting Crawler...'
handler.log('starting Crawler...')
while not queue.empty() and not pool.free_count() == workers_count:
gevent.sleep(0.8)
for x in xrange(0, min(queue.qsize(), pool.free_count())):
pool.spawn(crawler)
#wait for jobs to finish
pool.join()
print "Done"
handler.log("Done+\n")
print '\n'
print "collected %s imgs" % ITEMS_COUNT
handler.log("collected %s imgs" % ITEMS_COUNT)
print "see generated output and log files"
handler.close() #close the IO files
global crawled
while 1:
try:
u = queue.get(timeout=0)
response = requests.get(u)
print response.status_code, u
for link in re.findall('<a href="(http.*?)"', response.content):
if crawled < 10:
crawled += 1
queue.put(link)
except gevent.queue.Empty:
break
# Read the seed url from stdin
queue.put(sys.argv[1])
pool.spawn(crawler)
while not queue.empty() and not pool.free_count() == 5:
gevent.sleep(0.1)
for x in xrange(0, min(queue.qsize(), pool.free_count())):
pool.spawn(crawler)
# Wait for everything to complete
pool.join()
print datetime.now() - startTime # Took 5.943 seconds, varying
def crawler():
'''A very simple queued gevent web crawler'''
global crawled
while 1:
try:
u = queue.get(timeout=1)
response = requests.get(u)
print(response.status_code)
# Extract some links to follow
for link in re.findall('<a href="(http.*?)"', response.content):
# Limit to 10 pages (ignores links when the pool is already full)
if crawled < 10:
crawled += 1
queue.put(link)
except gevent.queue.Empty:
break
queue.put(sys.argv[1])
while not queue.empty() and not pool.free_count() == 5:
gevent.sleep(0.1)
for x in xrange(0, min(queue.qsize(), pool.free_count())):
pool.spawn(crawler)
# Wait for everything to complete
pool.join()
