def handle_fault_execution(request, schedule_time, exception):
unique_id = request["unique_id"]
job_url = request["url"]
project_name = request["project_name"]
job_name = request["job_name"]
user_id = request["user_id"]
crawler_name = request["crawler_name"]
schedule_category = request["schedule_category"]
# update relevant MongoDC entry in jobs collection with task_id and status
update_data = u'{ "unique_id": "' + unique_id + '", "url": "' + job_url + '", "project_name": "' \
+ project_name + '", "job_name": "' + job_name + '", "user_id": "' + user_id \
+ '", "crawler_name": "' + crawler_name \
+ '", "task_id": "Not Generated", "status": "FAILED" }'
data_item = json.loads(update_data)
data_item['schedule_category'] = schedule_category
data_item['schedule_time'] = schedule_time
mongo_connection = MongoConnection()
if schedule_category == "Instant":
# update relevant MongoDC entry in jobs collection with task_id and status
query = {'user_id': user_id, 'url': job_url, 'project_name': project_name, 'job_name': job_name,
'crawler_name': crawler_name}
mongo_connection.upsert_item(query, data_item, "jobs")
else:
# store job records in MongoDB database
mongo_connection.insert_item(data_item, "jobs")
logger.exception("Current can not schedule with invalid date or time format. "
"Hence, job execution failed. " + str(exception))
def project_create(request):
# take urls comes from client.
try:
json_data = json.loads(request.body)
user_id = json_data['user_id']
project_name = json_data['project_name']
if not user_id:
return JsonResponse(
{'Error': 'Request payload does not contain user_id'},
status=400)
if not project_name:
return JsonResponse(
{'Error': 'Request payload does not contain project_name'},
status=400)
except JSONDecodeError:
return JsonResponse(
{
'Error':
'Request payload does not contain required parameters or empty'
},
status=400)
# user Authorization
token_header = request.headers.get('Token')
auth = FirebaseAuth(token_header, user_id)
if not auth:
return JsonResponse(
{
'Error':
'User authentication failed. Please try again with a valid user login'
},
status=400)
try:
mongo_connection = MongoConnection()
data_item = dict(json_data)
query = {
'user_id': data_item['user_id'],
'project_name': data_item['project_name']
}
mongo_connection.upsert_item(query, data_item, "projects")
except Exception as e:
return JsonResponse(
{
'Error':
'Error while connecting to the MongoDB database, ' + str(e)
},
status=400)
return JsonResponse({
'status':
"SUCCESS",
'Message':
'Project:' + project_name + ' created successfully'
})
def crawl_new_job(request):
# Post requests are for new crawling tasks
if request.method == 'POST':
# take urls comes from client.
try:
json_data = json.loads(request.body)
url_data = json_data['urls']
job_name = json_data['job_name']
project_name = json_data['project_name']
user_id = json_data['user_id']
crawler_name = json_data['crawler_name']
schedule_type = json_data['schedule_type']
schedule_data = json_data['schedule_data']
except (JSONDecodeError, KeyError) as e:
return JsonResponse({'Error': 'Missing fields in the request payload or empty, ' + str(e)}, status=400)
if not user_id:
return JsonResponse({'Error': 'Missing user id key in the request payload'}, status=400)
if not job_name:
return JsonResponse({'Error': 'Missing job name key in the request payload'}, status=400)
if not url_data:
return JsonResponse({'Error': 'Missing urls key in the request payload'}, status=400)
if not project_name:
return JsonResponse({'Error': 'Missing project_name key in the request payload'}, status=400)
if not crawler_name:
return JsonResponse({'Error': 'Missing crawler_name key in the request payload'}, status=400)
if not schedule_type:
return JsonResponse({'Error': 'Missing schedule_type key in the request payload'}, status=400)
if (schedule_type != SCHEDULE_TASK_TYPE) \
and (schedule_type != INTERVAL_TASK_TYPE) and (schedule_type != HOT_TASK_TYPE):
return JsonResponse({'Error': 'Requested crawler_type:' + schedule_type + ' is not a valid type'},
status=400)
publish_url_ids = []
for url in url_data:
if not is_valid_url(url):
return JsonResponse({'Error': url + ' URL is invalid'}, status=400)
unique_id = str(uuid4()) # create a unique ID.
publish_data = u'{ "unique_id": "' + unique_id + '", "job_name": "' + job_name \
+ '", "url": "' + url + '", "project_name": "' \
+ project_name + '", "user_id": "' + user_id + '", "crawler_name": "' + crawler_name \
+ '", "task_id":"" }'
publish_data = json.loads(publish_data)
try:
# schedule data with celery task scheduler
if schedule_type == SCHEDULE_TASK_TYPE:
publish_data['schedule_data'] = schedule_data
publish_data['schedule_category'] = CRON
publish_data['status'] = "RUNNING"
celery_task = schedule_job_with_cron_tab(publish_data)
elif schedule_type == INTERVAL_TASK_TYPE:
publish_data['schedule_data'] = schedule_data
publish_data['schedule_category'] = INTERVAL
publish_data['status'] = "RUNNING"
celery_task = schedule_job_with_interval(publish_data)
else:
publish_data['schedule_category'] = INSTANT
publish_data['status'] = "PENDING"
celery_task = schedule_cron_job.delay(kwargs=json.dumps(publish_data))
if isinstance(celery_task, JsonResponse):
return celery_task
publish_url_ids.append(unique_id)
publish_data['celery_task_name'] = celery_task.name
try:
# store job records in MongoDB database
query = {'user_id': user_id, 'job_name': job_name, 'url': url,
'project_name': project_name, 'crawler_name': crawler_name}
mongo_connection = MongoConnection()
mongo_connection.upsert_item(query, publish_data, "jobs")
except Exception as e:
return JsonResponse({'Error': 'Error while connecting to the MongoDB database, ' + str(e)},
status=400)
except Exception as e:
return JsonResponse({'Status': "400 BAD",
'Error': 'Error occurred while scheduling the data with the Celery executor, '
+ str(e)}, status=400)
return JsonResponse({'status': "SUCCESS", 'Message': "Crawl job scheduled successfully.\n job_ids:"
+ str(publish_url_ids)})
def schedule_cron_job(self, **kwargs):
json_body = ""
schedule_time = str(time.time())
try:
json_body = kwargs
if "kwargs" in json_body:
json_body = json.loads(json_body['kwargs'])
unique_id = json_body['unique_id']
job_url = json_body["url"]
project_name = json_body["project_name"]
job_name = json_body["job_name"]
user_id = json_body["user_id"]
status = json_body["status"]
crawler_name = json_body["crawler_name"]
schedule_category = json_body["schedule_category"]
if not unique_id or not job_url or not project_name or not user_id or not status \
or not crawler_name or not job_name:
raise Exception('Required parameters are missing in the consumed message')
if check_url_gives_response(job_url):
job_domain = urlparse(job_url).netloc
try:
settings = {
'unique_id': unique_id,
'user_id': user_id,
'job_name': job_name,
'project_name': project_name,
'schedule_time': schedule_time,
'USER_AGENT': 'Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)'
}
# to list available spiders in the project
# print(scrapyd.list_spiders("crawlerx_project"))
# Schedule a crawl job with project and a specific spider
task_id = \
scrapy_daemon.schedule("crawlerx_project", crawler_name, settings=settings,
url=job_url, domain=job_domain)
mongo_connection = MongoConnection()
job_data = u'{ "unique_id": "' + unique_id + '", "url": "' + job_url + '", "project_name": "' \
+ project_name + '", "job_name": "' + job_name + '", "user_id": "' + user_id \
+ '", "crawler_name": "' + crawler_name \
+ '", "task_id": "' + task_id + '", "status": "RUNNING" }'
data_item = json.loads(job_data)
data_item['schedule_category'] = schedule_category
data_item['schedule_time'] = schedule_time
if schedule_category == "Instant":
# update relevant MongoDC entry in jobs collection with task_id and status
query = {'user_id': user_id, 'url': job_url, 'project_name': project_name, 'job_name': job_name,
'crawler_name': crawler_name}
mongo_connection.upsert_item(query, data_item, "jobs")
else:
# store job records in MongoDB database
mongo_connection.insert_item(data_item, "jobs")
# task id of the crawl job
logger.info("Crawling job has been started with ID - " + task_id)
except Exception as e:
handle_fault_execution(json_body, schedule_time, e)
else:
handle_fault_execution(json_body, schedule_time,
Exception("Current job URL does not seems available. Hence, job execution failed."))
except Exception as e:
handle_fault_execution(json_body, schedule_time, e)