def _getstatus(self): ''' Prepare and return the crawler's current status. ''' queued = QueuedURL.query().filter(QueuedURL.started == False) pending = QueuedURL.query().filter(QueuedURL.started == True).filter(QueuedURL.finished == False) finished = QueuedURL.query().filter(QueuedURL.finished == True) queued_count = queued.count() pending_count = pending.count() finished_count = finished.count() if queued_count > 0: queued_urls = queued.fetch(queued_count) else: queued_urls = [] if pending_count > 0: pending_urls = pending.fetch(pending_count) else: pending_urls = [] if finished_count > 0: finished_urls = finished.fetch(finished_count) else: finished_urls = [] status = CrawlStatus() status.total_queued = queued_count status.total_pending = pending_count status.total_finished = finished_count for status_mountpoint, group in ((status.queued_urls, queued_urls), (status.pending_urls, pending_urls), (status.finished_urls, finished_urls)): if len(group) > 0: for queued_url in group: ## Build response fragment qu = QueuedURLMessage(key=queued_url.key.urlsafe()) qu.url = queued_url.url qu.started = queued_url.started if queued_url.key.parent() is not None: qu.parent = queued_url.key.parent().urlsafe() qu.parent_url=queued_url.key.parent().id_or_name() qu.modified = str(queued_url.modified) qu.created = str(queued_url.created) if queued_url.status_code: qu.status_code = queued_url.status_code if queued_url.task_id: qu.task_id = queued_url.task_id status_mountpoint.append(qu) else: status_mountpoint = [] return status
def start(self, request):
''' Begin the crawl routine '''
global add_url
logging.critical('========== STARTING CRAWL ROUTINE ==========')
q = QueuedURL.query()
c = q.count()
qu = q.fetch(c)
logging.info('Found '+str(c)+' queued URLs. Kicking off tasks.')
crawl_tasks = []
for queued_url in qu:
logging.info('--Building task for queued URL at key "'+str(queued_url.key.urlsafe())+'"...')
task, qu = add_url(queued_url, None, False)
logging.info('------Built task: "'+str(task.name)+'".')
crawl_tasks.append(task)
logging.info('------Adding to deferred batch at index '+str(crawl_tasks.index(task))+'.')
try:
logging.info('Finished generating deferred batch. Kicking off tasks! :)')
if len(crawl_tasks) > 0:
logging.info('--Found '+str(len(crawl_tasks))+' crawltasks, which is more than 0.')
else:
logging.error('NO CRAWLTASKS :(')
raise remote.ApplicationError('No crawltasks could be generated :(')
except:
raise
else:
tasks = self.api.taskqueue.Queue('crawler').add(crawl_tasks)
logging.info('========== '+str(len(tasks))+' TASKS ENQUEUED '+'==========')
return self._getstatus()
def get(self):
queued_urls = QueuedURL.query()
scraped_pages = ScrapedPage.query()
qu_count = queued_urls.count()
if qu_count < 5:
qu_count = 5
sc_count = scraped_pages.count()
if sc_count < 5:
sc_count = 5
return self.render('main/urls.html', queued=queued_urls.fetch(qu_count), scraped=scraped_pages.fetch(sc_count))
def add_url(self, request):
''' Add a URL to the crawl routine '''
logging.info('========== ADD_URL ==========')
if request.parent is not None:
logging.info('URL has parent.')
p = ndb.Key(urlsafe=request.parent)
q = QueuedURL(key=ndb.Key(k, QueuedURL, request.url), url=request.url)
k = q.put()
else:
logging.info('URL has NO parent.')
q = QueuedURL(key=ndb.Key(QueuedURL, request.url), url=request.url)
k = q.put()
logging.info('QueuedURL key: '+k.urlsafe())
c = QueuedURL.query().count()
logging.info('New URL count: '+str(c))
## Generate response
qu = QueuedURLMessage(key=k.urlsafe())
qu.url = request.url
qu.started = q.started
if q.key.parent() is not None:
qu.parent = q.key.parent().urlsafe()
qu.parent_url=q.key.parent().id_or_name()
qu.modified = str(q.modified)
qu.created = str(q.created)
if q.status_code:
qu.status_code = q.status_code
if q.task_id:
qu.task_id = q.task_id
return AddURLResponse(added=True, newcount=c, queued_url=qu)