Git-Mirrors/brozzler
1
0
Fork 0
mirror of https://github.com/internetarchive/brozzler.git synced 2025-04-21 16:16:28 -04:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Merge branch 'max-claimed-sites' into qa

Browse Source 
* max-claimed-sites:
  fix timely time limit enforcement
  honor stop request before choosing proxy
  fix query to make tests pass?
This commit is contained in:
Noah Levitt 2018-03-05 17:11:10 -08:00
commit 0c9ebcff6e
2 changed files with 19 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
brozzler/frontier.py
View File

@ -98,7 +98,7 @@ class RethinkDbFrontier:
Returns a dictionary that looks like this:
{<job_id>: {'claimed_sites': 2, 'max_claimed_sites': 3}, ...}
'''
# js query: r.db('brozzler').table('sites').between(['ACTIVE', r.minval], ['ACTIVE', r.maxval], {'index': 'sites_last_disclaimed'}).eqJoin('job_id', r.db('brozzler').table('jobs')).group(function(x){return x('right')('id')}).ungroup().map(function(x){return {'job_id': x('group'), 'max_claimed_sites':x('reduction')('right')('max_claimed_sites')(0), 'claimed_sites':x('reduction')('left').filter({'claimed':true}).count()}})
# js query: r.db('brozzler').table('sites').between(['ACTIVE', r.minval], ['ACTIVE', r.maxval], {'index': 'sites_last_disclaimed'}).eqJoin('job_id', r.db('brozzler').table('jobs')).filter(function(x){return x('right').hasFields('max_claimed_sites')}).group(function(x){return x('right')('id')}).ungroup().map(function(x){return {'job_id': x('group'), 'max_claimed_sites':x('reduction')('right')('max_claimed_sites')(0), 'claimed_sites':x('reduction')('left').filter({'claimed':true}).count()}})
# returns results like:
# [{'max_claimed_sites': 2, 'claimed_sites': 0, 'job_id': 1234},
# {'max_claimed_sites': 3, 'claimed_sites': 3, 'job_id': 1235}]
@ -108,6 +108,7 @@ class RethinkDbFrontier:
['ACTIVE', r.minval], ['ACTIVE', r.maxval],
index='sites_last_disclaimed')
.eq_join('job_id', r.db(self.rr.dbname).table('jobs'))
.filter(lambda x: x['right'].has_fields('max_claimed_sites'))
.group(lambda x: x['right']['id'])
.ungroup()
.map(lambda x: {
@ -115,10 +116,14 @@ class RethinkDbFrontier:
'max_claimed_sites': x['reduction']['right']['max_claimed_sites'][0],
'claimed_sites': x['reduction']['left'].filter({'claimed':True}).count()
})).run()
xformed = {d['job_id']: {
'claimed_sites': d['claimed_sites'],
'max_claimed_sites': d['max_claimed_sites']
} for d in results}
xformed = {
d['job_id']: {
'claimed_sites': d['claimed_sites'],
'max_claimed_sites': d['max_claimed_sites']
}
for d in results
}
return xformed
def claim_sites(self, n=1):
@ -192,12 +197,13 @@ class RethinkDbFrontier:
else:
raise brozzler.NothingToClaim
def enforce_time_limit(self, site):
def enforce_time_limit(self, site, session_time=0):
'''
Raises `brozzler.ReachedTimeLimit` if appropriate.
'''
if (site.time_limit and site.time_limit > 0
and (site.active_brozzling_time or 0) > site.time_limit):
if (site.time_limit
and site.time_limit > 0
and (site.active_brozzling_time or 0) + session_time > site.time_limit):
self.logger.debug(
"site FINISHED_TIME_LIMIT! time_limit=%s "
"active_brozzling_time=%s %s", site.time_limit,

6
brozzler/worker.py
View File

@ -507,12 +507,16 @@ class BrozzlerWorker:
site.save()
start = time.time()
page = None
self._frontier.enforce_time_limit(site)
self._frontier.honor_stop_request(site)
# _proxy_for() call in log statement can raise brozzler.ProxyError
# which is why we honor time limit and stop request first☝🏻
self.logger.info(
"brozzling site (proxy=%r) %r",
self._proxy_for(site), site)
while time.time() - start < self.SITE_SESSION_MINUTES * 60:
site.refresh()
self._frontier.enforce_time_limit(site)
self._frontier.enforce_time_limit(site, time.time() - start)
self._frontier.honor_stop_request(site)
page = self._frontier.claim_page(site, "%s:%s" % (
socket.gethostname(), browser.chrome.port))