SickGear/sickbeard/search_queue.py

468 lines
17 KiB
Python
Raw Normal View History

# Author: Nic Wolfe <nic@wolfeden.ca>
# URL: http://code.google.com/p/sickbeard/
#
# This file is part of SickGear.
#
# SickGear is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# SickGear is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with SickGear. If not, see <http://www.gnu.org/licenses/>.
from __future__ import with_statement
import traceback
import threading
import datetime
import sickbeard
from sickbeard import db, logger, common, exceptions, helpers, network_timezones, generic_queue, search, \
failed_history, history, ui, properFinder
from sickbeard.search import wanted_episodes
search_queue_lock = threading.Lock()
BACKLOG_SEARCH = 10
RECENT_SEARCH = 20
FAILED_SEARCH = 30
MANUAL_SEARCH = 40
PROPER_SEARCH = 50
MANUAL_SEARCH_HISTORY = []
MANUAL_SEARCH_HISTORY_SIZE = 100
class SearchQueue(generic_queue.GenericQueue):
def __init__(self):
generic_queue.GenericQueue.__init__(self)
self.queue_name = 'SEARCHQUEUE'
def is_in_queue(self, show, segment):
with self.lock:
for cur_item in self.queue:
if isinstance(cur_item, BacklogQueueItem) and cur_item.show == show and cur_item.segment == segment:
return True
return False
def is_ep_in_queue(self, segment):
with self.lock:
for cur_item in self.queue:
if isinstance(cur_item, (ManualSearchQueueItem, FailedQueueItem)) and cur_item.segment == segment:
return True
return False
def is_show_in_queue(self, show):
with self.lock:
for cur_item in self.queue:
if isinstance(cur_item, (ManualSearchQueueItem, FailedQueueItem)) and cur_item.show.indexerid == show:
return True
return False
def get_all_ep_from_queue(self, show):
with self.lock:
ep_obj_list = []
for cur_item in self.queue:
if isinstance(cur_item, (ManualSearchQueueItem, FailedQueueItem)) and str(cur_item.show.indexerid) == show:
ep_obj_list.append(cur_item)
if ep_obj_list:
return ep_obj_list
return False
def pause_backlog(self):
with self.lock:
self.min_priority = generic_queue.QueuePriorities.HIGH
def unpause_backlog(self):
with self.lock:
self.min_priority = 0
def is_backlog_paused(self):
# backlog priorities are NORMAL, this should be done properly somewhere
with self.lock:
return self.min_priority >= generic_queue.QueuePriorities.NORMAL
def _is_in_progress(self, item_type):
with self.lock:
for cur_item in self.queue + [self.currentItem]:
if isinstance(cur_item, item_type):
return True
return False
def is_manualsearch_in_progress(self):
# Only referenced in webserve.py, only current running manualsearch or failedsearch is needed!!
return self._is_in_progress((ManualSearchQueueItem, FailedQueueItem))
def is_backlog_in_progress(self):
return self._is_in_progress(BacklogQueueItem)
def is_recentsearch_in_progress(self):
return self._is_in_progress(RecentSearchQueueItem)
def is_propersearch_in_progress(self):
return self._is_in_progress(ProperSearchQueueItem)
def is_standard_backlog_in_progress(self):
with self.lock:
for cur_item in self.queue + [self.currentItem]:
if isinstance(cur_item, BacklogQueueItem) and cur_item.standard_backlog:
return True
return False
def type_of_backlog_in_progress(self):
limited = full = other = False
with self.lock:
for cur_item in self.queue + [self.currentItem]:
if isinstance(cur_item, BacklogQueueItem):
if cur_item.standard_backlog:
if cur_item.limited_backlog:
limited = True
else:
full = True
else:
other = True
types = []
for msg, variant in ['Limited', limited], ['Full', full], ['On Demand', other]:
if variant:
types.append(msg)
message = 'None'
if types:
message = ', '.join(types)
return message
def queue_length(self):
length = {'backlog': [], 'recent': 0, 'manual': [], 'failed': [], 'proper': 0}
with self.lock:
for cur_item in [self.currentItem] + self.queue:
if isinstance(cur_item, RecentSearchQueueItem):
length['recent'] += 1
elif isinstance(cur_item, BacklogQueueItem):
length['backlog'].append([cur_item.show.indexerid, cur_item.show.name, cur_item.segment, cur_item.standard_backlog, cur_item.limited_backlog, cur_item.forced])
elif isinstance(cur_item, ProperSearchQueueItem):
length['proper'] += 1
elif isinstance(cur_item, ManualSearchQueueItem):
length['manual'].append([cur_item.show.indexerid, cur_item.show.name, cur_item.segment])
elif isinstance(cur_item, FailedQueueItem):
length['failed'].append([cur_item.show.indexerid, cur_item.show.name, cur_item.segment])
return length
def add_item(self, item):
if isinstance(item, (RecentSearchQueueItem, ProperSearchQueueItem)):
# recent and proper searches
generic_queue.GenericQueue.add_item(self, item)
elif isinstance(item, BacklogQueueItem) and not self.is_in_queue(item.show, item.segment):
# backlog searches
generic_queue.GenericQueue.add_item(self, item)
elif isinstance(item, (ManualSearchQueueItem, FailedQueueItem)) and not self.is_ep_in_queue(item.segment):
# manual and failed searches
generic_queue.GenericQueue.add_item(self, item)
else:
logger.log(u'Not adding item, it\'s already in the queue', logger.DEBUG)
class RecentSearchQueueItem(generic_queue.QueueItem):
def __init__(self):
2014-09-21 20:42:01 +00:00
self.success = None
self.episodes = []
generic_queue.QueueItem.__init__(self, 'Recent Search', RECENT_SEARCH)
Fixed issues with editing/saving custom scene exceptions. Fixed charmap issues for anime show names. Fixed issues with display show page and epCat key errors. Fixed duplicate log messages for clearing provider caches. Fixed issues with email notifier ep names not properly being encoded to UTF-8. TVDB<->TVRAGE Indexer ID mapping is now performed on demand to be used when needed such as newznab providers can be searched with tvrage_id's and some will return tvrage_id's that later can be used to create show objects from for faster and more accurate name parsing, mapping is done via Trakt API calls. Added stop event signals to schedualed tasks, SR now waits indefinate till task has been fully stopped before completing a restart or shutdown event. NameParserCache is now persistent and stores 200 parsed results at any given time for quicker lookups and better performance, this helps maintain results between updates or shutdown/startup events. Black and White lists for anime now only get used for anime shows as intended, performance gain for non-anime shows that dont need to load these lists. Internal name cache now builds it self on demand when needed per show request plus checks if show is already in cache and if true exits routine to save time. Schedualer and QueueItems classes are now a sub-class of threading.Thread and a stop threading event signal has been added to each. If I forgot to list something it doesn't mean its not fixed so please test and report back if anything is wrong or has been corrected by this new release.
2014-07-15 02:00:53 +00:00
def run(self):
generic_queue.QueueItem.run(self)
try:
self._change_missing_episodes()
self.update_providers()
show_list = sickbeard.showList
from_date = datetime.date.fromordinal(1)
for curShow in show_list:
if curShow.paused:
continue
self.episodes.extend(wanted_episodes(curShow, from_date))
if not self.episodes:
logger.log(u'No search of cache for episodes required')
self.success = True
else:
num_shows = len(set([ep.show.name for ep in self.episodes]))
logger.log(u'Found %d needed episode%s spanning %d show%s'
% (len(self.episodes), helpers.maybe_plural(len(self.episodes)),
num_shows, helpers.maybe_plural(num_shows)))
try:
logger.log(u'Beginning recent search for episodes')
found_results = search.search_for_needed_episodes(self.episodes)
if not len(found_results):
logger.log(u'No needed episodes found')
else:
for result in found_results:
# just use the first result for now
logger.log(u'Downloading %s from %s' % (result.name, result.provider.name))
self.success = search.snatch_episode(result)
helpers.cpu_sleep()
except Exception:
logger.log(traceback.format_exc(), logger.DEBUG)
if None is self.success:
self.success = False
2014-09-21 20:42:01 +00:00
finally:
self.finish()
@staticmethod
def _change_missing_episodes():
if not network_timezones.network_dict:
network_timezones.update_network_dict()
if network_timezones.network_dict:
cur_date = (datetime.date.today() + datetime.timedelta(days=1)).toordinal()
else:
cur_date = (datetime.date.today() - datetime.timedelta(days=2)).toordinal()
cur_time = datetime.datetime.now(network_timezones.sb_timezone)
my_db = db.DBConnection()
sql_results = my_db.select('SELECT * FROM tv_episodes WHERE status = ? AND season > 0 AND airdate <= ?',
[common.UNAIRED, cur_date])
sql_l = []
show = None
wanted = False
for sqlEp in sql_results:
try:
if not show or show.indexerid != int(sqlEp['showid']):
show = helpers.findCertainShow(sickbeard.showList, int(sqlEp['showid']))
# for when there is orphaned series in the database but not loaded into our showlist
if not show:
continue
except exceptions.MultipleShowObjectsException:
logger.log(u'ERROR: expected to find a single show matching %s' % sqlEp['showid'])
continue
try:
end_time = network_timezones.parse_date_time(sqlEp['airdate'], show.airs, show.network) + datetime.timedelta(minutes=helpers.tryInt(show.runtime, 60))
# filter out any episodes that haven't aired yet
if end_time > cur_time:
continue
except:
# if an error occurred assume the episode hasn't aired yet
continue
ep = show.getEpisode(int(sqlEp['season']), int(sqlEp['episode']))
with ep.lock:
# Now that it is time, change state of UNAIRED show into expected or skipped
ep.status = (common.WANTED, common.SKIPPED)[ep.show.paused]
result = ep.get_sql()
if None is not result:
sql_l.append(result)
wanted |= (False, True)[common.WANTED == ep.status]
else:
logger.log(u'No unaired episodes marked wanted')
if 0 < len(sql_l):
my_db = db.DBConnection()
my_db.mass_action(sql_l)
if wanted:
logger.log(u'Found new episodes marked wanted')
@staticmethod
def update_providers():
orig_thread_name = threading.currentThread().name
threads = []
providers = [x for x in sickbeard.providers.sortedProviderList() if x.is_active() and x.enable_recentsearch]
for cur_provider in providers:
if not cur_provider.cache.should_update():
continue
if not threads:
logger.log('Updating provider caches with recent upload data')
# spawn a thread for each provider to save time waiting for slow response providers
threads.append(threading.Thread(target=cur_provider.cache.updateCache,
name='%s :: [%s]' % (orig_thread_name, cur_provider.name)))
# start the thread we just created
threads[-1].start()
if not len(providers):
logger.log('No NZB/Torrent sources enabled in Search Provider options for cache update', logger.WARNING)
if threads:
# wait for all threads to finish
for t in threads:
t.join()
logger.log('Finished updating provider caches')
class ProperSearchQueueItem(generic_queue.QueueItem):
def __init__(self):
generic_queue.QueueItem.__init__(self, 'Proper Search', PROPER_SEARCH)
self.priority = generic_queue.QueuePriorities.HIGH
self.success = None
def run(self):
generic_queue.QueueItem.run(self)
try:
properFinder.search_propers()
finally:
self.finish()
class ManualSearchQueueItem(generic_queue.QueueItem):
def __init__(self, show, segment):
generic_queue.QueueItem.__init__(self, 'Manual Search', MANUAL_SEARCH)
self.priority = generic_queue.QueuePriorities.HIGH
self.name = 'MANUAL-%s' % show.indexerid
self.success = None
self.show = show
self.segment = segment
self.started = None
Fixed issues with editing/saving custom scene exceptions. Fixed charmap issues for anime show names. Fixed issues with display show page and epCat key errors. Fixed duplicate log messages for clearing provider caches. Fixed issues with email notifier ep names not properly being encoded to UTF-8. TVDB<->TVRAGE Indexer ID mapping is now performed on demand to be used when needed such as newznab providers can be searched with tvrage_id's and some will return tvrage_id's that later can be used to create show objects from for faster and more accurate name parsing, mapping is done via Trakt API calls. Added stop event signals to schedualed tasks, SR now waits indefinate till task has been fully stopped before completing a restart or shutdown event. NameParserCache is now persistent and stores 200 parsed results at any given time for quicker lookups and better performance, this helps maintain results between updates or shutdown/startup events. Black and White lists for anime now only get used for anime shows as intended, performance gain for non-anime shows that dont need to load these lists. Internal name cache now builds it self on demand when needed per show request plus checks if show is already in cache and if true exits routine to save time. Schedualer and QueueItems classes are now a sub-class of threading.Thread and a stop threading event signal has been added to each. If I forgot to list something it doesn't mean its not fixed so please test and report back if anything is wrong or has been corrected by this new release.
2014-07-15 02:00:53 +00:00
def run(self):
generic_queue.QueueItem.run(self)
try:
logger.log(u'Beginning manual search for: [%s]' % self.segment.prettyName())
self.started = True
search_result = search.search_providers(self.show, [self.segment], True)
if search_result:
# just use the first result for now
logger.log(u'Downloading %s from %s' % (search_result[0].name, search_result[0].provider.name))
self.success = search.snatch_episode(search_result[0])
helpers.cpu_sleep()
else:
ui.notifications.message('No downloads found',
u'Could not find a download for <i>%s</i>' % self.segment.prettyName())
logger.log(u'Unable to find a download for: [%s]' % self.segment.prettyName())
except Exception:
logger.log(traceback.format_exc(), logger.DEBUG)
finally:
# Keep a list with the 100 last executed searches
fifo(MANUAL_SEARCH_HISTORY, self, MANUAL_SEARCH_HISTORY_SIZE)
if self.success is None:
self.success = False
self.finish()
class BacklogQueueItem(generic_queue.QueueItem):
def __init__(self, show, segment, standard_backlog=False, limited_backlog=False, forced=False):
generic_queue.QueueItem.__init__(self, 'Backlog', BACKLOG_SEARCH)
self.priority = generic_queue.QueuePriorities.LOW
self.name = 'BACKLOG-%s' % show.indexerid
2014-05-12 17:49:59 +00:00
self.success = None
self.show = show
self.segment = segment
self.standard_backlog = standard_backlog
self.limited_backlog = limited_backlog
self.forced = forced
Fixed issues with editing/saving custom scene exceptions. Fixed charmap issues for anime show names. Fixed issues with display show page and epCat key errors. Fixed duplicate log messages for clearing provider caches. Fixed issues with email notifier ep names not properly being encoded to UTF-8. TVDB<->TVRAGE Indexer ID mapping is now performed on demand to be used when needed such as newznab providers can be searched with tvrage_id's and some will return tvrage_id's that later can be used to create show objects from for faster and more accurate name parsing, mapping is done via Trakt API calls. Added stop event signals to schedualed tasks, SR now waits indefinate till task has been fully stopped before completing a restart or shutdown event. NameParserCache is now persistent and stores 200 parsed results at any given time for quicker lookups and better performance, this helps maintain results between updates or shutdown/startup events. Black and White lists for anime now only get used for anime shows as intended, performance gain for non-anime shows that dont need to load these lists. Internal name cache now builds it self on demand when needed per show request plus checks if show is already in cache and if true exits routine to save time. Schedualer and QueueItems classes are now a sub-class of threading.Thread and a stop threading event signal has been added to each. If I forgot to list something it doesn't mean its not fixed so please test and report back if anything is wrong or has been corrected by this new release.
2014-07-15 02:00:53 +00:00
def run(self):
generic_queue.QueueItem.run(self)
try:
logger.log(u'Beginning backlog search for: [%s]' % self.show.name)
search_result = search.search_providers(self.show, self.segment, False)
if search_result:
for result in search_result:
# just use the first result for now
logger.log(u'Downloading %s from %s' % (result.name, result.provider.name))
search.snatch_episode(result)
helpers.cpu_sleep()
else:
logger.log(u'No needed episodes found during backlog search for: [%s]' % self.show.name)
except Exception:
logger.log(traceback.format_exc(), logger.DEBUG)
finally:
self.finish()
class FailedQueueItem(generic_queue.QueueItem):
def __init__(self, show, segment):
generic_queue.QueueItem.__init__(self, 'Retry', FAILED_SEARCH)
self.priority = generic_queue.QueuePriorities.HIGH
self.name = 'RETRY-%s' % show.indexerid
self.show = show
self.segment = segment
self.success = None
self.started = None
Fixed issues with editing/saving custom scene exceptions. Fixed charmap issues for anime show names. Fixed issues with display show page and epCat key errors. Fixed duplicate log messages for clearing provider caches. Fixed issues with email notifier ep names not properly being encoded to UTF-8. TVDB<->TVRAGE Indexer ID mapping is now performed on demand to be used when needed such as newznab providers can be searched with tvrage_id's and some will return tvrage_id's that later can be used to create show objects from for faster and more accurate name parsing, mapping is done via Trakt API calls. Added stop event signals to schedualed tasks, SR now waits indefinate till task has been fully stopped before completing a restart or shutdown event. NameParserCache is now persistent and stores 200 parsed results at any given time for quicker lookups and better performance, this helps maintain results between updates or shutdown/startup events. Black and White lists for anime now only get used for anime shows as intended, performance gain for non-anime shows that dont need to load these lists. Internal name cache now builds it self on demand when needed per show request plus checks if show is already in cache and if true exits routine to save time. Schedualer and QueueItems classes are now a sub-class of threading.Thread and a stop threading event signal has been added to each. If I forgot to list something it doesn't mean its not fixed so please test and report back if anything is wrong or has been corrected by this new release.
2014-07-15 02:00:53 +00:00
def run(self):
generic_queue.QueueItem.run(self)
self.started = True
try:
for epObj in self.segment:
logger.log(u'Marking episode as bad: [%s]' % epObj.prettyName())
failed_history.markFailed(epObj)
(release, provider) = failed_history.findRelease(epObj)
if release:
failed_history.logFailed(release)
history.logFailed(epObj, release, provider)
failed_history.revertEpisode(epObj)
logger.log(u'Beginning failed download search for: [%s]' % epObj.prettyName())
search_result = search.search_providers(self.show, self.segment, True)
if search_result:
for result in search_result:
# just use the first result for now
logger.log(u'Downloading %s from %s' % (result.name, result.provider.name))
search.snatch_episode(result)
helpers.cpu_sleep()
else:
pass
# logger.log(u'No valid episode found to retry for: [%s]' % self.segment.prettyName())
except Exception:
logger.log(traceback.format_exc(), logger.DEBUG)
finally:
# Keep a list with the 100 last executed searches
fifo(MANUAL_SEARCH_HISTORY, self, MANUAL_SEARCH_HISTORY_SIZE)
if self.success is None:
self.success = False
self.finish()
def fifo(my_list, item, max_size=100):
if len(my_list) >= max_size:
my_list.pop(0)
my_list.append(item)