def auto_history_purge(self):
""" Remove history items based on the configured history-retention """
if sabnzbd.cfg.history_retention() == "0":
return
if sabnzbd.cfg.history_retention() == "-1":
# Delete all non-failed ones
self.remove_completed()
if "d" in sabnzbd.cfg.history_retention():
# How many days to keep?
days_to_keep = int_conv(
sabnzbd.cfg.history_retention().strip()[:-1])
seconds_to_keep = int(time.time()) - days_to_keep * 86400
if days_to_keep > 0:
logging.info(
"Removing completed jobs older than %s days from history",
days_to_keep)
return self.execute(
"""DELETE FROM history WHERE status = ? AND completed < ?""",
(Status.COMPLETED, seconds_to_keep),
save=True,
)
else:
# How many to keep?
to_keep = int_conv(sabnzbd.cfg.history_retention())
if to_keep > 0:
logging.info(
"Removing all but last %s completed jobs from history",
to_keep)
return self.execute(
"""DELETE FROM history WHERE status = ? AND id NOT IN ( SELECT id FROM history WHERE status = ? ORDER BY completed DESC LIMIT ? )""",
(Status.COMPLETED, Status.COMPLETED, to_keep),
save=True,
)
def test_nntp_server_dict(kwargs):
# Grab the host/port/user/pass/connections/ssl
host = kwargs.get('host', '').strip()
if not host:
return False, T('The hostname is not set.')
username = kwargs.get('username', '').strip()
password = kwargs.get('password', '').strip()
server = kwargs.get('server', '').strip()
connections = int_conv(kwargs.get('connections', 0))
ssl = int_conv(kwargs.get('ssl', 0))
ssl_verify = int_conv(kwargs.get('ssl_verify', 1))
ssl_ciphers = kwargs.get('ssl_ciphers')
port = int_conv(kwargs.get('port', 0))
if not connections:
return False, T('There are no connections set. Please set at least one connection.')
if not port:
if ssl:
port = 563
else:
port = 119
return test_nntp_server(host, port, server, username=username, password=password,
ssl=ssl, ssl_verify=ssl_verify, ssl_ciphers=ssl_ciphers)
def test_nntp_server_dict(kwargs):
# Grab the host/port/user/pass/connections/ssl
host = kwargs.get("host", "").strip()
if not host:
return False, T("The hostname is not set.")
username = kwargs.get("username", "").strip()
password = kwargs.get("password", "").strip()
server = kwargs.get("server", "").strip()
connections = int_conv(kwargs.get("connections", 0))
if not connections:
return False, T(
"There are no connections set. Please set at least one connection."
)
ssl = int_conv(kwargs.get("ssl", 0))
ssl_verify = int_conv(kwargs.get("ssl_verify", 1))
ssl_ciphers = kwargs.get("ssl_ciphers")
port = int_conv(kwargs.get("port", 0))
if not port:
if ssl:
port = 563
else:
port = 119
return test_nntp_server(
host,
port,
server,
username=username,
password=password,
ssl=ssl,
ssl_verify=ssl_verify,
ssl_ciphers=ssl_ciphers,
)
def test_nntp_server_dict(kwargs):
# Grab the host/port/user/pass/connections/ssl
host = kwargs.get('host', '').strip()
if not host:
return False, T('The hostname is not set.')
username = kwargs.get('username', '').strip()
password = kwargs.get('password', '').strip()
server = kwargs.get('server', '').strip()
connections = int_conv(kwargs.get('connections', 0))
if not connections:
return False, T(
'There are no connections set. Please set at least one connection.'
)
ssl = int_conv(kwargs.get('ssl', 0))
ssl_verify = int_conv(kwargs.get('ssl_verify', 1))
port = int_conv(kwargs.get('port', 0))
if not port:
if ssl:
port = 563
else:
port = 119
return test_nntp_server(host,
port,
server,
username=username,
password=password,
ssl=ssl,
ssl_verify=ssl_verify)
def analyze_rar_filename(filename):
""" Extract volume number and setname from rar-filenames
Both ".part01.rar" and ".r01"
"""
m = RAR_NR.search(filename)
if m:
if m.group(4):
# Special since starts with ".rar", ".r00"
return m.group(1), int_conv(m.group(4)) + 2
return m.group(1), int_conv(m.group(3))
else:
# Detect if first of "rxx" set
if filename.endswith('.rar'):
return os.path.splitext(filename)[0], 1
return None, None
def analyze_rar_filename(filename):
"""Extract volume number and setname from rar-filenames
Both ".part01.rar" and ".r01"
"""
m = RAR_NR.search(filename)
if m:
if m.group(4):
# Special since starts with ".rar", ".r00"
return m.group(1), int_conv(m.group(4)) + 2
return m.group(1), int_conv(m.group(3))
else:
# Detect if first of "rxx" set
if filename.endswith(".rar"):
return os.path.splitext(filename)[0], 1
return None, None
def test_int_conv(self):
assert 0 == misc.int_conv("0")
assert 10 == misc.int_conv("10")
assert 10 == misc.int_conv(10)
assert 10 == misc.int_conv(10.0)
assert 0 == misc.int_conv(None)
assert 1 == misc.int_conv(True)
assert 0 == misc.int_conv(object)
def ep_match(season, episode, expr, title=None):
""" Return True if season, episode is at or above expected
Optionally `title` can be matched
"""
m = _RE_SP.search(expr)
if m:
# Make sure they are all integers for comparison
req_season = int(m.group(1))
req_episode = int(m.group(2))
season = int_conv(season)
episode = int_conv(episode)
if season > req_season or (season == req_season and episode >= req_episode):
if title:
show = expr[:m.start()].replace('.', ' ').replace('_', ' ').strip()
show = show.replace(' ', '[._ ]+')
return bool(re.search(show, title, re.I))
else:
return True
else:
return False
else:
return True
def ep_match(season, episode, expr, title=None):
""" Return True if season, episode is at or above expected
Optionally `title` can be matched
"""
m = _RE_SP.search(expr)
if m:
# Make sure they are all integers for comparison
req_season = int(m.group(1))
req_episode = int(m.group(2))
season = int_conv(season)
episode = int_conv(episode)
if season > req_season or (season == req_season and episode >= req_episode):
if title:
show = expr[: m.start()].replace(".", " ").replace("_", " ").strip()
show = show.replace(" ", "[._ ]+")
return bool(re.search(show, title, re.I))
else:
return True
else:
return False
else:
return True
def auto_history_purge(self):
""" Remove history items based on the configured history-retention """
if sabnzbd.cfg.history_retention() == "0":
return
if sabnzbd.cfg.history_retention() == "-1":
# Delete all non-failed ones
self.remove_completed()
if "d" in sabnzbd.cfg.history_retention():
# How many days to keep?
days_to_keep = int_conv(sabnzbd.cfg.history_retention().strip()[:-1])
seconds_to_keep = int(time.time()) - days_to_keep * 86400
if days_to_keep > 0:
logging.info('Removing completed jobs older than %s days from history', days_to_keep)
return self.execute("""DELETE FROM history WHERE status = 'Completed' AND completed < ?""", (seconds_to_keep,), save=True)
else:
# How many to keep?
to_keep = int_conv(sabnzbd.cfg.history_retention())
if to_keep > 0:
logging.info('Removing all but last %s completed jobs from history', to_keep)
return self.execute("""DELETE FROM history WHERE id NOT IN ( SELECT id FROM history WHERE status = 'Completed' ORDER BY completed DESC LIMIT ? )""", (to_keep,), save=True)
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
# Don't pound the website!
time.sleep(5.0)
(url, future_nzo) = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo and future_nzo.wait and future_nzo.wait > time.time():
# Re-queue when too early and still active
self.add(url, future_nzo)
continue
url = url.replace(' ', '')
try:
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing', url)
continue
logging.info('Grabbing URL %s', url)
req = urllib2.Request(url)
req.add_header('User-Agent', 'SABnzbd+/%s' % sabnzbd.version.__version__)
if not [True for item in _BAD_GZ_HOSTS if item in url]:
req.add_header('Accept-encoding', 'gzip')
filename = None
category = None
gzipped = False
nzo_info = {}
wait = 0
retry = True
fn = None
try:
fn = urllib2.urlopen(req)
except:
# Cannot list exceptions here, because of unpredictability over platforms
error0 = str(sys.exc_info()[0]).lower()
error1 = str(sys.exc_info()[1]).lower()
logging.debug('Error "%s" trying to get the url %s', error1, url)
if 'certificate_verify_failed' in error1 or 'certificateerror' in error0:
msg = T('Server %s uses an untrusted HTTPS certificate') % ''
retry = False
elif 'nodename nor servname provided' in error1:
msg = T('Server name does not resolve')
retry = False
elif '401' in error1 or 'unauthorized' in error1:
msg = T('Unauthorized access')
retry = False
elif '404' in error1:
msg = T('File not on server')
retry = False
new_url = dereferring(url, fn)
if new_url:
self.add(new_url, future_nzo)
continue
if fn:
for hdr in fn.headers:
try:
item = hdr.lower()
value = fn.headers[hdr]
except:
continue
if item in ('content-encoding',) and value == 'gzip':
gzipped = True
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo',):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name',):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item == 'x-dnzb-propername':
nzo_info['propername'] = value
elif item == 'x-dnzb-episodename':
nzo_info['episodename'] = value
elif item == 'x-dnzb-year':
nzo_info['year'] = value
elif item == 'x-dnzb-failure':
nzo_info['failure'] = value
elif item == 'x-dnzb-details':
nzo_info['details'] = value
elif item == 'retry-after':
# For NZBFinder
wait = misc.int_conv(value)
if not filename and "filename=" in value:
filename = value[value.index("filename=") + 9:].strip(';').strip('"')
if wait:
# For sites that have a rate-limiting attribute
msg = ''
retry = True
fn = None
elif retry:
fn, msg, retry, wait, data = _analyse(fn, url)
if not fn:
if retry:
logging.info('Retry URL %s', url)
self.add(url, future_nzo, wait)
else:
bad_fetch(future_nzo, url, msg)
continue
if not filename:
filename = os.path.basename(url) + '.nzb'
elif '&nzbname=' in filename:
# Sometimes the filename contains the full URL, duh!
filename = filename[filename.find('&nzbname=') + 9:]
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# process data
if gzipped:
filename = filename + '.gz'
if not data:
data = fn.read()
fn.close()
# Sanatize filename first
filename = misc.sanitize_filename(filename)
# Write data to temp file
path = os.path.join(cfg.admin_dir.get_path(), FUTURE_Q_FOLDER)
path = os.path.join(path, filename)
f = open(path, 'wb')
f.write(data)
f.close()
del data
# Check if nzb file
if os.path.splitext(filename)[1].lower() in ('.nzb', '.gz', 'bz2'):
res = dirscanner.ProcessSingleFile(filename, path, pp=pp, script=script, cat=cat, priority=priority,
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url, keep=False,
nzo_id=future_nzo.nzo_id)[0]
if res:
if res == -2:
logging.info('Incomplete NZB, retry after 5 min %s', url)
when = 300
elif res == -1:
# Error, but no reason to retry. Warning is already given
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
continue
else:
logging.info('Unknown error fetching NZB, retry after 2 min %s', url)
when = 120
self.add(url, future_nzo, when)
# Check if a supported archive
else:
if dirscanner.ProcessArchiveFile(filename, path, pp, script, cat, priority=priority,
nzbname=nzbname, url=future_nzo.url, keep=False,
nzo_id=future_nzo.nzo_id)[0]:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
logging.info('Unknown filetype when fetching NZB, retry after 30s %s', url)
self.add(url, future_nzo, 30)
except:
logging.error(T('URLGRABBER CRASHED'), exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
def __set_priority(self, nzo_id, priority):
""" Sets the priority on the nzo and places it in the queue at the appropriate position """
try:
priority = int_conv(priority)
nzo = self.__nzo_table[nzo_id]
nzo_id_pos1 = -1
pos = -1
# If priority == STOP_PRIORITY, then send to queue
if priority == STOP_PRIORITY:
self.end_job(nzo)
return
# Get the current position in the queue
for i in range(len(self.__nzo_list)):
if nzo_id == self.__nzo_list[i].nzo_id:
nzo_id_pos1 = i
break
# Don't change priority and order if priority is the same as asked
if priority == self.__nzo_list[nzo_id_pos1].priority:
return nzo_id_pos1
nzo.set_priority(priority)
if sabnzbd.scheduler.analyse(False, priority) and nzo.status in (
Status.CHECKING,
Status.DOWNLOADING,
Status.QUEUED,
):
nzo.status = Status.PAUSED
elif nzo.status == Status.PAUSED:
nzo.status = Status.QUEUED
nzo.save_to_disk()
if nzo_id_pos1 != -1:
del self.__nzo_list[nzo_id_pos1]
if priority == TOP_PRIORITY:
# A top priority item (usually a completed download fetching pars)
# is added to the top of the queue
self.__nzo_list.insert(0, nzo)
pos = 0
elif priority == LOW_PRIORITY:
pos = len(self.__nzo_list)
self.__nzo_list.append(nzo)
else:
# for high priority we need to add the item at the bottom
# of any other high priority items above the normal priority
# for normal priority we need to add the item at the bottom
# of the normal priority items above the low priority
if self.__nzo_list:
p = 0
added = False
for position in self.__nzo_list:
if position.priority < priority:
self.__nzo_list.insert(p, nzo)
pos = p
added = True
break
p += 1
if not added:
# if there are no other items classed as a lower priority
# then it will be added to the bottom of the queue
pos = len(self.__nzo_list)
self.__nzo_list.append(nzo)
else:
# if the queue is empty then simple append the item to the bottom
self.__nzo_list.append(nzo)
pos = 0
logging.info("Set priority=%s for job %s => position=%s ",
priority, self.__nzo_table[nzo_id].final_name, pos)
return pos
except:
return -1
def run(self):
# First check IPv6 connectivity
sabnzbd.EXTERNAL_IPV6 = sabnzbd.test_ipv6()
logging.debug("External IPv6 test result: %s", sabnzbd.EXTERNAL_IPV6)
# Then we check SSL certificate checking
sabnzbd.CERTIFICATE_VALIDATION = sabnzbd.test_cert_checking()
logging.debug("SSL verification test: %s", sabnzbd.CERTIFICATE_VALIDATION)
# Kick BPS-Meter to check quota
sabnzbd.BPSMeter.update()
# Check server expiration dates
check_server_expiration()
while 1:
now = time.time()
# Set Article to None so references from this
# thread do not keep the parent objects alive (see #1628)
article = None
for server in self.servers:
# Skip this server if there's no point searching for new stuff to do
if not server.busy_threads and server.next_article_search > now:
continue
for nw in server.busy_threads[:]:
if (nw.nntp and nw.nntp.error_msg) or (nw.timeout and now > nw.timeout):
if nw.nntp and nw.nntp.error_msg:
# Already showed error
self.__reset_nw(nw)
else:
self.__reset_nw(nw, "timed out", warn=True)
server.bad_cons += 1
self.maybe_block_server(server)
if server.restart:
if not server.busy_threads:
newid = server.newid
server.stop()
self.servers.remove(server)
if newid:
self.init_server(None, newid)
self.server_restarts -= 1
# Have to leave this loop, because we removed element
break
else:
# Restart pending, don't add new articles
continue
if (
not server.idle_threads
or server.restart
or self.is_paused()
or self.shutdown
or self.paused_for_postproc
or not server.active
):
continue
for nw in server.idle_threads[:]:
if nw.timeout:
if now < nw.timeout:
continue
else:
nw.timeout = None
if not server.info:
# Only request info if there's stuff in the queue
if not sabnzbd.NzbQueue.is_empty():
self.maybe_block_server(server)
server.request_info()
break
article = sabnzbd.NzbQueue.get_article(server, self.servers)
if not article:
# Skip this server for 0.5 second
server.next_article_search = now + 0.5
break
if server.retention and article.nzf.nzo.avg_stamp < now - server.retention:
# Let's get rid of all the articles for this server at once
logging.info("Job %s too old for %s, moving on", article.nzf.nzo.final_name, server.host)
while article:
self.decode(article, None)
article = article.nzf.nzo.get_article(server, self.servers)
break
server.idle_threads.remove(nw)
server.busy_threads.append(nw)
nw.article = article
if nw.connected:
self.__request_article(nw)
else:
try:
logging.info("%s@%s: Initiating connection", nw.thrdnum, server.host)
nw.init_connect()
except:
logging.error(
T("Failed to initialize %s@%s with reason: %s"),
nw.thrdnum,
server.host,
sys.exc_info()[1],
)
self.__reset_nw(nw, "failed to initialize", warn=True)
if self.force_disconnect or self.shutdown:
for server in self.servers:
for nw in server.idle_threads + server.busy_threads:
# Send goodbye if we have open socket
if nw.nntp:
self.__reset_nw(
nw,
"forcing disconnect",
wait=False,
count_article_try=False,
send_quit=True,
)
# Make sure server address resolution is refreshed
server.info = None
self.force_disconnect = False
# Exit-point
if self.shutdown:
logging.info("Shutting down")
break
# Use select to find sockets ready for reading/writing
readkeys = self.read_fds.keys()
if readkeys:
read, _, _ = select.select(readkeys, (), (), 1.0)
# Add a sleep if there are too few results compared to the number of active connections
if self.can_be_slowed and len(read) < 1 + len(readkeys) / 10:
time.sleep(self.sleep_time)
# Need to initialize the check during first 20 seconds
if self.can_be_slowed is None or self.can_be_slowed_timer:
# Wait for stable speed to start testing
if not self.can_be_slowed_timer and sabnzbd.BPSMeter.get_stable_speed(timespan=10):
self.can_be_slowed_timer = time.time()
# Check 10 seconds after enabling slowdown
if self.can_be_slowed_timer and time.time() > self.can_be_slowed_timer + 10:
# Now let's check if it was stable in the last 10 seconds
self.can_be_slowed = sabnzbd.BPSMeter.get_stable_speed(timespan=10)
self.can_be_slowed_timer = 0
logging.debug("Downloader-slowdown: %r", self.can_be_slowed)
else:
read = []
sabnzbd.BPSMeter.reset()
time.sleep(1.0)
with DOWNLOADER_CV:
while (
(sabnzbd.NzbQueue.is_empty() or self.is_paused() or self.paused_for_postproc)
and not self.shutdown
and not self.force_disconnect
and not self.server_restarts
):
DOWNLOADER_CV.wait()
if not read:
sabnzbd.BPSMeter.update()
continue
for selected in read:
nw = self.read_fds[selected]
article = nw.article
server = nw.server
try:
bytes_received, done, skip = nw.recv_chunk()
except:
bytes_received, done, skip = (0, False, False)
if skip:
sabnzbd.BPSMeter.update()
continue
if bytes_received < 1:
self.__reset_nw(nw, "server closed connection", wait=False)
continue
else:
try:
article.nzf.nzo.update_download_stats(sabnzbd.BPSMeter.bps, server.id, bytes_received)
except AttributeError:
# In case nzf has disappeared because the file was deleted before the update could happen
pass
if self.bandwidth_limit:
limit = self.bandwidth_limit
if bytes_received + sabnzbd.BPSMeter.bps > limit:
while sabnzbd.BPSMeter.bps > limit:
time.sleep(0.01)
sabnzbd.BPSMeter.update()
sabnzbd.BPSMeter.update(server.id, bytes_received)
if not done and nw.status_code != 222:
if not nw.connected or nw.status_code == 480:
done = False
try:
nw.finish_connect(nw.status_code)
if sabnzbd.LOG_ALL:
logging.debug(
"%s@%s last message -> %s", nw.thrdnum, nw.server.host, nntp_to_msg(nw.data)
)
nw.clear_data()
except NNTPPermanentError as error:
# Handle login problems
block = False
penalty = 0
msg = error.response
ecode = int_conv(msg[:3])
display_msg = " [%s]" % msg
logging.debug("Server login problem: %s, %s", ecode, msg)
if ecode in (502, 400, 481, 482) and clues_too_many(msg):
# Too many connections: remove this thread and reduce thread-setting for server
# Plan to go back to the full number after a penalty timeout
if server.active:
errormsg = T("Too many connections to server %s") % display_msg
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.warning(T("Too many connections to server %s"), server.host)
# Don't count this for the tries (max_art_tries) on this server
self.__reset_nw(nw, count_article_try=False, send_quit=True)
self.plan_server(server, _PENALTY_TOOMANY)
server.threads -= 1
elif ecode in (502, 481, 482) and clues_too_many_ip(msg):
# Account sharing?
if server.active:
errormsg = T("Probable account sharing") + display_msg
if server.errormsg != errormsg:
server.errormsg = errormsg
name = " (%s)" % server.host
logging.warning(T("Probable account sharing") + name)
penalty = _PENALTY_SHARE
block = True
elif ecode in (452, 481, 482, 381) or (ecode == 502 and clues_login(msg)):
# Cannot login, block this server
if server.active:
errormsg = T("Failed login for server %s") % display_msg
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.error(T("Failed login for server %s"), server.host)
penalty = _PENALTY_PERM
block = True
elif ecode in (502, 482):
# Cannot connect (other reasons), block this server
if server.active:
errormsg = T("Cannot connect to server %s [%s]") % ("", display_msg)
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.warning(T("Cannot connect to server %s [%s]"), server.host, msg)
if clues_pay(msg):
penalty = _PENALTY_PERM
else:
penalty = _PENALTY_502
block = True
elif ecode == 400:
# Temp connection problem?
if server.active:
logging.debug("Unspecified error 400 from server %s", server.host)
penalty = _PENALTY_VERYSHORT
block = True
else:
# Unknown error, just keep trying
if server.active:
errormsg = T("Cannot connect to server %s [%s]") % ("", display_msg)
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.warning(T("Cannot connect to server %s [%s]"), server.host, msg)
penalty = _PENALTY_UNKNOWN
block = True
if block or (penalty and server.optional):
if server.active:
server.active = False
if penalty and (block or server.optional):
self.plan_server(server, penalty)
# Note that this will count towards the tries (max_art_tries) on this server!
self.__reset_nw(nw, send_quit=True)
continue
except:
logging.error(
T("Connecting %s@%s failed, message=%s"),
nw.thrdnum,
nw.server.host,
nntp_to_msg(nw.data),
)
# No reset-warning needed, above logging is sufficient
self.__reset_nw(nw)
if nw.connected:
logging.info("Connecting %s@%s finished", nw.thrdnum, nw.server.host)
self.__request_article(nw)
elif nw.status_code == 223:
done = True
logging.debug("Article <%s> is present", article.article)
elif nw.status_code == 211:
done = False
logging.debug("group command ok -> %s", nntp_to_msg(nw.data))
nw.group = nw.article.nzf.nzo.group
nw.clear_data()
self.__request_article(nw)
elif nw.status_code in (411, 423, 430):
done = True
logging.debug(
"Thread %s@%s: Article %s missing (error=%s)",
nw.thrdnum,
nw.server.host,
article.article,
nw.status_code,
)
nw.clear_data()
elif nw.status_code == 500:
if article.nzf.nzo.precheck:
# Assume "STAT" command is not supported
server.have_stat = False
logging.debug("Server %s does not support STAT", server.host)
else:
# Assume "BODY" command is not supported
server.have_body = False
logging.debug("Server %s does not support BODY", server.host)
nw.clear_data()
self.__request_article(nw)
if done:
# Successful data, clear "bad" counter
server.bad_cons = 0
server.errormsg = server.warning = ""
if sabnzbd.LOG_ALL:
logging.debug("Thread %s@%s: %s done", nw.thrdnum, server.host, article.article)
self.decode(article, nw.data)
# Reset connection for new activity
nw.soft_reset()
server.busy_threads.remove(nw)
server.idle_threads.append(nw)
self.remove_socket(nw)
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
(url, future_nzo) = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo:
# Re-queue when too early and still active
if future_nzo.url_wait and future_nzo.url_wait > time.time():
self.add(url, future_nzo)
time.sleep(1.0)
continue
# Paused
if future_nzo.status == Status.PAUSED:
self.add(url, future_nzo)
time.sleep(1.0)
continue
url = url.replace(' ', '')
try:
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing', url)
continue
filename = None
category = None
gzipped = False
nzo_info = {}
wait = 0
retry = True
fetch_request = None
logging.info('Grabbing URL %s', url)
try:
fetch_request = _build_request(url)
except Exception, e:
# Cannot list exceptions here, because of unpredictability over platforms
error0 = str(sys.exc_info()[0]).lower()
error1 = str(sys.exc_info()[1]).lower()
logging.debug('Error "%s" trying to get the url %s', error1, url)
if 'certificate_verify_failed' in error1 or 'certificateerror' in error0:
msg = T('Server %s uses an untrusted HTTPS certificate') % ''
msg += ' - https://sabnzbd.org/certificate-errors'
retry = False
elif 'nodename nor servname provided' in error1:
msg = T('Server name does not resolve')
retry = False
elif '401' in error1 or 'unauthorized' in error1:
msg = T('Unauthorized access')
retry = False
elif '404' in error1:
msg = T('File not on server')
retry = False
elif hasattr(e, 'headers') and 'retry-after' in e.headers:
# Catch if the server send retry (e.headers is case-INsensitive)
wait = misc.int_conv(e.headers['retry-after'])
# Check if dereference is used
new_url = dereferring(url, fetch_request)
if new_url:
self.add(new_url, future_nzo)
continue
if fetch_request:
for hdr in fetch_request.headers:
try:
item = hdr.lower()
value = fetch_request.headers[hdr]
except:
continue
if item in ('content-encoding',) and value == 'gzip':
gzipped = True
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo',):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name',):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item == 'x-dnzb-propername':
nzo_info['propername'] = value
elif item == 'x-dnzb-episodename':
nzo_info['episodename'] = value
elif item == 'x-dnzb-year':
nzo_info['year'] = value
elif item == 'x-dnzb-failure':
nzo_info['failure'] = value
elif item == 'x-dnzb-details':
nzo_info['details'] = value
elif item == 'x-dnzb-password':
nzo_info['password'] = value
elif item == 'retry-after':
wait = misc.int_conv(value)
# Rating fields
if item in _RARTING_FIELDS:
nzo_info[item] = value
# Get filename from Content-Disposition header
if not filename and "filename=" in value:
filename = value[value.index("filename=") + 9:].strip(';').strip('"')
if wait:
# For sites that have a rate-limiting attribute
msg = ''
retry = True
fetch_request = None
elif retry:
fetch_request, msg, retry, wait, data = _analyse(fetch_request, future_nzo)
if not fetch_request:
if retry:
logging.info('Retry URL %s', url)
self.add(url, future_nzo, wait)
else:
self.fail_to_history(future_nzo, url, msg)
continue
if not filename:
filename = os.path.basename(urllib2.unquote(url))
# URL was redirected, maybe the redirect has better filename?
# Check if the original URL has extension
if url != fetch_request.url and misc.get_ext(filename) not in VALID_NZB_FILES:
filename = os.path.basename(urllib2.unquote(fetch_request.url))
elif '&nzbname=' in filename:
# Sometimes the filename contains the full URL, duh!
filename = filename[filename.find('&nzbname=') + 9:]
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# process data
if gzipped:
filename += '.gz'
if not data:
try:
data = fetch_request.read()
except (IncompleteRead, IOError):
self.fail_to_history(future_nzo, url, T('Server could not complete request'))
fetch_request.close()
continue
fetch_request.close()
if '<nzb' in data and misc.get_ext(filename) != '.nzb':
filename += '.nzb'
# Sanitize filename first (also removing forbidden Windows-names)
filename = misc.sanitize_filename(filename)
# Write data to temp file
path = os.path.join(cfg.admin_dir.get_path(), FUTURE_Q_FOLDER)
path = os.path.join(path, filename)
f = open(path, 'wb')
f.write(data)
f.close()
del data
# Check if nzb file
if misc.get_ext(filename) in VALID_NZB_FILES:
res = dirscanner.ProcessSingleFile(filename, path, pp=pp, script=script, cat=cat, priority=priority,
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url, keep=False,
nzo_id=future_nzo.nzo_id)[0]
if res:
if res == -2:
logging.info('Incomplete NZB, retry after 5 min %s', url)
when = 300
elif res == -1:
# Error, but no reason to retry. Warning is already given
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
continue
else:
logging.info('Unknown error fetching NZB, retry after 2 min %s', url)
when = 120
self.add(url, future_nzo, when)
else:
# Check if a supported archive
status, zf, exp_ext = dirscanner.is_archive(path)
if status == 0:
if misc.get_ext(filename) not in ('.rar', '.zip', '.7z'):
filename = filename + exp_ext
os.rename(path, path + exp_ext)
path = path + exp_ext
dirscanner.ProcessArchiveFile(filename, path, pp, script, cat, priority=priority,
nzbname=nzbname, url=future_nzo.url, keep=False,
nzo_id=future_nzo.nzo_id)
else:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fetch_request)
except:
pass
logging.info('Unknown filetype when fetching NZB, retry after 30s %s', url)
self.add(url, future_nzo, 30)
except:
def run(self):
self.shutdown = False
while not self.shutdown:
# Set NzbObject object to None so reference from this thread
# does not keep the object alive in the future (see #1628)
future_nzo = None
url, future_nzo = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo:
# Re-queue when too early and still active
if future_nzo.url_wait and future_nzo.url_wait > time.time():
self.add(url, future_nzo)
time.sleep(1.0)
continue
# Paused
if future_nzo.status == Status.PAUSED:
self.add(url, future_nzo)
time.sleep(1.0)
continue
url = url.replace(" ", "")
try:
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug("Dropping URL %s, job entry missing",
url)
continue
filename = None
category = None
nzo_info = {}
wait = 0
retry = True
fetch_request = None
logging.info("Grabbing URL %s", url)
try:
fetch_request = _build_request(url)
except Exception as e:
# Cannot list exceptions here, because of unpredictability over platforms
error0 = str(sys.exc_info()[0]).lower()
error1 = str(sys.exc_info()[1]).lower()
logging.debug('Error "%s" trying to get the url %s',
error1, url)
if "certificate_verify_failed" in error1 or "certificateerror" in error0:
msg = T("Server %s uses an untrusted HTTPS certificate"
) % ""
msg += " - https://sabnzbd.org/certificate-errors"
retry = False
elif "nodename nor servname provided" in error1:
msg = T("Server name does not resolve")
retry = False
elif "401" in error1 or "unauthorized" in error1:
msg = T("Unauthorized access")
retry = False
elif "404" in error1:
msg = T("File not on server")
retry = False
elif hasattr(e, "headers") and "retry-after" in e.headers:
# Catch if the server send retry (e.headers is case-INsensitive)
wait = misc.int_conv(e.headers["retry-after"])
if fetch_request:
for hdr in fetch_request.headers:
try:
item = hdr.lower()
value = fetch_request.headers[hdr]
except:
continue
if item in ("category_id", "x-dnzb-category"):
category = value
elif item in ("x-dnzb-moreinfo", ):
nzo_info["more_info"] = value
elif item in ("x-dnzb-name", ):
filename = value
if not filename.endswith(".nzb"):
filename += ".nzb"
elif item == "x-dnzb-propername":
nzo_info["propername"] = value
elif item == "x-dnzb-episodename":
nzo_info["episodename"] = value
elif item == "x-dnzb-year":
nzo_info["year"] = value
elif item == "x-dnzb-failure":
nzo_info["failure"] = value
elif item == "x-dnzb-details":
nzo_info["details"] = value
elif item == "x-dnzb-password":
nzo_info["password"] = value
elif item == "retry-after":
wait = misc.int_conv(value)
# Rating fields
if item in _RARTING_FIELDS:
nzo_info[item] = value
# Get filename from Content-Disposition header
if not filename and "filename=" in value:
filename = value[value.index("filename=") +
9:].strip(";").strip('"')
if wait:
# For sites that have a rate-limiting attribute
msg = ""
retry = True
fetch_request = None
elif retry:
fetch_request, msg, retry, wait, data = _analyse(
fetch_request, future_nzo)
if not fetch_request:
if retry:
logging.info("Retry URL %s", url)
self.add(url, future_nzo, wait)
else:
self.fail_to_history(future_nzo, url, msg)
continue
if not filename:
filename = os.path.basename(urllib.parse.unquote(url))
# URL was redirected, maybe the redirect has better filename?
# Check if the original URL has extension
if (url != fetch_request.geturl()
and sabnzbd.filesystem.get_ext(filename)
not in VALID_NZB_FILES + VALID_ARCHIVES):
filename = os.path.basename(
urllib.parse.unquote(fetch_request.geturl()))
elif "&nzbname=" in filename:
# Sometimes the filename contains the full URL, duh!
filename = filename[filename.find("&nzbname=") + 9:]
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == "*") and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# process data
if not data:
try:
data = fetch_request.read()
except (IncompleteRead, IOError):
self.fail_to_history(
future_nzo, url,
T("Server could not complete request"))
fetch_request.close()
continue
fetch_request.close()
if b"<nzb" in data and sabnzbd.filesystem.get_ext(
filename) != ".nzb":
filename += ".nzb"
# Sanitize filename first (also removing forbidden Windows-names)
filename = sabnzbd.filesystem.sanitize_filename(filename)
# If no filename, make one
if not filename:
filename = sabnzbd.get_new_id(
"url",
os.path.join(cfg.admin_dir.get_path(),
FUTURE_Q_FOLDER))
# Write data to temp file
path = os.path.join(cfg.admin_dir.get_path(), FUTURE_Q_FOLDER,
filename)
with open(path, "wb") as temp_nzb:
temp_nzb.write(data)
# Check if nzb file
if sabnzbd.filesystem.get_ext(
filename) in VALID_ARCHIVES + VALID_NZB_FILES:
res, _ = sabnzbd.add_nzbfile(
path,
pp=pp,
script=script,
cat=cat,
priority=priority,
nzbname=nzbname,
nzo_info=nzo_info,
url=future_nzo.url,
keep=False,
password=future_nzo.password,
nzo_id=future_nzo.nzo_id,
)
# -2==Error/retry, -1==Error, 0==OK, 1==Empty
if res == -2:
logging.info("Incomplete NZB, retry after 5 min %s",
url)
self.add(url, future_nzo, when=300)
elif res == -1:
# Error already thrown
self.fail_to_history(future_nzo, url)
elif res == 1:
# No NZB-files inside archive
self.fail_to_history(future_nzo, url,
T("Empty NZB file %s") % filename)
else:
logging.info(
"Unknown filetype when fetching NZB, retry after 30s %s",
url)
self.add(url, future_nzo, 30)
# Always clean up what we wrote to disk
try:
sabnzbd.filesystem.remove_file(path)
except:
pass
except:
logging.error(T("URLGRABBER CRASHED"), exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
# Don't pound the website!
time.sleep(5.0)
(url, future_nzo, retry_count) = self.queue.get()
if not url:
continue
url = url.replace(' ', '')
try:
del_bookmark = not future_nzo
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing', url)
continue
# Add nzbmatrix credentials if needed
url, matrix_id = _matrix_url(url)
# When still waiting for nzbmatrix wait period, requeue
if matrix_id and self.matrix_wait > time.time():
self.queue.put((url, future_nzo, retry_count))
continue
# _grab_url cannot reside in a function, because the tempfile
# would not survive the end of the function
if del_bookmark:
logging.info('Removing nzbmatrix bookmark %s', matrix_id)
else:
logging.info('Grabbing URL %s', url)
opener = urllib.FancyURLopener({})
opener.prompt_user_passwd = None
opener.addheaders = []
opener.addheader('User-Agent', 'SABnzbd+/%s' % sabnzbd.version.__version__)
opener.addheader('Accept-encoding','gzip')
filename = None
category = None
length = 0
nzo_info = {}
try:
fn, header = opener.retrieve(url)
except:
fn = None
if fn:
for tup in header.items():
try:
item = tup[0].lower()
value = tup[1].strip()
except:
continue
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo',):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name',):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item in ('content-length',):
length = misc.int_conv(value)
if not filename:
for item in tup:
if "filename=" in item:
filename = item[item.index("filename=") + 9:].strip(';').strip('"')
if matrix_id:
fn, msg, retry, wait = _analyse_matrix(fn, matrix_id)
if retry and wait > 0:
self.matrix_wait = time.time() + wait
logging.debug('Retry URL %s after waiting', url)
self.queue.put((url, future_nzo, retry_count))
continue
category = _MATRIX_MAP.get(category, category)
else:
msg = ''
retry = True
# Check if the filepath is specified, if not, check if a retry is allowed.
if not fn:
retry_count -= 1
if retry_count > 0 and retry:
logging.info('Retry URL %s', url)
self.queue.put((url, future_nzo, retry_count))
elif not del_bookmark:
misc.bad_fetch(future_nzo, url, msg, retry=True)
continue
if del_bookmark:
continue
if not filename:
filename = os.path.basename(url) + '.nzb'
# Sanitize and trim name, preserving the extension
filename, ext = os.path.splitext(filename)
filename = misc.sanitize_foldername(filename)
filename += '.' + misc.sanitize_foldername(ext)
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# Check if nzb file
if os.path.splitext(filename)[1].lower() in ('.nzb', '.gz'):
res = dirscanner.ProcessSingleFile(filename, fn, pp=pp, script=script, cat=cat, priority=priority, \
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url)
if res == 0:
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
elif res == -2:
retry_count -= 1
if retry_count > 0:
logging.info('Incomplete NZB, retry %s', url)
self.queue.put((url, future_nzo, retry_count))
else:
misc.bad_fetch(future_nzo, url, retry=True, content=True)
else:
misc.bad_fetch(future_nzo, url, retry=True, content=True)
# Check if a supported archive
else:
if dirscanner.ProcessArchiveFile(filename, fn, pp, script, cat, priority=priority, url=future_nzo.url) == 0:
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
else:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
misc.bad_fetch(future_nzo, url, retry=True, content=True)
except:
logging.error('URLGRABBER CRASHED', exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
def run_feed(self, feed=None, download=False, ignoreFirst=False, force=False, readout=True):
""" Run the query for one URI and apply filters """
self.shutdown = False
if not feed:
return 'No such feed'
newlinks = []
new_downloads = []
# Preparations, get options
try:
feeds = config.get_rss()[feed]
except KeyError:
logging.error(T('Incorrect RSS feed description "%s"'), feed)
logging.info("Traceback: ", exc_info=True)
return T('Incorrect RSS feed description "%s"') % feed
uris = feeds.uri()
defCat = feeds.cat()
import sabnzbd.api
if not notdefault(defCat) or defCat not in sabnzbd.api.list_cats(default=False):
defCat = None
defPP = feeds.pp()
if not notdefault(defPP):
defPP = None
defScript = feeds.script()
if not notdefault(defScript):
defScript = None
defPrio = feeds.priority()
if not notdefault(defPrio):
defPrio = None
# Preparations, convert filters to regex's
regexes = []
reTypes = []
reCats = []
rePPs = []
rePrios = []
reScripts = []
reEnabled = []
for filter in feeds.filters():
reCat = filter[0]
if defCat in ('', '*'):
reCat = None
reCats.append(reCat)
rePPs.append(filter[1])
reScripts.append(filter[2])
reTypes.append(filter[3])
if filter[3] in ('<', '>', 'F', 'S'):
regexes.append(filter[4])
else:
regexes.append(convert_filter(filter[4]))
rePrios.append(filter[5])
reEnabled.append(filter[6] != '0')
regcount = len(regexes)
# Set first if this is the very first scan of this URI
first = (feed not in self.jobs) and ignoreFirst
# Add sabnzbd's custom User Agent
feedparser.USER_AGENT = 'SABnzbd+/%s' % sabnzbd.version.__version__
# Read the RSS feed
msg = None
entries = None
if readout:
all_entries = []
for uri in uris:
uri = uri.replace(' ', '%20')
logging.debug("Running feedparser on %s", uri)
feed_parsed = feedparser.parse(uri.replace('feed://', 'http://'))
logging.debug("Done parsing %s", uri)
if not feed_parsed:
msg = T('Failed to retrieve RSS from %s: %s') % (uri, '?')
logging.info(msg)
status = feed_parsed.get('status', 999)
if status in (401, 402, 403):
msg = T('Do not have valid authentication for feed %s') % feed
logging.info(msg)
if status >= 500 and status <= 599:
msg = T('Server side error (server code %s); could not get %s on %s') % (status, feed, uri)
logging.info(msg)
entries = feed_parsed.get('entries')
if 'bozo_exception' in feed_parsed and not entries:
msg = str(feed_parsed['bozo_exception'])
if 'CERTIFICATE_VERIFY_FAILED' in msg:
msg = T('Server %s uses an untrusted HTTPS certificate') % get_urlbase(uri)
msg += ' - https://sabnzbd.org/certificate-errors'
logging.error(msg)
else:
msg = T('Failed to retrieve RSS from %s: %s') % (uri, xml_name(msg))
logging.info(msg)
if not entries:
msg = T('RSS Feed %s was empty') % uri
logging.info(msg)
all_entries.extend(entries)
entries = all_entries
if feed not in self.jobs:
self.jobs[feed] = {}
jobs = self.jobs[feed]
if readout:
if not entries:
return unicoder(msg)
else:
entries = jobs.keys()
# Sort in the order the jobs came from the feed
entries.sort(lambda x, y: jobs[x].get('order', 0) - jobs[y].get('order', 0))
order = 0
# Filter out valid new links
for entry in entries:
if self.shutdown:
return
if readout:
try:
link, category, size, age, season, episode = _get_link(uri, entry)
except (AttributeError, IndexError):
link = None
category = u''
size = 0L
age = None
logging.info(T('Incompatible feed') + ' ' + uri)
logging.info("Traceback: ", exc_info=True)
return T('Incompatible feed')
title = entry.title
# If there's multiple feeds, remove the duplicates based on title and size
if len(uris) > 1:
skip_job = False
for job_link, job in jobs.items():
# Allow 5% size deviation because indexers might have small differences for same release
if job.get('title') == title and link != job_link and (job.get('size')*0.95) < size < (job.get('size')*1.05):
logging.info("Ignoring job %s from other feed", title)
skip_job = True
break
if skip_job:
continue
else:
link = entry
category = jobs[link].get('orgcat', '')
if category in ('', '*'):
category = None
title = jobs[link].get('title', '')
size = jobs[link].get('size', 0L)
age = jobs[link].get('age')
season = jobs[link].get('season', 0)
episode = jobs[link].get('episode', 0)
if link:
# Make sure spaces are quoted in the URL
link = link.strip().replace(' ', '%20')
newlinks.append(link)
if link in jobs:
jobstat = jobs[link].get('status', ' ')[0]
else:
jobstat = 'N'
if jobstat in 'NGB' or (jobstat == 'X' and readout):
# Match this title against all filters
logging.debug('Trying title %s', title)
result = False
myCat = defCat
myPP = defPP
myScript = defScript
myPrio = defPrio
n = 0
if ('F' in reTypes or 'S' in reTypes) and (not season or not episode):
season, episode = sabnzbd.newsunpack.analyse_show(title)[1:3]
season = int_conv(season)
episode = int_conv(episode)
# Match against all filters until an positive or negative match
logging.debug('Size %s for %s', size, title)
for n in xrange(regcount):
if reEnabled[n]:
if category and reTypes[n] == 'C':
found = re.search(regexes[n], category)
if not found:
logging.debug("Filter rejected on rule %d", n)
result = False
break
elif reTypes[n] == '<' and size and from_units(regexes[n]) < size:
# "Size at most" : too large
logging.debug('Filter rejected on rule %d', n)
result = False
break
elif reTypes[n] == '>' and size and from_units(regexes[n]) > size:
# "Size at least" : too small
logging.debug('Filter rejected on rule %d', n)
result = False
break
elif reTypes[n] == 'F' and not ep_match(season, episode, regexes[n]):
# "Starting from SxxEyy", too early episode
logging.debug('Filter requirement match on rule %d', n)
result = False
break
elif reTypes[n] == 'S' and season and episode and ep_match(season, episode, regexes[n], title):
logging.debug('Filter matched on rule %d', n)
result = True
break
else:
if regexes[n]:
found = re.search(regexes[n], title)
else:
found = False
if reTypes[n] == 'M' and not found:
logging.debug("Filter rejected on rule %d", n)
result = False
break
if found and reTypes[n] == 'A':
logging.debug("Filter matched on rule %d", n)
result = True
break
if found and reTypes[n] == 'R':
logging.debug("Filter rejected on rule %d", n)
result = False
break
if len(reCats):
if not result and defCat:
# Apply Feed-category on non-matched items
myCat = defCat
elif result and notdefault(reCats[n]):
# Use the matched info
myCat = reCats[n]
elif category and not defCat:
# No result and no Feed-category
myCat = cat_convert(category)
if myCat:
myCat, catPP, catScript, catPrio = cat_to_opts(myCat)
else:
myCat = catPP = catScript = catPrio = None
if notdefault(rePPs[n]):
myPP = rePPs[n]
elif not (reCats[n] or category):
myPP = catPP
if notdefault(reScripts[n]):
myScript = reScripts[n]
elif not (notdefault(reCats[n]) or category):
myScript = catScript
if rePrios[n] not in (str(DEFAULT_PRIORITY), ''):
myPrio = rePrios[n]
elif not ((rePrios[n] != str(DEFAULT_PRIORITY)) or category):
myPrio = catPrio
if cfg.no_dupes() and self.check_duplicate(title):
if cfg.no_dupes() == 1:
# Dupe-detection: Discard
logging.info("Ignoring duplicate job %s", title)
continue
elif cfg.no_dupes() == 3:
# Dupe-detection: Fail
# We accept it so the Queue can send it to the History
logging.info("Found duplicate job %s", title)
else:
# Dupe-detection: Pause
myPrio = DUP_PRIORITY
act = download and not first
if link in jobs:
act = act and not jobs[link].get('status', '').endswith('*')
act = act or force
star = first or jobs[link].get('status', '').endswith('*')
else:
star = first
if result:
_HandleLink(jobs, link, title, size, age, season, episode, 'G', category, myCat, myPP, myScript,
act, star, order, priority=myPrio, rule=str(n))
if act:
new_downloads.append(title)
else:
_HandleLink(jobs, link, title, size, age, season, episode, 'B', category, myCat, myPP, myScript,
False, star, order, priority=myPrio, rule=str(n))
order += 1
# Send email if wanted and not "forced"
if new_downloads and cfg.email_rss() and not force:
emailer.rss_mail(feed, new_downloads)
remove_obsolete(jobs, newlinks)
return msg
def __set_priority(self, nzo_id, priority):
""" Sets the priority on the nzo and places it in the queue at the appropriate position """
try:
priority = int_conv(priority)
nzo = self.__nzo_table[nzo_id]
nzo_id_pos1 = -1
pos = -1
# If priority == STOP_PRIORITY, then send to queue
if priority == STOP_PRIORITY:
self.end_job(nzo)
return
# Get the current position in the queue
for i in xrange(len(self.__nzo_list)):
if nzo_id == self.__nzo_list[i].nzo_id:
nzo_id_pos1 = i
break
# Don't change priority and order if priority is the same as asked
if priority == self.__nzo_list[nzo_id_pos1].priority:
return nzo_id_pos1
nzo.set_priority(priority)
if sabnzbd.scheduler.analyse(False, priority) and \
nzo.status in (Status.CHECKING, Status.DOWNLOADING, Status.QUEUED):
nzo.status = Status.PAUSED
elif nzo.status == Status.PAUSED:
nzo.status = Status.QUEUED
nzo.save_to_disk()
if nzo_id_pos1 != -1:
del self.__nzo_list[nzo_id_pos1]
if priority == TOP_PRIORITY:
# A top priority item (usually a completed download fetching pars)
# is added to the top of the queue
self.__nzo_list.insert(0, nzo)
pos = 0
elif priority == LOW_PRIORITY:
pos = len(self.__nzo_list)
self.__nzo_list.append(nzo)
else:
# for high priority we need to add the item at the bottom
# of any other high priority items above the normal priority
# for normal priority we need to add the item at the bottom
# of the normal priority items above the low priority
if self.__nzo_list:
p = 0
added = False
for position in self.__nzo_list:
if position.priority < priority:
self.__nzo_list.insert(p, nzo)
pos = p
added = True
break
p += 1
if not added:
# if there are no other items classed as a lower priority
# then it will be added to the bottom of the queue
pos = len(self.__nzo_list)
self.__nzo_list.append(nzo)
else:
# if the queue is empty then simple append the item to the bottom
self.__nzo_list.append(nzo)
pos = 0
logging.info('Set priority=%s for job %s => position=%s ', priority, self.__nzo_table[nzo_id].final_name, pos)
return pos
except:
return -1
def run(self):
# First check IPv6 connectivity
sabnzbd.EXTERNAL_IPV6 = sabnzbd.test_ipv6()
logging.debug("External IPv6 test result: %s", sabnzbd.EXTERNAL_IPV6)
# Then we check SSL certificate checking
sabnzbd.CERTIFICATE_VALIDATION = sabnzbd.test_cert_checking()
logging.debug("SSL verification test: %s",
sabnzbd.CERTIFICATE_VALIDATION)
# Kick BPS-Meter to check quota
BPSMeter.do.update()
while 1:
for server in self.servers:
for nw in server.busy_threads[:]:
if (nw.nntp and nw.nntp.error_msg) or (
nw.timeout and time.time() > nw.timeout):
if nw.nntp and nw.nntp.error_msg:
self.__reset_nw(nw, "", warn=False)
else:
self.__reset_nw(nw, "timed out")
server.bad_cons += 1
self.maybe_block_server(server)
if server.restart:
if not server.busy_threads:
newid = server.newid
server.stop(self.read_fds, self.write_fds)
self.servers.remove(server)
if newid:
self.init_server(None, newid)
self.__restart -= 1
sabnzbd.nzbqueue.NzbQueue.do.reset_all_try_lists()
# Have to leave this loop, because we removed element
break
else:
# Restart pending, don't add new articles
continue
if not server.idle_threads or server.restart or self.is_paused(
) or self.shutdown or self.postproc:
continue
if not server.active:
continue
for nw in server.idle_threads[:]:
if nw.timeout:
if time.time() < nw.timeout:
continue
else:
nw.timeout = None
if not server.info:
# Only request info if there's stuff in the queue
if not sabnzbd.nzbqueue.NzbQueue.do.is_empty():
self.maybe_block_server(server)
request_server_info(server)
break
article = sabnzbd.nzbqueue.NzbQueue.do.get_article(
server, self.servers)
if not article:
break
if server.retention and article.nzf.nzo.avg_stamp < time.time(
) - server.retention:
# Let's get rid of all the articles for this server at once
logging.info("Job %s too old for %s, moving on",
article.nzf.nzo.final_name, server.host)
while article:
self.decode(article, None)
article = article.nzf.nzo.get_article(
server, self.servers)
break
server.idle_threads.remove(nw)
server.busy_threads.append(nw)
nw.article = article
if nw.connected:
self.__request_article(nw)
else:
try:
logging.info("%s@%s: Initiating connection",
nw.thrdnum, server.host)
nw.init_connect(self.write_fds)
except:
logging.error(
T("Failed to initialize %s@%s with reason: %s"
),
nw.thrdnum,
server.host,
sys.exc_info()[1],
)
self.__reset_nw(nw, "failed to initialize")
# Exit-point
if self.shutdown:
empty = True
for server in self.servers:
if server.busy_threads:
empty = False
break
if empty:
for server in self.servers:
server.stop(self.read_fds, self.write_fds)
logging.info("Shutting down")
break
if self.force_disconnect:
for server in self.servers:
for nw in server.idle_threads + server.busy_threads:
send_quit = nw.connected and server.active
self.__reset_nw(nw,
"forcing disconnect",
warn=False,
wait=False,
send_quit=send_quit)
# Make sure server address resolution is refreshed
server.info = None
self.force_disconnect = False
# Use select to find sockets ready for reading/writing
readkeys = self.read_fds.keys()
writekeys = self.write_fds.keys()
if readkeys or writekeys:
read, write, error = select.select(readkeys, writekeys, (),
1.0)
# Why check so often when so few things happened?
if self.can_be_slowed and len(readkeys) >= 8 and len(
read) <= 2:
time.sleep(0.01)
# Need to initialize the check during first 20 seconds
if self.can_be_slowed is None or self.can_be_slowed_timer:
# Wait for stable speed to start testing
if not self.can_be_slowed_timer and BPSMeter.do.get_stable_speed(
timespan=10):
self.can_be_slowed_timer = time.time()
# Check 10 seconds after enabling slowdown
if self.can_be_slowed_timer and time.time(
) > self.can_be_slowed_timer + 10:
# Now let's check if it was stable in the last 10 seconds
self.can_be_slowed = BPSMeter.do.get_stable_speed(
timespan=10)
self.can_be_slowed_timer = 0
logging.debug("Downloader-slowdown: %r",
self.can_be_slowed)
else:
read, write, error = ([], [], [])
BPSMeter.do.reset()
time.sleep(1.0)
DOWNLOADER_CV.acquire()
while ((sabnzbd.nzbqueue.NzbQueue.do.is_empty()
or self.is_paused() or self.postproc)
and not self.shutdown and not self.__restart):
DOWNLOADER_CV.wait()
DOWNLOADER_CV.release()
self.force_disconnect = False
for selected in write:
nw = self.write_fds[selected]
fileno = nw.nntp.sock.fileno()
if fileno not in self.read_fds:
self.read_fds[fileno] = nw
if fileno in self.write_fds:
self.write_fds.pop(fileno)
if not read:
BPSMeter.do.update()
continue
for selected in read:
nw = self.read_fds[selected]
article = nw.article
server = nw.server
if article:
nzo = article.nzf.nzo
try:
bytes_received, done, skip = nw.recv_chunk()
except:
bytes_received, done, skip = (0, False, False)
if skip:
BPSMeter.do.update()
continue
if bytes_received < 1:
self.__reset_nw(nw,
"server closed connection",
warn=False,
wait=False)
continue
else:
if self.bandwidth_limit:
limit = self.bandwidth_limit
if bytes_received + BPSMeter.do.bps > limit:
while BPSMeter.do.bps > limit:
time.sleep(0.05)
BPSMeter.do.update()
BPSMeter.do.update(server.id, bytes_received)
nzo.update_download_stats(BPSMeter.do.bps, server.id,
bytes_received)
if not done and nw.status_code != 222:
if not nw.connected or nw.status_code == 480:
done = False
try:
nw.finish_connect(nw.status_code)
if sabnzbd.LOG_ALL:
logging.debug("%s@%s last message -> %s",
nw.thrdnum, nw.server.host,
nntp_to_msg(nw.data))
nw.clear_data()
except NNTPPermanentError as error:
# Handle login problems
block = False
penalty = 0
msg = error.response
ecode = int_conv(msg[:3])
display_msg = " [%s]" % msg
logging.debug("Server login problem: %s, %s",
ecode, msg)
if ecode in (502, 400, 481,
482) and clues_too_many(msg):
# Too many connections: remove this thread and reduce thread-setting for server
# Plan to go back to the full number after a penalty timeout
if server.active:
errormsg = T(
"Too many connections to server %s"
) % display_msg
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.warning(
T("Too many connections to server %s"
), server.host)
self.__reset_nw(nw,
None,
warn=False,
destroy=True,
send_quit=True)
self.plan_server(server, _PENALTY_TOOMANY)
server.threads -= 1
elif ecode in (502, 481,
482) and clues_too_many_ip(msg):
# Account sharing?
if server.active:
errormsg = T("Probable account sharing"
) + display_msg
if server.errormsg != errormsg:
server.errormsg = errormsg
name = " (%s)" % server.host
logging.warning(
T("Probable account sharing") +
name)
penalty = _PENALTY_SHARE
block = True
elif ecode in (452, 481, 482,
381) or (ecode == 502
and clues_login(msg)):
# Cannot login, block this server
if server.active:
errormsg = T("Failed login for server %s"
) % display_msg
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.error(
T("Failed login for server %s"),
server.host)
penalty = _PENALTY_PERM
block = True
elif ecode in (502, 482):
# Cannot connect (other reasons), block this server
if server.active:
errormsg = T(
"Cannot connect to server %s [%s]") % (
"", display_msg)
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.warning(
T("Cannot connect to server %s [%s]"
), server.host, msg)
if clues_pay(msg):
penalty = _PENALTY_PERM
else:
penalty = _PENALTY_502
block = True
elif ecode == 400:
# Temp connection problem?
if server.active:
logging.debug(
"Unspecified error 400 from server %s",
server.host)
penalty = _PENALTY_VERYSHORT
block = True
else:
# Unknown error, just keep trying
if server.active:
errormsg = T(
"Cannot connect to server %s [%s]") % (
"", display_msg)
if server.errormsg != errormsg:
server.errormsg = errormsg
logging.warning(
T("Cannot connect to server %s [%s]"
), server.host, msg)
penalty = _PENALTY_UNKNOWN
block = True
if block or (penalty and server.optional):
if server.active:
server.active = False
if penalty and (block or server.optional):
self.plan_server(server, penalty)
sabnzbd.nzbqueue.NzbQueue.do.reset_all_try_lists(
)
self.__reset_nw(nw,
None,
warn=False,
send_quit=True)
continue
except:
logging.error(
T("Connecting %s@%s failed, message=%s"),
nw.thrdnum,
nw.server.host,
nntp_to_msg(nw.data),
)
# No reset-warning needed, above logging is sufficient
self.__reset_nw(nw, None, warn=False)
if nw.connected:
logging.info("Connecting %s@%s finished",
nw.thrdnum, nw.server.host)
self.__request_article(nw)
elif nw.status_code == 223:
done = True
logging.debug("Article <%s> is present",
article.article)
elif nw.status_code == 211:
done = False
logging.debug("group command ok -> %s",
nntp_to_msg(nw.data))
nw.group = nw.article.nzf.nzo.group
nw.clear_data()
self.__request_article(nw)
elif nw.status_code in (411, 423, 430):
done = True
logging.debug(
"Thread %s@%s: Article %s missing (error=%s)",
nw.thrdnum,
nw.server.host,
article.article,
nw.status_code,
)
nw.clear_data()
elif nw.status_code == 500:
if nzo.precheck:
# Assume "STAT" command is not supported
server.have_stat = False
logging.debug("Server %s does not support STAT",
server.host)
else:
# Assume "BODY" command is not supported
server.have_body = False
logging.debug("Server %s does not support BODY",
server.host)
nw.clear_data()
self.__request_article(nw)
if done:
server.bad_cons = 0 # Successful data, clear "bad" counter
server.errormsg = server.warning = ""
if sabnzbd.LOG_ALL:
logging.debug("Thread %s@%s: %s done", nw.thrdnum,
server.host, article.article)
self.decode(article, nw.data)
nw.soft_reset()
server.busy_threads.remove(nw)
server.idle_threads.append(nw)
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
# Don't pound the website!
time.sleep(5.0)
(url, future_nzo) = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo and future_nzo.wait and future_nzo.wait > time.time():
# Requeue when too early and still active
self.add(url, future_nzo)
continue
url = url.replace(' ', '')
try:
del_bookmark = not future_nzo
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing', url)
continue
# Add nzbmatrix credentials if needed
url, matrix_id = _matrix_url(url)
# _grab_url cannot reside in a function, because the tempfile
# would not survive the end of the function
if del_bookmark:
logging.info('Removing nzbmatrix bookmark %s', matrix_id)
else:
logging.info('Grabbing URL %s', url)
opener = urllib.FancyURLopener({})
opener.prompt_user_passwd = None
opener.addheaders = []
opener.addheader('User-Agent', 'SABnzbd+/%s' % sabnzbd.version.__version__)
if not [True for item in _BAD_GZ_HOSTS if item in url]:
opener.addheader('Accept-encoding','gzip')
filename = None
category = None
length = 0
nzo_info = {}
try:
fn, header = opener.retrieve(url)
except:
fn = None
if fn:
for tup in header.items():
try:
item = tup[0].lower()
value = tup[1].strip()
except:
continue
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo',):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name',):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item in ('content-length',):
length = misc.int_conv(value)
if not filename:
for item in tup:
if "filename=" in item:
filename = item[item.index("filename=") + 9:].strip(';').strip('"')
if matrix_id:
fn, msg, retry, wait = _analyse_matrix(fn, matrix_id)
if not fn:
if retry:
logging.info(msg)
logging.debug('Retry nzbmatrix item %s after waiting %s sec', matrix_id, wait)
self.add(url, future_nzo, wait)
else:
logging.error(msg)
misc.bad_fetch(future_nzo, clean_matrix_url(url), msg, retry=True)
continue
category = _MATRIX_MAP.get(category, category)
if del_bookmark:
# No retries of nzbmatrix bookmark removals
continue
else:
fn, msg, retry, wait = _analyse_others(fn, url)
if not fn:
if retry:
logging.info('Retry URL %s', url)
self.add(url, future_nzo, wait)
else:
misc.bad_fetch(future_nzo, url, msg, retry=True)
continue
if not filename:
filename = os.path.basename(url) + '.nzb'
# Sanitize and trim name, preserving the extension
filename, ext = os.path.splitext(filename)
filename = misc.sanitize_foldername(filename)
filename += '.' + misc.sanitize_foldername(ext)
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# Check if nzb file
if os.path.splitext(filename)[1].lower() in ('.nzb', '.gz'):
res, nzo_ids = dirscanner.ProcessSingleFile(filename, fn, pp=pp, script=script, cat=cat, priority=priority, \
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url)
if res == 0:
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
else:
if res == -2:
logging.info('Incomplete NZB, retry after 5 min %s', url)
when = 300
else:
logging.info('Unknown error fetching NZB, retry after 2 min %s', url)
when = 120
self.add(url, future_nzo, when)
# Check if a supported archive
else:
if dirscanner.ProcessArchiveFile(filename, fn, pp, script, cat, priority=priority, url=future_nzo.url)[0] == 0:
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
else:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
logging.info('Unknown filetype when fetching NZB, retry after 30s %s', url)
self.add(url, future_nzo, 30)
except:
logging.error('URLGRABBER CRASHED', exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
(url, future_nzo) = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo:
# Re-queue when too early and still active
if future_nzo.wait and future_nzo.wait > time.time():
self.add(url, future_nzo)
time.sleep(1.0)
continue
# Paused
if future_nzo.status == Status.PAUSED:
self.add(url, future_nzo)
time.sleep(1.0)
continue
url = url.replace(' ', '')
try:
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing', url)
continue
filename = None
category = None
gzipped = False
nzo_info = {}
wait = 0
retry = True
fn = None
logging.info('Grabbing URL %s', url)
try:
fn = _build_request(url)
except Exception, e:
# Cannot list exceptions here, because of unpredictability over platforms
error0 = str(sys.exc_info()[0]).lower()
error1 = str(sys.exc_info()[1]).lower()
logging.debug('Error "%s" trying to get the url %s', error1, url)
if 'certificate_verify_failed' in error1 or 'certificateerror' in error0:
msg = T('Server %s uses an untrusted HTTPS certificate') % ''
msg += ' - https://sabnzbd.org/certificate-errors'
retry = False
elif 'nodename nor servname provided' in error1:
msg = T('Server name does not resolve')
retry = False
elif '401' in error1 or 'unauthorized' in error1:
msg = T('Unauthorized access')
retry = False
elif '404' in error1:
msg = T('File not on server')
retry = False
elif hasattr(e, 'headers') and 'retry-after' in e.headers:
# Catch if the server send retry (e.headers is case-INsensitive)
wait = misc.int_conv(e.headers['retry-after'])
new_url = dereferring(url, fn)
if new_url:
self.add(new_url, future_nzo)
continue
if fn:
for hdr in fn.headers:
try:
item = hdr.lower()
value = fn.headers[hdr]
except:
continue
if item in ('content-encoding',) and value == 'gzip':
gzipped = True
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo',):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name',):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item == 'x-dnzb-propername':
nzo_info['propername'] = value
elif item == 'x-dnzb-episodename':
nzo_info['episodename'] = value
elif item == 'x-dnzb-year':
nzo_info['year'] = value
elif item == 'x-dnzb-failure':
nzo_info['failure'] = value
elif item == 'x-dnzb-details':
nzo_info['details'] = value
elif item == 'x-dnzb-password':
nzo_info['password'] = value
elif item == 'retry-after':
wait = misc.int_conv(value)
# Rating fields
if item in _RARTING_FIELDS:
nzo_info[item] = value
if not filename and "filename=" in value:
filename = value[value.index("filename=") + 9:].strip(';').strip('"')
if wait:
# For sites that have a rate-limiting attribute
msg = ''
retry = True
fn = None
elif retry:
fn, msg, retry, wait, data = _analyse(fn, url)
if not fn:
if retry:
logging.info('Retry URL %s', url)
self.add(url, future_nzo, wait)
else:
bad_fetch(future_nzo, url, msg)
continue
if not filename:
filename = os.path.basename(url)
elif '&nzbname=' in filename:
# Sometimes the filename contains the full URL, duh!
filename = filename[filename.find('&nzbname=') + 9:]
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# process data
if gzipped:
filename += '.gz'
if not data:
try:
data = fn.read()
except (IncompleteRead, IOError):
bad_fetch(future_nzo, url, T('Server could not complete request'))
fn.close()
continue
fn.close()
if '<nzb' in data and misc.get_ext(filename) != '.nzb':
filename += '.nzb'
# Sanitize filename first (also removing forbidden Windows-names)
filename = misc.sanitize_filename(filename)
# Write data to temp file
path = os.path.join(cfg.admin_dir.get_path(), FUTURE_Q_FOLDER)
path = os.path.join(path, filename)
f = open(path, 'wb')
f.write(data)
f.close()
del data
# Check if nzb file
if misc.get_ext(filename) in ('.nzb', '.gz', 'bz2'):
res = dirscanner.ProcessSingleFile(filename, path, pp=pp, script=script, cat=cat, priority=priority,
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url, keep=False,
nzo_id=future_nzo.nzo_id)[0]
if res:
if res == -2:
logging.info('Incomplete NZB, retry after 5 min %s', url)
when = 300
elif res == -1:
# Error, but no reason to retry. Warning is already given
NzbQueue.do.remove(future_nzo.nzo_id, add_to_history=False)
continue
else:
logging.info('Unknown error fetching NZB, retry after 2 min %s', url)
when = 120
self.add(url, future_nzo, when)
# Check if a supported archive
else:
status, zf, exp_ext = dirscanner.is_archive(path)
if status == 0:
if misc.get_ext(filename) not in ('.rar', '.zip', '.7z'):
filename = filename + exp_ext
os.rename(path, path + exp_ext)
path = path + exp_ext
dirscanner.ProcessArchiveFile(filename, path, pp, script, cat, priority=priority,
nzbname=nzbname, url=future_nzo.url, keep=False,
nzo_id=future_nzo.nzo_id)
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
logging.info('Unknown filetype when fetching NZB, retry after 30s %s', url)
self.add(url, future_nzo, 30)
except:
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
# Don't pound the website!
time.sleep(5.0)
(url, future_nzo) = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo and future_nzo.wait and future_nzo.wait > time.time(
):
# Re-queue when too early and still active
self.add(url, future_nzo)
continue
url = url.replace(' ', '')
try:
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing',
url)
continue
logging.info('Grabbing URL %s', url)
req = urllib2.Request(url)
req.add_header('User-Agent',
'SABnzbd+/%s' % sabnzbd.version.__version__)
if not [True for item in _BAD_GZ_HOSTS if item in url]:
req.add_header('Accept-encoding', 'gzip')
filename = None
category = None
gzipped = False
nzo_info = {}
wait = 0
retry = True
fn = None
try:
fn = urllib2.urlopen(req)
except:
# Cannot list exceptions here, because of unpredictability over platforms
error0 = str(sys.exc_info()[0]).lower()
error1 = str(sys.exc_info()[1]).lower()
logging.debug('Error "%s" trying to get the url %s',
error1, url)
if 'certificate_verify_failed' in error1 or 'certificateerror' in error0:
msg = T('Server %s uses an untrusted HTTPS certificate'
) % ''
retry = False
elif 'nodename nor servname provided' in error1:
msg = T('Server name does not resolve')
retry = False
elif '401' in error1 or 'unauthorized' in error1:
msg = T('Unauthorized access')
retry = False
elif '404' in error1:
msg = T('File not on server')
retry = False
new_url = dereferring(url, fn)
if new_url:
self.add(new_url, future_nzo)
continue
if fn:
for hdr in fn.headers:
try:
item = hdr.lower()
value = fn.headers[hdr]
except:
continue
if item in ('content-encoding', ) and value == 'gzip':
gzipped = True
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo', ):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name', ):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item == 'x-dnzb-propername':
nzo_info['propername'] = value
elif item == 'x-dnzb-episodename':
nzo_info['episodename'] = value
elif item == 'x-dnzb-year':
nzo_info['year'] = value
elif item == 'x-dnzb-failure':
nzo_info['failure'] = value
elif item == 'x-dnzb-details':
nzo_info['details'] = value
elif item == 'x-dnzb-password':
nzo_info['password'] = value
elif item == 'retry-after':
# For NZBFinder
wait = misc.int_conv(value)
if not filename and "filename=" in value:
filename = value[value.index("filename=") +
9:].strip(';').strip('"')
if wait:
# For sites that have a rate-limiting attribute
msg = ''
retry = True
fn = None
elif retry:
fn, msg, retry, wait, data = _analyse(fn, url)
if not fn:
if retry:
logging.info('Retry URL %s', url)
self.add(url, future_nzo, wait)
else:
bad_fetch(future_nzo, url, msg)
continue
if not filename:
filename = os.path.basename(url) + '.nzb'
elif '&nzbname=' in filename:
# Sometimes the filename contains the full URL, duh!
filename = filename[filename.find('&nzbname=') + 9:]
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# process data
if gzipped:
filename = filename + '.gz'
if not data:
data = fn.read()
fn.close()
# Sanatize filename first
filename = misc.sanitize_filename(filename)
# Write data to temp file
path = os.path.join(cfg.admin_dir.get_path(), FUTURE_Q_FOLDER)
path = os.path.join(path, filename)
f = open(path, 'wb')
f.write(data)
f.close()
del data
# Check if nzb file
if os.path.splitext(filename)[1].lower() in ('.nzb', '.gz',
'bz2'):
res = dirscanner.ProcessSingleFile(
filename,
path,
pp=pp,
script=script,
cat=cat,
priority=priority,
nzbname=nzbname,
nzo_info=nzo_info,
url=future_nzo.url,
keep=False,
nzo_id=future_nzo.nzo_id)[0]
if res:
if res == -2:
logging.info(
'Incomplete NZB, retry after 5 min %s', url)
when = 300
elif res == -1:
# Error, but no reason to retry. Warning is already given
NzbQueue.do.remove(future_nzo.nzo_id,
add_to_history=False)
continue
else:
logging.info(
'Unknown error fetching NZB, retry after 2 min %s',
url)
when = 120
self.add(url, future_nzo, when)
# Check if a supported archive
else:
if dirscanner.ProcessArchiveFile(
filename,
path,
pp,
script,
cat,
priority=priority,
nzbname=nzbname,
url=future_nzo.url,
keep=False,
nzo_id=future_nzo.nzo_id)[0]:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
logging.info(
'Unknown filetype when fetching NZB, retry after 30s %s',
url)
self.add(url, future_nzo, 30)
except:
logging.error(T('URLGRABBER CRASHED'), exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
# Don't pound the website!
time.sleep(5.0)
(url, future_nzo) = self.queue.get()
if not url:
# stop signal, go test self.shutdown
continue
if future_nzo and future_nzo.wait and future_nzo.wait > time.time(
):
# Requeue when too early and still active
self.add(url, future_nzo)
continue
url = url.replace(' ', '')
try:
del_bookmark = not future_nzo
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except AttributeError:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing',
url)
continue
# Add nzbmatrix credentials if needed
url, matrix_id = _matrix_url(url)
# _grab_url cannot reside in a function, because the tempfile
# would not survive the end of the function
if del_bookmark:
logging.info('Removing nzbmatrix bookmark %s', matrix_id)
else:
logging.info('Grabbing URL %s', url)
if '.nzbsrus.' in url:
opener = urllib.URLopener({})
else:
opener = urllib.FancyURLopener({})
opener.prompt_user_passwd = None
opener.addheaders = []
opener.addheader('User-Agent',
'SABnzbd+/%s' % sabnzbd.version.__version__)
if not [True for item in _BAD_GZ_HOSTS if item in url]:
opener.addheader('Accept-encoding', 'gzip')
filename = None
category = None
length = 0
nzo_info = {}
wait = 0
try:
fn, header = opener.retrieve(url)
except:
fn = None
if fn:
for tup in header.items():
try:
item = tup[0].lower()
value = tup[1].strip()
except:
continue
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo', ):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name', ):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item == 'x-dnzb-propername':
nzo_info['propername'] = value
elif item == 'x-dnzb-episodename':
nzo_info['episodename'] = value
elif item == 'x-dnzb-year':
nzo_info['year'] = value
elif item == 'x-dnzb-failure':
nzo_info['failure'] = value
elif item == 'x-dnzb-details':
nzo_info['details'] = value
elif item in ('content-length', ):
length = misc.int_conv(value)
elif item == 'retry-after':
# For NZBFinder
wait = misc.int_conv(value)
if not filename:
for item in tup:
if "filename=" in item:
filename = item[item.index("filename=") +
9:].strip(';').strip('"')
if matrix_id:
fn, msg, retry, wait = _analyse_matrix(fn, matrix_id)
if not fn:
if retry:
logging.info(msg)
logging.debug(
'Retry nzbmatrix item %s after waiting %s sec',
matrix_id, wait)
self.add(url, future_nzo, wait)
else:
logging.error(msg)
misc.bad_fetch(future_nzo,
clean_matrix_url(url),
msg,
retry=True)
continue
category = get_matrix_category(url, category)
if del_bookmark:
# No retries of nzbmatrix bookmark removals
continue
else:
if wait:
# For sites that have a rate-limiting attribute
msg = ''
retry = True
fn = None
else:
fn, msg, retry, wait = _analyse_others(fn, url)
if not fn:
if retry:
logging.info('Retry URL %s', url)
self.add(url, future_nzo, wait)
else:
misc.bad_fetch(future_nzo, url, msg, retry=True)
continue
if not filename:
filename = os.path.basename(url) + '.nzb'
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# Check if nzb file
if os.path.splitext(filename)[1].lower() in ('.nzb', '.gz'):
res, nzo_ids = dirscanner.ProcessSingleFile(filename, fn, pp=pp, script=script, cat=cat, priority=priority, \
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url)
if res == 0:
NzbQueue.do.remove(future_nzo.nzo_id,
add_to_history=False)
else:
if res == -2:
logging.info(
'Incomplete NZB, retry after 5 min %s', url)
when = 300
elif res == -1:
# Error, but no reason to retry. Warning is already given
NzbQueue.do.remove(future_nzo.nzo_id,
add_to_history=False)
continue
else:
logging.info(
'Unknown error fetching NZB, retry after 2 min %s',
url)
when = 120
self.add(url, future_nzo, when)
# Check if a supported archive
else:
if dirscanner.ProcessArchiveFile(
filename,
fn,
pp,
script,
cat,
priority=priority,
nzbname=nzbname,
url=future_nzo.url)[0] == 0:
NzbQueue.do.remove(future_nzo.nzo_id,
add_to_history=False)
else:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
logging.info(
'Unknown filetype when fetching NZB, retry after 30s %s',
url)
self.add(url, future_nzo, 30)
except:
logging.error('URLGRABBER CRASHED', exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
def run_feed(self, feed=None, download=False, ignoreFirst=False, force=False, readout=True):
""" Run the query for one URI and apply filters """
self.shutdown = False
def dup_title(title):
""" Check if this title was in this or other feeds
Return matching feed name
"""
title = title.lower()
for fd in self.jobs:
for lk in self.jobs[fd]:
item = self.jobs[fd][lk]
if item.get('status', ' ')[0] == 'D' and \
item.get('title', '').lower() == title:
return fd
return ''
if not feed:
return 'No such feed'
newlinks = []
new_downloads = []
# Preparations, get options
try:
feeds = config.get_rss()[feed]
except KeyError:
logging.error(T('Incorrect RSS feed description "%s"'), feed)
logging.info("Traceback: ", exc_info=True)
return T('Incorrect RSS feed description "%s"') % feed
uri = feeds.uri()
defCat = feeds.cat()
import sabnzbd.api
if not notdefault(defCat) or defCat not in sabnzbd.api.list_cats(default=False):
defCat = None
defPP = feeds.pp()
if not notdefault(defPP):
defPP = None
defScript = feeds.script()
if not notdefault(defScript):
defScript = None
defPrio = feeds.priority()
if not notdefault(defPrio):
defPrio = None
# Preparations, convert filters to regex's
regexes = []
reTypes = []
reCats = []
rePPs = []
rePrios = []
reScripts = []
reEnabled = []
for filter in feeds.filters():
reCat = filter[0]
if defCat in ('', '*'):
reCat = None
reCats.append(reCat)
rePPs.append(filter[1])
reScripts.append(filter[2])
reTypes.append(filter[3])
if filter[3] in ('<', '>', 'F'):
regexes.append(filter[4])
else:
regexes.append(convert_filter(filter[4]))
rePrios.append(filter[5])
reEnabled.append(filter[6] != '0')
regcount = len(regexes)
# Set first if this is the very first scan of this URI
first = (feed not in self.jobs) and ignoreFirst
# Add sabnzbd's custom User Agent
feedparser.USER_AGENT = 'SABnzbd+/%s' % sabnzbd.version.__version__
# Check for nzbs.org
if 'nzbs.org/' in uri and '&dl=1' not in uri:
uri += '&dl=1'
# Read the RSS feed
msg = None
entries = None
if readout:
uri = uri.replace(' ', '%20')
logging.debug("Running feedparser on %s", uri)
d = feedparser.parse(uri.replace('feed://', 'http://'))
logging.debug("Done parsing %s", uri)
if not d:
msg = T('Failed to retrieve RSS from %s: %s') % (uri, '?')
logging.info(msg)
return unicoder(msg)
status = d.get('status', 999)
if status in (401, 402, 403):
msg = T('Do not have valid authentication for feed %s') % feed
logging.info(msg)
return unicoder(msg)
if status >= 500 and status <= 599:
msg = T('Server side error (server code %s); could not get %s on %s') % (status, feed, uri)
logging.info(msg)
return unicoder(msg)
entries = d.get('entries')
if 'bozo_exception' in d and not entries:
msg = str(d['bozo_exception'])
if 'CERTIFICATE_VERIFY_FAILED' in msg:
msg = T('Server %s uses an untrusted HTTPS certificate') % get_urlbase(uri)
logging.error(msg)
else:
msg = T('Failed to retrieve RSS from %s: %s') % (uri, xml_name(msg))
logging.info(msg)
return unicoder(msg)
if not entries:
msg = T('RSS Feed %s was empty') % uri
logging.info(msg)
if feed not in self.jobs:
self.jobs[feed] = {}
jobs = self.jobs[feed]
if readout:
if not entries:
return unicoder(msg)
else:
entries = jobs.keys()
# Sort in the order the jobs came from the feed
entries.sort(lambda x, y: jobs[x].get('order', 0) - jobs[y].get('order', 0))
order = 0
# Filter out valid new links
for entry in entries:
if self.shutdown:
return
if readout:
try:
link, category, size = _get_link(uri, entry)
except (AttributeError, IndexError):
link = None
category = u''
size = 0L
logging.info(T('Incompatible feed') + ' ' + uri)
logging.info("Traceback: ", exc_info=True)
return T('Incompatible feed')
title = entry.title
else:
link = entry
category = jobs[link].get('orgcat', '')
if category in ('', '*'):
category = None
title = jobs[link].get('title', '')
size = jobs[link].get('size', 0L)
if link:
# Make sure spaces are quoted in the URL
link = link.strip().replace(' ', '%20')
newlinks.append(link)
if link in jobs:
jobstat = jobs[link].get('status', ' ')[0]
else:
jobstat = 'N'
if jobstat in 'NGB' or (jobstat == 'X' and readout):
# Match this title against all filters
logging.debug('Trying title %s', title)
result = False
myCat = defCat
myPP = defPP
myScript = defScript
myPrio = defPrio
n = 0
if 'F' in reTypes:
season, episode = sabnzbd.newsunpack.analyse_show(title)[1:3]
season = int_conv(season)
episode = int_conv(episode)
else:
season = episode = 0
# Match against all filters until an positive or negative match
logging.debug('Size %s for %s', size, title)
for n in xrange(regcount):
if reEnabled[n]:
if category and reTypes[n] == 'C':
found = re.search(regexes[n], category)
if not found:
logging.debug("Filter rejected on rule %d", n)
result = False
break
elif reTypes[n] == '<' and size and from_units(regexes[n]) < size:
# "Size at most" : too large
logging.debug('Filter rejected on rule %d', n)
result = False
break
elif reTypes[n] == '>' and size and from_units(regexes[n]) > size:
# "Size at least" : too small
logging.debug('Filter rejected on rule %d', n)
result = False
break
elif reTypes[n] == 'F' and not ep_match(season, episode, regexes[n]):
# "Starting from SxxEyy", too early episode
logging.debug('Filter requirement match on rule %d', n)
result = False
break
else:
if regexes[n]:
found = re.search(regexes[n], title)
else:
found = False
if reTypes[n] == 'M' and not found:
logging.debug("Filter rejected on rule %d", n)
result = False
break
if found and reTypes[n] == 'A':
logging.debug("Filter matched on rule %d", n)
result = True
break
if found and reTypes[n] == 'R':
logging.debug("Filter rejected on rule %d", n)
result = False
break
if len(reCats):
if notdefault(reCats[n]):
myCat = reCats[n]
elif category and not defCat:
myCat = cat_convert(category)
if myCat:
myCat, catPP, catScript, catPrio = cat_to_opts(myCat)
else:
myCat = catPP = catScript = catPrio = None
if notdefault(rePPs[n]):
myPP = rePPs[n]
elif not (reCats[n] or category):
myPP = catPP
if notdefault(reScripts[n]):
myScript = reScripts[n]
elif not (notdefault(reCats[n]) or category):
myScript = catScript
if rePrios[n] not in (str(DEFAULT_PRIORITY), ''):
myPrio = rePrios[n]
elif not ((rePrios[n] != str(DEFAULT_PRIORITY)) or category):
myPrio = catPrio
if cfg.no_dupes() and dup_title(title):
if cfg.no_dupes() == 1:
logging.info("Ignoring duplicate job %s", title)
continue
else:
myPrio = DUP_PRIORITY
act = download and not first
if link in jobs:
act = act and not jobs[link].get('status', '').endswith('*')
act = act or force
star = first or jobs[link].get('status', '').endswith('*')
else:
star = first
if result:
_HandleLink(jobs, link, title, size, 'G', category, myCat, myPP, myScript,
act, star, order, priority=myPrio, rule=str(n))
if act:
new_downloads.append(title)
else:
_HandleLink(jobs, link, title, size, 'B', category, myCat, myPP, myScript,
False, star, order, priority=myPrio, rule=str(n))
order += 1
# Send email if wanted and not "forced"
if new_downloads and cfg.email_rss() and not force:
emailer.rss_mail(feed, new_downloads)
remove_obsolete(jobs, newlinks)
return ''
def run(self):
logging.info('URLGrabber starting up')
self.shutdown = False
while not self.shutdown:
# Don't pound the website!
time.sleep(5.0)
(url, future_nzo, retry_count) = self.queue.get()
if not url:
continue
url = url.replace(' ', '')
try:
del_bookmark = not future_nzo
if future_nzo:
# If nzo entry deleted, give up
try:
deleted = future_nzo.deleted
except:
deleted = True
if deleted:
logging.debug('Dropping URL %s, job entry missing',
url)
continue
# Add nzbmatrix credentials if needed
url, matrix_id = _matrix_url(url)
# When still waiting for nzbmatrix wait period, requeue
if matrix_id and self.matrix_wait > time.time():
self.queue.put((url, future_nzo, retry_count))
continue
# _grab_url cannot reside in a function, because the tempfile
# would not survive the end of the function
if del_bookmark:
logging.info('Removing nzbmatrix bookmark %s', matrix_id)
else:
logging.info('Grabbing URL %s', url)
opener = urllib.FancyURLopener({})
opener.prompt_user_passwd = None
opener.addheaders = []
opener.addheader('User-Agent',
'SABnzbd+/%s' % sabnzbd.version.__version__)
opener.addheader('Accept-encoding', 'gzip')
filename = None
category = None
length = 0
nzo_info = {}
try:
fn, header = opener.retrieve(url)
except:
fn = None
if fn:
for tup in header.items():
try:
item = tup[0].lower()
value = tup[1].strip()
except:
continue
if item in ('category_id', 'x-dnzb-category'):
category = value
elif item in ('x-dnzb-moreinfo', ):
nzo_info['more_info'] = value
elif item in ('x-dnzb-name', ):
filename = value
if not filename.endswith('.nzb'):
filename += '.nzb'
elif item in ('content-length', ):
length = misc.int_conv(value)
if not filename:
for item in tup:
if "filename=" in item:
filename = item[item.index("filename=") +
9:].strip(';').strip('"')
if matrix_id:
fn, msg, retry, wait = _analyse_matrix(fn, matrix_id)
if retry and wait > 0:
self.matrix_wait = time.time() + wait
logging.debug('Retry URL %s after waiting', url)
self.queue.put((url, future_nzo, retry_count))
continue
category = _MATRIX_MAP.get(category, category)
else:
msg = ''
retry = True
# Check if the filepath is specified, if not, check if a retry is allowed.
if not fn:
retry_count -= 1
if retry_count > 0 and retry:
logging.info('Retry URL %s', url)
self.queue.put((url, future_nzo, retry_count))
elif not del_bookmark:
misc.bad_fetch(future_nzo, url, msg, retry=True)
continue
if del_bookmark:
continue
if not filename:
filename = os.path.basename(url) + '.nzb'
# Sanitize and trim name, preserving the extension
filename, ext = os.path.splitext(filename)
filename = misc.sanitize_foldername(filename)
filename += '.' + misc.sanitize_foldername(ext)
pp = future_nzo.pp
script = future_nzo.script
cat = future_nzo.cat
if (cat is None or cat == '*') and category:
cat = misc.cat_convert(category)
priority = future_nzo.priority
nzbname = future_nzo.custom_name
# Check if nzb file
if os.path.splitext(filename)[1].lower() in ('.nzb', '.gz'):
res = dirscanner.ProcessSingleFile(filename, fn, pp=pp, script=script, cat=cat, priority=priority, \
nzbname=nzbname, nzo_info=nzo_info, url=future_nzo.url)
if res == 0:
NzbQueue.do.remove(future_nzo.nzo_id,
add_to_history=False)
elif res == -2:
retry_count -= 1
if retry_count > 0:
logging.info('Incomplete NZB, retry %s', url)
self.queue.put((url, future_nzo, retry_count))
else:
misc.bad_fetch(future_nzo,
url,
retry=True,
content=True)
else:
misc.bad_fetch(future_nzo,
url,
retry=True,
content=True)
# Check if a supported archive
else:
if dirscanner.ProcessArchiveFile(filename,
fn,
pp,
script,
cat,
priority=priority,
url=future_nzo.url) == 0:
NzbQueue.do.remove(future_nzo.nzo_id,
add_to_history=False)
else:
# Not a supported filetype, not an nzb (text/html ect)
try:
os.remove(fn)
except:
pass
misc.bad_fetch(future_nzo,
url,
retry=True,
content=True)
except:
logging.error('URLGRABBER CRASHED', exc_info=True)
logging.debug("URLGRABBER Traceback: ", exc_info=True)
