def upgrade():
with db_session() as db:
db.query(Release).\
filter(Release.tvshow_metablack_id == MetaBlack.id).\
filter(MetaBlack.status=='IMPOSSIBLE').\
update({Release.tvshow_metablack_id: None}, synchronize_session='fetch')
db.commit()
def remove_group(group_name):
with db_session() as db:
deleted = db.query(Group).filter(Group.name == group_name).delete()
if deleted:
db.commit()
return True
return False
def convert_users(mysql):
"""Converts Newznab users table into Pynab format. More or less
of this may be necessary depending on what people want. I'm pretty
much just after bare API access, so we only really need rsstoken."""
print('Converting users...')
with db_session() as db:
from_query = """
SELECT username, email, password, rsstoken, userseed, grabs
FROM users
ORDER BY id;
"""
cursor = mysql.cursor()
cursor.execute(from_query)
db.query(User).delete()
for r in cursor.fetchall():
u = User(
email=r[1],
api_key=r[3],
grabs=r[5]
)
db.add(u)
def convert_imdb(mysql):
"""Converts Newznab imdb table into Pynab format."""
print('Converting imdb...')
with db_session() as db:
from_query = """
SELECT imdbID, title, year, language, genre
FROM movieinfo
WHERE imdbID > 0
ORDER BY imdbID
"""
cursor = mysql.cursor()
cursor.execute(from_query)
for r in cursor.fetchall():
if not r[2]:
# Blank years do not work, skip them
continue
movie = db.query(Movie).filter(Movie.id==str(r[0])).first()
if not movie:
movie = Movie(
id=r[0],
name=r[1],
year=r[2],
genre=r[4]
)
db.add(movie)
def convert_groups(mysql):
"""Converts Newznab groups table into Pynab. Only really
copies backfill records and status."""
# removed minsize/minfiles, since we're not really using them
# most of the groups I index don't come up with too many stupid
# releases, so if anyone has problem groups they can re-add it
print('Converting groups...')
with db_session() as db:
from_query = """
SELECT name, first_record, last_record, active
FROM groups;
"""
cursor = mysql.cursor()
cursor.execute(from_query)
for r in cursor.fetchall():
g = db.query(Group).filter(Group.name==r[0]).first()
if not g:
g = Group(name=r[0])
g.first = r[1]
g.last = r[2]
g.active = bool(r[3])
db.add(g)
def get_nzb(dataset=None):
user = auth()
if user:
id = request.query.guid or None
if not id:
id = request.query.id or None
if id:
with db_session() as db:
release = db.query(Release).join(NZB).join(Category).filter(Release.id == id).one()
if release:
release.grabs += 1
user.grabs += 1
db.merge(release)
db.merge(user)
db.commit()
data = release.nzb.data
response.set_header('Content-type', 'application/x-nzb-compressed-gzip')
response.set_header('X-DNZB-Name', release.search_name)
response.set_header('X-DNZB-Category', release.category.name)
response.set_header('Content-Disposition', 'attachment; filename="{0}"'
.format(release.search_name.replace(' ', '_') + '.nzb.gz')
)
return gzip.decompress(data)
else:
return api_error(300)
else:
return api_error(200)
else:
return api_error(100)
def remove_group(group_name):
with db_session() as db:
deleted = db.query(Group).filter(Group.name == group_name).delete()
if deleted:
db.commit()
return True
return False
def convert_groups(mysql):
"""Converts Newznab groups table into Pynab. Only really
copies backfill records and status."""
# removed minsize/minfiles, since we're not really using them
# most of the groups I index don't come up with too many stupid
# releases, so if anyone has problem groups they can re-add it
print('Converting groups...')
with db_session() as db:
from_query = """
SELECT name, first_record, last_record, active
FROM groups;
"""
cursor = mysql.cursor()
cursor.execute(from_query)
for r in cursor.fetchall():
g = db.query(Group).filter(Group.name == r[0]).first()
if not g:
g = Group(name=r[0])
g.first = r[1]
g.last = r[2]
g.active = bool(r[3])
db.add(g)
def details(dataset=None):
if auth():
if request.query.id:
with db_session() as db:
release = db.query(Release).filter(
Release.id == request.query.id).first()
if release:
dataset['releases'] = [release]
dataset['detail'] = True
dataset['api_key'] = request.query.apikey
try:
tmpl = Template(filename=os.path.join(
root_dir, 'templates/api/result.mako'))
return tmpl.render(**dataset)
except:
log.error('Failed to deliver page: {0}'.format(
exceptions.text_error_template().render()))
return None
else:
return api_error(300)
else:
return api_error(200)
else:
return api_error(100)
def rename_pre_releases():
count = 0
with db_session() as db:
query = db.query(Release).filter(Release.pre_id != None)
query = query.outerjoin(
Pre,
Pre.id == Release.pre_id).filter((Release.name != Pre.name) | (
Release.search_name != Pre.searchname))
for release in query.all():
old_category_id = release.category_id
release.name = release.pre.name
release.search_name = release.pre.searchname
release.category_id = pynab.categories.determine_category(
release.search_name, release.group.name)
db.merge(release)
count += 1
log.info('rename: [{}] -> [{}]'.format(release.search_name,
release.pre.searchname))
db.commit()
log.info('rename: successfully renamed {} releases'.format(count))
def upgrade():
with db_session() as db:
db.query(Release).\
filter(Release.tvshow_metablack_id == MetaBlack.id).\
filter(MetaBlack.status=='IMPOSSIBLE').\
update({Release.tvshow_metablack_id: None}, synchronize_session='fetch')
db.commit()
def test_load_and_categorise(self):
from pynab.db import db_session, Release, Group, windowed_query
from pickle import load
with open('release_categoriser.pkl', 'rb') as cat_file:
categoriser = load(cat_file)
with db_session() as db:
errors = []
i = 0
query = db.query(Release).join(Group)
count = query.count()
for result in windowed_query(query, Release.id, 500):
features = extract_features(result.name)
features['group'] = result.group.name
features['name'] = result.name
guess = categoriser.classify(features)
if guess[:2] != str(result.category_id)[:2]:
errors.append((result.category_id, guess, features))
i += 1
if i % 500 == 0:
print('{} - {:.3f}%'.format((i/count)*100, (1 - (len(errors) / i)) * 100))
for tag, guess, features in errors:
print('correct={} guess={} name={}'.format(tag, guess, features['name'].encode('utf-8')))
print('accuracy={}'.format(1 - (len(errors)/i)))
def local_postprocess():
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone
db.query(MetaBlack).filter(MetaBlack.status=='IMPOSSIBLE').filter((MetaBlack.movie!=None)|(MetaBlack.tvshow!=None)).delete(synchronize_session=False)
pynab.ids.process('movie', online=False)
pynab.ids.process('tv', online=False)
def process(limit=None, category=0):
"""Process releases for NFO parts and download them."""
with Server() as server:
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone
query = db.query(Release).join(Group).join(NZB).filter(
Release.nfo == None).filter(Release.nfo_metablack_id == None)
if category:
query = query.filter(Release.category_id == int(category))
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
for release in releases:
found = False
nzb = pynab.nzbs.get_nzb_details(release.nzb)
if nzb:
nfos = []
for nfo in nzb['nfos']:
for part in nfo['segments']:
if int(part['size']) > NFO_MAX_FILESIZE:
continue
nfos.append(part)
for nfo in nfos:
try:
article = server.get(release.group.name, [
nfo['message_id'],
])
except Exception as e:
# if usenet's not accessible, don't block it forever
log.error('nfo: unable to get nfo: {}'.format(e))
continue
if article:
data = gzip.compress(article.encode('utf-8'))
nfo = NFO(data=data)
db.add(nfo)
release.nfo = nfo
release.nfo_metablack_id = None
db.add(release)
log.debug('nfo: [{}] - nfo added'.format(
release.search_name))
found = True
break
if not found:
log.debug(
'nfo: [{}] - [{}] - no nfos in release'.format(
release.id, release.search_name))
mb = MetaBlack(nfo=release, status='IMPOSSIBLE')
db.add(mb)
db.commit()
def process(limit=None, category=0):
"""Process releases for NFO parts and download them."""
with Server() as server:
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone
query = db.query(Release).join(Group).join(NZB).filter(Release.nfo == None).filter(
Release.nfo_metablack_id == None)
if category:
query = query.filter(Release.category_id == int(category))
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
for release in releases:
found = False
nzb = pynab.nzbs.get_nzb_details(release.nzb)
if nzb:
nfos = []
for nfo in nzb['nfos']:
for part in nfo['segments']:
if int(part['size']) > NFO_MAX_FILESIZE:
continue
nfos.append(part)
for nfo in nfos:
try:
article = server.get(release.group.name, [nfo['message_id'], ])
except Exception as e:
# if usenet's not accessible, don't block it forever
log.error('nfo: unable to get nfo: {}'.format(e))
continue
if article:
data = gzip.compress(article.encode('utf-8'))
nfo = NFO(data=data)
db.add(nfo)
release.nfo = nfo
release.nfo_metablack_id = None
db.add(release)
log.debug('nfo: [{}] - nfo added'.format(
release.search_name
))
found = True
break
if not found:
log.debug('nfo: [{}] - [{}] - no nfos in release'.format(
release.id,
release.search_name
))
mb = MetaBlack(nfo=release, status='IMPOSSIBLE')
db.add(mb)
db.commit()
def group_list():
with db_session() as db:
groups = db.query(Group).order_by(Group.name)
group_list = []
for group in groups:
group_list.append(group)
return group_list
def group_list():
with db_session() as db:
groups = db.query(Group).order_by(Group.name)
group_list = []
for group in groups:
group_list.append(group)
return group_list
def test_search_releases(self):
from sqlalchemy_searchable import search
from pynab.db import Release
with db_session() as db:
q = db.query(Release)
q = search(q, 'engaged e06')
print(q.first().search_name)
def test_nzb_parse(self):
import pynab.nzbs
from pynab.db import NZB
with db_session() as db:
nzb = db.query(NZB).filter(NZB.id==1).one()
import pprint
pprint.pprint(pynab.nzbs.get_nzb_details(nzb))
def info(email):
"""Information about a specific email."""
with db_session() as db:
user = db.query(User).filter(User.email == email).first()
if user:
return [user.email, user.api_key, user.grabs]
else:
return None
def get_categories(self):
if self.categories:
return self.categories
else:
self.categories = {}
with db_session() as db:
for category in db.query(Category).all():
self.categories[int(category.id)] = category.name
return self.categories
def local_postprocess():
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone
db.query(MetaBlack).filter(MetaBlack.status == 'IMPOSSIBLE').filter((
MetaBlack.movie != None) | (MetaBlack.tvshow != None)).delete(
synchronize_session=False)
pynab.ids.process('movie', online=False)
pynab.ids.process('tv', online=False)
def list():
"""List all users."""
with db_session() as db:
users = db.query(User).order_by(User.email)
user_list = []
for user in users:
user_list.append([user.email, user.api_key, user.grabs])
return user_list
def auth():
api_key = request.query.apikey or ''
with db_session() as db:
user = db.query(User).filter(User.api_key == api_key).first()
if user:
return user
else:
return None
def get_categories(self):
if self.categories:
return self.categories
else:
self.categories = {}
with db_session() as db:
for category in db.query(Category).all():
self.categories[int(category.id)] = category.name
return self.categories
def disable_group(group_name):
with db_session() as db:
group = db.query(Group).filter(Group.name == group_name).first()
if group:
group.active = False
db.add(group)
db.commit()
return True
return False
def auth():
api_key = request.query.apikey or ''
with db_session() as db:
user = db.query(User).filter(User.api_key == api_key).first()
if user:
return user
else:
return None
def disable_group(group_name):
with db_session() as db:
group = db.query(Group).filter(Group.name == group_name).first()
if group:
group.active = False
db.add(group)
db.commit()
return True
return False
def delete(email):
"""Deletes a user by email."""
with db_session() as db:
deleted = db.query(User).filter(User.email == email).delete()
if deleted:
db.commit()
return True
return False
def reset_group(group_name):
with db_session() as db:
group = db.query(Group).filter(Group.name == group_name).first()
if group:
group.first = 0
group.last = 0
db.add(group)
db.commit()
return True
return False
def reset_group(group_name):
with db_session() as db:
group = db.query(Group).filter(Group.name == group_name).first()
if group:
group.first = 0
group.last = 0
db.add(group)
db.commit()
return True
return False
def process(limit=None, category=0):
"""Process releases for SFV parts and download them."""
with Server() as server:
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone
query = db.query(Release).join(Group).join(NZB).filter(Release.sfv == None).filter(
Release.sfv_metablack_id == None)
if category:
query = query.filter(Release.category_id == int(category))
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
for release in releases:
found = False
nzb = pynab.nzbs.get_nzb_details(release.nzb)
if nzb:
sfvs = []
for sfv in nzb['sfvs']:
for part in sfv['segments']:
if int(part['size']) > SFV_MAX_FILESIZE:
continue
sfvs.append(part)
for sfv in sfvs:
try:
article = server.get(release.group.name, [sfv['message_id'], ])
except:
article = None
if article:
data = gzip.compress(article.encode('utf-8'))
sfv = SFV(data=data)
db.add(sfv)
release.sfv = sfv
release.sfv_metablack_id = None
db.add(release)
log.info('sfv: [{}] - sfv added'.format(
release.search_name
))
found = True
break
if not found:
log.debug('sfv: [{}] - no sfvs in release'.format(
release.search_name
))
mb = MetaBlack(sfv=release, status='IMPOSSIBLE')
db.add(mb)
db.commit()
def process(limit=None, category=0):
"""Process releases for SFV parts and download them."""
with Server() as server:
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone
query = db.query(Release).join(Group).join(NZB).filter(
Release.sfv == None).filter(Release.sfv_metablack_id == None)
if category:
query = query.filter(Release.category_id == int(category))
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
for release in releases:
found = False
nzb = pynab.nzbs.get_nzb_details(release.nzb)
if nzb:
sfvs = []
for sfv in nzb['sfvs']:
for part in sfv['segments']:
if int(part['size']) > SFV_MAX_FILESIZE:
continue
sfvs.append(part)
for sfv in sfvs:
try:
article = server.get(release.group.name, [
sfv['message_id'],
])
except:
article = None
if article:
data = gzip.compress(article.encode('utf-8'))
sfv = SFV(data=data)
db.add(sfv)
release.sfv = sfv
release.sfv_metablack_id = None
db.add(release)
log.info('sfv: [{}] - sfv added'.format(
release.search_name))
found = True
break
if not found:
log.debug('sfv: [{}] - no sfvs in release'.format(
release.search_name))
mb = MetaBlack(sfv=release, status='IMPOSSIBLE')
db.add(mb)
db.commit()
def process(limit=None, category=0):
"""Processes release rarfiles to check for passwords and filecounts."""
with Server() as server:
with db_session() as db:
# noinspection PyComparisonWithNone
query = db.query(Release).join(Group).join(NZB).filter(~Release.files.any()). \
filter(Release.passworded == 'UNKNOWN').filter(Release.rar_metablack_id == None)
if category:
query = query.filter(Release.category_id == int(category))
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
for release in releases:
log.debug('rar: processing {}'.format(release.search_name))
nzb = pynab.nzbs.get_nzb_details(release.nzb)
if nzb and nzb['rars']:
try:
passworded, info = check_release_files(server, release.group.name, nzb)
except Exception as e:
# if usenet isn't accessible, we don't want to blacklist it
log.error('rar: file info failed: {}'.format(e))
continue
if info:
log.info('rar: file info add [{}]'.format(
release.search_name
))
release.passworded = passworded
size = 0
for file in info:
f = File(name=file['name'][:512],
size=file['size'])
f.release = release
size += file['size']
db.add(f)
if size != 0:
release.size = size
release.rar_metablack_id = None
db.add(release)
db.commit()
continue
log.debug('rar: [{}] - file info: no readable rars in release'.format(
release.search_name
))
mb = MetaBlack(rar=release, status='IMPOSSIBLE')
db.add(mb)
db.commit()
def process(limit=None, category=0):
"""Processes release rarfiles to check for passwords and filecounts."""
with Server() as server:
with db_session() as db:
# noinspection PyComparisonWithNone
query = db.query(Release).join(Group).join(NZB).filter(~Release.files.any()). \
filter(Release.passworded == 'UNKNOWN').filter(Release.rar_metablack_id == None)
if category:
query = query.filter(Release.category_id == int(category))
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
for release in releases:
log.debug('rar: processing {}'.format(release.search_name))
nzb = pynab.nzbs.get_nzb_details(release.nzb)
if nzb and nzb['rars']:
try:
passworded, info = check_release_files(
server, release.group.name, nzb)
except Exception as e:
# if usenet isn't accessible, we don't want to blacklist it
log.error('rar: file info failed: {}'.format(e))
continue
if info:
log.info('rar: file info add [{}]'.format(
release.search_name))
release.passworded = passworded
size = 0
for file in info:
f = File(name=file['name'][:512],
size=file['size'])
f.release = release
size += file['size']
db.add(f)
if size != 0:
release.size = size
release.rar_metablack_id = None
db.add(release)
db.commit()
continue
log.debug('rar: [{}] - file info: no readable rars in release'.
format(release.search_name))
mb = MetaBlack(rar=release, status='IMPOSSIBLE')
db.add(mb)
db.commit()
def process(limit=None):
"""Process releases for requests"""
with db_session() as db:
query = db.query(Release).join(Group).filter(Release.name.like('REQ:%')).filter(Release.pre_id == None).filter(
Release.category_id == '8010')
if limit:
releases = query.order_by(Release.posted.desc()).limit(limit)
else:
releases = query.order_by(Release.posted.desc()).all()
# create a dict of request id's and releases
requests = {}
if releases:
for release in releases:
# check if it's aliased
if release.group.name in GROUP_ALIASES:
group_name = GROUP_ALIASES[release.group.name]
else:
group_name = release.group.name
if group_name not in requests:
requests[group_name] = {}
try:
requests[group_name][int(release.name.split(': ')[1])] = release
except ValueError:
# request hash?
continue
else:
log.info("requests: no release requests to process")
# per-group
for group_name, group_requests in requests.items():
# query for the requestids
if requests:
pres = db.query(Pre).filter(Pre.requestgroup==group_name).filter(Pre.requestid.in_(group_requests.keys())).all()
else:
log.info("requests: no pre requests found")
pres = []
# loop through and associate pres with their requests
for pre in pres:
# no longer need to check group
updated_release = group_requests.get(pre.requestid)
updated_release.pre_id = pre.id
db.merge(updated_release)
log.info("requests: found pre request id {} ({}) for {}".format(pre.requestid, group_name,
updated_release.name))
db.commit()
def rename_bad_releases(category):
count = 0
s_count = 0
for_deletion = []
with db_session() as db:
# noinspection PyComparisonWithNone,PyComparisonWithNone,PyComparisonWithNone,PyComparisonWithNone
query = db.query(Release).filter(Release.category_id==int(category)).filter(
(Release.files.any())|(Release.nfo_id!=None)|(Release.sfv_id!=None)|(Release.pre_id!=None)
).filter((Release.status!=1)|(Release.status==None)).filter(Release.unwanted==False)
for release in windowed_query(query, Release.id, config.scan.get('binary_process_chunk_size', 1000)):
count += 1
name, category_id = pynab.releases.discover_name(release)
if not name and category_id:
# don't change the name, but the category might need changing
release.category_id = category_id
# we're done with this release
release.status = 1
db.merge(release)
elif name and category_id:
# only add it if it doesn't exist already
existing = db.query(Release).filter(Release.name==name,
Release.group_id==release.group_id,
Release.posted==release.posted).first()
if existing:
# if it does, delete this one
for_deletion.append(release.id)
db.expunge(release)
else:
# we found a new name!
s_count += 1
release.name = name
release.search_name = pynab.releases.clean_release_name(name)
release.category_id = category_id
# we're done with this release
release.status = 1
db.merge(release)
else:
# nein
release.status = 0
release.unwanted = True
db.commit()
if for_deletion:
deleted = db.query(Release).filter(Release.id.in_(for_deletion)).delete(synchronize_session=False)
else:
deleted = 0
log.info('rename: successfully renamed {} of {} releases and deleted {} duplicates'.format(s_count, count, deleted))
def get_nzb(dataset=None):
user = auth()
if user:
id = request.query.guid or None
if not id:
id = request.query.id or None
# couchpotato doesn't support nzb.gzs, so decompress them
decompress = 'CouchPotato' in request.headers.get('User-Agent')
if id:
with db_session() as db:
release = db.query(Release).join(NZB).join(Category).filter(
Release.id == id).first()
if release:
release.grabs += 1
user.grabs += 1
db.merge(release)
db.merge(user)
db.commit()
if decompress:
data = release.nzb.data
response.set_header('Content-type',
'application/x-nzb')
response.set_header('X-DNZB-Name', release.search_name)
response.set_header('X-DNZB-Category',
release.category.name)
response.set_header(
'Content-Disposition',
'attachment; filename="{0}"'.format(
release.search_name.replace(' ', '_') +
'.nzb'))
return gzip.decompress(data)
else:
data = release.nzb.data
response.set_header(
'Content-type',
'application/x-nzb-compressed-gzip')
response.set_header('X-DNZB-Name', release.search_name)
response.set_header('X-DNZB-Category',
release.category.name)
response.set_header(
'Content-Disposition',
'attachment; filename="{0}"'.format(
release.search_name.replace(' ', '_') +
'.nzb.gz'))
return data
else:
return api_error(300)
else:
return api_error(200)
else:
return api_error(100)
def recategorise():
with db_session() as db:
i = 0
for release in db.query(Release).join(Group).all():
category_id = pynab.categories.determine_category(release.search_name, release.group.name)
release.category_id = category_id
db.merge(release)
i += 1
# commit every 50k rows
if i == 50000:
db.commit()
i = 0
def create(email):
"""Creates a user by email with a random API key."""
api_key = hashlib.md5(uuid.uuid4().bytes).hexdigest()
with db_session() as db:
user = User()
user.email = email
user.api_key = api_key
user.grabs = 0
db.merge(user)
return api_key
def get_stats():
"""
Retrieve relevant stats for display.
"""
with db_session() as db:
parts = db.query(Part).count()
binaries = db.query(Binary).count()
releases = db.query(Release).count()
# backlog = db.query(Release).filter(Release.passworded=='UNKNOWN').count()
# other-misc releases, ie. hashed, yet-to-be renamed or just bad releases
others = db.query(Release).filter(Release.category_id==8010).count()
return parts, binaries, releases, others
def get_stats():
"""
Retrieve relevant stats for display.
"""
with db_session() as db:
parts = db.query(Part).count()
binaries = db.query(Binary).count()
releases = db.query(Release).count()
# backlog = db.query(Release).filter(Release.passworded=='UNKNOWN').count()
# other-misc releases, ie. hashed, yet-to-be renamed or just bad releases
others = db.query(Release).filter(Release.category_id == 8010).count()
return parts, binaries, releases, others
def recategorise():
with db_session() as db:
i = 0
for release in db.query(Release).join(Group).all():
category_id = pynab.categories.determine_category(
release.search_name, release.group.name)
release.category_id = category_id
db.merge(release)
i += 1
# commit every 50k rows
if i == 50000:
db.commit()
i = 0
def process(limit=None):
"""Process releases for requests"""
with db_session() as db:
requests = {}
for group, reg in GROUP_REQUEST_REGEXES.items():
# noinspection PyComparisonWithNone
query = db.query(Release).join(Group).filter(Group.name==group).filter(Release.pre_id == None).\
filter(Release.category_id == '8010').filter("releases.name ~ '{}'".format(reg))
for release in windowed_query(
query, Release.id,
config.scan.get('binary_process_chunk_size')):
# check if it's aliased
if release.group.name in GROUP_ALIASES:
group_name = GROUP_ALIASES[release.group.name]
else:
group_name = release.group.name
if group_name not in requests:
requests[group_name] = {}
result = regex.search(reg, release.name)
if result:
requests[group_name][result.group(0)] = release
else:
log.info("requests: no release requests to process")
# per-group
for group_name, group_requests in requests.items():
# query for the requestids
if requests:
pres = db.query(Pre).filter(
Pre.requestgroup == group_name).filter(
Pre.requestid.in_(group_requests.keys())).all()
else:
log.info("requests: no pre requests found")
pres = []
# loop through and associate pres with their requests
for pre in pres:
# no longer need to check group
updated_release = group_requests.get(str(pre.requestid))
updated_release.pre_id = pre.id
db.merge(updated_release)
log.info(
"requests: found pre request id {} ({}) for {}".format(
pre.requestid, group_name, updated_release.name))
db.commit()
def scan_missing_segments(group_name):
"""Scan for previously missed segments."""
log.info('missing: checking for missed segments')
with db_session() as db:
# recheck for anything to delete
expired = db.query(Miss).filter(
Miss.attempts >= config.scan.get('miss_retry_limit')).filter(
Miss.group_name == group_name).delete()
db.commit()
if expired:
log.info('missing: deleted {} expired misses'.format(expired))
# get missing articles for this group
missing_messages = [
r for r, in db.query(Miss.message).filter(
Miss.group_name == group_name).all()
]
if missing_messages:
# mash it into ranges
missing_ranges = intspan(missing_messages).ranges()
server = Server()
server.connect()
status, parts, messages, missed = server.scan(
group_name, message_ranges=missing_ranges)
# if we got some missing parts, save them
if parts:
pynab.parts.save_all(parts)
# even if they got blacklisted, delete the ones we got from the misses
if messages:
db.query(Miss).filter(Miss.message.in_(messages)).filter(
Miss.group_name == group_name).delete(False)
db.commit()
if missed:
# clear up those we didn't get
save_missing_segments(group_name, missed)
if server.connection:
try:
server.connection.quit()
except:
pass
def fill_sizes():
with db_session() as db:
# noinspection PyComparisonWithNone
for release in db.query(Release).filter((Release.size == 0) | (
Release.size == None)).yield_per(500):
size = pynab.nzbs.get_size(release.nzb)
if size != 0:
log.debug('fill_size: [{}] - [{}] - added size: {}'.format(
release.id, release.search_name, size))
release.size = size
db.add(release)
db.commit()
def stats(dataset=None):
if not dataset:
dataset = {}
with db_session() as db:
tv_totals = db.query(func.count(Release.tvshow_id), func.count(Release.tvshow_metablack_id),
func.count(Release.id)).join(Category).filter(Category.parent_id == 5000).one()
movie_totals = db.query(func.count(Release.movie_id), func.count(Release.movie_metablack_id),
func.count(Release.id)).join(Category).filter(Category.parent_id == 2000).one()
nfo_total = db.query(func.count(Release.nfo_id), func.count(Release.nfo_metablack_id)).one()
file_total = db.query(Release.id).filter((Release.files.any()) | (Release.passworded != 'UNKNOWN')).count()
file_failed_total = db.query(func.count(Release.rar_metablack_id)).one()
release_total = db.query(Release.id).count()
dataset['totals'] = {
'TV': {
'processed': tv_totals[0],
'failed': tv_totals[1],
'total': tv_totals[2]
},
'Movies': {
'processed': movie_totals[0],
'failed': movie_totals[1],
'total': movie_totals[2]
},
'NFOs': {
'processed': nfo_total[0],
'failed': nfo_total[1],
'total': release_total
},
'File Info': {
'processed': file_total,
'failed': file_failed_total[0],
'total': release_total
}
}
dataset['categories'] = db.query(Category, func.count(Release.id)).join(Release).group_by(Category).order_by(
desc(func.count(Release.id))).all()
dataset['groups'] = db.query(Group, func.min(Release.posted), func.count(Release.id)).join(Release).group_by(Group).order_by(desc(func.count(Release.id))).all()
try:
tmpl = Template(
filename=os.path.join(root_dir, 'templates/api/stats.mako'))
return tmpl.render(**dataset)
except:
log.error('Failed to deliver page: {0}'.format(exceptions.text_error_template().render()))
return None
def fill_sizes():
with db_session() as db:
for release in db.query(Release).filter((Release.size==0)|(Release.size==None)).yield_per(500):
size = pynab.nzbs.get_size(release.nzb)
if size != 0:
log.debug('fill_size: [{}] - [{}] - added size: {}'.format(
release.id,
release.search_name,
size
))
release.size = size
db.add(release)
db.commit()
def process(limit=None):
"""Process releases for requests"""
with db_session() as db:
requests = {}
for group, reg in GROUP_REQUEST_REGEXES.items():
# noinspection PyComparisonWithNone
query = db.query(Release).join(Group).filter(Group.name==group).filter(Release.pre_id == None).\
filter(Release.category_id == '8010').filter("releases.name ~ '{}'".format(reg))
for release in windowed_query(query, Release.id, config.scan.get('binary_process_chunk_size')):
# check if it's aliased
if release.group.name in GROUP_ALIASES:
group_name = GROUP_ALIASES[release.group.name]
else:
group_name = release.group.name
if group_name not in requests:
requests[group_name] = {}
result = regex.search(reg, release.name)
if result:
requests[group_name][result.group(0)] = release
else:
log.info("requests: no release requests to process")
# per-group
for group_name, group_requests in requests.items():
# query for the requestids
if requests:
pres = db.query(Pre).filter(Pre.requestgroup==group_name).filter(Pre.requestid.in_(group_requests.keys())).all()
else:
log.info("requests: no pre requests found")
pres = []
# loop through and associate pres with their requests
for pre in pres:
# no longer need to check group
updated_release = group_requests.get(str(pre.requestid))
updated_release.pre_id = pre.id
updated_release.name = pre.name
updated_release.search_name = pre.searchname
db.merge(updated_release)
log.info("requests: found pre request id {} ({}) for {}".format(pre.requestid, group_name,
updated_release.name))
db.commit()
def upgrade():
with db_session() as db:
for release in db.query(Release). \
filter(Release.added >= dateutil.parser.parse('2014/11/12 16:37 GMT+8')). \
filter(Release.added <= dateutil.parser.parse('2014/11/13 16:20 GMT+8')). \
order_by(Release.added). \
all():
nzb = gzip.decompress(release.nzb.data).decode('utf-8')
if '<?xml' not in nzb:
nzb = ('<?xml version="1.0" encoding="UTF-8"?>\n'
'<!DOCTYPE nzb PUBLIC "-//newzBin//DTD NZB 1.1//EN" "http://www.newzbin.com/DTD/nzb/nzb-1.1.dtd">\n'
'<nzb>\n'
'<head><meta type="category">{}</meta><meta type="name">{}</meta></head>\n'.format(release.category.name, escape(release.search_name))) + \
nzb
release.nzb.data = gzip.compress(nzb.encode('utf-8'))
db.commit()
def upgrade():
with db_session() as db:
for release in db.query(Release). \
filter(Release.added >= dateutil.parser.parse('2014/11/12 16:37 GMT+8')). \
filter(Release.added <= dateutil.parser.parse('2014/11/13 16:20 GMT+8')). \
order_by(Release.added). \
all():
nzb = gzip.decompress(release.nzb.data).decode('utf-8')
if '<?xml' not in nzb:
nzb = ('<?xml version="1.0" encoding="UTF-8"?>\n'
'<!DOCTYPE nzb PUBLIC "-//newzBin//DTD NZB 1.1//EN" "http://www.newzbin.com/DTD/nzb/nzb-1.1.dtd">\n'
'<nzb>\n'
'<head><meta type="category">{}</meta><meta type="name">{}</meta></head>\n'.format(release.category.name, escape(release.search_name))) + \
nzb
release.nzb.data = gzip.compress(nzb.encode('utf-8'))
db.commit()
def nzedbirc(unformattedPre):
formattedPre = parseNzedbirc(unformattedPre)
with db_session() as db:
p = db.query(Pre).filter(Pre.name == formattedPre['name']).first()
if not p:
p = Pre(**formattedPre)
else:
for k, v in formattedPre.items():
setattr(p, k, v)
try:
db.add(p)
log.info("pre: Inserted/Updated - {}".format(formattedPre["name"]))
except Exception as e:
log.debug("pre: Error - {}".format(e))
def save_missing_segments(group_name, missing_segments):
"""Handles any missing segments by mashing them into ranges
and saving them to the db for later checking."""
with db_session() as db:
# we don't want to get the whole db's worth of segments
# just get the ones in the range we need
first, last = min(missing_segments), max(missing_segments)
# get previously-missed parts
previous_misses = [
r for r, in db.query(Miss.message).filter(
Miss.message >= first).filter(Miss.message <= last).filter(
Miss.group_name == group_name).all()
]
# find any messages we're trying to get again
repeats = list(set(previous_misses) & set(missing_segments))
# update the repeats to include the new attempt
if repeats:
stmt = Miss.__table__.update().where(
Miss.__table__.c.message == bindparam('m')).values(
attempts=Miss.__table__.c.attempts + 1)
db.execute(stmt, [{'m': m} for m in repeats if m])
# subtract the repeats from our new list
new_misses = list(set(missing_segments) - set(repeats))
# batch-insert the missing messages
if new_misses:
db.execute(Miss.__table__.insert(), [{
'message': m,
'group_name': group_name,
'attempts': 1
} for m in new_misses])
# delete anything that's been attempted enough
expired = db.query(Miss).filter(
Miss.attempts >= config.scan.get('miss_retry_limit')).filter(
Miss.group_name == group_name).delete()
db.commit()
log.info(
'missing: saved {} misses and deleted {} expired misses'.format(
len(new_misses), expired))
