def dump(files: iter, output_dir: str, unique: bool, update_all: bool):
"""
Parse MEDLINE XML files into tabular flat-files for each DB table.
In addition, a ``delete.txt`` file is generated, containing the PMIDs
that should first be deleted from the DB before copying the dump.
:param files: a list of XML files to parse (optionally, gzipped)
:param output_dir: path to the output directory for the dump
:param unique: if ``True`` only VersionId == "1" records are dumped
:param update_all: if ``True`` the PMIDs of all parsed records are
added to the list of PMIDs for deletion
"""
out_stream = {
Citation.__tablename__:
open(join(output_dir, "citations.tab"), "wt"),
Abstract.__tablename__:
open(join(output_dir, "abstracts.tab"), "wt"),
Section.__tablename__:
open(join(output_dir, "sections.tab"), "wt"),
Descriptor.__tablename__:
open(join(output_dir, "descriptors.tab"), "wt"),
Qualifier.__tablename__:
open(join(output_dir, "qualifiers.tab"), "wt"),
Author.__tablename__:
open(join(output_dir, "authors.tab"), "wt"),
Identifier.__tablename__:
open(join(output_dir, "identifiers.tab"), "wt"),
Database.__tablename__:
open(join(output_dir, "databases.tab"), "wt"),
PublicationType.__tablename__:
open(join(output_dir, "publication_types.tab"), "wt"),
Chemical.__tablename__:
open(join(output_dir, "chemicals.tab"), "wt"),
Keyword.__tablename__:
open(join(output_dir, "keywords.tab"), "wt"),
'delete':
open(join(output_dir, "delete.txt"), "wt"),
}
count = 0
parser = MedlineXMLParser(unique)
for f in files:
logger.info('dumping %s', f)
if f.lower().endswith('.gz'):
in_stream = gunzip(f, 'rb')
else:
in_stream = open(f)
count += _dump(in_stream, out_stream, parser, update_all)
for stream in out_stream.values():
if stream.tell() == 0:
stream.close()
remove(join(output_dir, stream.name))
else:
stream.close()
logger.info("parsed %i records", count)
def dump(files: iter, output_dir: str, unique: bool, update_all: bool):
"""
Parse MEDLINE XML files into tabular flat-files for each DB table.
In addition, a ``delete.txt`` file is generated, containing the PMIDs
that should first be deleted from the DB before copying the dump.
:param files: a list of XML files to parse (optionally, gzipped)
:param output_dir: path to the output directory for the dump
:param unique: if ``True`` only VersionId == "1" records are dumped
:param update_all: if ``True`` the PMIDs of all parsed records are
added to the list of PMIDs for deletion
"""
out_stream = {
Citation.__tablename__:
open(join(output_dir, "citations.tab"), "wt"),
Abstract.__tablename__:
open(join(output_dir, "abstracts.tab"), "wt"),
Section.__tablename__:
open(join(output_dir, "sections.tab"), "wt"),
Descriptor.__tablename__:
open(join(output_dir, "descriptors.tab"), "wt"),
Qualifier.__tablename__:
open(join(output_dir, "qualifiers.tab"), "wt"),
Author.__tablename__:
open(join(output_dir, "authors.tab"), "wt"),
Identifier.__tablename__:
open(join(output_dir, "identifiers.tab"), "wt"),
Database.__tablename__:
open(join(output_dir, "databases.tab"), "wt"),
PublicationType.__tablename__:
open(join(output_dir, "publication_types.tab"), "wt"),
Chemical.__tablename__:
open(join(output_dir, "chemicals.tab"), "wt"),
Keyword.__tablename__:
open(join(output_dir, "keywords.tab"), "wt"),
'delete': open(join(output_dir, "delete.txt"), "wt"),
}
count = 0
parser = MedlineXMLParser(unique)
for f in files:
logger.info('dumping %s', f)
if f.lower().endswith('.gz'):
in_stream = gunzip(f, 'rb')
else:
in_stream = open(f)
count += _dump(in_stream, out_stream, parser, update_all)
for stream in out_stream.values():
if stream.tell() == 0:
stream.close()
remove(join(output_dir, stream.name))
else:
stream.close()
logger.info("parsed %i records", count)
def _openFile(name):
if name.lower().endswith('.gz'):
# use wrapper to support pre-3.3
return gunzip(name, 'rb')
else:
return open(name)
def run_module():
# seed the result dict in the object
result = dict(
changed=True,
)
# define available arguments/parameters a user can pass to the module
module = AnsibleModule(
argument_spec=dict(
state=dict(
type='str',
default='enabled',
required=False,
choices=[
'present',
'absent',
'enabled',
'disabled'
]
),
name=dict(
type='str',
required=True
),
kodi_user=dict(
type='str',
required=False,
default='kodi'
),
kodi_home=dict(
type='str',
required=False,
default=''
),
kodi_release=dict(
type='str',
required=True
),
),
supports_check_mode=True
)
# We only work with releases that have the database version 27
# See: https://kodi.wiki/view/Databases#Database_Versions
if not SUPPORTED_RELEASES.__contains__(module.params['kodi_release']):
module.fail_json(
msg='Unsupported kodi release. Supported: %s' % (SUPPORTED_RELEASES),
**result
)
# Seed the default kodi home if not supplied
if module.params['kodi_home']:
kodi_home = module.params['kodi_home']
else:
kodi_home = "%s/.kodi" % (getpwnam(module.params['kodi_user']).pw_dir)
# Seed the default kodi repository
kodi_repo = 'http://mirrors.kodi.tv/addons/%s/addons.xml.gz' % (module.params['kodi_release'])
# Shortcut for the actual repository, computed removing 'addons.xml(.gz)'
repo_base = '/'.join(kodi_repo.split('/')[0:-1])
if module.params['state'] == 'absent':
if module.check_mode:
# We would only remove the addon if it's in the directory,
# or if it has traces in the db
result['changed'] = (
exists("%s/addons/%s" % (kodi_home, module.params['name']))
or is_in_db(module.params['name'], kodi_home)
)
module.exit_json(**result)
result['changed'] = remove_addon(kodi_home, module.params['name'])
module.exit_json(**result)
elif module.params['state'] == 'disabled':
enabled = False
elif module.params['state'] in ['present', 'enabled']:
enabled = True
# If the directory already exists and the enabled status is as desired, then we have nothing to do
if exists("%s/addons/%s" % (kodi_home, module.params['name'])):
if is_enabled(module.params['name'], kodi_home) == enabled:
result['changed'] = False
module.exit_json(**result)
addons_xml = mktemp()
# Download the repository definition, uncompressing it if necessary
# This check is left from an attempt to support multiple repositories which might be retried, no need to delete it.
if kodi_repo.split('.')[-1] == 'gz':
addons_xml_gz = mktemp()
download(kodi_repo, addons_xml_gz.name)
addons_xml.seek(0)
addons_xml.truncate()
addons_xml.write(gunzip(addons_xml_gz.name).read())
addons_xml.flush()
else:
download(kodi_repo, addons_xml.name)
# Load the actual xml
addons = etree.parse(addons_xml.name)
install_addon(repo_base, addons, module.params['name'], module.params['kodi_user'], kodi_home, module.params['kodi_release'], enabled)
result['changed'] = True
module.exit_json(**result)
def _openFile(name):
if name.lower().endswith('.gz'):
# use wrapper to support pre-3.3
return gunzip(name, 'rb')
else:
return open(name)
#!/usr/bin/env python3
from collections import Counter
import csv
from gzip import open as gunzip
import json
import string
import re
separators = '[{}]'.format(string.punctuation + string.whitespace)
queries = set()
for i in range(10):
filename = 'data/aol/aol-{}.txt.gz'.format(str(i + 1).zfill(2))
with gunzip(filename, 'rt') as file:
reader = csv.reader(file, delimiter='\t')
_header = next(reader)
queries |= set(line[1] for line in reader)
data = Counter(item for query in queries
for item in re.split(separators, query))
data = dict(data)
# filter out stopwords
with open('data/stopwords.txt') as f:
for word in f:
word = word.strip()
if word in data:
del data[word]
type=float,
metavar='min. proportion alt alleles',
required=True,
help='minimum proportion of alternative alleles to allow')
parser.add_argument(
'-ly',
type=str,
metavar='lyrata_only?',
required=False,
default='false',
help='do you want to include lyrata only (true) or not (false)?')
args = parser.parse_args()
if args.gz == 'true' and args.v[-3:] == '.gz':
gzip.gunzip(args.v)
lookup_table_file = open(
args.v + args.o + "repolarized.lookupKey.minAlleles_" + str(args.mi) +
".txt", 'w')
lookup_table_file = open(
args.o + "repolarized.lookupKey.minInd_" + str(args.mi) + ".txt", 'w')
if args.ly == 'true':
args.mi = 2 # args.mi must = 2, since there are only two lyrata samples
args.mp = 1.0
count = 0
type_counts = [0, 0, 0, 0]
count_file = open(args.o + "counts.txt", 'w')
