def get_response(self,
path="",
q=None,
format=None,
http_method='get',
**params):
url = self.url_prefix + path
if q is not None:
url += "/" + q
if format is not None:
url = url + "." + format
if http_method == 'get':
request = requests.get(url,
params=params,
headers={
'User-Agent':
get_user_agent(modules=[requests],
caller_name=__name__)
})
elif http_method == 'post':
request = requests.post(url,
data=params,
headers={
'User-Agent':
get_user_agent(modules=[requests],
caller_name=__name__)
})
else:
raise RequestException
return request
def _ensure_file(self, file):
logging.info("Ensure file: {}".format(file))
if isinstance(file,str):
# TODO Let's fix this if/elseif chain.
if file.startswith("ftp"):
f = tempfile.NamedTemporaryFile()
fn = f.name
cmd = ['wget',file,'-O',fn]
subprocess.run(cmd, check=True)
return open(fn,"r")
elif file.startswith("http"):
url = file
with closing(requests.get(url, stream=False, headers={'User-Agent': get_user_agent(modules=[requests], caller_name=__name__)})) as resp:
logging.info("URL: {} STATUS: {} ".format(url, resp.status_code))
ok = resp.status_code == 200
if ok:
logging.debug("HEADER: {}".format(resp.headers))
if file.endswith(".gz"):
return io.StringIO(str(gzip.decompress(resp.content),'utf-8'))
else:
out = io.StringIO(resp.content)
return out
else:
return None
else:
logging.info("Testing suffix of {}".format(file))
if file.endswith(".gz"):
return gzip.open(file, "rt")
else:
return open(file, "r")
else:
return file
def download_source_gafs(group_metadata, target_dir, exclusions=[]):
gaf_urls = { data["dataset"]: data["source"] for data in group_metadata["datasets"] if data["type"] == "gaf" and data["dataset"] not in exclusions }
click.echo("Found {}".format(", ".join(gaf_urls.keys())))
downloaded_paths = {}
for dataset, gaf_url in gaf_urls.items():
path = os.path.join(target_dir, "groups", group_metadata["id"], "{}-src.gaf.gz".format(dataset))
os.makedirs(os.path.split(path)[0], exist_ok=True)
click.echo("Downloading source gaf to {}".format(path))
if urllib.parse.urlparse(gaf_url)[0] in ["ftp", "file"]:
urllib.request.urlretrieve(gaf_url, path)
else:
response = requests.get(gaf_url, stream=True, headers={'User-Agent': get_user_agent(modules=[requests], caller_name=__name__)})
content_length = int(response.headers.get("Content-Length", None))
with open(path, "wb") as downloaded:
with click.progressbar(iterable=response.iter_content(chunk_size=512 * 1024), length=content_length, show_percent=True) as chunks:
for chunk in chunks:
if chunk:
downloaded.write(chunk)
downloaded_paths[dataset] = path
return downloaded_paths
def get_response(self, path="", q=None, format=None, **params):
url = self.url_prefix + path
if q is not None:
url += "/" +q
if format is not None:
url = url + "." + format
r = requests.get(url, params=params, headers={'User-Agent': get_user_agent(name=NAME, version=VERSION, modules=[requests], caller_name=__name__)})
return r
def create_from_remote_file(self, group, snapshot=True, **args):
"""
Creates from remote GAF
"""
import requests
url = "http://snapshot.geneontology.org/annotations/{}.gaf.gz".format(group)
r = requests.get(url, stream=True, headers={'User-Agent': get_user_agent(modules=[requests], caller_name=__name__)})
p = GafParser()
results = p.skim(r.raw)
return self.create_from_tuples(results, **args)
def get_ecomap_str(url):
logger.info("Fetching ecomap from {}".format(url))
with closing(
requests.get(url,
stream=False,
headers={
'User-Agent':
get_user_agent(modules=[requests],
caller_name=__name__)
})) as resp:
# TODO: redirects
if resp.status_code == 200:
return resp.text
def _get_response(self, path="", q=None, format=None, **params):
url = self.url
if not url.endswith("/"):
url += "/"
url += path
if q is not None:
url += "/" + q
if format is not None:
url = url + "." + format
r = requests.get(url,
params=params,
headers={
'User-Agent':
get_user_agent(modules=[requests],
caller_name=__name__)
})
return r
def search_associations_go(
subject_category=None,
object_category=None,
relation=None,
subject=None,
**kwargs):
"""
Perform association search using Monarch golr
"""
go_golr_url = "http://golr.geneontology.org/solr/"
go_solr = pysolr.Solr(go_golr_url, timeout=5)
go_solr.get_session().headers['User-Agent'] = get_user_agent(caller_name=__name__)
return search_associations(subject_category,
object_category,
relation,
subject,
solr=go_solr,
field_mapping=goassoc_fieldmap(),
**kwargs)
- mapping to and from CURIEs used in Monarch
- providing simple to call methods for common queries
TODO:
Return objects following the biolink/OBAN association model
"""
import SPARQLWrapper, logging
from biolink import NAME, VERSION
from ontobio.util.user_agent import get_user_agent
USER_AGENT = get_user_agent(name=NAME,
version=VERSION,
modules=[SPARQLWrapper],
caller_name=__name__)
sparql = SPARQLWrapper.SPARQLWrapper("http://query.wikidata.org/sparql",
agent=USER_AGENT)
class PrefixMap:
"""
Common SPARQL prefixes used by wikidata.
Note we use the "trick" whereby an entire property URI can be encoded as a prefix.
"""
def prefixes(self):
return [
attr for attr in dir(self)
if not callable(getattr(self, attr)) and not attr.startswith("__")
import git from ontobio.util.user_agent import get_user_agent NAME = "biolink-api" VERSION = git.Repo(search_parent_directories=True).head.object.hexsha[:7] USER_AGENT = get_user_agent(name=NAME, version=VERSION)
