def _get_table(self, table):
logger.info(self.organism)
s = get_session(self.baseurl, self.login)
query_payload = {
"answerSpec": {
"questionName": "GeneQuestions.GenesByTaxonGene",
"parameters": {"organism": self.organism},
"viewFilters": [],
"filters": []
},
"formatting": {
"formatConfig": {
"tables": [table],
"includeHeader": True,
"attachmentType": "plain"
},
"format": "tableTabular"
}
}
r = s.post(self.baseurl + "/a/service/answer/report",
data=json.dumps(query_payload),
headers={'Content-Type': 'application/json'})
r.raise_for_status()
return self._parse_table(r.text, table)
def _get_json(self, url, params=None):
logger.info(' retrieving %s' % url)
s = get_session(self.baseurl, self.login)
res = s.get(url, verify=True, params=params)
if "autologin" in res.text or 'Login</title>' in res.text:
raise RuntimeError(
"Login failed -- please check user credentials.")
if (res.ok):
return res.json()
else:
res.raise_for_status()
def _get_json(self, url, params=None):
logger.info(' retrieving %s' % url)
s = get_session(self.baseurl, self.login)
# workaround for problematic urlencoding if organism contains '#'
params = "?organism=" + params['organism'] + \
"&o-fields=" + params['o-fields']
res = s.get(url + params, verify=True)
if "autologin" in res.text or 'Login</title>' in res.text:
raise RuntimeError("Login failed -- please check user credentials.")
if(res.ok):
return res.json()
else:
res.raise_for_status()
def __init__(self, baseurl, login=None):
self.baseurl = baseurl
self.login = parse_login(login)
self.fields = [
"organism", "ncbi_tax_id", "is_reference_strain",
"is_annotated_genome", "URLGenomeFasta", "URLgff"
]
url = (
'{0}/webservices/OrganismQuestions/GeneMetrics.json?o-fields={1}'
).format(self.baseurl, ','.join(self.fields))
s = get_session(self.baseurl, self.login)
res = s.get(url, verify=True)
self.orgs = collections.deque()
if (res.ok):
j = res.json()
else:
res.raise_for_status()
for v in j['response']['recordset']['records']:
item = {}
for f in v['fields']:
item[f['name']] = f['value']
self.orgs.append(item)
def __init__(self, baseurl, organism, login=None):
self.login = parse_login(login)
self.baseurl = baseurl
url = ('{0}/webservices/GenomicSequenceQuestions/' +
'SequencesByTaxon.json' +
'?organism={1}&o-fields=primary_key').format(
baseurl, organism.replace('#', "%23"))
res = self._get_json(url)
seqids = []
for v in res['response']['recordset']['records']:
seqids.append(v['id'])
logger.info(' needing to retrieve %s seqs for organism %s' %
(len(seqids), organism))
parsed_url = urlparse(baseurl)
self.baseurl = "%s://%s" % (parsed_url.scheme, parsed_url.netloc)
project_id = parsed_url.netloc.split('.')[0].capitalize()
project_id = re.sub('db$', 'DB', project_id)
payload = {'project_id': project_id, 'ids': '\n'.join(seqids)}
url = ('{0}/cgi-bin/contigSrt').format(self.baseurl)
s = get_session(self.baseurl, self.login)
res = s.post(url, data=payload)
if not res.ok:
res.raise_for_status()
self.out = res.text