def codsearch_composition(composition, tolerance):
from diffpy.pdfgetx.functs import composition_analysis
from elasticsearch import Elasticsearch
from elasticsearch.helpers import scan
es = Elasticsearch(ELASTICHOST)
smbls, counts = composition_analysis(composition)
if tolerance == 0:
mustterms = [{
'term': {
("composition." + s): c
}
} for s, c in zip(smbls, counts)]
q = {"bool": {"must": mustterms}}
else:
rangeterms = [{
'range': {
("composition." + s): {
"gte": c - tolerance,
"lte": c + tolerance,
}
}
} for s, c in zip(smbls, counts)]
q = {"bool": {"must": rangeterms}}
gscan = scan(es,
query={'query': q},
index='cod',
doc_type='cif',
_source=False)
for e in gscan:
codid = normcodid(e['_id'])
yield codid
pass
def normalized_formula(formula):
from diffpy.pdfgetx.functs import composition_analysis
smbls, counts = composition_analysis(formula)
totalcount = sum(counts)
rv = dict.fromkeys(smbls, 0.0)
for s, c in zip(smbls, counts):
rv[s] += c
for s in rv:
rv[s] /= totalcount
return rv
def cifsearch(q=None, composition=None, tol=None, fields=None, **kwargs):
"""
Execute search for CIF structures using Lucene query string syntax.
Parameters
----------
q : str, optional
The string search query in Lucene syntax.
query : dict, optional, keyword-only
The search definition using the Query DSL.
composition : str, optional
Normalized chemical stoichiometry to be matched.
tol : float, optional
Maximum allowed difference from stoichiometry.
fields : list or str, optional
Name of CIF fields to be returned.
kwargs : misc, optional
Extra arguments passed to the `Elasticsearch.search` function.
Returns
-------
databroker.Results
Iterable object encapsulating the matching databroker Headers.
"""
from diffpy.pdfgetx.functs import composition_analysis
from elasticsearch import Elasticsearch
es = Elasticsearch(ELASTICHOST)
kw = dict(q=q, index='cod')
if 'query' in kwargs:
kw['body'] = kwargs.pop('query')
if composition:
smbls, counts = composition_analysis(composition)
if not tol:
mustterms = [{
'term': {
("composition." + s): c
}
} for s, c in zip(smbls, counts)]
cq = {"bool": {"must": mustterms}}
else:
rangeterms = [{
'range': {
("composition." + s): {
"gte": c - tol,
"lte": c + tol,
}
}
} for s, c in zip(smbls, counts)]
cq = {"bool": {"must": rangeterms}}
kw['body'] = {'query': cq}
kw.update(**kwargs)
if isinstance(fields, str):
fields = fields.replace(',', ' ').split()
if fields:
kw['_source'] = fields
res = es.search(**kw)
rv = res
if fields:
rv = [
tuple(hit['_source'].get(n) for n in fields)
for hit in res['hits']['hits']
]
if len(fields) == 1:
rv = [x[0] for x in rv]
return rv