def get_case_hits(size, order, f, cy):
hits = []
cquery = ""
if cy == "":
order = order.split(":")
retval = "RETURN DISTINCT Project.name,Study.subtype,Sample.id,Project.subtype,Sample.fma_body_site ORDER BY %s %s SKIP %s LIMIT %s"
cquery = "%s %s" % (full_traversal, retval)
if f != 0:
f = f - 1
cquery = cquery % (order[0], order[1].upper(), f, size)
elif '"op"' in cy:
cquery = build_cypher(match, cy, order, f, size, "cases")
else:
cquery = build_adv_cypher(match, cy, order, f, size, "cases")
res = graph.data(cquery)
for x in range(0, len(res)):
cur = CaseHits(project=Project(
projectId=res[x]['Project.subtype'],
primarySite=res[x]['Sample.fma_body_site'],
name=res[x]['Project.name'],
studyName=res[x]['Study.name'],
studyFullName=res[x]['Study.full_name']),
caseId=res[x]['Sample.id'])
hits.append(cur)
return hits
def count_props_and_files(node, prop, cy):
cquery, with_distinct = ("" for i in range(2))
if cy == "":
retval = ("WITH (COUNT(DISTINCT(Sample))) as ccounts, "
"COUNT(DISTINCT(File)) AS dcounts, %s.%s AS prop, "
"collect(DISTINCT File) AS f UNWIND f AS fs "
"RETURN prop,ccounts,dcounts,SUM(toInt(fs.size)) as tot")
mod_retval = retval % (node, prop)
cquery = "%s %s" % (full_traversal, mod_retval)
else:
if node == 'Study' and prop == 'name':
prop = 'sname'
prop_detailed = "%s_detailed" % (prop)
if "op" in cy:
cquery = build_cypher(match, cy, "null", "null", "null",
prop_detailed)
else:
cquery = build_adv_cypher(match, cy, "null", "null", "null",
prop_detailed)
return graph.data(cquery)
def get_total_file_size(cy):
cquery = ""
if cy == "":
cquery = "MATCH (File) WHERE NOT File.node_type=~'.*prep' RETURN SUM(toInt(File.size)) AS tot"
elif '"op"' in cy:
cquery = build_cypher(match, cy, "null", "null", "null", "size")
else:
cquery = build_adv_cypher(match, cy, "null", "null", "null", "size")
res = graph.data(cquery)
return res[0]['tot']
def get_pagination(cy, size, f, c_or_f):
cquery = ""
if cy == "":
if c_or_f == 'c':
cquery = "MATCH (n:Case {node_type:'sample'}) RETURN count(n) AS tot"
else:
cquery = "MATCH (n:File) WHERE NOT n.node_type=~'.*prep' RETURN count(n) AS tot"
res = graph.data(cquery)
calcs = pagination_calcs(res[0]['tot'], f, size, c_or_f)
return Pagination(count=calcs[2],
sort=calcs[4],
fromNum=f,
page=calcs[1],
total=calcs[3],
pages=calcs[0],
size=size)
else:
if '"op"' in cy:
if c_or_f == 'c':
cquery = build_cypher(match, cy, "null", "null", "null",
"c_pagination")
else:
cquery = build_cypher(match, cy, "null", "null", "null",
"f_pagination")
else:
if c_or_f == 'c':
cquery = build_adv_cypher(match, cy, "null", "null", "null",
"c_pagination")
else:
cquery = build_adv_cypher(match, cy, "null", "null", "null",
"f_pagination")
res = graph.data(cquery)
calcs = pagination_calcs(res[0]['tot'], f, size, c_or_f)
return Pagination(count=calcs[2],
sort=calcs[4],
fromNum=f,
page=calcs[1],
total=calcs[3],
pages=calcs[0],
size=size)
def count_props(node, prop, cy):
cquery = ""
if cy == "":
if node in count_props_dict:
cquery = count_props_dict[node] % (prop)
elif node == 'sample':
cquery = "MATCH (n:Case{node_type:'sample'})<-[:PREPARED_FROM]-(pf)<-[:SHORTCUT]-(File) WITH DISTINCT n RETURN n.%s AS prop, COUNT(n.%s) as counts" % (
prop, prop)
else:
cquery = "Match (n:File) WHERE NOT n.node_type=~'.*prep' RETURN n.%s as prop, count(n.%s) as counts" % (
prop, prop)
else:
cquery = build_cypher(match, cy, "null", "null", "null", prop)
return graph.data(cquery)
def get_file_hits(size, order, f, cy):
hits = []
cquery = ""
if cy == "":
order = order.split(":")
retval = "RETURN DISTINCT Project,File,Sample.id ORDER BY %s %s SKIP %s LIMIT %s"
cquery = "%s %s" % (full_traversal, retval)
if f != 0:
f = f - 1
cquery = cquery % (order[0], order[1].upper(), f, size)
elif '"op"' in cy:
cquery = build_cypher(match, cy, order, f, size, "files")
else:
cquery = build_adv_cypher(match, cy, order, f, size, "files")
res = graph.data(cquery)
for x in range(0, len(res)):
case_hits = [] # reinit each iteration
cur_case = CaseHits(project=Project(
projectId=res[x]['Project']['subtype'],
name=res[x]['Project']['name']),
caseId=res[x]['Sample.id'])
case_hits.append(cur_case)
furl = extract_url(res[x]['File']) # File name is our URL
if '.hmpdacc' in furl: # HMP endpoint
furl = re.search(r'/data/(.*)', furl).group(1)
elif '.ihmpdcc':
furl = re.search(r'.org/(.*)', furl).group(1)
cur_file = FileHits(dataType=res[x]['File']['subtype'],
fileName=furl,
dataFormat=res[x]['File']['format'],
submitterId="null",
access="open",
state="submitted",
fileId=res[x]['File']['id'],
dataCategory=res[x]['File']['node_type'],
experimentalStrategy=res[x]['File']['subtype'],
fileSize=res[x]['File']['size'],
cases=case_hits)
hits.append(cur_file)
return hits