def output_query(self, out_file, cntl, topk, query_time_ms):
out_file.write("\n")
out_file.write("QUERY\t" + self.name + "\t" + str(query_time_ms) +
"\n")
if len(self.sorted_docs) == 0:
# no results? nothing can be output
return
"""
for mquery in self.mqueries:
out_file.write("E\t" + mquery.expression + "\n")
if self.tquery:
for keyword in self.tquery.keywords:
out_file.write("P\t" + keyword + "\n")
"""
d = MathDocument(cntl)
min_score = self.sorted_docs[len(self.sorted_docs) - 1].final_score
if len(self.sorted_docs) < topk:
print("Warning: Query produced less than " + str(topk) +
" documents. Results will be repeated",
flush=True)
# force output topk results
for idx in range(topk):
doc = self.sorted_docs[idx % len(self.sorted_docs)]
positions = self.get_math_pos_with_score(doc)
try:
exprids = list(
map(
lambda pos:
(d.find_mathml_id(doc.doc_id, pos[0]), pos[1]),
positions))
except (IOError): # cannot read ids
exprids = positions
#out_file.write("R\t" + str(doc.doc_name) + "\t" + str(doc.final_score) + "\t(at: " + str(exprids) + str(self.get_text_pos(doc))+ ")\n")
row_elements = [str(idx + 1)]
if idx < len(self.sorted_docs):
# use original score
doc_score = doc.final_score
else:
doc_score = min_score
if isinstance(doc_score, list):
row_elements.append(str(doc_score[0]))
else:
row_elements.append(str(doc_score))
row_elements.append(doc.doc_name)
# add formulas *M*
for exprid, mscore in exprids:
row_elements += ["*M*", str(exprid), str(mscore[0])]
# add Keywords *W*
if self.tquery:
for keyword in self.tquery.keywords:
row_elements += ["*W*", keyword, str(doc.tscore[1])]
out_file.write("\t".join(row_elements) + "\n")
