def __init__(self, user_loc_string, debug=False):
analyzer = StopAnalyzer()
fields = ['name', 'alternate_names', 'state']
directory = FSDirectory.getDirectory("index")
self.DEBUG = debug
self.searcher = IndexSearcher(directory)
self.parser = MultiFieldQueryParser(fields, analyzer)
self.user_location = self.doSearch(user_loc_string)
def search_lucene_index(search_params, index_dir, index_metadata,
records_per_page):
"""
Uses the query term provided to search the disease ontology lucene index
"""
results = []
index_dir = SimpleFSDirectory(File(index_dir))
analyzer = build_perfield_analyzer(index_metadata)
searcher = IndexSearcher(index_dir)
index_fields = index_metadata.keys()
# Since we are paging results we want to grab what page we are on
page = (int(search_params.get('page', 1))) - 1
# Doing something pretty hacky here since we are trying to move from 0-based to 1
# based indexing to match our pagingation display
offset = int(page) * records_per_page
# If we are executing an advanced search we will be building a BooleanQuery
# in parts as opposed to the one MultiFieldQueryParser when doing a basic
# serach
query = None
if search_params.get('adv_search') == "True":
query = build_advanced_search_query(search_params,
search_params.get('operator'),
analyzer)
else:
parser = MultiFieldQueryParser(Version.LUCENE_CURRENT, index_fields,
analyzer)
query = MultiFieldQueryParser.parse(
parser, process_query_param((search_params.get('q'))))
# Parse through our hits
hits = searcher.search(query, 10000)
total_hits = hits.totalHits
count = min(hits.totalHits - offset, records_per_page)
for i in xrange(0, count):
score_doc = hits.scoreDocs[offset + i]
doc = searcher.doc(score_doc.doc)
term_id = doc.get('term id')
name = doc.get('name')
explain = searcher.explain(query, score_doc.doc)
match_fields = get_field_matches(explain.toString(), index_fields)
results.append((term_id, name, list(match_fields)))
searcher.close()
return (results, total_hits)
