hans/Nominatim
1
0
Fork 1
mirror of https://github.com/osm-search/Nominatim.git synced 2026-03-13 06:14:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

introduce name analyzer

Browse Source 
The name analyzer is the actual work horse of the tokenizer. It
is instantiated on a thread-base and provides all functions for
analysing names and queries.
This commit is contained in:
Sarah Hoffmann
2021-04-24 22:35:46 +02:00
parent e1c5673ac3
commit fa2bc60468
5 changed files with 113 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
nominatim/indexer/indexer.py
View File

@@ -124,8 +124,9 @@ class Indexer:
LOG.warning("Starting indexing boundaries using %s threads",
self.num_threads)
for rank in range(max(minrank, 4), min(maxrank, 26)):
self._index(runners.BoundaryRunner(rank))
with self.tokenizer.name_analyzer() as analyzer:
for rank in range(max(minrank, 4), min(maxrank, 26)):
self._index(runners.BoundaryRunner(rank, analyzer))
def index_by_rank(self, minrank, maxrank):
""" Index all entries of placex in the given rank range (inclusive)
@@ -138,15 +139,16 @@ class Indexer:
LOG.warning("Starting indexing rank (%i to %i) using %i threads",
minrank, maxrank, self.num_threads)
for rank in range(max(1, minrank), maxrank):
self._index(runners.RankRunner(rank))
with self.tokenizer.name_analyzer() as analyzer:
for rank in range(max(1, minrank), maxrank):
self._index(runners.RankRunner(rank, analyzer))
if maxrank == 30:
self._index(runners.RankRunner(0))
self._index(runners.InterpolationRunner(), 20)
self._index(runners.RankRunner(30), 20)
else:
self._index(runners.RankRunner(maxrank))
if maxrank == 30:
self._index(runners.RankRunner(0, analyzer))
self._index(runners.InterpolationRunner(), 20)
self._index(runners.RankRunner(30, analyzer), 20)
else:
self._index(runners.RankRunner(maxrank, analyzer))
def index_postcodes(self):