]> git.openstreetmap.org Git - nominatim.git/blobdiff - nominatim/indexer/indexer.py
indexer: ANALYSE must be run outside transactions
[nominatim.git] / nominatim / indexer / indexer.py
index 520464565bebc96f04769b336c04604a531bab8f..06c05e1d5a49e32a7e2c44290339d19c1344b548 100644 (file)
@@ -5,8 +5,10 @@ Main work horse for indexing (computing addresses) the database.
 import logging
 import select
 
+import psycopg2
+
 from .progress import ProgressLogger
-from db.async_connection import DBConnection, make_connection
+from ..db.async_connection import DBConnection
 
 LOG = logging.getLogger()
 
@@ -59,8 +61,8 @@ class InterpolationRunner:
     @staticmethod
     def sql_index_place(ids):
         return """UPDATE location_property_osmline
-                  SET indexed_status = 0 WHERE place_id IN ({})"""\
-               .format(','.join((str(i) for i in ids)))
+                  SET indexed_status = 0 WHERE place_id IN ({})
+               """.format(','.join((str(i) for i in ids)))
 
 class BoundaryRunner:
     """ Returns SQL commands for indexing the administrative boundaries
@@ -77,51 +79,172 @@ class BoundaryRunner:
         return """SELECT count(*) FROM placex
                   WHERE indexed_status > 0
                     AND rank_search = {}
-                    AND class = 'boundary' and type = 'administrative'""".format(self.rank)
+                    AND class = 'boundary' and type = 'administrative'
+               """.format(self.rank)
 
     def sql_get_objects(self):
         return """SELECT place_id FROM placex
                   WHERE indexed_status > 0 and rank_search = {}
                         and class = 'boundary' and type = 'administrative'
-                  ORDER BY partition, admin_level""".format(self.rank)
+                  ORDER BY partition, admin_level
+               """.format(self.rank)
 
     @staticmethod
     def sql_index_place(ids):
         return "UPDATE placex SET indexed_status = 0 WHERE place_id IN ({})"\
                .format(','.join((str(i) for i in ids)))
 
+
+class PostcodeRunner:
+    """ Provides the SQL commands for indexing the location_postcode table.
+    """
+
+    @staticmethod
+    def name():
+        return "postcodes (location_postcode)"
+
+    @staticmethod
+    def sql_count_objects():
+        return 'SELECT count(*) FROM location_postcode WHERE indexed_status > 0'
+
+    @staticmethod
+    def sql_get_objects():
+        return """SELECT place_id FROM location_postcode
+                  WHERE indexed_status > 0
+                  ORDER BY country_code, postcode"""
+
+    @staticmethod
+    def sql_index_place(ids):
+        return """UPDATE location_postcode SET indexed_status = 0
+                  WHERE place_id IN ({})
+               """.format(','.join((str(i) for i in ids)))
+
+
+def _analyse_db_if(conn, condition):
+    if condition:
+        with conn.cursor() as cur:
+            cur.execute('ANALYSE')
+
+
 class Indexer:
     """ Main indexing routine.
     """
 
-    def __init__(self, opts):
-        self.minrank = max(1, opts.minrank)
-        self.maxrank = min(30, opts.maxrank)
-        self.conn = make_connection(opts)
-        self.threads = [DBConnection(opts) for _ in range(opts.threads)]
+    def __init__(self, dsn, num_threads):
+        self.dsn = dsn
+        self.num_threads = num_threads
+        self.conn = None
+        self.threads = []
+
+
+    def _setup_connections(self):
+        self.conn = psycopg2.connect(self.dsn)
+        self.threads = [DBConnection(self.dsn) for _ in range(self.num_threads)]
+
 
-    def index_boundaries(self):
+    def _close_connections(self):
+        if self.conn:
+            self.conn.close()
+            self.conn = None
+
+        for thread in self.threads:
+            thread.close()
+        self.threads = []
+
+
+    def index_full(self, analyse=True):
+        """ Index the complete database. This will first index boudnaries
+            followed by all other objects. When `analyse` is True, then the
+            database will be analysed at the appropriate places to
+            ensure that database statistics are updated.
+        """
+        conn = psycopg2.connect(self.dsn)
+        conn.autocommit = True
+
+        try:
+            self.index_by_rank(0, 4)
+            _analyse_db_if(conn, analyse)
+
+            self.index_boundaries(0, 30)
+            _analyse_db_if(conn, analyse)
+
+            self.index_by_rank(5, 25)
+            _analyse_db_if(conn, analyse)
+
+            self.index_by_rank(26, 30)
+            _analyse_db_if(conn, analyse)
+
+            self.index_postcodes()
+            _analyse_db_if(conn, analyse)
+        finally:
+            conn.close()
+
+
+    def index_boundaries(self, minrank, maxrank):
+        """ Index only administrative boundaries within the given rank range.
+        """
         LOG.warning("Starting indexing boundaries using %s threads",
-                    len(self.threads))
+                    self.num_threads)
+
+        self._setup_connections()
 
-        for rank in range(max(self.minrank, 5), min(self.maxrank, 26)):
-            self.index(BoundaryRunner(rank))
+        try:
+            for rank in range(max(minrank, 4), min(maxrank, 26)):
+                self.index(BoundaryRunner(rank))
+        finally:
+            self._close_connections()
 
-    def index_by_rank(self):
-        """ Run classic indexing by rank.
+    def index_by_rank(self, minrank, maxrank):
+        """ Index all entries of placex in the given rank range (inclusive)
+            in order of their address rank.
+
+            When rank 30 is requested then also interpolations and
+            places with address rank 0 will be indexed.
         """
+        maxrank = min(maxrank, 30)
         LOG.warning("Starting indexing rank (%i to %i) using %i threads",
-                    self.minrank, self.maxrank, len(self.threads))
+                    minrank, maxrank, self.num_threads)
+
+        self._setup_connections()
+
+        try:
+            for rank in range(max(1, minrank), maxrank):
+                self.index(RankRunner(rank))
+
+            if maxrank == 30:
+                self.index(RankRunner(0))
+                self.index(InterpolationRunner(), 20)
+                self.index(RankRunner(30), 20)
+            else:
+                self.index(RankRunner(maxrank))
+        finally:
+            self._close_connections()
+
+
+    def index_postcodes(self):
+        """Index the entries ofthe location_postcode table.
+        """
+        LOG.warning("Starting indexing postcodes using %s threads", self.num_threads)
+
+        self._setup_connections()
+
+        try:
+            self.index(PostcodeRunner(), 20)
+        finally:
+            self._close_connections()
+
+    def update_status_table(self):
+        """ Update the status in the status table to 'indexed'.
+        """
+        conn = psycopg2.connect(self.dsn)
 
-        for rank in range(max(1, self.minrank), self.maxrank):
-            self.index(RankRunner(rank))
+        try:
+            with conn.cursor() as cur:
+                cur.execute('UPDATE import_status SET indexed = true')
 
-        if self.maxrank == 30:
-            self.index(RankRunner(0))
-            self.index(InterpolationRunner(), 20)
-            self.index(RankRunner(self.maxrank), 20)
-        else:
-            self.index(RankRunner(self.maxrank))
+            conn.commit()
+        finally:
+            conn.close()
 
     def index(self, obj, batch=1):
         """ Index a single rank or table. `obj` describes the SQL to use