Merge pull request #3108 from mtmail/remove-legacy-wikipedia-tag-syntax

[nominatim.git] / nominatim / db / sql_preprocessor.py

diff --git a/nominatim/db/sql_preprocessor.py b/nominatim/db/sql_preprocessor.py
index b450422da09396cb687cfe50750490311f82a31d..2e11f57195323e2e53aec9dc77d541f81b6522e6 100644 (file)
--- a/nominatim/db/sql_preprocessor.py
+++ b/nominatim/db/sql_preprocessor.py
@@ -11,6 +11,7 @@ from typing import Set, Dict, Any
 import jinja2
 
 from nominatim.db.connection import Connection
+from nominatim.db.async_connection import WorkerPool
 from nominatim.config import Configuration
 
 def _get_partitions(conn: Connection) -> Set[int]:
@@ -56,9 +57,11 @@ def _setup_postgresql_features(conn: Connection) -> Dict[str, Any]:
     """
     pg_version = conn.server_version_tuple()
     postgis_version = conn.postgis_version_tuple()
+    pg11plus = pg_version >= (11, 0, 0)
+    ps3 = postgis_version >= (3, 0)
     return {
-        'has_index_non_key_column': pg_version >= (11, 0, 0),
-        'spgist_geom' : 'SPGIST' if postgis_version >= (3, 0) else 'GIST'
+        'has_index_non_key_column': pg11plus,
+        'spgist_geom' : 'SPGIST' if pg11plus and ps3 else 'GIST'
     }
 
 class SQLPreprocessor:
@@ -96,3 +99,21 @@ class SQLPreprocessor:
         with conn.cursor() as cur:
             cur.execute(sql)
         conn.commit()
+
+
+    def run_parallel_sql_file(self, dsn: str, name: str, num_threads: int = 1,
+                              **kwargs: Any) -> None:
+        """ Execure the given SQL files using parallel asynchronous connections.
+            The keyword arguments may supply additional parameters for
+            preprocessing.
+
+            After preprocessing the SQL code is cut at lines containing only
+            '---'. Each chunk is sent to one of the `num_threads` workers.
+        """
+        sql = self.env.get_template(name).render(**kwargs)
+
+        parts = sql.split('\n---\n')
+
+        with WorkerPool(dsn, num_threads) as pool:
+            for part in parts:
+                pool.next_free_worker().perform(part)