Merge remote-tracking branch 'upstream/master'

[nominatim.git] / nominatim / tools / tiger_data.py

diff --git a/nominatim/tools/tiger_data.py b/nominatim/tools/tiger_data.py
index 19a1268253feaa7ff1e2e6de20be3c43f1d74025..4a32bb1ebf93848b0126d0dfd31b5b0bd472c858 100644 (file)
--- a/nominatim/tools/tiger_data.py
+++ b/nominatim/tools/tiger_data.py
@@ -1,50 +1,88 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# This file is part of Nominatim. (https://nominatim.org)
+#
+# Copyright (C) 2022 by the Nominatim developer community.
+# For a full list of authors see the git log.

 """
 Functions for importing tiger data and handling tarbar and directory files
 """
 """
 Functions for importing tiger data and handling tarbar and directory files
 """

+from typing import Any, TextIO, List, Union, cast

 import csv
 import io
 import logging
 import os
 import tarfile
 
 import csv
 import io
 import logging
 import os
 import tarfile
 

+from psycopg2.extras import Json
+
+from nominatim.config import Configuration

 from nominatim.db.connection import connect
 from nominatim.db.async_connection import WorkerPool
 from nominatim.db.sql_preprocessor import SQLPreprocessor
 from nominatim.errors import UsageError
 from nominatim.db.connection import connect
 from nominatim.db.async_connection import WorkerPool
 from nominatim.db.sql_preprocessor import SQLPreprocessor
 from nominatim.errors import UsageError

-from nominatim.indexer.place_info import PlaceInfo
+from nominatim.data.place_info import PlaceInfo
+from nominatim.tokenizer.base import AbstractAnalyzer, AbstractTokenizer

 
 LOG = logging.getLogger()
 
 
 LOG = logging.getLogger()
 

-
-def handle_tarfile_or_directory(data_dir):
-    """ Handles tarfile or directory for importing tiger data
+class TigerInput:
+    """ Context manager that goes through Tiger input files which may
+        either be in a directory or gzipped together in a tar file.

     """
 
     """
 

-    tar = None
-    if data_dir.endswith('.tar.gz'):
-        try:
-            tar = tarfile.open(data_dir)
-        except tarfile.ReadError as err:
-            LOG.fatal("Cannot open '%s'. Is this a tar file?", data_dir)
-            raise UsageError("Cannot open Tiger data file.") from err
-
-        csv_files = [i for i in tar.getmembers() if i.name.endswith('.csv')]
-        LOG.warning("Found %d CSV files in tarfile with path %s", len(csv_files), data_dir)
-        if not csv_files:
-            LOG.warning("Tiger data import selected but no files in tarfile's path %s", data_dir)
-            return None, None
-    else:
-        files = os.listdir(data_dir)
-        csv_files = [os.path.join(data_dir, i) for i in files if i.endswith('.csv')]
-        LOG.warning("Found %d CSV files in path %s", len(csv_files), data_dir)
-        if not csv_files:
-            LOG.warning("Tiger data import selected but no files found in path %s", data_dir)
-            return None, None
-
-    return csv_files, tar
-
-
-def handle_threaded_sql_statements(pool, fd, analyzer):
+    def __init__(self, data_dir: str) -> None:
+        self.tar_handle = None
+        self.files: List[Union[str, tarfile.TarInfo]] = []
+
+        if data_dir.endswith('.tar.gz'):
+            try:
+                self.tar_handle = tarfile.open(data_dir) # pylint: disable=consider-using-with
+            except tarfile.ReadError as err:
+                LOG.fatal("Cannot open '%s'. Is this a tar file?", data_dir)
+                raise UsageError("Cannot open Tiger data file.") from err
+
+            self.files = [i for i in self.tar_handle.getmembers() if i.name.endswith('.csv')]
+            LOG.warning("Found %d CSV files in tarfile with path %s", len(self.files), data_dir)
+        else:
+            files = os.listdir(data_dir)
+            self.files = [os.path.join(data_dir, i) for i in files if i.endswith('.csv')]
+            LOG.warning("Found %d CSV files in path %s", len(self.files), data_dir)
+
+        if not self.files:
+            LOG.warning("Tiger data import selected but no files found at %s", data_dir)
+
+
+    def __enter__(self) -> 'TigerInput':
+        return self
+
+
+    def __exit__(self, exc_type: Any, exc_val: Any, exc_tb: Any) -> None:
+        if self.tar_handle:
+            self.tar_handle.close()
+            self.tar_handle = None
+
+
+    def next_file(self) -> TextIO:
+        """ Return a file handle to the next file to be processed.
+            Raises an IndexError if there is no file left.
+        """
+        fname = self.files.pop(0)
+
+        if self.tar_handle is not None:
+            extracted = self.tar_handle.extractfile(fname)
+            assert extracted is not None
+            return io.TextIOWrapper(extracted)
+
+        return open(cast(str, fname), encoding='utf-8')
+
+
+    def __len__(self) -> int:
+        return len(self.files)
+
+
+def handle_threaded_sql_statements(pool: WorkerPool, fd: TextIO,
+                                   analyzer: AbstractAnalyzer) -> None:

     """ Handles sql statement with multiplexing
     """
     lines = 0
     """ Handles sql statement with multiplexing
     """
     lines = 0


@@ -57,7 +95,7 @@ def handle_threaded_sql_statements(pool, fd, analyzer):
             address = dict(street=row['street'], postcode=row['postcode'])
             args = ('SRID=4326;' + row['geometry'],
                     int(row['from']), int(row['to']), row['interpolation'],
             address = dict(street=row['street'], postcode=row['postcode'])
             args = ('SRID=4326;' + row['geometry'],
                     int(row['from']), int(row['to']), row['interpolation'],

-                    PlaceInfo({'address': address}).analyze(analyzer),
+                    Json(analyzer.process_place(PlaceInfo({'address': address}))),

                     analyzer.normalize_postcode(row['postcode']))
         except ValueError:
             continue
                     analyzer.normalize_postcode(row['postcode']))
         except ValueError:
             continue


@@ -69,39 +107,35 @@ def handle_threaded_sql_statements(pool, fd, analyzer):
             lines = 0
 
 
             lines = 0
 
 

-def add_tiger_data(data_dir, config, threads, tokenizer):
+def add_tiger_data(data_dir: str, config: Configuration, threads: int,
+                   tokenizer: AbstractTokenizer) -> int:

     """ Import tiger data from directory or tar file `data dir`.
     """
     dsn = config.get_libpq_dsn()
     """ Import tiger data from directory or tar file `data dir`.
     """
     dsn = config.get_libpq_dsn()

-    files, tar = handle_tarfile_or_directory(data_dir)

 
 

-    if not files:
-        return
+    with TigerInput(data_dir) as tar:
+        if not tar:
+            return 1

 
 

-    with connect(dsn) as conn:
-        sql = SQLPreprocessor(conn, config)
-        sql.run_sql_file(conn, 'tiger_import_start.sql')
+        with connect(dsn) as conn:
+            sql = SQLPreprocessor(conn, config)
+            sql.run_sql_file(conn, 'tiger_import_start.sql')

 
 

-    # Reading files and then for each file line handling
-    # sql_query in <threads - 1> chunks.
-    place_threads = max(1, threads - 1)
+        # Reading files and then for each file line handling
+        # sql_query in <threads - 1> chunks.
+        place_threads = max(1, threads - 1)

 
 

-    with WorkerPool(dsn, place_threads, ignore_sql_errors=True) as pool:
-        with tokenizer.name_analyzer() as analyzer:
-            for fname in files:
-                if not tar:
-                    fd = open(fname)
-                else:
-                    fd = io.TextIOWrapper(tar.extractfile(fname))
+        with WorkerPool(dsn, place_threads, ignore_sql_errors=True) as pool:
+            with tokenizer.name_analyzer() as analyzer:
+                while tar:
+                    with tar.next_file() as fd:
+                        handle_threaded_sql_statements(pool, fd, analyzer)

 
 

-                handle_threaded_sql_statements(pool, fd, analyzer)
+        print('\n')

 
 

-                fd.close()
-
-    if tar:
-        tar.close()
-    print('\n')

     LOG.warning("Creating indexes on Tiger data")
     with connect(dsn) as conn:
         sql = SQLPreprocessor(conn, config)
         sql.run_sql_file(conn, 'tiger_import_finish.sql')
     LOG.warning("Creating indexes on Tiger data")
     with connect(dsn) as conn:
         sql = SQLPreprocessor(conn, config)
         sql.run_sql_file(conn, 'tiger_import_finish.sql')

+
+    return 0