Merge pull request #2708 from lonvia/use-format-literals

Assorted fixes for new pylint warnings

.pylintrc

@@ -11,6 +11,6 @@ ignored-modules=icu,datrie
 # 'with' statements.
 ignored-classes=NominatimArgs,closing
 # 'too-many-ancestors' is triggered already by deriving from UserDict
-disable=too-few-public-methods,duplicate-code,too-many-ancestors
+disable=too-few-public-methods,duplicate-code,too-many-ancestors,bad-option-value,no-self-use
 
 good-names=i,x,y,fd,db

nominatim/cli.py

@@ -60,9 +60,9 @@ class CommandlineParser:
     def nominatim_version_text():
         """ Program name and version number as string
         """
-        text = 'Nominatim version %s.%s.%s.%s' % version.NOMINATIM_VERSION
+        text = f'Nominatim version {version.version_str()}'
         if version.GIT_COMMIT_HASH is not None:
-            text += ' (%s)' % version.GIT_COMMIT_HASH
+            text += f' ({version.GIT_COMMIT_HASH})'
         return text
 
     def add_subcommand(self, name, cmd):

nominatim/clicmd/replication.py

@@ -21,7 +21,7 @@ LOG = logging.getLogger()
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111
 # Using non-top-level imports to make pyosmium optional for replication only.
-# pylint: disable=E0012,C0415
+# pylint: disable=C0415
 
 class UpdateReplication:
     """\
@@ -96,8 +96,7 @@ class UpdateReplication:
         end = dt.datetime.now(dt.timezone.utc)
         LOG.warning("Update completed. Import: %s. %sTotal: %s. Remaining backlog: %s.",
                     round_time((start_index or end) - start_import),
-                    "Indexing: {} ".format(round_time(end - start_index))
+                    f"Indexing: {round_time(end - start_index)} " if start_index else '',
-                    if start_index else '',
                     round_time(end - start_import),
                     round_time(end - batchdate))
 

nominatim/clicmd/setup.py

@@ -14,12 +14,12 @@ import psutil
 
 from nominatim.db.connection import connect
 from nominatim.db import status, properties
-from nominatim.version import NOMINATIM_VERSION
+from nominatim.version import version_str
 
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111
 # Using non-top-level imports to avoid eventually unused imports.
-# pylint: disable=E0012,C0415
+# pylint: disable=C0415
 
 LOG = logging.getLogger()
 
@@ -194,10 +194,10 @@ class SetupAll:
             LOG.warning('Creating support index')
             if tablespace:
                 tablespace = 'TABLESPACE ' + tablespace
-            cur.execute("""CREATE INDEX idx_placex_pendingsector
+            cur.execute(f"""CREATE INDEX idx_placex_pendingsector
                             ON placex USING BTREE (rank_address,geometry_sector)
-                           {} WHERE indexed_status > 0
+                            {tablespace} WHERE indexed_status > 0
-                        """.format(tablespace))
+                         """)
         conn.commit()
 
 
@@ -213,5 +213,4 @@ class SetupAll:
             except Exception as exc: # pylint: disable=broad-except
                 LOG.error('Cannot determine date of database: %s', exc)
 
-            properties.set_property(conn, 'database_version',
+            properties.set_property(conn, 'database_version', version_str())
-                                    '{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(NOMINATIM_VERSION))

nominatim/config.py

@@ -144,7 +144,7 @@ class Configuration:
         style = self.__getattr__('IMPORT_STYLE')
 
         if style in ('admin', 'street', 'address', 'full', 'extratags'):
-            return self.config_dir / 'import-{}.style'.format(style)
+            return self.config_dir / f'import-{style}.style'
 
         return self.find_config_file('', 'IMPORT_STYLE')
 

nominatim/db/connection.py

@@ -25,7 +25,8 @@ class _Cursor(psycopg2.extras.DictCursor):
         execution functions.
     """
 
-    def execute(self, query, args=None): # pylint: disable=W0221
+    # pylint: disable=arguments-renamed,arguments-differ
+    def execute(self, query, args=None):
         """ Query execution that logs the SQL query when debugging is enabled.
         """
         LOG.debug(self.mogrify(query, args).decode('utf-8'))
@@ -163,7 +164,7 @@ def connect(dsn):
         ctxmgr.connection = conn
         return ctxmgr
     except psycopg2.OperationalError as err:
-        raise UsageError("Cannot connect to database: {}".format(err)) from err
+        raise UsageError(f"Cannot connect to database: {err}") from err
 
 
 # Translation from PG connection string parameters to PG environment variables.

nominatim/db/sql_preprocessor.py

@@ -39,10 +39,10 @@ def _setup_tablespace_sql(config):
     out = {}
     for subset in ('ADDRESS', 'SEARCH', 'AUX'):
         for kind in ('DATA', 'INDEX'):
-            tspace = getattr(config, 'TABLESPACE_{}_{}'.format(subset, kind))
+            tspace = getattr(config, f'TABLESPACE_{subset}_{kind}')
             if tspace:
-                tspace = 'TABLESPACE "{}"'.format(tspace)
+                tspace = f'TABLESPACE "{tspace}"'
-            out['{}_{}'.format(subset.lower(), kind.lower())] = tspace
+            out[f'{subset.lower()}_{kind.lower()}'] = tspace
 
     return out
 

nominatim/db/status.py

@@ -34,7 +34,7 @@ def compute_database_date(conn):
 
     LOG.info("Using node id %d for timestamp lookup", osmid)
     # Get the node from the API to find the timestamp when it was created.
-    node_url = 'https://www.openstreetmap.org/api/0.6/node/{}/1'.format(osmid)
+    node_url = f'https://www.openstreetmap.org/api/0.6/node/{osmid}/1'
     data = get_url(node_url)
 
     match = re.search(r'timestamp="((\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2}):(\d{2}))Z"', data)

nominatim/db/utils.py

@@ -40,8 +40,8 @@ def execute_file(dsn, fname, ignore_errors=False, pre_code=None, post_code=None)
         cmd.extend(('-v', 'ON_ERROR_STOP=1'))
     if not LOG.isEnabledFor(logging.INFO):
         cmd.append('--quiet')
-    proc = subprocess.Popen(cmd, env=get_pg_env(dsn), stdin=subprocess.PIPE)
 
+    with subprocess.Popen(cmd, env=get_pg_env(dsn), stdin=subprocess.PIPE) as proc:
         try:
             if not LOG.isEnabledFor(logging.INFO):
                 proc.stdin.write('set client_min_messages to WARNING;'.encode('utf-8'))
@@ -67,9 +67,9 @@ def execute_file(dsn, fname, ignore_errors=False, pre_code=None, post_code=None)
 
 
 # List of characters that need to be quoted for the copy command.
-_SQL_TRANSLATION = {ord(u'\\'): u'\\\\',
+_SQL_TRANSLATION = {ord('\\'): '\\\\',
-                    ord(u'\t'): u'\\t',
+                    ord('\t'): '\\t',
-                    ord(u'\n'): u'\\n'}
+                    ord('\n'): '\\n'}
 
 
 class CopyBuffer:

nominatim/indexer/runners.py

@@ -66,7 +66,7 @@ class RankRunner(AbstractPlacexRunner):
     """
 
     def name(self):
-        return "rank {}".format(self.rank)
+        return f"rank {self.rank}"
 
     def sql_count_objects(self):
         return pysql.SQL("""SELECT count(*) FROM placex
@@ -86,7 +86,7 @@ class BoundaryRunner(AbstractPlacexRunner):
     """
 
     def name(self):
-        return "boundaries rank {}".format(self.rank)
+        return f"boundaries rank {self.rank}"
 
     def sql_count_objects(self):
         return pysql.SQL("""SELECT count(*) FROM placex

nominatim/tokenizer/icu_tokenizer.py

@@ -278,8 +278,7 @@ class LegacyICUNameAnalyzer(AbstractAnalyzer):
                + [(k, v, part_ids.get(v, None)) for k, v in partial_tokens.items()]
 
 
-    @staticmethod
+    def normalize_postcode(self, postcode):
-    def normalize_postcode(postcode):
         """ Convert the postcode to a standardized form.
 
             This function must yield exactly the same result as the SQL function

nominatim/tokenizer/legacy_tokenizer.py

@@ -74,10 +74,10 @@ def _check_module(module_dir, conn):
     with conn.cursor() as cur:
         try:
             cur.execute("""CREATE FUNCTION nominatim_test_import_func(text)
-                           RETURNS text AS '{}/nominatim.so', 'transliteration'
+                           RETURNS text AS %s, 'transliteration'
                            LANGUAGE c IMMUTABLE STRICT;
                            DROP FUNCTION nominatim_test_import_func(text)
-                        """.format(module_dir))
+                        """, (f'{module_dir}/nominatim.so', ))
         except psycopg2.DatabaseError as err:
             LOG.fatal("Error accessing database module: %s", err)
             raise UsageError("Database module cannot be accessed.") from err
@@ -250,12 +250,12 @@ class LegacyTokenizer(AbstractTokenizer):
         php_file = self.data_dir / "tokenizer.php"
 
         if not php_file.exists() or overwrite:
-            php_file.write_text(dedent("""\
+            php_file.write_text(dedent(f"""\
                 <?php
-                @define('CONST_Max_Word_Frequency', {0.MAX_WORD_FREQUENCY});
+                @define('CONST_Max_Word_Frequency', {config.MAX_WORD_FREQUENCY});
-                @define('CONST_Term_Normalization_Rules', "{0.TERM_NORMALIZATION}");
+                @define('CONST_Term_Normalization_Rules', "{config.TERM_NORMALIZATION}");
-                require_once('{0.lib_dir.php}/tokenizer/legacy_tokenizer.php');
+                require_once('{config.lib_dir.php}/tokenizer/legacy_tokenizer.php');
-                """.format(config)), encoding='utf-8')
+                """), encoding='utf-8')
 
 
     def _init_db_tables(self, config):
@@ -337,8 +337,7 @@ class LegacyNameAnalyzer(AbstractAnalyzer):
         return self.normalizer.transliterate(phrase)
 
 
-    @staticmethod
+    def normalize_postcode(self, postcode):
-    def normalize_postcode(postcode):
         """ Convert the postcode to a standardized form.
 
             This function must yield exactly the same result as the SQL function

nominatim/tokenizer/token_analysis/config_variants.py

@@ -66,7 +66,7 @@ class _VariantMaker:
         """
         parts = re.split(r'(\|)?([=-])>', rule)
         if len(parts) != 4:
-            raise UsageError("Syntax error in variant rule: " + rule)
+            raise UsageError(f"Syntax error in variant rule: {rule}")
 
         decompose = parts[1] is None
         src_terms = [self._parse_variant_word(t) for t in parts[0].split(',')]
@@ -89,7 +89,7 @@ class _VariantMaker:
         name = name.strip()
         match = re.fullmatch(r'([~^]?)([^~$^]*)([~$]?)', name)
         if match is None or (match.group(1) == '~' and match.group(3) == '~'):
-            raise UsageError("Invalid variant word descriptor '{}'".format(name))
+            raise UsageError(f"Invalid variant word descriptor '{name}'")
         norm_name = self.norm.transliterate(match.group(2)).strip()
         if not norm_name:
             return None

nominatim/tools/database_import.py

@@ -234,7 +234,7 @@ def create_search_indices(conn, config, drop=False):
         bad_indices = [row[0] for row in list(cur)]
         for idx in bad_indices:
             LOG.info("Drop invalid index %s.", idx)
-            cur.execute('DROP INDEX "{}"'.format(idx))
+            cur.execute(pysql.SQL('DROP INDEX {}').format(pysql.Identifier(idx)))
     conn.commit()
 
     sql = SQLPreprocessor(conn, config)

nominatim/tools/exec_utils.py

@@ -12,7 +12,7 @@ import subprocess
 import urllib.request as urlrequest
 from urllib.parse import urlencode
 
-from nominatim.version import NOMINATIM_VERSION
+from nominatim.version import version_str
 from nominatim.db.connection import get_pg_env
 
 LOG = logging.getLogger()
@@ -55,10 +55,10 @@ def run_api_script(endpoint, project_dir, extra_env=None, phpcgi_bin=None,
     query_string = urlencode(params or {})
 
     env = dict(QUERY_STRING=query_string,
-               SCRIPT_NAME='/{}.php'.format(endpoint),
+               SCRIPT_NAME=f'/{endpoint}.php',
-               REQUEST_URI='/{}.php?{}'.format(endpoint, query_string),
+               REQUEST_URI=f'/{endpoint}.php?{query_string}',
                CONTEXT_DOCUMENT_ROOT=webdir,
-               SCRIPT_FILENAME='{}/{}.php'.format(webdir, endpoint),
+               SCRIPT_FILENAME=f'{webdir}/{endpoint}.php',
                HTTP_HOST='localhost',
                HTTP_USER_AGENT='nominatim-tool',
                REMOTE_ADDR='0.0.0.0',
@@ -150,7 +150,7 @@ def run_osm2pgsql(options):
 def get_url(url):
     """ Get the contents from the given URL and return it as a UTF-8 string.
     """
-    headers = {"User-Agent": "Nominatim/{0[0]}.{0[1]}.{0[2]}-{0[3]}".format(NOMINATIM_VERSION)}
+    headers = {"User-Agent": f"Nominatim/{version_str()}"}
 
     try:
         with urlrequest.urlopen(urlrequest.Request(url, headers=headers)) as response:

nominatim/tools/migration.py

@@ -9,9 +9,11 @@ Functions for database migration to newer software versions.
 """
 import logging
 
+from psycopg2 import sql as pysql
+
 from nominatim.db import properties
 from nominatim.db.connection import connect
-from nominatim.version import NOMINATIM_VERSION
+from nominatim.version import NOMINATIM_VERSION, version_str
 from nominatim.tools import refresh
 from nominatim.tokenizer import factory as tokenizer_factory
 from nominatim.errors import UsageError
@@ -47,7 +49,7 @@ def migrate(config, paths):
         for version, func in _MIGRATION_FUNCTIONS:
             if db_version <= version:
                 LOG.warning("Runnning: %s (%s)", func.__doc__.split('\n', 1)[0],
-                            '{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(version))
+                            version_str(version))
                 kwargs = dict(conn=conn, config=config, paths=paths)
                 func(**kwargs)
                 conn.commit()
@@ -59,8 +61,7 @@ def migrate(config, paths):
             tokenizer = tokenizer_factory.get_tokenizer_for_db(config)
             tokenizer.update_sql_functions(config)
 
-        properties.set_property(conn, 'database_version',
+        properties.set_property(conn, 'database_version', version_str())
-                                '{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(NOMINATIM_VERSION))
 
         conn.commit()
 
@@ -125,11 +126,11 @@ def add_nominatim_property_table(conn, config, **_):
     """
     if not conn.table_exists('nominatim_properties'):
         with conn.cursor() as cur:
-            cur.execute("""CREATE TABLE nominatim_properties (
+            cur.execute(pysql.SQL("""CREATE TABLE nominatim_properties (
                                         property TEXT,
                                         value TEXT);
-                           GRANT SELECT ON TABLE nominatim_properties TO "{}";
+                                     GRANT SELECT ON TABLE nominatim_properties TO {};
-                        """.format(config.DATABASE_WEBUSER))
+                                  """).format(pysql.Identifier(config.DATABASE_WEBUSER)))
 
 @_migration(3, 6, 0, 0)
 def change_housenumber_transliteration(conn, **_):
@@ -194,7 +195,8 @@ def install_legacy_tokenizer(conn, config, **_):
                                            and column_name = 'token_info'""",
                                         (table, ))
                 if has_column == 0:
-                    cur.execute('ALTER TABLE {} ADD COLUMN token_info JSONB'.format(table))
+                    cur.execute(pysql.SQL('ALTER TABLE {} ADD COLUMN token_info JSONB')
+                                .format(pysql.Identifier(table)))
         tokenizer = tokenizer_factory.create_tokenizer(config, init_db=False,
                                                        module_name='legacy')
 

nominatim/tools/postcodes.py

@@ -36,7 +36,7 @@ class _CountryPostcodesCollector:
 
     def __init__(self, country):
         self.country = country
-        self.collected = dict()
+        self.collected = {}
 
 
     def add(self, postcode, x, y):
@@ -136,13 +136,13 @@ class _CountryPostcodesCollector:
 
 
     def _open_external(self, project_dir):
-        fname = project_dir / '{}_postcodes.csv'.format(self.country)
+        fname = project_dir / f'{self.country}_postcodes.csv'
 
         if fname.is_file():
             LOG.info("Using external postcode file '%s'.", fname)
-            return open(fname, 'r')
+            return open(fname, 'r', encoding='utf-8')
 
-        fname = project_dir / '{}_postcodes.csv.gz'.format(self.country)
+        fname = project_dir / f'{self.country}_postcodes.csv.gz'
 
         if fname.is_file():
             LOG.info("Using external postcode file '%s'.", fname)

nominatim/tools/refresh.py

@@ -15,7 +15,7 @@ from psycopg2 import sql as pysql
 
 from nominatim.db.utils import execute_file
 from nominatim.db.sql_preprocessor import SQLPreprocessor
-from nominatim.version import NOMINATIM_VERSION
+from nominatim.version import version_str
 
 LOG = logging.getLogger()
 
@@ -52,16 +52,19 @@ def load_address_levels(conn, table, levels):
     with conn.cursor() as cur:
         cur.drop_table(table)
 
-        cur.execute("""CREATE TABLE {} (country_code varchar(2),
+        cur.execute(pysql.SQL("""CREATE TABLE {} (
+                                        country_code varchar(2),
                                         class TEXT,
                                         type TEXT,
                                         rank_search SMALLINT,
-                                        rank_address SMALLINT)""".format(table))
+                                        rank_address SMALLINT)
+                              """).format(pysql.Identifier(table)))
 
         cur.execute_values(pysql.SQL("INSERT INTO {} VALUES %s")
                            .format(pysql.Identifier(table)), rows)
 
-        cur.execute('CREATE UNIQUE INDEX ON {} (country_code, class, type)'.format(table))
+        cur.execute(pysql.SQL('CREATE UNIQUE INDEX ON {} (country_code, class, type)')
+                    .format(pysql.Identifier(table)))
 
     conn.commit()
 
@@ -186,16 +189,15 @@ def setup_website(basedir, config, conn):
         LOG.info('Creating website directory.')
         basedir.mkdir()
 
-    template = dedent("""\
+    template = dedent(f"""\
                       <?php
 
                       @define('CONST_Debug', $_GET['debug'] ?? false);
-                      @define('CONST_LibDir', '{0}');
+                      @define('CONST_LibDir', '{config.lib_dir.php}');
-                      @define('CONST_TokenizerDir', '{2}');
+                      @define('CONST_TokenizerDir', '{config.project_dir / 'tokenizer'}');
-                      @define('CONST_NominatimVersion', '{1[0]}.{1[1]}.{1[2]}-{1[3]}');
+                      @define('CONST_NominatimVersion', '{version_str()}');
 
-                      """.format(config.lib_dir.php, NOMINATIM_VERSION,
+                      """)
-                                 config.project_dir / 'tokenizer'))
 
     for php_name, conf_name, var_type in PHP_CONST_DEFS:
         varout = _quote_php_variable(var_type, config, conf_name)

nominatim/tools/special_phrases/sp_csv_loader.py

@@ -39,8 +39,8 @@ class SPCsvLoader(Iterator):
         """
         phrases = set()
 
-        with open(self.csv_path) as file:
+        with open(self.csv_path, encoding='utf-8') as fd:
-            reader = csv.DictReader(file, delimiter=',')
+            reader = csv.DictReader(fd, delimiter=',')
             for row in reader:
                 phrases.add(
                     SpecialPhrase(row['phrase'], row['class'], row['type'], row['operator'])
@@ -54,4 +54,4 @@ class SPCsvLoader(Iterator):
         _, extension = os.path.splitext(self.csv_path)
 
         if extension != '.csv':
-            raise UsageError('The file {} is not a csv file.'.format(self.csv_path))
+            raise UsageError(f'The file {self.csv_path} is not a csv file.')

nominatim/tools/special_phrases/sp_importer.py

@@ -16,7 +16,7 @@
 import logging
 import re
 
-from psycopg2.sql import Identifier, Literal, SQL
+from psycopg2.sql import Identifier, SQL
 from nominatim.tools.special_phrases.importer_statistics import SpecialPhrasesImporterStatistics
 
 LOG = logging.getLogger()
@@ -191,39 +191,41 @@ class SPImporter():
 
     def _create_place_classtype_table(self, sql_tablespace, phrase_class, phrase_type):
         """
-            Create table place_classtype of the given phrase_class/phrase_type if doesn't exit.
+            Create table place_classtype of the given phrase_class/phrase_type
+            if doesn't exit.
         """
         table_name = _classtype_table(phrase_class, phrase_type)
-        with self.db_connection.cursor() as db_cursor:
+        with self.db_connection.cursor() as cur:
-            db_cursor.execute(SQL("""
+            cur.execute(SQL("""CREATE TABLE IF NOT EXISTS {} {} AS
-                    CREATE TABLE IF NOT EXISTS {{}} {}
+                                 SELECT place_id AS place_id,
-                    AS SELECT place_id AS place_id,st_centroid(geometry) AS centroid FROM placex
+                                        st_centroid(geometry) AS centroid
-                    WHERE class = {{}} AND type = {{}}""".format(sql_tablespace))
+                                 FROM placex
-                              .format(Identifier(table_name), Literal(phrase_class),
+                                 WHERE class = %s AND type = %s
-                                      Literal(phrase_type)))
+                             """).format(Identifier(table_name), SQL(sql_tablespace)),
+                        (phrase_class, phrase_type))
 
 
     def _create_place_classtype_indexes(self, sql_tablespace, phrase_class, phrase_type):
         """
             Create indexes on centroid and place_id for the place_classtype table.
         """
-        index_prefix = 'idx_place_classtype_{}_{}_'.format(phrase_class, phrase_type)
+        index_prefix = f'idx_place_classtype_{phrase_class}_{phrase_type}_'
         base_table = _classtype_table(phrase_class, phrase_type)
         # Index on centroid
         if not self.db_connection.index_exists(index_prefix + 'centroid'):
             with self.db_connection.cursor() as db_cursor:
-                db_cursor.execute(SQL("""
+                db_cursor.execute(SQL("CREATE INDEX {} ON {} USING GIST (centroid) {}")
-                    CREATE INDEX {{}} ON {{}} USING GIST (centroid) {}""".format(sql_tablespace))
                                   .format(Identifier(index_prefix + 'centroid'),
-                                          Identifier(base_table)), sql_tablespace)
+                                          Identifier(base_table),
+                                          SQL(sql_tablespace)))
 
         # Index on place_id
         if not self.db_connection.index_exists(index_prefix + 'place_id'):
             with self.db_connection.cursor() as db_cursor:
-                db_cursor.execute(SQL(
+                db_cursor.execute(SQL("CREATE INDEX {} ON {} USING btree(place_id) {}")
-                    """CREATE INDEX {{}} ON {{}} USING btree(place_id) {}""".format(sql_tablespace))
                                   .format(Identifier(index_prefix + 'place_id'),
-                                          Identifier(base_table)))
+                                          Identifier(base_table),
+                                          SQL(sql_tablespace)))
 
 
     def _grant_access_to_webuser(self, phrase_class, phrase_type):

nominatim/tools/tiger_data.py

@@ -21,33 +21,57 @@ from nominatim.indexer.place_info import PlaceInfo
 
 LOG = logging.getLogger()
 
-
+class TigerInput:
-def handle_tarfile_or_directory(data_dir):
+    """ Context manager that goes through Tiger input files which may
-    """ Handles tarfile or directory for importing tiger data
+        either be in a directory or gzipped together in a tar file.
     """
 
-    tar = None
+    def __init__(self, data_dir):
+        self.tar_handle = None
+        self.files = []
+
         if data_dir.endswith('.tar.gz'):
             try:
-            tar = tarfile.open(data_dir)
+                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
 
-        csv_files = [i for i in tar.getmembers() if i.name.endswith('.csv')]
+            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(csv_files), data_dir)
+            LOG.warning("Found %d CSV files in tarfile with path %s", len(self.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')]
+            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(csv_files), data_dir)
+            LOG.warning("Found %d CSV files in path %s", len(self.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
+        if not self.files:
+            LOG.warning("Tiger data import selected but no files found at %s", data_dir)
+
+
+    def __enter__(self):
+        return self
+
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        if self.tar_handle:
+            self.tar_handle.close()
+            self.tar_handle = None
+
+
+    def next_file(self):
+        """ 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:
+            return io.TextIOWrapper(self.tar_handle.extractfile(fname))
+
+        return open(fname, encoding='utf-8')
+
+
+    def __len__(self):
+        return len(self.files)
 
 
 def handle_threaded_sql_statements(pool, fd, analyzer):
@@ -79,9 +103,9 @@ def add_tiger_data(data_dir, config, threads, tokenizer):
     """ 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:
+    with TigerInput(data_dir) as tar:
+        if not tar:
             return
 
         with connect(dsn) as conn:
@@ -94,19 +118,12 @@ def add_tiger_data(data_dir, config, threads, tokenizer):
 
         with WorkerPool(dsn, place_threads, ignore_sql_errors=True) as pool:
             with tokenizer.name_analyzer() as analyzer:
-            for fname in files:
+                while tar:
-                if not tar:
+                    with tar.next_file() as fd:
-                    fd = open(fname)
-                else:
-                    fd = io.TextIOWrapper(tar.extractfile(fname))
-
                         handle_threaded_sql_statements(pool, fd, analyzer)
 
-                fd.close()
-
-    if tar:
-        tar.close()
         print('\n')
+
     LOG.warning("Creating indexes on Tiger data")
     with connect(dsn) as conn:
         sql = SQLPreprocessor(conn, config)

nominatim/version.py

@@ -34,3 +34,11 @@ POSTGIS_REQUIRED_VERSION = (2, 2)
 # cmake/tool-installed.tmpl is used to build the binary 'nominatim'. Inside
 # there is a call to set the variable value below.
 GIT_COMMIT_HASH = None
+
+
+# pylint: disable=consider-using-f-string
+def version_str(version=NOMINATIM_VERSION):
+    """
+    Return a human-readable string of the version.
+    """
+    return '{}.{}.{}-{}'.format(*version)