bdd: extend reverse API tests for format checks

[nominatim.git] / test / bdd / steps / steps_db_ops.py

diff --git a/test/bdd/steps/steps_db_ops.py b/test/bdd/steps/steps_db_ops.py
index 5ce416b013678a741f064cbbe2b19d30040e4b42..14ae5d520684eb060ac9a9f6d5632de1399fde09 100644 (file)
--- a/test/bdd/steps/steps_db_ops.py
+++ b/test/bdd/steps/steps_db_ops.py
@@ -18,13 +18,19 @@ from nominatim.tokenizer import factory as tokenizer_factory
 def check_database_integrity(context):
     """ Check some generic constraints on the tables.
     """
 def check_database_integrity(context):
     """ Check some generic constraints on the tables.
     """

-    # place_addressline should not have duplicate (place_id, address_place_id)
-    cur = context.db.cursor()
-    cur.execute("""SELECT count(*) FROM
-                    (SELECT place_id, address_place_id, count(*) as c
-                     FROM place_addressline GROUP BY place_id, address_place_id) x
-                   WHERE c > 1""")
-    assert cur.fetchone()[0] == 0, "Duplicates found in place_addressline"
+    with context.db.cursor() as cur:
+        # place_addressline should not have duplicate (place_id, address_place_id)
+        cur.execute("""SELECT count(*) FROM
+                        (SELECT place_id, address_place_id, count(*) as c
+                         FROM place_addressline GROUP BY place_id, address_place_id) x
+                       WHERE c > 1""")
+        assert cur.fetchone()[0] == 0, "Duplicates found in place_addressline"
+
+        # word table must not have empty word_tokens
+        if context.nominatim.tokenizer != 'legacy':
+            cur.execute("SELECT count(*) FROM word WHERE word_token = ''")
+            assert cur.fetchone()[0] == 0, "Empty word tokens found in word table"
+

 
 
 ################################ GIVEN ##################################
 
 
 ################################ GIVEN ##################################


@@ -93,13 +99,17 @@ def add_data_to_planet_ways(context):
 def import_and_index_data_from_place_table(context):
     """ Import data previously set up in the place table.
     """
 def import_and_index_data_from_place_table(context):
     """ Import data previously set up in the place table.
     """

-    context.nominatim.run_nominatim('refresh', '--functions')

     context.nominatim.run_nominatim('import', '--continue', 'load-data',
                                               '--index-noanalyse', '-q',
                                               '--offline')
 
     check_database_integrity(context)
 
     context.nominatim.run_nominatim('import', '--continue', 'load-data',
                                               '--index-noanalyse', '-q',
                                               '--offline')
 
     check_database_integrity(context)
 

+    # Remove the output of the input, when all was right. Otherwise it will be
+    # output when there are errors that had nothing to do with the import
+    # itself.
+    context.log_capture.buffer.clear()
+

 @when("updating places")
 def update_place_table(context):
     """ Update the place table with the given data. Also runs all triggers
 @when("updating places")
 def update_place_table(context):
     """ Update the place table with the given data. Also runs all triggers


@@ -108,11 +118,20 @@ def update_place_table(context):
     context.nominatim.run_nominatim('refresh', '--functions')
     with context.db.cursor() as cur:
         for row in context.table:
     context.nominatim.run_nominatim('refresh', '--functions')
     with context.db.cursor() as cur:
         for row in context.table:

-            PlaceColumn(context).add_row(row, False).db_insert(cur)
+            col = PlaceColumn(context).add_row(row, False)
+            col.db_delete(cur)
+            col.db_insert(cur)
+        cur.execute('SELECT flush_deleted_places()')

 
     context.nominatim.reindex_placex(context.db)
     check_database_integrity(context)
 
 
     context.nominatim.reindex_placex(context.db)
     check_database_integrity(context)
 

+    # Remove the output of the input, when all was right. Otherwise it will be
+    # output when there are errors that had nothing to do with the import
+    # itself.
+    context.log_capture.buffer.clear()
+
+

 @when("updating postcodes")
 def update_postcodes(context):
     """ Rerun the calculation of postcodes.
 @when("updating postcodes")
 def update_postcodes(context):
     """ Rerun the calculation of postcodes.


@@ -127,11 +146,18 @@ def delete_places(context, oids):
     """
     context.nominatim.run_nominatim('refresh', '--functions')
     with context.db.cursor() as cur:
     """
     context.nominatim.run_nominatim('refresh', '--functions')
     with context.db.cursor() as cur:

+        cur.execute('TRUNCATE place_to_be_deleted')

         for oid in oids.split(','):
             NominatimID(oid).query_osm_id(cur, 'DELETE FROM place WHERE {}')
         for oid in oids.split(','):
             NominatimID(oid).query_osm_id(cur, 'DELETE FROM place WHERE {}')

+        cur.execute('SELECT flush_deleted_places()')

 
     context.nominatim.reindex_placex(context.db)
 
 
     context.nominatim.reindex_placex(context.db)
 

+    # Remove the output of the input, when all was right. Otherwise it will be
+    # output when there are errors that had nothing to do with the import
+    # itself.
+    context.log_capture.buffer.clear()
+

 ################################ THEN ##################################
 
 @then("(?P<table>placex|place) contains(?P<exact> exactly)?")
 ################################ THEN ##################################
 
 @then("(?P<table>placex|place) contains(?P<exact> exactly)?")


@@ -164,7 +190,10 @@ def check_place_contents(context, table, exact):
 
         if exact:
             cur.execute('SELECT osm_type, osm_id, class from {}'.format(table))
 
         if exact:
             cur.execute('SELECT osm_type, osm_id, class from {}'.format(table))

-            assert expected_content == set([(r[0], r[1], r[2]) for r in cur])
+            actual = set([(r[0], r[1], r[2]) for r in cur])
+            assert expected_content == actual, \
+                   f"Missing entries: {expected_content - actual}\n" \
+                   f"Not expected in table: {actual - expected_content}"

 
 
 @then("(?P<table>placex|place) has no entry for (?P<oid>.*)")
 
 
 @then("(?P<table>placex|place) has no entry for (?P<oid>.*)")


@@ -349,6 +378,51 @@ def check_location_property_osmline(context, oid, neg):
 
             DBRow(oid, res, context).assert_row(row, ('start', 'end'))
 
 
             DBRow(oid, res, context).assert_row(row, ('start', 'end'))
 

-        assert not todo
+        assert not todo, f"Unmatched lines in table: {list(context.table[i] for i in todo)}"
+
+@then("location_property_osmline contains(?P<exact> exactly)?")
+def check_place_contents(context, exact):
+    """ Check contents of the interpolation table. Each row represents a table row
+        and all data must match. Data not present in the expected table, may
+        be arbitry. The rows are identified via the 'object' column which must
+        have an identifier of the form '<osm id>[:<startnumber>]'. When multiple
+        rows match (for example because 'startnumber' was left out and there are
+        multiple entries for the given OSM object) then all must match. All
+        expected rows are expected to be present with at least one database row.
+        When 'exactly' is given, there must not be additional rows in the database.
+    """
+    with context.db.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
+        expected_content = set()
+        for row in context.table:
+            if ':' in row['object']:
+                nid, start = row['object'].split(':', 2)
+                start = int(start)
+            else:
+                nid, start = row['object'], None
+
+            query = """SELECT *, ST_AsText(linegeo) as geomtxt,
+                              ST_GeometryType(linegeo) as geometrytype
+                       FROM location_property_osmline WHERE osm_id=%s"""
+
+            if ':' in row['object']:
+                query += ' and startnumber = %s'
+                params = [int(val) for val in row['object'].split(':', 2)]
+            else:
+                params = (int(row['object']), )

 
 

+            cur.execute(query, params)
+            assert cur.rowcount > 0, "No rows found for " + row['object']
+
+            for res in cur:
+                if exact:
+                    expected_content.add((res['osm_id'], res['startnumber']))
+
+                DBRow(nid, res, context).assert_row(row, ['object'])
+
+        if exact:
+            cur.execute('SELECT osm_id, startnumber from location_property_osmline')
+            actual = set([(r[0], r[1]) for r in cur])
+            assert expected_content == actual, \
+                   f"Missing entries: {expected_content - actual}\n" \
+                   f"Not expected in table: {actual - expected_content}"