chore: added keep_all_tags parameter with a test

quackosm/functions.py

@@ -21,6 +21,7 @@ def convert_pbf_to_gpq(
     tags_filter: Optional[Union[OsmTagsFilter, GroupedOsmTagsFilter]] = None,
     geometry_filter: Optional[BaseGeometry] = None,
     result_file_path: Optional[Union[str, Path]] = None,
+    keep_all_tags: bool = False,
     explode_tags: Optional[bool] = None,
     ignore_cache: bool = False,
     filter_osm_ids: Optional[list[str]] = None,
@@ -47,10 +48,14 @@ def convert_pbf_to_gpq(
         result_file_path (Union[str, Path], optional): Where to save
             the geoparquet file. If not provided, will be generated based on hashes
             from provided tags filter and geometry filter. Defaults to `None`.
+        keep_all_tags (bool, optional): Works only with the `tags_filter` parameter.
+            Whether to keep all tags related to the element, or return only those defined
+            in the `tags_filter`. When `True`, will override the optional grouping defined
+            in the `tags_filter`. Defaults to `False`.
         explode_tags (bool, optional): Whether to split tags into columns based on OSM tag keys.
-            If `None`, will be set based on `tags_filter` parameter. If no tags filter is provided,
-            then `explode_tags` will set to `False`, if there is tags filter it will set to `True`.
-            Defaults to `None`.
+            If `None`, will be set based on `tags_filter` and `keep_all_tags` parameters.
+            If there is tags filter defined and `keep_all_tags` is set to `False`, then it will
+            be set to `True`. Otherwise it will be set to `False`. Defaults to `None`.
         ignore_cache (bool, optional): Whether to ignore precalculated geoparquet files or not.
             Defaults to False.
         filter_osm_ids: (list[str], optional): List of OSM features ids to read from the file.
@@ -220,6 +225,7 @@ def convert_pbf_to_gpq(
     ).convert_pbf_to_gpq(
         pbf_path=pbf_path,
         result_file_path=result_file_path,
+        keep_all_tags=keep_all_tags,
         explode_tags=explode_tags,
         ignore_cache=ignore_cache,
         filter_osm_ids=filter_osm_ids,
@@ -230,6 +236,7 @@ def get_features_gdf(
     file_paths: Union[str, Path, Iterable[Union[str, Path]]],
     tags_filter: Optional[Union[OsmTagsFilter, GroupedOsmTagsFilter]] = None,
     geometry_filter: Optional[BaseGeometry] = None,
+    keep_all_tags: bool = False,
     explode_tags: Optional[bool] = None,
     ignore_cache: bool = False,
     filter_osm_ids: Optional[list[str]] = None,
@@ -257,10 +264,14 @@ def get_features_gdf(
             If `None`, handler will allow all of the tags to be parsed. Defaults to `None`.
         geometry_filter (BaseGeometry, optional): Region which can be used to filter only
             intersecting OSM objects. Defaults to `None`.
+        keep_all_tags (bool, optional): Works only with the `tags_filter` parameter.
+            Whether to keep all tags related to the element, or return only those defined
+            in the `tags_filter`. When `True`, will override the optional grouping defined
+            in the `tags_filter`. Defaults to `False`.
         explode_tags (bool, optional): Whether to split tags into columns based on OSM tag keys.
-            If `None`, will be set based on `tags_filter` parameter. If no tags filter is provided,
-            then `explode_tags` will set to `False`, if there is tags filter it will set to `True`.
-            Defaults to `None`.
+            If `None`, will be set based on `tags_filter` and `keep_all_tags` parameters.
+            If there is tags filter defined and `keep_all_tags` is set to `False`, then it will
+            be set to `True`. Otherwise it will be set to `False`. Defaults to `None`.
         ignore_cache: (bool, optional): Whether to ignore precalculated geoparquet files or not.
             Defaults to False.
         filter_osm_ids: (list[str], optional): List of OSM features ids to read from the file.
@@ -392,6 +403,7 @@ def get_features_gdf(
         osm_way_polygon_features_config=osm_way_polygon_features_config,
     ).get_features_gdf(
         file_paths=file_paths,
+        keep_all_tags=keep_all_tags,
         explode_tags=explode_tags,
         ignore_cache=ignore_cache,
         filter_osm_ids=filter_osm_ids,

quackosm/pbf_file_reader.py

@@ -142,6 +142,7 @@ def __init__(
     def get_features_gdf(
         self,
         file_paths: Union[str, Path, Iterable[Union[str, Path]]],
+        keep_all_tags: bool = False,
         explode_tags: Optional[bool] = None,
         ignore_cache: bool = False,
         filter_osm_ids: Optional[list[str]] = None,
@@ -155,10 +156,14 @@ def get_features_gdf(
         Args:
             file_paths (Union[str, Path, Iterable[Union[str, Path]]]):
                 Path or list of paths of `*.osm.pbf` files to be parsed.
+            keep_all_tags (bool, optional): Works only with the `tags_filter` parameter.
+                Whether to keep all tags related to the element, or return only those defined
+                in the `tags_filter`. When `True`, will override the optional grouping defined
+                in the `tags_filter`. Defaults to `False`.
             explode_tags (bool, optional): Whether to split tags into columns based on OSM tag keys.
-                If `None`, will be set based on `tags_filter` parameter.
-                If no tags filter is provided, then `explode_tags` will set to `False`,
-                if there is tags filter it will set to `True`. Defaults to `None`.
+                If `None`, will be set based on `tags_filter` and `keep_all_tags` parameters.
+                If there is tags filter defined and `keep_all_tags` is set to `False`, then it will
+                be set to `True`. Otherwise it will be set to `False`. Defaults to `None`.
             ignore_cache: (bool, optional): Whether to ignore precalculated geoparquet files or not.
                 Defaults to False.
             filter_osm_ids: (list[str], optional): List of OSM features ids to read from the file.
@@ -203,6 +208,7 @@ def convert_pbf_to_gpq(
         self,
         pbf_path: Union[str, Path],
         result_file_path: Optional[Union[str, Path]] = None,
+        keep_all_tags: bool = False,
         explode_tags: Optional[bool] = None,
         ignore_cache: bool = False,
         filter_osm_ids: Optional[list[str]] = None,
@@ -215,10 +221,14 @@ def convert_pbf_to_gpq(
             result_file_path (Union[str, Path], optional): Where to save
                 the geoparquet file. If not provided, will be generated based on hashes
                 from provided tags filter and geometry filter. Defaults to `None`.
+            keep_all_tags (bool, optional): Works only with the `tags_filter` parameter.
+                Whether to keep all tags related to the element, or return only those defined
+                in the `tags_filter`. When `True`, will override the optional grouping defined
+                in the `tags_filter`. Defaults to `False`.
             explode_tags (bool, optional): Whether to split tags into columns based on OSM tag keys.
-                If `None`, will be set based on `tags_filter` parameter.
-                If no tags filter is provided, then `explode_tags` will set to `False`,
-                if there is tags filter it will set to `True`. Defaults to `None`.
+                If `None`, will be set based on `tags_filter` and `keep_all_tags` parameters.
+                If there is tags filter defined and `keep_all_tags` is set to `False`, then it will
+                be set to `True`. Otherwise it will be set to `False`. Defaults to `None`.
             ignore_cache (bool, optional): Whether to ignore precalculated geoparquet files or not.
                 Defaults to False.
             filter_osm_ids: (list[str], optional): List of OSM features ids to read from the file.

tests/base/test_pbf_file_reader.py

@@ -25,7 +25,7 @@
 from srai.loaders.osm_loaders.filters import GEOFABRIK_LAYERS, HEX2VEC_FILTER
 
 from quackosm._constants import FEATURES_INDEX
-from quackosm._osm_tags_filters import OsmTagsFilter
+from quackosm._osm_tags_filters import GroupedOsmTagsFilter, OsmTagsFilter
 from quackosm.pbf_file_reader import PbfFileReader
 
 ut = TestCase()
@@ -140,6 +140,46 @@ def test_pbf_reader_features_ids_filtering(filter_osm_ids: list[str], expected_r
     assert len(features_gdf) == expected_result_length
 
 
+@pytest.mark.parametrize(  # type: ignore
+    "filter_osm_id,osm_tags_filter,keep_all_tags,expected_tags_keys",
+    [
+        ("way/389888402", {"building": "apartments"}, False, ["building"]),
+        (
+            "way/389888402",
+            {"building": "apartments"},
+            True,
+            [
+                "addr:city",
+                "addr:country",
+                "addr:housenumber",
+                "addr:postcode",
+                "addr:street",
+                "building",
+                "building:levels",
+            ],
+        ),
+    ],
+)
+def test_pbf_reader_proper_tags_reading(
+    filter_osm_id: str,
+    osm_tags_filter: Union[OsmTagsFilter, GroupedOsmTagsFilter],
+    keep_all_tags: bool,
+    expected_tags_keys: list[str],
+):
+    """Test proper tags tags reading in `PbfFileReader`."""
+    file_name = "monaco.osm.pbf"
+    features_gdf = PbfFileReader(tags_filter=osm_tags_filter).get_features_gdf(
+        file_paths=[Path(__file__).parent.parent / "test_files" / file_name],
+        ignore_cache=True,
+        filter_osm_ids=[filter_osm_id],
+        explode_tags=False,
+        keep_all_tags=keep_all_tags,
+    )
+    assert len(features_gdf) == 1
+    returned_tags_keys = list(features_gdf.iloc[0].tags.keys())
+    ut.assertListEqual(returned_tags_keys, expected_tags_keys)
+
+
 # Copyright (C) 2011 by Hong Minhee <http://dahlia.kr/>,
 #                       Robert Kajic <http://github.com/kajic>
 # Copyright (C) 2020 by Salesforce.com, Inc