Resolve "Introduce a new filter for when rules should be run based on the source ID of a building."

Closes #21 (closed)

\approve @laurens @ds @chengzhi

rulelib/abstractrule.py

@@ -18,7 +18,6 @@
 
 
 import abc
-from typing import Union
 
 import shapely
 
@@ -29,22 +28,26 @@ class AbstractRule(abc.ABC):
     structure that is or could be necessary for all rules.
 
     Args:
-        filter_string (str):
-            String representing the source_id_filter for the rule. May contain more than one filter
-            input, such as the `source_id` and the `geographic_filter_boundary`.
+        rule_source_ids (list):
+            A list of all valid source IDs for the rule. This is provided by the rule's XML file
+            if relevant.
+        geographic_filter_boundary (str):
+            Boundary polygon in WKT format defining the geographic boundary of the rule.
     """
 
-    def __init__(self, filter_string:str = None):
-
-        rule_source_id, geographic_filter_boundary = self.parse_filters(filter_string)
+    def __init__(
+        self,
+        rule_source_ids: list | None = None,
+        geographic_filter_boundary: str | None = None,
+    ):
 
         self.geographic_filter_boundary_geometry = None
         if geographic_filter_boundary is not None:
-            self.geographic_filter_boundary_geometry = self.read_geometry_from_wkt(
+            self.geographic_filter_boundary_geometry = shapely.from_wkt(
                 geographic_filter_boundary
             )
 
-        self.rule_source_id = rule_source_id
+        self.rule_source_ids = rule_source_ids
 
     @abc.abstractmethod
     def __call__(self, *args, **kwargs):
@@ -52,7 +55,7 @@ class AbstractRule(abc.ABC):
         The `__call__` method implements the core process of the rule.
         """
 
-    def parse_data(self, data: Union[str, bytes], source_file: str):
+    def parse_data(self, data: str | bytes, source_file: str):
         """
         The `parse_data` method can be used to parse files that are attached to the
         rule. As file types can widely vary, this method needs to be implemented for each
@@ -73,52 +76,18 @@ class AbstractRule(abc.ABC):
             + "please implement this class first."
         )
 
-    @staticmethod
-    def parse_filters(filter_string: str) -> tuple | None:
-        """
-        Parses the source_id_filter string input from the rule's XML file or the file attached to it.
-        Takes a single string separated by semicolons and splits it into the different filter
-        inputs it includes.
-
-        Args:
-            filter_string (str):
-                String with all the filter inputs, separated by semicolons.
-
-        Returns:
-            tuple | None:
-                A list of one or more strings each of which represents a different filter, or
-                None.
-        """
-
-        if filter_string is not None:
-            source_id, geographic_filter_boundary = filter_string.split(';')
-            return (int(source_id) if source_id != "" else None), geographic_filter_boundary
-
-        return None, None
-
-
-    @staticmethod
-    def read_geometry_from_wkt(geographic_filter_boundary):
-        """
-        Converts a WKT boundary into a Shapely polygon.
-
-        Args:
-            geographic_filter_boundary (str):
-                Boundary polygon in WKT format defining the geographic boundary of the rule.
-
-        Returns:
-            shapely.geometry.polygon.Polygon:
-                A Shapely polygon based on the WKT string geometry.
-        """
-
-        return shapely.from_wkt(geographic_filter_boundary)
-
-    def filter(self, longitude:float =None, latitude:float =None, source_id:int =None, *args,
-               **kwargs):
+    def filter(
+        self,
+        longitude: float | None = None,
+        latitude: float | None = None,
+        source_id: int | None = None,
+        *args,
+        **kwargs
+    ):
         """
         Applies a spatial filter to ensure that only buildings within the provided geographic
-        polygon are processed and a source filter to ensure source specific rules only run
-        for inputs with a matching source ID.
+        polygon are processed and a source filter to ensure source-specific rules only run
+        for buildings with a matching source ID.
 
         Args:
             longitude (float):
@@ -126,34 +95,24 @@ class AbstractRule(abc.ABC):
             latitude (float):
                 Latitude of the building being processed.
             source_id (int):
-                Integer representing the data source for which a rule is relevant.
+                Integer representing the building-source ID.
 
         Returns:
             bool:
-               True if all tests pass and False if at least one doesn't pass.
+               True if all tests pass and False if at least one does not pass.
         """
 
-        filter_list = []
-
-        # Geographic filter, passes if the inputs coordinates lie within the boundary geometry.
-        if self.geographic_filter_boundary_geometry is None:
-            filter_list.append(True)
-        elif self.geographic_filter_boundary_geometry.contains(
-            shapely.geometry.Point(longitude, latitude)
+        # Geographic filter, passes if the building's coordinates lie within the boundary
+        # geometry.
+        if (
+            self.geographic_filter_boundary_geometry is not None
+            and self.geographic_filter_boundary_geometry.disjoint(
+                shapely.geometry.Point(longitude, latitude)
+            )
         ):
-            filter_list.append(True)
-        else:
-            filter_list.append(False)
-
-        # Source ID filter, passes if the rules and the inputs source ID match.
-        if self.rule_source_id is None:
-            filter_list.append(True)
-        elif self.rule_source_id == source_id:
-            filter_list.append(True)
-        else:
-            filter_list.append(False)
-
-        if all(filter_list):
-            return True
-
-        return False
\ No newline at end of file
+            return False
+
+        # Source ID filter, passes if the rule's source ID and the building-source ID match.
+        if self.rule_source_ids is not None and source_id not in self.rule_source_ids:
+            return False
+        return True