Merge pull request #29 from OnroerendErfgoed/actions_update

actions update
OnroerendErfgoed · Jun 26, 2024 · 6c36a84 · 6c36a84
2 parents f9028d5 + 1bd414e
commit 6c36a84
Show file tree

Hide file tree

Showing 26 changed files with 794 additions and 404 deletions.
diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml
@@ -6,35 +6,45 @@ name: Python package
 on:
   push:
     branches: [ "main" ]
+
   pull_request:
     branches: [ "main" ]
 
 jobs:
   build:
 
     runs-on: ubuntu-latest
+
     strategy:
-      fail-fast: false
       matrix:
         python-version: ["3.9", "3.10", "3.11", "3.12"]
 
     steps:
     - uses: actions/checkout@v4
+
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v3
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
+
     - name: Install dependencies
       run: |
-        python -m pip install --upgrade pip
-        python -m pip install flake8 pytest
-        if [ -f requirements-dev.txt ]; then pip install -r requirements-dev.txt; fi
+        pip install --upgrade pip
+        pip install pip-tools
+        pip-sync requirements-dev.txt
+
+    - name: Black code style check
+      run: |
+        # stop the build if there are files that need to reformatted
+        black --check . 
+
     - name: Lint with flake8
       run: |
         # stop the build if there are Python syntax errors or undefined names
         flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
         # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
         flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
+
     - name: Test with pytest
       run: |
         pytest
diff --git a/brdr/aligner.py b/brdr/aligner.py
@@ -42,9 +42,16 @@
 from brdr.geometry_utils import safe_intersection
 from brdr.geometry_utils import safe_symmetric_difference
 from brdr.geometry_utils import safe_union
-from brdr.utils import diffs_from_dict_series, get_breakpoints_zerostreak, \
-    filter_resulting_series_by_key, get_collection, geojson_tuple_from_series, write_geojson, \
-    merge_geometries_by_theme_id, geojson_from_dict
+from brdr.utils import (
+    diffs_from_dict_series,
+    get_breakpoints_zerostreak,
+    filter_resulting_series_by_key,
+    get_collection,
+    geojson_tuple_from_series,
+    write_geojson,
+    merge_geometries_by_theme_id,
+    geojson_from_dict,
+)
 
 logging.basicConfig(
     level=logging.INFO, format="%(asctime)s - %(message)s", datefmt="%d-%b-%y %H:%M:%S"
@@ -69,7 +76,7 @@ def __init__(
         relevant_distance=1,
         threshold_overlap_percentage=50,
         od_strategy=OpenbaarDomeinStrategy.SNAP_SINGLE_SIDE,
-        crs=DEFAULT_CRS
+        crs=DEFAULT_CRS,
     ):
         """
         Initializes the Aligner object
@@ -104,16 +111,22 @@ def __init__(
         # reference
         self.reference_input = None  # to save the initially loaded geojson
 
-        self.name_reference_id = "ref_identifier" # name of the identifier-field of the reference data (id has to be unique,f.e CAPAKEY for GRB-parcels)
+        self.name_reference_id = "ref_identifier"  # name of the identifier-field of the reference data (id has to be unique,f.e CAPAKEY for GRB-parcels)
         self.dict_reference = {}  # dictionary to store all reference geometries
-        self.reference_union = None # to save a unioned geometry of all reference polygons; needed for calculation in most OD-strategies
+        self.reference_union = None  # to save a unioned geometry of all reference polygons; needed for calculation in most OD-strategies
 
         # output-dictionaries (when processing dict_thematic)
         self.dict_result = None  # dictionary to save resulting geometries
         self.dict_result_diff = None  # dictionary to save global resulting differences
-        self.dict_result_diff_plus = None  # dictionary to save positive resulting differences
-        self.dict_result_diff_min = None  # dictionary to save negative resulting differences
-        self.dict_relevant_intersection = None  # dictionary to save relevant_intersections
+        self.dict_result_diff_plus = (
+            None  # dictionary to save positive resulting differences
+        )
+        self.dict_result_diff_min = (
+            None  # dictionary to save negative resulting differences
+        )
+        self.dict_relevant_intersection = (
+            None  # dictionary to save relevant_intersections
+        )
         self.dict_relevant_difference = None  # dictionary to save relevant_differences
 
         # Coordinate reference system
@@ -445,7 +458,9 @@ def process_dict_thematic(
         self.dict_result_diff = merge_geometries_by_theme_id(dict_result_diff)
         self.dict_result_diff_plus = merge_geometries_by_theme_id(dict_result_diff_plus)
         self.dict_result_diff_min = merge_geometries_by_theme_id(dict_result_diff_min)
-        self.dict_relevant_intersection = merge_geometries_by_theme_id(dict_relevant_intersection)
+        self.dict_relevant_intersection = merge_geometries_by_theme_id(
+            dict_relevant_intersection
+        )
         self.dict_relevant_difference = merge_geometries_by_theme_id(dict_relevant_diff)
         self.feedback_info("thematic dictionary processed")
         return (
@@ -458,23 +473,23 @@ def process_dict_thematic(
         )
 
     def predictor(
-            self,
-            relevant_distances=np.arange(0, 300, 10, dtype=int)/100,
-            od_strategy=OpenbaarDomeinStrategy.SNAP_SINGLE_SIDE,
-            treshold_overlap_percentage=50,
+        self,
+        relevant_distances=np.arange(0, 300, 10, dtype=int) / 100,
+        od_strategy=OpenbaarDomeinStrategy.SNAP_SINGLE_SIDE,
+        treshold_overlap_percentage=50,
     ):
         """
         Predicts the 'most interesting' relevant distances for changes in thematic elements based on a distance series.
 
-        This function analyzes a set of thematic geometries (`self.dict_thematic`) to identify potentially 
+        This function analyzes a set of thematic geometries (`self.dict_thematic`) to identify potentially
         interesting distances where changes occur. It performs the following steps:
 
         1. **Process Distance Series:**
             - Calculates a series of results for different distances specified by `relevant_distances`.
             - This calculation might involve functions like `self.process_series` (implementation details likely depend on your specific code).
 
         2. **Calculate Difference Metrics:**
-            - Analyzes the results from the distance series to compute difference metrics 
+            - Analyzes the results from the distance series to compute difference metrics
               between thematic elements at each distance (using `diffs_from_dict_series`).
 
         3. **Identify Breakpoints and Zero-Streaks:**
@@ -506,17 +521,25 @@ def predictor(
         """
         dict_predicted = {}
         for key in self.dict_thematic.keys():
-            dict_predicted[key]={}
-        dict_series = self.process_series(relevant_distances=relevant_distances,od_strategy=od_strategy,treshold_overlap_percentage=treshold_overlap_percentage)
+            dict_predicted[key] = {}
+        dict_series = self.process_series(
+            relevant_distances=relevant_distances,
+            od_strategy=od_strategy,
+            treshold_overlap_percentage=treshold_overlap_percentage,
+        )
         diffs = diffs_from_dict_series(dict_series, self.dict_thematic)
         for key in diffs:
             if len(diffs[key]) == len(relevant_distances):
                 lst_diffs = list(diffs[key].values())
-                breakpoints, zero_streaks = get_breakpoints_zerostreak(relevant_distances, lst_diffs)
+                breakpoints, zero_streaks = get_breakpoints_zerostreak(
+                    relevant_distances, lst_diffs
+                )
                 logging.debug(str(key))
                 for zs in zero_streaks:
                     dict_predicted[key][zs[0]] = dict_series[zs[0]]
-                dict_predicted[key] = filter_resulting_series_by_key(dict_predicted[key],key)
+                dict_predicted[key] = filter_resulting_series_by_key(
+                    dict_predicted[key], key
+                )
         return dict_predicted, diffs
 
     def process_series(
@@ -549,7 +572,7 @@ def process_series(
         self.feedback_debug("Process series" + str(relevant_distances))
         self.od_strategy = od_strategy
         self.threshold_overlap_percentage = treshold_overlap_percentage
-        #self._prepare_thematic_data() #not necessary? Assumed that dict_thematic is already loaded
+        # self._prepare_thematic_data() #not necessary? Assumed that dict_thematic is already loaded
         dict_series = {}
         for s in relevant_distances:
             self.feedback_info(
@@ -649,7 +672,7 @@ def get_last_version_date(self, geometry, grb_type=GRBType.ADP):
             "https://geo.api.vlaanderen.be/GRB/ogc/features/collections/"
             + grb_type
             + "/items?"
-            "limit=" + str(limit) + "&crs=" + crs + "&bbox-crs=" + crs +"&bbox=" + bbox
+            "limit=" + str(limit) + "&crs=" + crs + "&bbox-crs=" + crs + "&bbox=" + bbox
         )
         update_dates = []
         collection = get_collection(actual_url, limit)
@@ -664,48 +687,60 @@ def get_results_as_dict(self):
         """
         get a dict-tuple of the results
         """
-        return (self.dict_result, self.dict_result_diff, self.dict_result_diff_plus, self.dict_result_diff_min,
-                self.dict_relevant_intersection, self.dict_relevant_difference)
+        return (
+            self.dict_result,
+            self.dict_result_diff,
+            self.dict_result_diff_plus,
+            self.dict_result_diff_min,
+            self.dict_relevant_intersection,
+            self.dict_relevant_difference,
+        )
 
     def get_results_as_geojson(self):
         """
         get a geojson-tuple of the results
         """
-        return geojson_tuple_from_series({self.relevant_distance: self.get_results_as_dict()}, self.CRS,
-                                         self.name_thematic_id)
+        return geojson_tuple_from_series(
+            {self.relevant_distance: self.get_results_as_dict()},
+            self.CRS,
+            self.name_thematic_id,
+        )
 
     def get_reference_as_geojson(self):
         """
         get a geojson of the reference polygons
         """
-        return geojson_from_dict(self.dict_reference, self.CRS, self.name_reference_id,geom_attributes=False)
+        return geojson_from_dict(
+            self.dict_reference, self.CRS, self.name_reference_id, geom_attributes=False
+        )
+
     def export_results(self, path, multi_to_single=True):
         """
-            Exports analysis results as GeoJSON files.
-
-            This function exports 6 GeoJSON files containing the analysis results to the specified `path`.
-
-            Args:
-                path (str): The path to the directory where the GeoJSON files will be saved.
-                multi_to_single (bool, optional): If True (default), converts MultiPolygon geometries to single Polygons
-                                                    in the exported GeoJSON files. This can be useful for visualization purposes.
-
-            Details of exported files:
-                - result.geojson: Contains the original thematic data from `self.dict_result`.
-                - result_diff.geojson: Contains the difference between the original and predicted data from `self.dict_result_diff`.
-                - result_diff_plus.geojson: Contains results for areas that are added (increased area).
-                - result_diff_min.geojson: Contains results for areas that are removed (decreased area).
-                - result_relevant_intersection.geojson: Contains the areas with relevant intersection that has to be included in the result.
-                - result_relevant_difference.geojson: Contains the areas with relevant difference that has to be excluded from the result.
-            """
+        Exports analysis results as GeoJSON files.
+
+        This function exports 6 GeoJSON files containing the analysis results to the specified `path`.
+
+        Args:
+            path (str): The path to the directory where the GeoJSON files will be saved.
+            multi_to_single (bool, optional): If True (default), converts MultiPolygon geometries to single Polygons
+                                                in the exported GeoJSON files. This can be useful for visualization purposes.
+
+        Details of exported files:
+            - result.geojson: Contains the original thematic data from `self.dict_result`.
+            - result_diff.geojson: Contains the difference between the original and predicted data from `self.dict_result_diff`.
+            - result_diff_plus.geojson: Contains results for areas that are added (increased area).
+            - result_diff_min.geojson: Contains results for areas that are removed (decreased area).
+            - result_relevant_intersection.geojson: Contains the areas with relevant intersection that has to be included in the result.
+            - result_relevant_difference.geojson: Contains the areas with relevant difference that has to be excluded from the result.
+        """
         fcs = self.get_results_as_geojson()
         resultnames = [
             "result.geojson",
             "result_diff.geojson",
             "result_diff_plus.geojson",
             "result_diff_min.geojson",
             "result_relevant_intersection.geojson",
-            "result_relevant_difference.geojson"
+            "result_relevant_difference.geojson",
         ]
         for count, fc in enumerate(fcs):
             write_geojson(os.path.join(path, resultnames[count]), fcs[count])
@@ -1007,14 +1042,14 @@ def _calculate_geom_by_intersection_and_reference(
                     self.buffer_distance(),
                 ),
             )
-            #TODO BEGIN: experimental fix - check if it is ok in all cases?
-            #when calculating for OD, we create a 'virtual parcel'. When calculating this virtual parcel, it is buffered to take outer boundaries into account.
-            #This results in a side-effect that there are extra non-logical parts included in the result. The function below tries to exclude these non-logica parts.
+            # TODO BEGIN: experimental fix - check if it is ok in all cases?
+            # when calculating for OD, we create a 'virtual parcel'. When calculating this virtual parcel, it is buffered to take outer boundaries into account.
+            # This results in a side-effect that there are extra non-logical parts included in the result. The function below tries to exclude these non-logica parts.
             # see eo_id 206363 with relevant distance=0.2m and SNAP_ALL_SIDE
             if is_openbaar_domein:
-                #geom = buffer_neg_pos(geom, self.buffer_distance())
-                geom = self.get_relevant_polygons_from_geom (geom)
-            #TODO END
+                # geom = buffer_neg_pos(geom, self.buffer_distance())
+                geom = self.get_relevant_polygons_from_geom(geom)
+            # TODO END
         elif (
             not geom_relevant_intersection.is_empty
             and geom_relevant_difference.is_empty
@@ -1045,7 +1080,6 @@ def _calculate_geom_by_intersection_and_reference(
                 geom = geom_relevant_intersection  # (=empty geometry)
         return geom, geom_relevant_intersection, geom_relevant_difference
 
-
     def get_relevant_polygons_from_geom(self, geom):
         """
         Get only the relevant parts (polygon) from a geometry.
@@ -1058,12 +1092,12 @@ def get_relevant_polygons_from_geom(self, geom):
             geom = make_valid(unary_union(geom))
             # Create a GeometryCollection from the input geometry.
             geometry_collection = GeometryCollection(geom)
-            array=[]
+            array = []
             for g in geometry_collection.geoms:
                 # Ensure each sub-geometry is valid.
                 g = make_valid(g)
                 if str(g.geom_type) in ["Polygon", "MultiPolygon"]:
-                    relevant_geom = buffer_neg(g,self.buffer_distance())
+                    relevant_geom = buffer_neg(g, self.buffer_distance())
                     if relevant_geom != None and not relevant_geom.is_empty:
                         array.append(g)
         return make_valid(unary_union(array))

diff --git a/brdr/constants.py b/brdr/constants.py
@@ -32,8 +32,8 @@
 # Limit used when extracting features by URL, using the feature API (fe from GRB)
 DOWNLOAD_LIMIT = 10000
 
-#default CRS:
-DEFAULT_CRS= "EPSG:31370"
+# default CRS:
+DEFAULT_CRS = "EPSG:31370"
 
-#MULTI_SINGLE_ID_SEPARATOR #separator to split multipolygon_ids to single polygons
-MULTI_SINGLE_ID_SEPARATOR = '*$*'
+# MULTI_SINGLE_ID_SEPARATOR #separator to split multipolygon_ids to single polygons
+MULTI_SINGLE_ID_SEPARATOR = "*$*"
diff --git a/brdr/geometry_utils.py b/brdr/geometry_utils.py
@@ -168,7 +168,9 @@ def safe_union(geom_a: BaseGeometry, geom_b: BaseGeometry) -> BaseGeometry:
         geom = union(geom_a, geom_b)
     except GEOSException:
         try:
-            logging.warning("union_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt)
+            logging.warning(
+                "union_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt
+            )
             geom = union(buffer(geom_a, 0.0000001), buffer(geom_b, 0.0000001))
         except Exception:
             logging.error("error: empty geometry returned")
@@ -206,7 +208,9 @@ def safe_intersection(geom_a: BaseGeometry, geom_b: BaseGeometry) -> BaseGeometr
         geom = intersection(geom_a, geom_b)
     except GEOSException:
         try:
-            logging.warning("intersection_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt)
+            logging.warning(
+                "intersection_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt
+            )
             geom = intersection(buffer(geom_a, 0.0000001), buffer(geom_b, 0.0000001))
         except Exception:
             logging.error("error: empty geometry returned")
@@ -244,7 +248,9 @@ def safe_difference(geom_a, geom_b):
         geom = difference(geom_a, geom_b)
     except GEOSException:
         try:
-            logging.warning("difference_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt)
+            logging.warning(
+                "difference_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt
+            )
             geom = difference(buffer(geom_a, 0.0000001), buffer(geom_b, 0.0000001))
         except Exception:
             logging.error("error: empty geometry returned")
@@ -282,7 +288,12 @@ def safe_symmetric_difference(geom_a, geom_b):
         geom = symmetric_difference(geom_a, geom_b)
     except GEOSException:
         try:
-            logging.warning("symmetric_difference_error for geoms:" + geom_a.wkt + " and " + geom_b.wkt)
+            logging.warning(
+                "symmetric_difference_error for geoms:"
+                + geom_a.wkt
+                + " and "
+                + geom_b.wkt
+            )
             geom = symmetric_difference(
                 buffer(geom_a, 0.0000001), buffer(geom_b, 0.0000001)
             )
@@ -316,7 +327,7 @@ def grid_bounds(geom: BaseGeometry, delta: float):
         nx = 2
     if ny < 2:
         ny = 2
-    gx, gy = np.linspace(min_x, max_x, nx+1), np.linspace(min_y, max_y, ny+1)
+    gx, gy = np.linspace(min_x, max_x, nx + 1), np.linspace(min_y, max_y, ny + 1)
     grid = []
     for i in range(len(gx) - 1):
         for j in range(len(gy) - 1):