Adding a dialog to choose the column that is used as join when mergin…

…g tables. (#23)

* GUI question to choose the column on which the table join will be based

* Adding little improvements

* Fix broken CI pipeline

othello/gis/util.py

@@ -2,6 +2,7 @@
 import functools
 from typing import Dict, List
 
+import pandas
 from geopandas import GeoDataFrame
 
 from othello.gis import io
@@ -19,7 +20,17 @@ def make_dataframe_with_common_columns(dfs: List[GeoDataFrame], common_columns:
     return df
 
 
-def add_weighted_columns_to_dataframe(df: GeoDataFrame, criteria_information: List[Dict]) -> GeoDataFrame:
+def add_weighted_columns_to_dataframe(df: GeoDataFrame, criteria_information: List[Dict], join_on: str) -> GeoDataFrame:
+    """Add the weighted criteria to the 'df' dataframe by joining data on the join_on column name.
+
+    Args:
+        df: Base GeoDataFrame
+        criteria_information: Criteria data that allows to retrieve the exact column from a table from a layer from aGDB file
+        join_on: Column name on which the table join will be based
+
+    Returns:
+        GeoDataFrame with the added criteria and the final score
+    """
     weighted_columns = []
 
     for criterion_information in criteria_information:
@@ -29,11 +40,16 @@ def add_weighted_columns_to_dataframe(df: GeoDataFrame, criteria_information: Li
         weight = criterion_information['weight']
         criterion_name = criterion_information['criterion_name']
 
-        criterion_geoseries = io.read(filepath, layer=layer)[field]
-        df[criterion_name + '_np'] = criterion_geoseries
-        df[criterion_name + '_p'] = weight * criterion_geoseries
+        criterion_df = io.read(filepath, layer=layer)[[join_on, field]]
+        criterion_df.columns = [join_on, criterion_name + '_np']
+        criterion_df[criterion_name + '_p'] = weight * criterion_df[criterion_name + '_np']
+
+        df = pandas.merge(left=df, right=criterion_df, on=join_on)
         weighted_columns.append(criterion_name + '_p')
 
+        # df[criterion_name + '_np'] = criterion_df
+        # df[criterion_name + '_p'] = weight * criterion_df
+
     # Final score in a new series
     df['FinalScore'] = [0 for _ in range(len(df))]
     for column in weighted_columns:

othello/ui/aggregate.py

@@ -2,7 +2,7 @@
 
 import geopandas
 from PySide2 import QtCore, QtWidgets
-from PySide2.QtWidgets import QMessageBox
+from PySide2.QtWidgets import QInputDialog, QMessageBox
 
 from othello import gis
 from othello.ui import errors
@@ -35,7 +35,7 @@ def __init__(self, parent):
         self.table.setGeometry(QtCore.QRect(20, 80, 731, 400))
         self.table.setRowCount(0)
         self.table.setColumnCount(5)
-        self.table.setHorizontalHeaderLabels(['Criterion', 'Layer', 'Field', 'Weight [0-1]', 'criterion name'])
+        self.table.setHorizontalHeaderLabels(['Criterion', 'Layer', 'Field', 'Weight [0-1]', 'Criterion name'])
 
         header = self.table.horizontalHeader()
         header.setSectionResizeMode(0, QtWidgets.QHeaderView.Stretch)
@@ -71,10 +71,23 @@ def aggregate(self, to_new_file: bool, force_normalized_weights: bool = True):
                 self.assert_weights_are_normalized()
 
             common_columns = gis.util.find_common_columns(self.dfs)
+
+            # Choose the column on which the table join will be based
+            join_on_column, has_not_failed = QInputDialog.getItem(
+                self,
+                'Question',
+                'Choose the column on which the table join will be based.',
+                common_columns,
+                0,
+                False
+            )
+            if not (has_not_failed and join_on_column):
+                return
+
             df = gis.util.make_dataframe_with_common_columns(self.dfs, common_columns)
 
             if to_new_file:
-                filepath = QtWidgets.QFileDialog.getSaveFileName(self)[0]
+                filepath = QtWidgets.QFileDialog.getSaveFileName(self, 'Save to file', '', 'GDB (*.gdb)')[0]
             else:
                 filepath = QtWidgets.QFileDialog.getExistingDirectory(self)
 
@@ -85,7 +98,7 @@ def aggregate(self, to_new_file: bool, force_normalized_weights: bool = True):
                 popup.show()
                 return
 
-            df = self.add_weighted_columns(df)
+            df = self.add_weighted_columns(df, join_on_column)
             gis.io.write(df, filepath, layer=f'FinalLayer-{datetime.now().strftime("%Y-%m-%d-%H:%M:%S")}')
 
             popup = Popup(f'The file "{filepath}" have been written.', self)
@@ -154,7 +167,7 @@ def assert_no_duplicates_in_criterion_names(self):
 
             criterion_names.add(criterion_name)
 
-    def add_weighted_columns(self, df: geopandas.GeoDataFrame) -> geopandas.GeoDataFrame:
+    def add_weighted_columns(self, df: geopandas.GeoDataFrame, join_on: str) -> geopandas.GeoDataFrame:
         criteria_information = []
 
         for row_index in range(self.table.rowCount()):
@@ -166,7 +179,7 @@ def add_weighted_columns(self, df: geopandas.GeoDataFrame) -> geopandas.GeoDataF
                 'criterion_name': self.table.item(row_index, 4).text(),
             })
 
-        df = gis.util.add_weighted_columns_to_dataframe(df, criteria_information)
+        df = gis.util.add_weighted_columns_to_dataframe(df, criteria_information, join_on=join_on)
 
         return df
 

requirements.txt

@@ -7,9 +7,9 @@ click-plugins==1.1.1
 cligj==0.7.1
 colorama==0.4.4
 cx-Freeze==6.5.3
-Fiona @ https://download.lfd.uci.edu/pythonlibs/r4tycu3t/Fiona-1.8.20-cp38-cp38-win_amd64.whl
+Fiona @ https://download.lfd.uci.edu/pythonlibs/y2rycu7g/Fiona-1.8.20-cp38-cp38-win_amd64.whl
 future==0.18.2
-GDAL @ https://download.lfd.uci.edu/pythonlibs/r4tycu3t/GDAL-3.3.1-cp38-cp38-win_amd64.whl
+GDAL @ https://download.lfd.uci.edu/pythonlibs/y2rycu7g/GDAL-3.3.1-cp38-cp38-win_amd64.whl
 geopandas==0.9.0
 importlib-metadata==3.7.3
 iniconfig==1.1.1
@@ -21,14 +21,14 @@ pefile==2019.4.18
 pluggy==0.13.1
 py==1.10.0
 pyparsing==2.4.7
-pyproj @ https://download.lfd.uci.edu/pythonlibs/r4tycu3t/pyproj-3.0.1-cp38-cp38-win_amd64.whl
+pyproj==3.0.1
 PySide2==5.15.2
 pytest==6.2.2
 python-dateutil==2.8.1
 pytz==2021.1
 pywin32-ctypes==0.2.0
-Rtree @ https://download.lfd.uci.edu/pythonlibs/r4tycu3t/Rtree-0.9.7-cp38-cp38-win_amd64.whl
-Shapely @ https://download.lfd.uci.edu/pythonlibs/r4tycu3t/Shapely-1.7.1-cp38-cp38-win_amd64.whl
+Rtree==0.9.7
+Shapely==1.7.1
 shiboken2==5.15.2
 six==1.15.0
 toml==0.10.2

tests/gis/test_util.py

@@ -48,7 +48,7 @@ def test_make_dataframe_with_common_columns(self, dfs, common_columns, expected_
 
         assert list(result.columns) == expected_columns
 
-    @pytest.mark.parametrize('df, criteria_info, expected_first_criterion_score, expected_second_criterion_score, expected_first_final_score', [(
+    @pytest.mark.parametrize('df, criteria_info, join_on_column, expected_first_criterion_score, expected_second_criterion_score, expected_first_final_score', [(
             geopandas.read_file(GDB_FILEPATH, driver='FileGDB', layer='ArretsTEST_ON'),
             [{
                 'filepath': GDB_FILEPATH,
@@ -63,12 +63,13 @@ def test_make_dataframe_with_common_columns(self, dfs, common_columns, expected_
                 'weight': .5,
                 'criterion_name': 'another-criterion',
             }],
+            'ID',
             (8, 8 * .5),
             (-46.67, -46.67 * .5),
             8 * .5 - 46.67 * .5,
     )])
-    def test_add_weighted_columns_to_dataframe(self, df, criteria_info, expected_first_criterion_score, expected_second_criterion_score, expected_first_final_score):
-        result = gis.util.add_weighted_columns_to_dataframe(df, criteria_info)
+    def test_add_weighted_columns_to_dataframe(self, df, criteria_info, join_on_column, expected_first_criterion_score, expected_second_criterion_score, expected_first_final_score):
+        result = gis.util.add_weighted_columns_to_dataframe(df, criteria_info, join_on_column)
 
         assert 'FinalScore' in result.columns
         assert result['a-criterion_np'][0] == expected_first_criterion_score[0]