Merge branch 'dev' into f-fix-windows-unit-tests

src/caosadvancedtools/table_json_conversion/convert.py

@@ -478,6 +478,12 @@ class XLSXConverter:
             if isinstance(value, datetime.date) and (
                     {'type': 'string', 'format': 'date'} in subschema["anyOf"]):
                 return value
+        # booleans might be retrieved as an integer or formula
+        if subschema.get('type') == 'boolean':
+            if value == 0 or isinstance(value, str) and '=false()' == value.lower():
+                value = False
+            if value == 1 or isinstance(value, str) and '=true()' == value.lower():
+                value = True
         jsonschema.validate(value, subschema)
 
         # Finally: convert to target type

unittests/table_json_conversion/data/simple_data_booleans.json

+{
+  "Training": [
+    {
+      "date": "2023-01-01",
+      "url": "www.indiscale.com",
+      "coach": [
+        {
+          "family_name": "Sky",
+          "given_name": "Max",
+          "Organisation": "ECB"
+        },
+        {
+          "family_name": "Sky",
+          "given_name": "Min",
+          "Organisation": "ECB"
+        }
+      ],
+      "supervisor": {
+        "family_name": "Steve",
+        "given_name": "Stevie",
+              "Organisation": "IMF"
+      },
+      "duration": 1.0,
+      "participants": 1,
+      "subjects": ["Math", "Physics"],
+      "remote": false
+    },
+    {
+      "date": "2023-01-02",
+      "url": "www.indiscale.com",
+      "supervisor": {
+        "family_name": "Steve",
+        "given_name": "Stevie",
+              "Organisation": "IMF"
+      },
+      "duration": 1.0,
+      "participants": 1,
+      "subjects": ["Math", "Physics"],
+      "remote": true
+    }
+  ],
+  "Person": [{
+    "family_name": "Steve",
+    "given_name": "Stevie",
+    "Organisation": "IMF"
+  }]
+}

unittests/table_json_conversion/test_read_xlsx.py

@@ -26,6 +26,7 @@ import os
 import re
 
 from types import SimpleNamespace
+from typing import Optional
 
 import jsonschema
 import pytest
@@ -41,7 +42,7 @@ def rfp(*pathcomponents):
 
 
 def convert_and_compare(xlsx_file: str, schema_file: str, known_good_file: str,
-                        known_good_data: dict = None, strict: bool = False,
+                        known_good_data: Optional[dict] = None, strict: bool = False,
                         validate: bool = True) -> dict:
     """Convert an XLSX file and compare to a known result.
 
@@ -77,6 +78,9 @@ def test_conversions():
                         schema_file=rfp("data/multiple_choice_schema.json"),
                         known_good_file=rfp("data/multiple_choice_data.json"),
                         strict=True)
+    convert_and_compare(xlsx_file=rfp("data/simple_data_booleans.xlsx"),
+                        schema_file=rfp("data/simple_schema.json"),
+                        known_good_file=rfp("data/simple_data_booleans.json"))
 
     with open(rfp("data/simple_data.json"), encoding="utf-8") as myfile:
         expected_datetime = json.load(myfile)
@@ -126,29 +130,40 @@ def test_error_table():
     assert "'There is no entry in the schema" in str(caught.value)
     assert "'Not an enum' is not one of [" in str(caught.value)
     # Correct Locations
+    matches = set()
     for line in str(caught.value).split('\n'):
         if "'Not a num' is not of type 'number'" in line:
             assert "J7" in line
+            matches.add("J7")
         if "'Yes a number?' is not of type 'number'" in line:
             assert "J8" in line
+            matches.add("J8")
         if "1.5 is not of type 'integer'" in line:
             assert "K7" in line
+            matches.add("K7")
         if "1.2345 is not of type 'integer'" in line:
             assert "K8" in line
+            matches.add("K8")
         if "'There is no entry in the schema" in line:
             assert "Column M" in line
+            matches.add("Col M")
         if "'Not an enum' is not one of [" in line:
             assert "G8" in line
+            matches.add("K8")
+        # The next two tests could potentially be removed in the future, once we evaluate formulas.
+        if "'=NOT(FALSE())' is not of type 'boolean'" in line:
+            assert "L9" in line
+            matches.add("L9")
+        if "'=NOT(TRUE())' is not of type 'boolean'" in line:
+            assert "L10" in line
+            matches.add("L10")
+    assert matches == {"J7", "J8", "K7", "K8", "Col M", "K8", "L9", "L10"}
+
     # No additional errors
     assert str(caught.value).count("Malformed metadata: Cannot parse paths in worksheet") == 1
     assert str(caught.value).count("There is no entry in the schema") == 1
     assert str(caught.value).count("is not one of") == 1
-    # FIXME ToDo: Remove when boolean is fixed / when everything works as
-    #             expected, set correct number.
-    if "is not of type 'boolean'" in str(caught.value):
-        assert str(caught.value).count("is not of type") == 6
-    else:
-        assert str(caught.value).count("is not of type") == 4
+    assert str(caught.value).count("is not of type") == 6
     # Check correct error message for completely unknown path
     with pytest.raises(jsonschema.ValidationError) as caught:
         convert.to_dict(xlsx=rfp("data/simple_data_broken_paths.xlsx"),

unittests/table_json_conversion/utils.py

@@ -58,7 +58,8 @@ Raise an assertion exception if they are not equal."""
                 "the other.")
         return
     assert isinstance(json1, list) and isinstance(json2, list), f"Is not a list, path: {path}"
-    assert len(json1) == len(json2), f"Lists must have equal length, path: {path}"
+    assert len(json1) == len(json2), (f"Lists must have equal length, path: {path}\n"
+                                      f"{json1}\n ---\n{json2}")
     for idx, (el1, el2) in enumerate(zip(json1, json2)):
         this_path = path + [idx]
         if isinstance(el1, dict):