Revert "Relax dependency constraints (#1713)" (#1754)

This reverts commit 21f7b3a.

Author: SFJohnson24

.github/workflows/lint-format.yml

@@ -49,7 +49,7 @@ jobs:
           python-version: "3.12"
       - name: Install linters
         run: |
-          pip install black flake8
+          pip install black flake8 -c requirements-dev.txt
       - name: Run flake8
         run: |
           flake8 ${{needs.get_changed_files.outputs.py}} --count --select=E9,F63,F7,F82 --show-source --statistics

.github/workflows/test-unit.yml

@@ -16,8 +16,8 @@ jobs:
           python-version: "3.12"
       - name: Install requirements
         run: |
-          pip install -r requirements.txt
-          pip install --group dev -e .
+          pip install -r requirements-dev.txt
+          pip install -e .
       - name: Running Tests
         env:
           CDISC_LIBRARY_API_KEY: fakekey12341234

cdisc_rules_engine/dataset_builders/contents_define_vlm_dataset_builder.py

@@ -66,7 +66,7 @@ def build(self):
         data_contents_with_vlm["variable_value_length"] = data_contents_with_vlm.data[
             ["variable_value", "define_vlm_data_type"]
         ].apply(
-            lambda row: ValuesDatasetBuilder.calculate_variable_value_length(
+            lambda row: self.calculate_variable_value_length(
                 row["variable_value"], row["define_vlm_data_type"]
             ),
             axis=1,

cdisc_rules_engine/models/dataset/dask_dataset.py

@@ -81,9 +81,9 @@ def __setitem__(self, key, value):
             array_values = da.from_array(value, chunks=tuple(chunks))
             self._data[key] = array_values
         elif isinstance(value, pd.Series):
-            pdf = self._data.compute()
-            pdf[key] = value.values
-            self._data = dd.from_pandas(pdf, npartitions=1)
+            self._data = self._data.reset_index()
+            self._data = self._data.set_index("index")
+            self._data[key] = value
         elif isinstance(value, dd.DataFrame):
             for column in value:
                 self._data[column] = value[column]

cdisc_rules_engine/operations/codelist_extensible.py

@@ -32,10 +32,6 @@ def _handle_multiple_versions(self) -> pd.Series:
                 "codelist_code": "string",
             }
         )
-        cast_cols = {self.params.ct_version: "string"}
-        if self.params.codelist_code in self.evaluation_dataset.columns:
-            cast_cols[self.params.codelist_code] = "string"
-        self.evaluation_dataset = self.evaluation_dataset.astype(cast_cols)
         if self.params.codelist_code in self.evaluation_dataset.columns:
             is_extensible = self.evaluation_dataset.merge(
                 ct_df.data,

cdisc_rules_engine/operations/codelist_terms.py

@@ -64,10 +64,6 @@ def _handle_multiple_versions(self) -> pd.Series:
                 "codelist_code": "string",
             }
         )
-        cast_cols = {self.params.ct_version: "string"}
-        if self.params.codelist_code in self.evaluation_dataset.columns:
-            cast_cols[self.params.codelist_code] = "string"
-        self.evaluation_dataset = self.evaluation_dataset.astype(cast_cols)
         if self.params.codelist_code in self.evaluation_dataset.columns:
             result = self.evaluation_dataset.merge(
                 ct_df.data,

cdisc_rules_engine/services/data_readers/csv_reader.py

@@ -1,6 +1,6 @@
 import tempfile
 
-import dask.dataframe as dd
+from dask.dataframe import dd
 
 from cdisc_rules_engine.exceptions.custom_exceptions import InvalidCSVFile
 from cdisc_rules_engine.interfaces import DataReaderInterface

cdisc_rules_engine/services/data_services/usdm_data_service.py

@@ -399,19 +399,8 @@ def __read_node_metadata(
         }
 
     @staticmethod
-    def __get_full_path(node: DatumInContext) -> str:
-        parts = []
-        current = node
-        while current is not None and current.context is not None:
-            parts.append(str(current.path))
-            current = current.context
-        result = ""
-        for part in reversed(parts):
-            if part.startswith("["):
-                result += part
-            else:
-                result = (result + "." if result else "") + part
-        return result
+    def __get_full_path(node: DatumInContext):
+        return f"{node.full_path}".replace(".[", "[")
 
     def __get_datasets_content_index(self) -> List[dict]:
         """

cdisc_rules_engine/services/datasetxpt_metadata_reader.py

@@ -61,7 +61,7 @@ def read(self) -> dict:
             "variable_labels": list(metadata.column_labels),
             "variable_names": list(metadata.column_names),
             "variable_formats": [
-                "" if (data_type == "NULL" or data_type is None) else data_type
+                "" if data_type == "NULL" else data_type
                 for data_type in metadata.original_variable_types.values()
             ],
             "variable_name_to_label_map": metadata.column_names_to_labels,

core.py

@@ -357,13 +357,15 @@ def load_custom_dotenv_from_data_options(_ctx, _param, value):
     "-s",
     "--standard",
     required=True,
+    default=None,
     help="CDISC standard to validate against",
     envvar="PRODUCT",
 )
 @click.option(
     "-v",
     "--version",
     required=True,
+    default=None,
     help="Standard version to validate against",
     envvar="VERSION",
 )