cdiscbuilder 1.1.5__tar.gz → 1.1.7__tar.gz

{cdiscbuilder-1.1.5/src/cdiscbuilder.egg-info → cdiscbuilder-1.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cdiscbuilder
-Version: 1.1.5
+Version: 1.1.7
 Summary: A package to convert ODM XML to SDTM/ADaM Datasets
 Author-email: Ming-Chun Chen <hellomingchun@gmail.com>
 Requires-Python: >=3.8

{cdiscbuilder-1.1.5 → cdiscbuilder-1.1.7}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "cdiscbuilder"
-version = "1.1.5"
+version = "1.1.7"
 description = "A package to convert ODM XML to SDTM/ADaM Datasets"
 authors = [{name = "Ming-Chun Chen", email = "hellomingchun@gmail.com"}]
 dependencies = [

{cdiscbuilder-1.1.5 → cdiscbuilder-1.1.7}/src/cdiscbuilder/adam/adam_derivation/derivations/sql_derivation.py

@@ -149,9 +149,13 @@ class SQLDerivation(BaseDerivation):
                 # The original code raised ValueError here.
                 raise ValueError(f"Column {source} not found in target dataset")
 
-            # Apply mapping if present (legacy / optimization for local cols)
-            if "mapping" in derivation:
-                series = self._apply_mapping(series, derivation["mapping"])
+        # Auto-Strip Whitespace
+        if series.dtype == pl.Utf8:
+             series = series.str.strip_chars()
+
+        # Apply mapping if present (legacy / optimization for local cols)
+        if "mapping" in derivation:
+            series = self._apply_mapping(series, derivation["mapping"])
 
         return series
 
@@ -425,16 +429,22 @@ class SQLDerivation(BaseDerivation):
                 # Calculate distance to target date
                 if target_col in subject_data.columns and date_col in subject_data.columns:
                     # Get target date (should be same for all rows of this subject)
-                    target_date = subject_data[target_col][0]
-
                     # Calculate date differences and find closest
-                    # Handle partial dates by using strptime with appropriate format
+                    # Handle multiple formats: "YYYY-MM-DD HH:MM:SS", "YYYY-MM-DDTHH:MM:SS" (ISO), or "YYYY-MM-DD"
+                    # We convert everything to Datetime for precision comparison (ms level)
+                    def parse_to_datetime(expr: pl.Expr) -> pl.Expr:
+                        return (
+                            expr.str.strptime(pl.Datetime, "%Y-%m-%d %H:%M:%S", strict=False)
+                            .fill_null(expr.str.strptime(pl.Datetime, "%Y-%m-%dT%H:%M:%S", strict=False))
+                            .fill_null(expr.str.strptime(pl.Date, "%Y-%m-%d", strict=False).cast(pl.Datetime))
+                        )
+
                     with_diff = subject_data.with_columns(
                         (
-                            pl.col(date_col).str.strptime(pl.Date, "%Y-%m-%d", strict=False)
-                            - pl.lit(target_date).str.strptime(pl.Date, "%Y-%m-%d", strict=False)
+                            parse_to_datetime(pl.col(date_col))
+                            - parse_to_datetime(pl.lit(target_date))
                         )
-                        .dt.total_days()
+                        .dt.total_milliseconds()
                         .abs()
                         .alias("date_diff")
                     )

{cdiscbuilder-1.1.5 → cdiscbuilder-1.1.7}/src/cdiscbuilder/sdtm/engine/classes/finding.py

@@ -111,6 +111,10 @@ class FindingProcessor:
                     elif source_expr in source_df.columns:
                          series = source_df[source_expr]
                     
+                    # Auto-Strip
+                    if series is not None and pd.api.types.is_object_dtype(series):
+                         series = series.astype(str).str.strip().replace('nan', None)
+                    
                     if regex_extract and series is not None:
                          # Extract group 1
                          series = series.astype(str).str.extract(regex_extract)[0]

{cdiscbuilder-1.1.5 → cdiscbuilder-1.1.7}/src/cdiscbuilder/sdtm/engine/classes/general.py

@@ -185,6 +185,13 @@ class GeneralProcessor:
                         # Source defined but not found.
                         print(f"Warning: Source column '{source_expr}' not found for '{domain_name}.{target_col}'. Filling with NaN.")
                         series = pd.Series([None] * len(pivoted))
+                    
+                    # Auto-Strip Whitespace for strings
+                    if series is not None and pd.api.types.is_object_dtype(series):
+                        try:
+                            series = series.astype(str).str.strip().replace('nan', None)
+                        except:
+                            pass
                 else:
                     print(f"Warning: No source or literal defined for '{domain_name}.{target_col}'. Filling with NaN.")
                     series = pd.Series([None] * len(pivoted))

{cdiscbuilder-1.1.5 → cdiscbuilder-1.1.7/src/cdiscbuilder.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cdiscbuilder
-Version: 1.1.5
+Version: 1.1.7
 Summary: A package to convert ODM XML to SDTM/ADaM Datasets
 Author-email: Ming-Chun Chen <hellomingchun@gmail.com>
 Requires-Python: >=3.8