Add dataset loading functionality for Litroacp and XACBench

- Introduced DatasetReader to load datasets from specified directories.
- Implemented loading for Litroacp datasets and printed the loaded dataset keys.
- Added loading for XACBench datasets with corresponding output of loaded dataset keys.

Author: NisonChrist

datasets/process.py

@@ -2,8 +2,10 @@
 Dataset Processing Module - Handles loading and processing datasets
 """
 
+from calendar import c
 from pathlib import Path
 import os
+import re
 import pandas as pd
 
 
@@ -20,33 +22,37 @@ def read_csv(abs_path: str) -> pd.DataFrame:
         """Read CSV format file"""
         return pd.read_csv(abs_path)
 
-
-def load_litroacp_datasets() -> dict[str, pd.DataFrame]:
-    """
-    Load all datasets from litroacp folder
-
-    Returns:
-        Dictionary with dataset names as keys and DataFrames as values
-    """
-    datasets = {}
-    litroacp_path = Path(__file__).parent / "litroacp"
-
-    for file_name in sorted(os.listdir(litroacp_path)):
-        file_path = litroacp_path / file_name
-
-        try:
-            if file_name.endswith(".jsonl"):
-                df = DatasetReader.read_jsonl(str(file_path))
-                datasets[file_name] = df
-                print(f"Loaded {file_name} with shape: {df.shape}")
-            elif file_name.endswith(".csv"):
-                df = DatasetReader.read_csv(str(file_path))
-                datasets[file_name] = df
-                print(f"Loaded {file_name} with shape: {df.shape}")
-        except Exception as e:
-            print(f"Error loading {file_name}: {e}")
-
-    return datasets
-
-
-__all__ = ["DatasetReader", "load_litroacp_datasets"]
+    def load_datasets(self, datasets_dir_path: Path) -> dict[str, pd.DataFrame]:
+        """
+        Load all datasets from litroacp folder
+
+        Returns:
+            Dictionary with dataset names as keys and DataFrames as values
+        """
+        datasets = {}
+        # litroacp_path = Path(__file__).parent / "litroacp"
+
+        for file_name in sorted(os.listdir(datasets_dir_path)):
+            file_path = datasets_dir_path / file_name
+            try:
+                if file_name.endswith(".jsonl"):
+                    df = DatasetReader.read_jsonl(str(file_path))
+                    datasets[file_name] = df
+                    print(f"Loaded {file_name} with shape: {df.shape}")
+                elif file_name.endswith(".xml"):
+                    with open(file_path, "r", encoding="utf-8") as file:
+                        policy_pattern = re.compile(r"<Policy[\s\S]*?<\/Policy>")
+                        xacml_content = file.read()
+                        policies = policy_pattern.findall(xacml_content)
+                        datasets[file_name] = pd.DataFrame({"policy": policies})
+                        print(f"Loaded {file_name} with {len(policies)} policies.")
+                else:
+                    # print(f"Unsupported file format for {file_name}, skipping.")
+                    continue
+            except Exception as e:
+                print(f"Error loading {file_name}: {e}")
+
+        return datasets
+
+
+__all__ = ["DatasetReader"]