refactor: use shared module for comprehensive ArXiv categories

Author: Goziee-git

dev/create_arxiv_category_map.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+"""
+Create arxiv_category_map.yaml with essential ArXiv category mappings
+based on common categories found in ArXiv CC licensed papers.
+"""
+# Standard library
+import os
+import sys
+
+# Third-party
+import yaml
+
+# Add parent directory for shared imports
+sys.path.append(os.path.join(os.path.dirname(__file__), "..", "scripts"))
+
+# First-party/Local
+import shared  # noqa: E402
+
+
+def create_arxiv_category_mapping():
+    """Fetch ArXiv category mappings from API."""
+
+    # Import shared module to use the comprehensive mapping
+    try:
+        categories = shared.get_arxiv_categories()
+        if categories:
+            return categories
+    except Exception as e:
+        print(f"Failed to get categories: {e}")
+
+    # Fallback to essential categories if API fails 
+    # Covering the most common ArXiv categories 
+    return {
+        "cs.AI": "Computer Science - Artificial Intelligence",
+        "cs.CL": "Computer Science - Computation and Language",
+        "cs.CV": "Computer Science - Computer Vision and Pattern Recognition",
+        "cs.LG": "Computer Science - Machine Learning",
+        "physics.comp-ph": "Physics - Computational Physics",
+        "math.CO": "Mathematics - Combinatorics",
+        "stat.AP": "Statistics - Applications",
+    }
+
+
+def main():
+    """Main function."""
+    logger, paths = shared.setup(__file__)
+
+    # Create category mapping
+    categories = create_arxiv_category_mapping()
+
+    # Output path
+    output_path = os.path.join(paths["data"], "arxiv_category_map.yaml")
+
+    try:
+        # Add header comment
+        header = [
+            "# ArXiv category code to name mappings",
+            "# Generated by dev/create_arxiv_category_map.py",
+            "# Comprehensive ArXiv category taxonomy",
+            "",
+        ]
+
+        with open(output_path, "w", encoding="utf-8") as f:
+            f.write("\n".join(header))
+            yaml.dump(categories, f, default_flow_style=False, sort_keys=True)
+
+        logger.info(
+            f"Created ArXiv category mapping: {output_path} "
+            f"with {len(categories)} categories"
+        )
+
+    except Exception as e:
+        logger.error(f"Failed to create category mapping: {e}")
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()