Add rpy2 dependency, refactor process_pdfs method, and implement unit…

… tests for OddpubWrapper

dsst_etl/oddpub_wrapper.py

@@ -105,7 +105,7 @@ def _cleanup_output_folder(self, output_folder: str) -> None:
             logger.error(f"Error cleaning up output folder: {str(e)}")
             raise
 
-    def process_pdfs(self, pdf_folder: str, output_folder: str) -> OddpubMetrics:
+    def process_pdfs(self, pdf_folder: str) -> OddpubMetrics:
         """
         Process PDFs through the complete ODDPub workflow and store results in database.
 
@@ -118,6 +118,7 @@ def process_pdfs(self, pdf_folder: str, output_folder: str) -> OddpubMetrics:
         """
         try:
             # Create output directory if it doesn't exist
+            output_folder = "oddpub_output"
             Path(output_folder).mkdir(parents=True, exist_ok=True)
 
             # Execute the workflow
@@ -130,9 +131,6 @@ def process_pdfs(self, pdf_folder: str, output_folder: str) -> OddpubMetrics:
             self.db.commit()
             logger.info("Successfully stored results in database")
 
-            # Cleanup
-            self._cleanup_output_folder(output_folder)
-
             return result
         except Exception as e:
             logger.error(f"Error in PDF processing workflow: {str(e)}")

pyproject.toml

@@ -27,6 +27,7 @@ dependencies = [
     "pdf2doi",
     "tqdm",
     "pypdf",
+    "rpy2"
 ]
 
 [project.optional-dependencies]

tests/test_oddpub_wrapper.py

@@ -0,0 +1,34 @@
+import unittest
+from unittest.mock import patch, MagicMock
+from sqlalchemy import inspect
+from dsst_etl.oddpub_wrapper import OddpubWrapper
+from dsst_etl.models import OddpubMetrics
+from dsst_etl.db import get_db_session, init_db
+from dsst_etl import get_db_engine
+
+
+class TestOddpubWrapper(unittest.TestCase):
+
+    def setUp(self):
+        self.engine = get_db_engine(is_test=True)
+        init_db(self.engine)
+        self.session = get_db_session(self.engine)
+        self.wrapper = OddpubWrapper(self.session)
+
+    def tearDown(self):
+        self.session.close()
+        self.engine.dispose()
+
+    def test_oddpub_wrapper(self):
+        self.wrapper.process_pdfs("test_pdfs", "test_output")
+
+        # Check if the OddpubMetrics table exists
+        inspector = inspect(self.session.bind)
+        self.assertTrue("oddpub_metrics" in inspector.get_table_names())
+
+        # Check if the data was inserted correctly
+        result = self.session.query(OddpubMetrics).first()
+        self.assertIsNotNone(result)
+
+if __name__ == '__main__':
+    unittest.main()