fixed pdf issue

app.py

@@ -7,7 +7,7 @@ import io
 import json
 import logging
 import os
-from typing import Any, List, Tuple
+from typing import Any, List, Tuple, Union
 
 import gradio as gr
 import openai
@@ -83,7 +83,7 @@ def get_rag_pipeline(study_name: str) -> RAGPipeline:
     return rag_cache[study_name]
 
 
-def get_study_info(study_name: str | list) -> str:
+def get_study_info(study_name: Union[str, list]) -> str:
     """Retrieve information about the specified study."""
     if isinstance(study_name, list):
         study_name = study_name[0] if study_name else None
@@ -546,19 +546,25 @@ def create_gr_interface() -> gr.Blocks:
                 return "Please select PDF files", None
 
             try:
-                result = process_pdf_uploads(files, name)
+                processor = PDFProcessor()
+                # Process PDFs
+                output_path = processor.process_pdfs(files, name)
                 collection_id = f"pdf_{slugify(name)}"
-                return result, collection_id
+                
+                # Add to study files JSON
+                append_to_study_files("study_files.json", collection_id, output_path)
+                
+                # Add to ChromaDB
+                add_study_files_to_chromadb("study_files.json", "study_files_collection")
+                
+                # Add to SQLite database - this is the crucial missing step
+                add_study_files_to_db("study_files.json", "local")  # Add library_id parameter
+                
+                return f"Successfully processed PDFs into collection: {collection_id}", collection_id
             except Exception as e:
                 logger.error(f"Error in handle_pdf_upload: {str(e)}")
                 return f"Error: {str(e)}", None
 
-        upload_btn.click(
-            handle_pdf_upload,
-            inputs=[pdf_files, collection_name],
-            outputs=[pdf_status, current_collection],
-        )
-
         def add_message(history, message):
             """Add user message to chat history."""
             if not message.strip():

study_files_backup.json

@@ -1,13 +0,0 @@
-{
-    "Vaccine coverage": "data/vaccine_coverage_zotero_items.json",
-    "Ebola Virus": "data/ebola_virus_zotero_items.json",
-    "GeneXpert": "data/gene_xpert_zotero_items.json",
-    "Zotero Collection Pastan": "data/zotero-collection-pastan_zotero_items.json",
-    "EBSCOhost": "data/ebscohost_zotero_items.json",
-    "ref BMGF": "data/ref-bmgf_zotero_items.json",
-    "scholar (29)": "data/scholar-29_zotero_items.json",
-    "iom": "data/iom_zotero_items.json",
-    "ExportedRis_file_1_of_1 (1)": "data/exportedris-file-1-of-1-1_zotero_items.json",
-    "wb_1813-9450-6689": "data/wb-1813-9450-6689_zotero_items.json",
-    "kayongo papers": "data/kayongo-papers_zotero_items.json"
-}

utils/db.py

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fc6c599c827559f1eb0b001f4a132109b004ae3d12851ac2e2327492a323e44
-size 4968
+oid sha256:e165937f30a52640195e2a6ee5a71bcf980f0037e901658ebf813d794332ca6c
+size 5045