Update app.py

app.py

@@ -118,13 +118,29 @@ def get_file(temp_dir, file_path):
     full_path = os.path.join(temp_dir, file_path)
     return full_path
 
+
+
+def process_directory(directory):
+    code_partial_paths = ['kadi_apy/lib/resources/']
+    
+    zip_file_path = os.listdir(directory) if file.endswith('.zip')
+    print(zip_file_path)
+ #   with tempfile.TemporaryDirectory() as tmpdirname:
+    # Unzip the file into the temporary directory
+ #   with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:
+ #       zip_ref.extractall(tmpdirname)
+
+
+  #      unzipped_root = os.listdir(tmpdirname)
     
 def process_directory(directory, partial_paths=None, file_paths=None):
     all_texts = []
     file_references = []
 
-    zip_files = [file for file in os.listdir(directory) if file.endswith('.zip')]
+    zip_file = next((file for file in os.listdir(directory) if file.endswith('.zip')), None)
 
+
+    
     if not zip_files:
         print("No zip file found in the directory.")
         return all_texts, file_references
@@ -379,24 +395,27 @@ def rag_workflow(query):
 def initialize():
     global docstore, codestore, chunks, llm
     #download_gitlab_repo()
-    #code_partial_paths = ['kadi_apy/lib/']
-    #code_file_path = []
-    doc_partial_paths = []
+    code_partial_paths = ['kadi_apy/lib/resources/']
+    code_file_path = []
+    #doc_partial_paths = []
     #doc_partial_paths = ['docs/source/setup/']
-    doc_file_paths = ['docs/source/usage/lib.rst']
+    #doc_file_paths = ['docs/source/usage/lib.rst']
+
+
 
+    code_files = process_directory(DATA_DIR)
     
     #code_files, code_file_references = process_directory(DATA_DIR, code_partial_paths, code_file_path)
     
-    doc_files, doc_file_references = process_directory(DATA_DIR, doc_partial_paths, doc_file_paths)
+    #doc_files, doc_file_references = process_directory(DATA_DIR, doc_partial_paths, doc_file_paths)
     
     #code_chunks = split_pythoncode_into_chunks(code_files, code_file_references, 1500, 0)
-    doc_chunks = split_into_chunks(doc_files, doc_file_references, CHUNK_SIZE, CHUNK_OVERLAP)
+    #doc_chunks = split_into_chunks(doc_files, doc_file_references, CHUNK_SIZE, CHUNK_OVERLAP)
 
     #print(f"Total number of code_chunks: {len(code_chunks)}")
-    print(f"Total number of doc_chunks: {len(doc_chunks)}")
+    #print(f"Total number of doc_chunks: {len(doc_chunks)}")
 
-    docstore = setup_vectorstore(doc_chunks, EMBEDDING_MODEL_NAME, PERSIST_DOC_DIRECTORY)
+    #docstore = setup_vectorstore(doc_chunks, EMBEDDING_MODEL_NAME, PERSIST_DOC_DIRECTORY)
     #codestore = setup_vectorstore(code_chunks, EMBEDDING_MODEL_NAME, PERSIST_CODE_DIRECTORY)
     
     #llm = setup_llm(LLM_MODEL_NAME, LLM_TEMPERATURE, GROQ_API_KEY)