There are great TikTok / Instagram scrapers on Apify, check them out

I already fixed that. Couldn’t get it in ts, but did in python

@Dmitry Avramenko this function does all the work. append it to the normal run script: def show_file_search_results(run_id): try: # Get all run steps run_steps = client.beta.threads.runs.steps.list( thread_id=thread.id, run_id=run_id ) # Find and show details for each step for step in run_steps.data: # Retrieve detailed step information with file search content step_details = client.beta.threads.runs.steps.retrieve( thread_id=thread.id, run_id=run_id, step_id=step.id, extra_query={ "include": ["step_details.tool_calls[*].file_search.results[*].content"] } ) if hasattr(step_details.step_details, 'tool_calls'): for tool_call in step_details.step_details.tool_calls: if tool_call.type == 'file_search': print("\n=== File Search Chunks Used ===") for result in tool_call.file_search.results: print(f"\nFile ID: {result.file_id}") print(f"Similarity Score: {result.score}") print(f"Content Chunk:\n{result.content}") print("=" * 50) except Exception as e: print(f"Error in showing file search results: {str(e)}") print("Step details:", step_details) # This will help us debug the structure

I recommend just writing a pyproject.toml at the very beginning and keep updating it through your dev process. Cannot really go wrong using poetry

@Wahib Abk is the magician

@Wahib Abk Really? I think the quality is not good enough to be published. But I’ll try to talk to some profs about it. Need to find them first, never was in the IT department XD And thank you!

@Wahib Abk Update: just talked with my statistics professor who has research works in ML and AI. He said it’s cool and he can help, so we’ll meet mid November to polish it