Forum in maintenance, we will back soon 🙂
Notifications
Clear all
instagram post genatate tool error, Fast API
Building Online Tools
7
Posts
3
Users
1
Reactions
150
Views
Topic starter
Mr Hassan, I start to create a tool to generate different Instagram posts using live blog posts. So my Idea is to read the blog post from the given link and change that blog post according to the given Instagram content type. I use ChatGPT to generate this. but it gives me an error. please check the below code...
Sample prompt for testing:
generate_instagram_posts = """As an expert social media content creator, your task is to analyze the given blog post content: {blog_content} and generate Instagram posts. Each post should be tailored to the specified content type: {content_type}. The output should be ONLY valid JSON as follows:
[
 {{
  "content_type": "{content_type}",
  "content": "Generated Instagram post content based on the blog content."
 }}
]
The JSON object:\n\n
"""
class InstagramPost(BaseModel): content_type: str content: str class InstagramPostGeneratorResponse(BaseModel): success: bool message: str posts: list[InstagramPost]
@router.get("/instagram/generate") async def generate_instagram_posts(url: str, content_type: str): start_time = time.time() max_retries = 5 headers = { "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3" } for retry_count in range(max_retries): try: response = requests.get(url, headers=headers) if response.status_code != 200: logger.error(f"Failed to fetch blog post content, HTTP status code: {response.status_code}") raise HTTPException(status_code=response.status_code, detail="Failed to fetch blog post content") blog_content = response.text if await llm.is_text_flagged(blog_content) or await llm.is_text_flagged(content_type): return { "success": False, "message": "Input Not Allowed", "result": None } prompt = pr.generate_instagram_posts.format(blog_content=blog_content, content_type=content_type) # Assuming llm.generate_with_response_model generates Instagram posts directly # Replace with appropriate logic based on llm functionality generated_posts = await llm.generate_with_response_model(prompt, 3, InstagramPostGeneratorResponse) if generated_posts: return { "success": True, "message": "Generated Instagram Posts Successfully", "result": generated_posts.posts } else: raise Exception("Failed to generate Instagram posts") except requests.exceptions.RequestException as e: logger.error(f"An HTTP error occurred: {e}") raise HTTPException(status_code=500, detail=f"An HTTP error occurred: {e}") except (json.JSONDecodeError, ValidationError) as e: logger.warning(f"Failed during JSON decoding or validation. Retry count: {retry_count + 1}. Error: {e}") except KeyError as e: logger.warning(f"Missing key in JSON: {e}") except Exception as e: logger.error(e) continue finally: elapsed_time = time.time() - start_time # do something with the elapsed time return { "success": False, "message": "Failed to generate Instagram posts", "result": None }
Posted : 06/23/2024 2:48 am
Topic Tags
@dassa what's the error?
Regards,
Earnie Boyd, CEO
Seasoned Solutions Advisor LLC
Schedule 1-on-1 help
Join me on Slack
Posted : 06/23/2024 12:42 pm
sorry, what is the error?Â
did you run locally?
is this my template?
Posted : 06/24/2024 5:00 pm
Topic starter
Yes, this is your template. and I did run locally.
Â
Posted : 06/26/2024 2:29 am
I see an error in the Mongo connection. Do you have a Mongo database?
Posted : 06/26/2024 3:25 am
Forum Information
Our newest member: Shaik Abdul Raheem
Latest Post: Affiliate tool.
Forum Icons:
Forum contains no unread posts
Forum contains unread posts
Topic Icons:
Not Replied
Replied
Active
Hot
Sticky
Unapproved
Solved
Private
Closed
[the_ad_group id="312"]