excute(update_query) return influenced rows

[66dexmg]

https://github.com/PacktPublishing/Building-Data-Science-Applications-with-FastAPI/blob/da186755fba69a56917a222caa19661a8c19fc8b/chapter6/sqlalchemy/app.py#L91 I have tried in my code.Maybe the method 'execute' for 'update' return influenced rows not the primary-key id.And I know that in the 'insert' condition new id was returned. So maybe 'post.id' should be the parameter instead of 'post_id' in next function.

[frankie567]

Hi @66dexmg 👋

Do you get a specific error when doing this?

[66dexmg]

@frankie567 Yes,it may be a small mistake.In next row,i can only get first post which was not the post i updated because the function 'execute' always return the number 1.

[frankie567]

You're perfectly right @66dexmg, that's actually quite a dumb mistake from my part. You can find the correct, fixed code here: https://github.com/PacktPublishing/Building-Data-Science-Applications-with-FastAPI/commit/e95e35f484809e937071e45a3c1f6fcab83fd8f7

I'll now contact Packt to coordinate a proper erratum. Thank you for raising this 🙏

[66dexmg]

@frankie567 You are welcome.You have written a fantastic book although I had waitted its Chinese translation for a long time😀.Any oriented advice for me after I finish it?

[frankie567]

Thanks!

Well, my main advice to fellow developers is usually: practice. It's by experimenting a lot, building new things, observe what others are doing in their code that you'll make progress and gain experience.

That said, there are a few subjects I didn't have the opportunity to show in the book that are interesting:

• Worker process: when you have long-running operations to perform, it's usually a good idea to offload it to a background process, different from your web server. This way, you keep your API fast and responsive while being able to perform large jobs (data extraction, heavy machine learning...) in the background. In Python, this can be achieved with Celery or Dramatiq.
  • I have a quick introduction to those concepts in this blog post: https://www.francoisvoron.com/blog/create-deploy-reliable-data-ingestion-service-fastapi-sqlmodel-dramatiq
• Structured logging: for robust production systems, it's important to have clear logs to monitor what's going on. An interesting pattern is to structure those logs using JSON objects rather than just free-text logs. It'll help you search through them or index them correctly into a log ingestion platform like Logstash. Libraries to help you with this: structlog or loguru.
• Monitoring: it can also be valuable to have relevant metrics about your system (speed, number of operations...). A good approach is to set-up Prometheus, which is a lightweight but powerful way to define and report metrics. It's quite simple to add it in your Python codebase and then be able to properly query and display this data in a Grafana dashboard.

[66dexmg]

Thanks a lot!