Ponicode DogString automatically adds docstrings to all undocumented Python code - essentially explaining its behaviour using human language. This AI-based tech is available as a Github Action and on our interactive playground. Let us know what you think!

Hello guys, Very excited to share with you that developer’s life is going to get a whole lot easier today. Our magic pony made a new AI-powered Github Action generating docstring for all your undocumented Python functions. We are going public with our beta today, it’s available as a Github Action so you can directly generate docstrings in your repo and have a pull request with 100% docstring coverage ready to merge. At this stage we would love to get your feedback! What do you like? Do you get any errors we haven’t spotted? What you wish to see in the final version? Is there another language you’d love us to tackle next? Please share with us your feelings about this Github Action. For those who want a 30 seconds test, we also have made a little playground, where you can get a feel for the tech and train our AI with your functions. Why? Remove collaboration constraints. Don’t compromise quality for efficiency Save time and money on your next development projects Code cleanliness is a must; it shows to your clients, your team members, your management and all stakeholders in your projects that you are committed to making your code shareable and reader friendly, while also helping to reduce legacy. How do get started? You want to try on one function? Just check out our Ponicode dogstring playground here: https://api.ponicode.com/playgro... You want to reach 100% docstring coverage? Generate docstrings in bulk right where they belong in your repo using our Github Action. The ReadMe will guide you step by step with the Github Action onboarding Thanks for checking it out, and let us know what you think! We love feedback!

@bbouffaut This is really cool. And I have questions. :) 1. Which languages does it support? 2. Is there any specific formatting it's looking for in terms of naming? Or is it trying to analyze the code of the function itself? Will a complex function cause it to fail? 3. Is there a CLI planned to run on files locally? Or even an IDE plugin?

Hi @ethan_glover. thanks for the feedback. It supports python only so far, but other languages will come soon (JS, TS). Since it is based on both semantic and computer science technics, the better the variables and functions are named, the best it is. So far we might be a bit weak for very complex functions, but we will improve continuously. Regarding the UX, yes we are thinking to have an IDE extension and a CLI as well => Would you find this interesting ? Many thanks again for your feedback

Great to see the team launch this new product which I've been testing since beta, hope lots of people will use it to generate their docs as well 😊