Inspiration
We all make mistakes when we start something new. The same applies for coding. Even though we all strive to get the right output, sometimes we make mistakes that go unnoticed. We wanted to make something that would help beginners (and others) become better coders by catching those mistakes and providing code analysis that goes beyond just the compilation stage.
What it does
Our application takes in code snippets, either as images or as text, and analyzes the code based on certain specifications and parameters such as run-time errors, warnings, and stylistic proficiency (redundancy, unused variables, etc.).
How we built it
We used C++ to create some executable files that analyze the user's code snippets. We then called the executables and bash scripts in our Node.js backend. We built our front-end using HTML, CSS, vanilla JavaScript, and BootStrap 4. Finally, we used Google Cloud Vision API to help with text recognition in images.
Challenges we ran into
One of our major challenges was the accuracy of the image analysis. We found out pretty early on that there were many bugs with converting handwritten text into executable code. We compensated for this by focusing on images that had really neat handwriting or printed code.
Accomplishments that we're proud of
We are extremely proud of the fact that a user can take a picture, upload it into our application, and instantly receive analysis and evaluation.
What we learned
Since none of us were too familiar with Node.js, we all felt that creating a web application with a Node.js back-end was a valuable learning experience. We were also able to gain a better understanding of using APIs since we had to use them to connect the front-end and the back-end, as well as to use Google Cloud Vision.
What's next for CodeVision
We hope to support more languages, have more accuracy with hand-written code, and provide more extensive analysis for our users.
Log in or sign up for Devpost to join the conversation.