Code OCR

Anant Patni

View on GitHub →

Problem Statement

Allowing developers to do extract code from screenshots or image url to a new VS Code file. Useful for people who are following YouTube/Udemy/Any other video based tutorial.

Methodology

Used VS Code WebView API to allow users to either upload an image or enter an image URL followed by which the image gets processed using free OCR API (ocr.space). Once the image has been processed, a new tab in VS Code is opened with contents as that of what was read from the image.

Scope

Can be used by anyone using video/image sources to learn programming.

Future Prospects

  1. Auto detecting only code portion from image uploaded.
  2. Allowing users to put a YouTube link at a specified timestamp and extracting code from that particular timestamp.
  3. Auto detecting language used in code so that the new file created is in correct format which enables auto syntax highlighting.