image_to_text_decoder is a simple flutter application that calls an AWS lambda function, which uses aws textract itself, to decode screenshots of code/text and turn them to usable text.
- Supports windows/linux
- Select a folder to track screenshot creation.
- Append decoded text to a file.
- Ability to check if text is palagrized using the Standford university palegrization checker*.
- Typewrite decoded text/text from clipboard**
- speed of typewriting is controllable
- delay to start decoding can be set
- Audiable cues are fired on success/failure (on windows only).
- Remove extra bracket pairs while typewriting on editor auto-insert closing pairs by default
1- Download a suitable binary zip file for your system from the downloads section.
2- Run the application as you'd run any executable.
- Inorder for the decoding to work you'll need to set a valid screenshotting path;this folder is tracked for creation changes so make sure it's the same directory/folder where your screenshotting program sends its captures and make sure you have read/write permissions.
*If you want to use MOSS functionality you'll have to supply your own .perl script in the root folder of where you extracted your binaries, get MOSS from this link https://theory.stanford.edu/~aiken/moss/ and you need to make sure perl is installed on your system.
-
**Typewriting unfortunately works on windows only right now.
-
Typewriting is useful when copy/paste are not available.
-
Once an image is decoded the option to typewrite its content becomes available (you can typewrite from the clipboard regradless).
1- Adust the typewriting delay, which is the delay before the system registers key events, to your desire. 2- Adjust the typewriting speed, which is the time between each individual keyevent. 3- Start the typewriter by clicking the relative button. 4- Type writing can be stopped once started.
1- Install flutter
2- Clone repo
3- flutter build [platform]
- Better documentation.
- Integrate MOSS as a dart/python script rather than using system calls.
