Improvements
TonioGela opened this issue · comments
It will be great to have a help menu using -h and the possibility to pass an image file from command line like ocr my_image.png
Thanks for the speedy PR @adam-zethraeus . I’m currently in bed about to fall asleep. Will review first thing tomorrow.
@adam-zethraeus just noticed that the target is now macOS 11 due to the use of the Scene (guessing SwiftUI).
I'm not sure this is necessary for a CLI first app. Thoughts?
🤔 sounds very unnecessary
Oh it does target 11. No use of scene or 11-oriented APIs though, it's just the platform I'm on. Sorry!
Feel free to grab from by fork and change the API target if you'd like! I made more changes to accept stdin (and to remove the pasteboard functionality—pbcopy is fine for me!).
fork: https://github.com/adam-zethraeus/macOCR
behavior:
❯ ocr --help
USAGE: ocr [--capture] [--stdin] [-i <i>]
OPTIONS:
-c, --capture Capture screenshot.
-s, --stdin Read stdin binary data.
-i <i> Path to input image.
-h, --help Show help information.
Another option that would be really useful - and help in Alfred workflows etc
ocr --clipboard
Use image or path to image from clipboard.
Suggested:
❯ ocr --help
USAGE: ocr [--capture] [--stdin] [-i <i>]
OPTIONS:
-b, --clipboard Use image or path from clipboard
-c, --capture Capture screenshot.
-s, --stdin Read stdin binary data
-i <i> Path to input image.
-h, --help Show help information.
Another way would be ocr --source <clipboard|capture|path> [url|-] with default being capture (for compat). Stdin can just be the special path '-'.