I want to use image OCR with Retool mobile.
I read topic below, I decided to use Free OCR API instead of tesseract.js.
For the API limitation, I should compress image within 1MB.
So I wrote codes to compress image with the both way of use external library and not.
The app I created works fine in both editor mode and preview mode on desktop (both Chrome and Safari) but it doesn't work in the iOS app.
I know that debugging tools for mobile will be released in the near future.
I hope I can use it to debug my app and get it to work fine, but I would report it as a issue.
The relevant part of the app attached below as json.
test_camera_share.json (22.3 KB)
↓compress working on Preview on Chrome
↓compress doesn't working on iOS
Thanks for surfacing this and providing a good repro app as well. I've let the devs know so that we can investigate what it would mean to expand the capability for JS libraries and Web APIs within the mobile environment!
Not at all smart, but I found one temporary solution.
To open the app itself in web view and compress the image within the app displayed in web view.
There is also a bug in the web version where the camera button is very unresponsive.
Since we don't yet have a debugging tool on the actual device, I debugged it using
utils.shownotification and I've tracked it down to the point where it is caused by
load event on
new Image() not being called.
Hopefully there will be a fix for this problem someday.