In this post I will describe the work I did to add auto-zoom functionality to Android-Image-Cropper library.
The goal is a smooth zoom-in/out experience affected by the size of the cropping window. When the user adjusts the crop window to cover less than 50% of the image current total visible sub-area, auto-zoom-in is triggered to zoom-in and center on the relevant cropping image zoomed sub-area so the crop window will cover 64% of it. Similarly, when the crop rectangle is above 65% auto-zoom-out is triggered so zoomed sub-area will be 51% covered.
- Using matrix scale type for the cropping image in image view .
- Handle image rotation via matrix transformation.
- Adding auto-zoom-in/out to matrix transformation.
- Smoothing transitions using animations.
- Cropping using invert matrix transformation.
It's embarrassing really, soon after creating my Android-Image-Cropper fork as I wrote about in my previous post, a GitHub user noted that I have a slowness issue caused by loading/decoding image URI on main thread potentially causing an ANR, really a newbie mistake.
Unfortunately starting a new startup I didn't have the time to fix it then, and no one come forward with a pull request. Fortunately the company is more mature now so I can commit a bit of my free time for the project.
What we will have here:
- A few words on the implementation.
- Overriding default progress bar UI using
showProgressBar view attribute and listeners.
TL;DR See the gist.
Working on Android open-source I needed to publish two libraries (Android-Image-Cropper and Android-Fast-Image-Loader) to JCenter.
What I will touch in this post:
- Publishing AAR
- "Sources not found" issue with AAR
- AAR vs. JAR
- Publishing JAR
- Handling optional dependency in Gradle
TL;DR see the release gradle script.
Recently I needed to implement avatar image upload from an Android app, I didn't found a library that did all that I needed so I forked a pretty good one and made it better, check it out: Android Image Cropper.
See also the followup post: Android Image Cropper async support and custom progress UI.
- Pick image from camera or gallery using single chooser.
- Select circular crop window in the image for the avatar.
- Limit output avatar image to 500×500 pixels.
- Efficient memory usage.