How can I install apps with uiautomator? - android-uiautomator

My app to test depends on Facebook. On such devices which does not have Facebook installed, I needs to install Facebook first.
The first way I have tried is to install it from an APK file. However, I cannot find any APIs to achieve this from uiautomator developer docs. I guess this is determined by the design of uiautomator.
The second way I have tried is to install it from Google Play store. However, on the devices which has no Google account signed in, I need to sign in Google account first. I tried to sign in using uiautomator but unfortunately the sign-in page is NAF (not accessibility friendly).
Does anyone know if there is any other ways? Thanks in advance.

AndroidViewClient/culebra is a pure python tool that automates or simplifies test script creation. The scripts use UiAutomator as a backend.
Culebra GUI can generate tests that include the installation of an APK as a precondition.
culebra --install-apk=file.apk --unit-test-class --gui
EDIT
culebra only generates python, so if you want to generate Java or Kotlin you may need CulebraTester.
However, due to the limited connectivity and capabilities when you run those tests in Firebase Test Lab it might be challenging to find a way of installing dependencies.

Related

How can I test an older version of my app in Google Play Store?

I'm using Google Play Store to distribute an app.
For a certain issue, I would like to install an OLDER version of my app to one of the client's device.
I'm struggling to do so, is there any way to do that?
No matter how and which "testing" method I apply, Google Play automatically shows the HIGHEST version of the app which is the "release" version ignoring the old test versions.
Creating an APK would be NO way because I need Google Play's own signature in the apk to make things work (to test the things we would like to)
Is there any way at all to distribute my OLDER app version from Google Play store?
(By the way Google's testing methods are inexpressibly unusable even apart from this problem. You never see what really happens, you cannot really manage the testing channels the GUI is miserable the notification email is late the testing is slow and painful you don't see the version number in the Google Play Store, there are like 3 methods to do so and the difference between them is not self explaining, I give 1 start to the developers who made these parts of Google Play I hope they read this.)
Thanks in advance.
From Google Play Console go to App bundle explorer
Then choose the version you want to install
Go to Downloads tab
click on Copy sharable link
install it from you device using this link
Note: the device's Google Play account should be an internal tester and enable internal testing from Google Play on this device.

How can I run React Native code in real iPhone, if I can not login Apple develop account?

I am a fresher in React native development
and I have a stupid question
How can I install app into real iphone
If I can not login to Apple develop account, but i can ask project owner to download and send me some file in bellow photo (certificate, profile, key,...)
Thanks, thanks
Attempt to build and run using Expo instead of native method.
Start application:
expo start
Document
From real Iphone, install Expo Go application and use it to run your build.
I recommend using Expo. Here is a good walkthrough that will help you a lot. In addition, you should also have the idea about the 2 Expo Project Workflows.
With Expo, sharing prototypes, for example, will be easier.

Is fastlane capable of publishing to Mac App Store

We're trying to automate the release of our app to the Mac App Store. I've been trying to achieve this using fastlane. I managed to get this tool working for submitting the apps' metadata (release notes, screenshots etc).
I can also build the app using the gym module, however this results in a .app and not in a .pkg that is needed for submission to the App Store. Even when I use
export_method: "package"
I cannot imagine I'm the only one trying to achieve this, but I cannot find an answer if it is possible at all. And if possible, how to achieve this.
I have a (old) work in progress PR made for this - https://github.com/fastlane/fastlane/pull/12195. Majority of our users are mobile so it has been hard for me to get testing on this :). If you'd like to comment on that PR I would be happy to revive it and work with you on making this work for you!

Is it possible to keep an a custom Google Actions app (Built into Google Assistant SDK) active all the time?

I am building a Google Assistant application but would like to restrict the app from ever exiting. Meaning I just want the experience to be my app. This isn't a public facing app, its personal and on a custom device, like a Raspberry Pi.
As far as I know there's no official way to do it that way but there is a trick you can use using the gActions CLI by running:
$ gactions preview --action_package mypackage.json --preview_mins 420
I'm not sure how far you can push the preview time though. You can find more details here

after using capdetection tool application stopped opening

i used capdetection tool for generating capabilities of of my wp7 app. when i pasted the capabilities generated by the tool, application build was successful but after installing in device application was not opening, after that i reverted back the previously present capabilities in manifest file and then app was launching fine. Is there any problem with capdetection tool. Can some one tell me the problem about the tool, Is that a certification requirement using capdetection tool for declaring capabilities in manifest file?
The tool exists so you can test your app with the capabilities it will be given when available through the marketplace. If you ignore the situation you have described you could end up failing marketplace certification or end up with an app that is unusable.
You are likely relying on a capability that the tool can't detect through static analysis.
I woudl strongly recommend you work out what this capability is and use it in a way that the tool can detect.
You do no need to run the tool for certification, they run it during submission and replace the contents of the file regardless of what you put in there.
I believe the main purpose of it is so you can know in advance what services are going to be requested, giving you a chance to check they are correct.
For example some ad controls include location tracking and if you unexpectedly get that capability added on submission, you will probably fail as you need to ask the user and show privacy statements to have that capability.

Resources