At Dropbox, we are focused on creating a robust and easy-to-test architecture that reliably stores our users' information, learns quickly and requires minimal maintenance. At MBLTdev 16 I’ll give a talk on “Architecting for the Win” and tell you how Dropbox makes complex decisions within Swift and Objective-C through consistently designed data exchange. It doesn't matter if you work with MVVM, MVC, Rx or others, the presentation will definitely be useful for developers who want to create a solid architecture for working with an abundance of data that must be shared.
Let's compare cloud testing technologies using the best practices as an example - all from the point of view of a developer, not a tester. The Android ecosystem is very fragmented, all devices have different screens. It seems that it is impossible to conduct testing at all at once. But if we write automated tests, we will be able to run them on hundreds of real devices in the cloud. Let's compare Expresso with Appium and Amazon Device Farm with Firebase Device Lab and see which ones we like more.
Google recommends using snippets to build a flexible UI. In cases when we need to optimize the interface for tablets or implement the work with ViewPager, this is really the simplest solution. At the same time, the integration of fragments is not such an easy task. Anyone who has ever seen a diagram of their life cycle understands the seriousness of what has been said. Potential memory leaks, complex APIs and bugs (the number of questions on StackOverflow exceeds 100,000). And all this no longer seems such a good decision. Let's discuss what are the alternatives.
My report on MBLTdev will be devoted to Robolectric, a framework for running tests on a local Android-specific JVM. We will look under the hood of this tool and find out its internal structure in detail, examine the existing Shadow implementations and write our own. We also estimate the performance of the Robolectric compared to alternative solutions.
The report on MBLTdev 16 will be about how to build applications on the Realm Mobile Platform. I will also tell you how to use it in already existing applications, about various features and nontrivial scenarios. There will be many examples.
On MBLTdev 16, I’ll tell you how we are measuring the performance of mobile applications in Yandex: which metrics and how we collect, how we run test scripts, what our reports look like. And there will be another great story about how we drilled the iPhone in order to measure the current consumption of the battery.
Source: https://habr.com/ru/post/312314/
All Articles