3 min read

Last week, Michael Thomsen, the Project Manager for Dart announced the stable release of Dart 2.5 SDK (Software Development Kit). This release includes two technical previews of ML Complete and the dart:ffi foreign function interface to be used for calling C code directly from Dart. Dart 2.5 also brings improved support for constant expressions.

Preview of ML Complete

In his blog, Thomsen has regarded ML Complete as a “powerful addition” to their existing suite of productivity tools like hot reload, customizable static analysis, and Dart DevTools. It works by training a model of possible member occurrences in a given context. The possible occurrences can be analyzed from the available open-source Dart code on Github. The training model uses TensorFlow Lite tools to predict the next probable symbol, while the developer is editing.

As ML Complete is built directly into the Dart analyzer, it is available on all Dart-enabled editors including Android Studio, IntelliJ, and VS Code. Since it is still in preview, developers are advised to use the Flutter dev channel or the Dart dev channel for previewing this feature.

Preview of the dart:ffi foreign function interface

The dart:ffi feature enables users to take advantage of the existing native APIs, where Dart code is already running. Users can also utilize the existing cross-platform native libraries written in C. Currently, the support for calling C directly from Dart is limited to the Dart VM deep integration which uses native extensions.

The new dart:ffi foreign function interface will function on a new mechanism, offering great performance, easy approach, and will work across many Dart supported platforms and compilers.

Dart-C interop works on two main scenarios:

  • Calling a C-based system API on the host operating system (OS)

For calling a C-based system API, the Linux command system is used. The system command allows the execution of any system command. It also allows the argument to be essentially passed to the shell/terminal and also run there. For implementing the dart:ffi, the Dart code needs to represent the C function and the types of its arguments and return type. It also needs to represent the corresponding Dart function, and its types. Both the representations are done by defining two typedefs in the C header for the command.

  • Calling a C-based library for a single OS or cross-platform

The dart:ffi feature is also used to implore C-based frameworks and components. It will allow the user to run TensorFlow across all the operating systems where code completion is needed. It also offers high performance of the native TensorFlow implementation.

Thomsen adds, “We also expect that the ability to call C-based libraries will be of great use to Flutter apps. You can imagine calling native libraries such as Realm or SQLite, and we think dart:ffi will be valuable for enabling plugins for Flutter desktop.”

Developers are advised to use the Flutter master channel or a Dart dev channel to quickly learn about the changes and improvements in the dart:ffi feature.

Read Also: Dart 2.2 is out with support for set literals and more!

Improvements in constant expressions

In the earlier versions, the abilities of constant expressions were limited, however, Dart 2.5 includes many new changes. In Dart 2.5, constant expressions can be defined using many ways, which includes the ability to use casts and the new control flow and collection spread features.

Image Source: Medium

Users love Dart 2.5 features.

To know more about this announcement in detail, visit Michael Thomsen’s blog on Medium.

Other Interesting News in Programming

Microsoft releases Cascadia Code version 1909.16, the latest monospaced font for Windows Terminal and Visual Studio Code

Microsoft open-sources its C++ Standard Library (STL) used by MSVC tool-chain and Visual Studio

Linux 5.3 releases with support for AMD Navi GPUs, Zhaoxin x86 CPUs and power usage improvements

Vincy Davis
A born storyteller turned writer!