News

Copilot AI Gets Smarter, More Secure in Visual Studio Previews

Microsoft's Visual Studio 2022 dev team has been focusing on the AI-powered GitHub Copilot coding assistant as it ramps up the next major release, v17.11.

After the recent Preview 1, which the company said is all about "quality of life," the company published a Preview 2 and Preview 3, the latter just out today (July 9).

As the release notes show, much attention has been paid to Copilot, which debuted in 2021 as a technical preview for Visual Studio Code and basically kicked off the current generative AI craze.

Here's a rundown of what's new for the "AI pair programmer" in the most recent two previews:

  • Understand symbols: The tool helps devs understand symbols at different invocations without a code base, being now integrated into the tooltip on hover over symbols to provide AI-generated summaries of the selected symbol. This animated GIF, for example, shows a dev hovering over a symbol to get a summary of the symbol's definition and usage and then asking the tool to "tell me more" to see generated documentation for the selected symbol:
    Understand Your Symbols with GitHub Copilot in Animated Action
    [Click on image for larger, animated GIF view.] Understand Your Symbols with GitHub Copilot in Animated Action (source: Microsoft).
  • Getting smarter: It now includes context from an entire repository and can search the web, as GitHub Copilot Enterprise subscribers in Visual Studio can now use GitHub Copilot Chat to get answers enriched with context from their repo and Bing search results.
  • More secure: GitHub Copilot Business and Enterprise customers can prevent specified files or repositories from being used to inform code completion suggestions made by the tool, with GitHub Copilot Content Exclusion used to prevent unwanted sources from being used.
  • Refine suggestions: Two items of note:
    • Refine Completions with Inline Chat: GitHub Copilot now allows devs to have more control over its suggestions. Instead of just accepting or ignoring a suggestion, devs can retry and modify it. This feature lets users curate the proactive suggestions by adding context or tweaking the completions.
    • Promote Inline Chat to the Chat Window: Devs can promote their inline chat to the chat window, preserving the conversation history. This lets them continue the chat on a larger screen at their convenience.

    Devs can select "Continue in chat window" to explore these features further.

  • Naming things made easy: Devs can use GitHub Copilot to generate naming suggestions for identifiers in C++, specifically suggesting names for identifiers (variables, methods, or classes) based on how the identifier is being used and the style of code.
  • AI-generated breakpoint expressions: Also in C++, dev can use AI-generated expressions to insert conditional breakpoints or tracepoints. GitHub Copilot analyzes code and offers insightful breakpoint expressions, streamlining the debugging process.

Outside of the AI space, Microsoft highlighted many other tweaks, enhancements and new features in the most recent two previews affecting:

About the Author

David Ramel is an editor and writer at Converge 360.

comments powered by Disqus

Featured

  • Windows Community Toolkit v8.2 Adds Native AOT Support

    Microsoft shipped Windows Community Toolkit v8.2, an incremental update to the open-source collection of helper functions and other resources designed to simplify the development of Windows applications. The main new feature is support for native ahead-of-time (AOT) compilation.

  • New 'Visual Studio Hub' 1-Stop-Shop for GitHub Copilot Resources, More

    Unsurprisingly, GitHub Copilot resources are front-and-center in Microsoft's new Visual Studio Hub, a one-stop-shop for all things concerning your favorite IDE.

  • Mastering Blazor Authentication and Authorization

    At the Visual Studio Live! @ Microsoft HQ developer conference set for August, Rockford Lhotka will explain the ins and outs of authentication across Blazor Server, WebAssembly, and .NET MAUI Hybrid apps, and show how to use identity and claims to customize application behavior through fine-grained authorization.

  • Linear Support Vector Regression from Scratch Using C# with Evolutionary Training

    Dr. James McCaffrey from Microsoft Research presents a complete end-to-end demonstration of the linear support vector regression (linear SVR) technique, where the goal is to predict a single numeric value. A linear SVR model uses an unusual error/loss function and cannot be trained using standard simple techniques, and so evolutionary optimization training is used.

  • Low-Code Report Says AI Will Enhance, Not Replace DIY Dev Tools

    Along with replacing software developers and possibly killing humanity, advanced AI is seen by many as a death knell for the do-it-yourself, low-code/no-code tooling industry, but a new report belies that notion.

Subscribe on YouTube