anycode

A Tree-sitter-based language extension that inaccurately implements popular features like "Outline & Breadcrumbs", "Go to Symbol in Workspace", "Document Highlights" and more. This extension should be used when running in enviroments that don't allow for running actual language services, like https://github.dev or https://vscode.dev.

The features provided by this extension are meant to be better than full-text search, but fall short when compared to real language services. We refer to this as "partial mode" and anycode will show a language status item when inaccurate language support is active.

Language Status

The table below shows what features have been implemented for what language. The following paragraph outlines how things are implemented.

	`rust`	`go`	`csharp`	`java`	`php`	`c`	`cpp`
Outline	✅	✅	✅	✅	✅	✅	✅
Go to Symbol	✅	✅	✅	✅	✅	✅	✅
Go to Definition	✅	✅	✅	✅	✅	✅	✅
Completions	✅	✅	✅	✅	✅	✅	✅
Expand/Shrink Selection	✅	✅	✅	✅	✅	✅	✅
Syntax Validation	✅	✅	✅	✅	✅	✅	✅
Document Highlights	✅	✅	✅	✅	✅	✅¹	✅¹
Reference Search	✅	✅	✅	✅	⭕️	⭕️	⭕️

Anycode can be thought of as being full text search along Tree-sitter parse trees. All features are based on parse trees and there is no semantic information - that means there is no guarantee for correctness. Parse trees allow to identify declarations and usages, like "these lines define a function named foo" or "this is an invocation of a function named bar". In essence, anycode identifies and compares usages and declarations. Looking at each feature individually:

Outline - A parse tree itself is an outline. However, it is too detailed and anycode selects declarations like functions, classes with its memebers, interfaces etc.
Go to Symbol - Anything that would show in outline (declarations) is stored in a trie so that it can be found by typing names or partial names of declarations. How well this works across file depends on the anycode.symbolIndexSize-setting - it defauls to 100 and defines how many files are eagerly fetched and analyzed.
Go to Definition - Works just like "Go to Symbol" but it uses the current identifier as the name to find declarations for. This isn't semantic and depending on your project yields false positives, e.g it will find all run-methods and not just Runnable#run etc
Completions - Any identifier of the current document and the names of all known declarations are always suggested as completions. So, no precise member-completion but identifiers.
Expand/Shrink Selection - Solely based on parse trees. It finds the node at the current position and returns all parents as selection.
Syntax Validation - Tree-sitter parsers can recover from malformed source code. In case error nodes are inserted into the parse tree and anycode can report them as diagnostics. Note that validation is off by default and that it must be enabled via the anycode.language.features#diagnostics-setting.
Document Highlights - For local variables and function arguments usages in their scope are computed and highlighted. If that's not possible, all identifiers with the same name are highlighted - ¹ only identifier based highlights implemented
Reference Search - All usages, such as function-calls or member-accesses, are kept in a trie which is later used for reference search. Like "Go to Symbol" its result depend on the anycode.symbolIndexSize-setting.

Last, there is the anycode.language.features-setting which controls what feature is enabled or disabled - either for all or just specific languages.

Development

To compile tree-sitter languages you need docker or emscripten, follow these steps:

have emcc on your path or docker running
run npm install

There is a watch task to build TS, either npm run watch or "F1 > Run Task > npm: watch". The task will auto launch on next open if desired.

Push a tag to publish a new version to the marketplace:

run npm version path and then
run git push --follow-tags

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.