-
Notifications
You must be signed in to change notification settings - Fork 2.5k
Support VB.NET (.vb) language — files skipped during indexing #1073
Copy link
Copy link
Open
Labels
enhancementNew feature or requestNew feature or requestlanguage-requestRequest for new language supportRequest for new language supportparsing/qualityGraph extraction bugs, false positives, missing edgesGraph extraction bugs, false positives, missing edgespriority/backlogValuable contribution, lower scheduling urgency; review when maintainer capacity opens.Valuable contribution, lower scheduling urgency; review when maintainer capacity opens.windowsWindows-specific issuesWindows-specific issues
Milestone
Description
Metadata
Metadata
Assignees
Labels
enhancementNew feature or requestNew feature or requestlanguage-requestRequest for new language supportRequest for new language supportparsing/qualityGraph extraction bugs, false positives, missing edgesGraph extraction bugs, false positives, missing edgespriority/backlogValuable contribution, lower scheduling urgency; review when maintainer capacity opens.Valuable contribution, lower scheduling urgency; review when maintainer capacity opens.windowsWindows-specific issuesWindows-specific issues
What problem does this solve?
VB.NET (.vb) project files are completely skipped during indexing. The indexer reports 0 nodes for all .vb source files despite the project containing ~200+ .vb files with classes, functions, interfaces, and events. The project compiles and runs fine — the issue is purely on the indexing side: codebase-memory-mcp does not recognize .vb as a parseable language.
This means users with VB.NET codebases (still actively used in Windows desktop / WPF / WinForms / enterprise environments) cannot benefit from any of the 14 MCP tools — search_graph, trace_path, get_architecture, etc. — for their VB.NET source files.
Proposed solution
Add VB.NET to the supported languages list by vendoring a tree-sitter grammar for Visual Basic and wiring a VB extraction pass.
Tree-sitter grammar: https://github.com/CodeAnt-AI/tree-sitter-vb-dotnet
Required capabilities:
No Hybrid LSP pass is expected initially — syntactic AST extraction would already be a massive improvement over the current zero-node outcome.
Alternatives considered
Users can fall back to grep/glob for searching .vb files, but this loses all structural querying (call graphs, trace_path, dead code detection, architecture overview) that makes codebase-memory-mcp valuable. Converting the project to C# is not a practical workaround for existing codebases.
Confirmations