@mjbvz Hello.. Need some clarification

This is because the extension markdown-language-features lacks the workspace symbol provider similar to workspace provider for typescript

Am i right or am going in a wrong direction ?

Yes, the WorkspaceSymbolProvider is what powers this feature. The high level flow of what provideWorkspaceSymbols should do for markdown:

1. Find all markdown files in the current workspace
2. For each file, use markdown's TableOfContentsProvider to extract out the basic heading info for each document
3. Convert the heading info into SymbolInformation

We'll have to see how poorly this performs in a large workspace. We may need cache the table of contents for each file so that we don't have to reparse the files each time

Let me know if you run into any problems looking into this or have any questions

Hi @mjbvz ,

I tried to implement the above and its partially working. The only problem i face is I am using

workspace.textDocuments

which just fetches the opened docs in the current worksapce. Is there a way to fetch all the text documents ?

Fixed by #47610