neovim

commit ca26ec34386dfe98b0edf3de9aeb7b66f40d5efd
parent 21187e1c73e6887ce407164c98320555374387a9
Author: Gregory Anders <8965202+gpanders@users.noreply.github.com>
Date:   Mon,  5 Jun 2023 08:21:23 -0500

fix(lsp): use only utf-16 in default client positionEncodings (#23903)

The Nvim client does not yet support multiple offset encodings for
clients in the same buffer. Until it does, stick to utf-16 by default.
Diffstat:
M
runtime/doc/news.txt
 | 
8
+++-----
M
runtime/lua/vim/lsp/protocol.lua
 | 
2
--

2 files changed, 3 insertions(+), 7 deletions(-)
diff --git a/runtime/doc/news.txt b/runtime/doc/news.txt
@@ -44,11 +44,9 @@ The following new APIs or features were added.
 • Neovim's LSP client now always saves and restores named buffer marks when
   applying text edits.
 
-• Nvim's LSP client now advertises the general.positionEncodings client
-  capability to indicate to servers that it supports utf-8, utf-16, and utf-32
-  encodings. If the server responds with the positionEncoding capability in
-  its initialization response, Nvim automatically sets the client's
-  `offset_encoding` field.
+• Nvim now supports the `positionEncoding` server capability. If a server
+  responds with the `positionEncoding` capability in its initialization
+  response, Nvim automatically sets the client's `offset_encoding` field.
 
 • Dynamic registration of LSP capabilities. An implication of this change is
   that checking a client's `server_capabilities` is no longer a sufficient
diff --git a/runtime/lua/vim/lsp/protocol.lua b/runtime/lua/vim/lsp/protocol.lua
@@ -636,9 +636,7 @@ function protocol.make_client_capabilities()
   return {
     general = {
       positionEncodings = {
-        'utf-8',
         'utf-16',
-        'utf-32',
       },
     },
     textDocument = {