bug#81637: Regression: Tree-sitter font-lock drops highlighting in indirect buffers
James Cherti <[email protected]>
| Newsgroups | gmane.emacs.bugs |
|---|---|
| Message-ID | <[email protected]> |
On 2026-08-19 01:40, Yuan Fu wrote: > > >> On Aug 17, 2026, at 3:59 AM, Eli Zaretskii <[email protected]> wrote: >> >>> Date: Sun, 16 Aug 2026 18:09:52 -0400 >>> From: James Cherti <[email protected]> >>> >>> There is a regression in tree-sitter where syntax >>> highlighting is completely lost when creating an indirect >>> buffer (via clone-indirect-buffer or C-x 4 c) from a base >>> buffer running a tree-sitter major mode. >>> >>> This worked correctly in Emacs 29 but broke in the recent >>> Emacs 31 and 32 changes. >>> >>> To reproduce: >>> >>> - Open a source file that uses a tree-sitter major mode. >>> - Create an indirect buffer using >>> M-x clone-indirect-buffer (or 'C-x 4 c'). >>> - Observe that the syntax highlighting is entirely or >>> partially missing in the new indirect buffer. >> >> Thanks. Yuan, could you please look into this? > > Thanks for the CC. James, are you able to reproduce it with emacs -Q? Because I can’t reproduce it with either my personal Emacs session or emacs -Q. 🤔 Hello Yuan, Yes, I can reproduce it with emacs -Q. Please see the attached screenshot. I used python-ts-mode in this example, but the bug also occurs with other modes like yaml-ts-mode. Here are the steps to reproduce the issue shown in the screenshot: - Open a Python file. - Switch to python-ts-mode. - Split the window to the right. - Run M-x clone-indirect-buffer. - Important: After cloning the file using clone-indirect-buffer, the initially displayed text is highlighted correctly. However, if you scroll down one to two pages, you will notice that the rest of the indirect buffer isn't highlighted. > > Yuan > > -- James Cherti GitHub: https://github.com/jamescherti Website: https://www.jamescherti.com/
screenshot-emacs-q.png
(image/png, 313.7 KB) - not displayed