emacs-elpa-diffs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[nongnu] externals/sly ae4bbca 45/47: Don't force mREPL to use channel t


From: ELPA Syncer
Subject: [nongnu] externals/sly ae4bbca 45/47: Don't force mREPL to use channel thread for everything
Date: Thu, 17 Dec 2020 18:57:22 -0500 (EST)

branch: externals/sly
commit ae4bbcaadc2b65d4f64c99b956c739372eeba1cc
Author: João Távora <joaotavora@gmail.com>
Commit: João Távora <joaotavora@gmail.com>

    Don't force mREPL to use channel thread for everything
    
    An mREPL thread's main function (if it exists) is to serve
    :emacs-channel-send and reply with :channel-send.  It can also serve
    :emacs-rex requests, but those can normally be spawned in a new worker
    thread.
    
    Most implementations don't care, but Clasp, in particular, has what I
    suspect to be a broken SLYNK-BACKEND:THREAD-ID implementation, and so
    finding a specific thread by ID doesn't work so well.  Since this
    shouldn't hurt other implementations, stop setting the Elisp variable
    sly-current-thread in the REPL, thus letting it have its default value
    of t there.
    
    * contrib/sly-mrepl.el (sly-mrepl-new): Don't set
    sly-current-thread.
---
 contrib/sly-mrepl.el | 1 -
 1 file changed, 1 deletion(-)

diff --git a/contrib/sly-mrepl.el b/contrib/sly-mrepl.el
index 90c1397..9c412bf 100644
--- a/contrib/sly-mrepl.el
+++ b/contrib/sly-mrepl.el
@@ -1013,7 +1013,6 @@ handle to distinguish the new buffer from the existing."
             (with-current-buffer buffer
               (sly-mrepl--read-input-ring)
               (setq header-line-format nil)
-              (setq sly-current-thread thread-id)
               (set (make-local-variable 'sly-mrepl--remote-channel) remote)
               (unwind-protect
                   (run-hooks 'sly-mrepl-hook 'sly-mrepl-runonce-hook)



reply via email to

[Prev in Thread] Current Thread [Next in Thread]