Hi all:

I have now read through the recent messages more carefully, duplicated the hang in c++-mode using pillowtrucker's init.el and his drm_formats.cpp file simply by invoking M->.  I then added both Alan's cc-defs.el fix and his new c++-defun-prompt-regexp and that resolved the hang.  If I add the new c++ regexp without the cc-defs.el fix, that also seems to solve the hang.  So I retract my prior message and acknowledge one source of the problem has been the existing Hyperbole regexps.  We'll update Hyperbole to use Allen's new regexps and hopefully this will close this issue after pillowtrucker does some testing.  (I think Allen's other fixes should be applied to and that they will help performance).

Regards,

Bob

On Sat, Apr 20, 2024 at 9:14 AM Robert Weiner <rswgnu@gmail.com> wrote:

> On Apr 20, 2024, at 5:22 AM, Alan Mackenzie <acm@muc.de> wrote:
>
> Hello, M.
>
> Yes, I can understand that. As I said, I'll fix c-beginning-of-defun-1
> so that that can't possibly happen again, regardless of what happens to
> these regexps.

I think this is the crux of the fix needed, not a change in Hyperbole regexps. Alan, Hyperbole had your new defun regexp for Java and Emacs still would hang until we applied your recent fix to cc-defs.el. With that in place and no change to the c++ defun regex in Hyperbole, I have not been able to trigger a hang. So let’s just add your additional fix described above, skip the c++ defun rewrite and see if any other issues appear.

Cheers,

Bob