bug#81602: 32.0.50; Is this a bug? setopt on a not-yet-loaded option breaks loading the file that defines it
Rahul Martim Juliato <[email protected]>
| Newsgroups | gmane.emacs.bugs |
|---|---|
| Message-ID | <[email protected]> |
Michael Heerdegen <[email protected]> writes: > Rahul Martim Juliato <[email protected]> writes: > >> I noticed some strange errors after trying to compile from master today. >> >> I'm not sure whether this is a bug, although I suspect it might be, so I >> thought I'd ask and share my findings. > > It is not only you: my Emacs also doesn't start any more after > rebuilding today because of an analogue problem. > > Should we revert the crucial commit for now? Maybe it is something easily fixable by Stefan. In the meantime, I patched this hacky hack into the top of my init.el. See if it helps in your case too: (when (and (fboundp 'setopt--set) (let ((probe (make-symbol "emacs-solo--setopt-probe"))) (setopt--set probe "probe") (not (listp (car (get probe 'custom-check-values)))))) (defun setopt--set (variable value) (custom-load-symbol variable) (let ((type (get variable 'custom-type))) (if (not (or type (get variable 'standard-value))) (push (list value) (get variable 'custom-check-values)) (unless (widget-apply (widget-convert type) :match value) (warn "Value does not match %S's type `%S': %S" variable type value)))) (funcall (or (get variable 'custom-set) #'set-default) variable value))) -- Rahul Martim Juliato