Daniel Mendler

@artemkovalyov What's going on in the second screenshot of https://github.com/oantolin/embark/pull/560#issuecomment-1328360294? Why do we see the results of the embark export buffer? It shouldn't be backed by a file.

@artemkovalyov Thanks for the refactoring. From my side this looks good. However I haven't tested this yet. @oantolin can surely give you some additional comments.

Good idea. You can use the respective code from consult which is based on isearch/org-fold. https://github.com/minad/consult/blob/b404039f567ce296823425b641b2348482a1750e/consult.el#L1344-L1354

It seems one needs a special EXWM kill command which can be bound as kill action in the embark-buffer-map? See https://github.com/ch11ng/exwm/issues/576#issuecomment-543201407. In any case, this should not be part of...

I meant something more primitive, just marking/removing the candidate directly to indicate the manipulation, but only in collect buffers.

I think it is okay to keep the candidates selected, in order to act again for example. It is also easy to unmark all via `C-. A SPC`. What about...

Better variant: ~~~ elisp (advice-add #'embark-select :after (lambda (&rest _) (force-mode-line-update t))) (defun embark--selected-mode-line () (when-let ((sel (buffer-local-value 'embark--selected (or (when-let ((win (active-minibuffer-window))) (window-buffer win)) (current-buffer))))) (format #(" Embark:%s "...

I think showing this information always is warranted since the selection is some state which is there even if you are not inside the `embark-act` context. But of course this...

`embark--selected-mode-line` is maybe simple enough for the wiki or only of wiki-quality anyway. In my setup I try to keep as few custom additions as possible, so if it would...

> OK, that sounds reasonable, but I think the mode line indicator should be wrapped up in a minor mode, no? Automatically messing with people's mode lines is a sure-fire...