On 04/16/2018 09:10 PM, Joel Kulesza wrote:
> On Mon, Apr 16, 2018 at 7:06 PM, Richard Kimberly Heck
> <rikih...@lyx.org <mailto:rikih...@lyx.org>> wrote:
>
>     On 04/16/2018 08:56 PM, Scott Kostyshak wrote:
>     > On Tue, Apr 17, 2018 at 12:45:39AM +0000, Joel Kulesza wrote:
>     >> On Mon, Apr 16, 2018 at 12:08 PM, Richard Kimberly Heck
>     <rikih...@lyx.org <mailto:rikih...@lyx.org>>
>     >> wrote:
>     >>
>     >>> One polishing thing I've considered is to allow this to be done by
>     >>> clicking on the spinning wheel thing that pops up during
>     compilation (or
>     >>> maybe double clicking), and possibly to bind this to some key. Any
>     >>> thoughts?
>     >>>
>     >> Regarding polishing, and spoken in ignorance because I haven't
>     yet tested,
>     >> one issue I'm often caught by is beginning a compile, spotting
>     an error,
>     >> and correcting it before the compile finishes.  Does a check
>     for document
>     >> consistency during the render make sense?  For example, a
>     dialog to ask if
>     >> the compilation should be restarted if changes are detected? 
>     Perhaps this
>     >> is best set to be user-configurable so that the dialog doesn't
>     become a
>     >> nuisance with current behavior kept as the default?
>     > So suppose I'm compiling, and then I start typing. If this
>     feature is
>     > enabled, a dialog would pop up when I press a key? Perhaps
>     instead, we
>     > could add a menu entry next to "Cancel Background Process" along the
>     > lines of "Restart Compilation", that is also only visible when
>     > compiling? Or we could list it as an example in the LFUN doc. e.g.
>     > "consider making a keyboard shortcut for <<command-sequence
>     > export-cancel; buffer-export>> to restart compilation".
>
>     I often compile to check something, and keep typing on purpose.
>     Indeed,
>     part of the point
>     of background compilation is to allow that sort of thing. So I'd be
>     inclined not to do this
>     kind of thing. It's easy enough to cancel manually in that case and
>     re-start---though we could
>     make canceling easier (which was the original question).
>
>
> Sorry I didn't clarify, as I noted in my reply to Scott, I'd imagine
> the query would be triggered by saving the document with the changes
> during the ongoing compile.

I do that, too, since I save very often.

Riki

Reply via email to