On 10/01/2013 05:27 PM, Scott Kostyshak wrote:
On Sun, Sep 29, 2013 at 10:53 AM, Jean-Marc Lasgouttes
<lasgout...@lyx.org> wrote:
Le 28/09/13 22:56, Scott Kostyshak a écrit :
Ah, you mean that something could go wrong if someone tries to define
an inset with a name that already exists? Do we check for this? Is
inset-forall the only place where there could be such a problem?
No, what happens now is that InsetLayout can change the way tabulars are
displayed or their LaTeX output. In general it is nice (for example one can
do nice tricks with branches this way), but it can give surprising results.
Ah, I didn't know we allowed overriding of built-in insets. This is nice.
In this case, though, I tend to think that there is no problem. And a nice
way to find out is to enable it :)
I still don't understand the risks. Should I wait for 2.2 to commit this?
I'd commit immediately after 2.1.0 is out. It's not a format change, so
it would be fine for 2.1.1.
rh
