On Mon, Feb 6, 2017 at 3:17 PM, Stefan Monnier <monnier@iro.umontreal.ca>
wrote:

> >> I'm not sure what you're referring to here.  Do some of the existing
> >> mixed mode things piggyback on the existing html mode?
> > Both https://github.com/purcell/mmm-mode/ and
> > https://github.com/vspinu/polymode/ do.
>
> BTW: any chance to see some of those in elpa.git some day?
>
> >> I guess my view is that it is always ok to make things in-tree work
> >> better with each other, even at the expense of some code that is
> >> out-of-tree and presumably relying on implementation details to do its
> >> work.
>
> What Dmitry is saying is that it's OK to define an ad-hoc multi-mode for
> html+css+js (after all, they're all part of the HTML standard, oh and we
> should also add SVG in there), but it should be structured as a separate
> mode on top of html-mode, css-mode, and js-mode.
>
> > That seems orthogonal to the possibility of breaking the cache after
> calling
> > syntax-ppss while a different syntax table is in effect.
>
> BTW, the better solution might be to extend syntax.el to provide some
> hooks for that.  Not sure what's the best functionality to export, but
> maybe a good solution is a way to manage several caches.  Or maybe we
> can just have syntax-propertize-function place ad-hoc cache entries
> (via some new syntax-ppss-add-to-cache function) at every
> major-mode boundary.
>
>
>         Stefan
>
>
> Excuse me for not taking part here, but I simply do not have time now. As
you probably know I tried to do these things in MuMaMo (part of nXhtml).
Unfortunately I found that it can not be done without some rather deep
changes to Emacs.

The problem is that basic functions which are used in simple language
parsing in Emacs can't easily be restricted to a region. At least not as
far as I can see. If I am right these basic functions (in C) must be
enhanced to allow this. If that is done then MuMaMo can do the work. (With
some fixes now I guess for changes at the elisp level which I have not had
time to do.)