>>>>"JV" == Jan Vroonhof
<vroonhof(a)math.ethz.ch> writes:
JV> Christoph Wedler <wedler(a)fmi.uni-passau.de> writes:
> That's exactly what font-lock is supposed to do: attaching
faces to
> characters matched by specific (parts of) font-lock keywords.
JV> Look at what the font-lock does in the end: it ataches faces to
JV> regions of the buffer. That sounds like extents to me.
JV> That may be implemented now as "regions of chars with certain
JV> properties".
JV> (Note that the 'add-non-duplicable-text-properties' in questions is
JV> called on regions!)
Should that mean, whenever regions come into the game, extents are more
appropriate? Text props don't have any chance then... ;-)
JV> $ a = \alpha + \int d f $
JV> [ -- math face --- ]
JV> [ sy ] [sy]
I admit it would be nice to have a font-lock like package that handles
syntactical structures in this way (we could also set highlight extents
and maps and so on...), but the point is: font-lock does not work this
way.
JV> I am a bit worried about being able to support the 'append keywords
JV> efficiently.
Exactly, the same for 'prepend. BTW, you get also problems for OVERRIDE
= nil, t or keep.
Conclusion: using ("real") extents would be nice, but would change
font-lock fundamentally.
- Christoph