[PATCH] Check 'x-resource is a specifier tag before passing it to map-specifier

[Ben Wing]

Stephen J. Turnbull wrote:
> Aidan Kehoe writes:
>
>  > (I dislike the design of erroring when a specifier tag hasn't been
>  > seen; this breaks run-time checks for, say, msprinter, for
>  > example. But that's the way the design is right now.)
>
> Ben?
>
> Aidan, what do you think is sane behavior in that case?  It seems to
> me that it's not obvious how matches with '(x not-yet-defined) should
> be done.  In some cases you would want the match to succeed (because
> the semantic requirement can't fail if the option is not there), in
> others, to fail (because the semantic requirement can't succeed), no?
>
> _______________________________________________
> XEmacs-Beta mailing list
> XEmacs-Beta at xemacs.org
> http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
>
>   
Hmmm ...  seems to me that the sane behavior with an unrecognized tag is 
not to match on it.  that way, an instantiator tagged with "msprinter" 
will never apply on X windows, which is probably what you want.

i'm not totally averse to removing the error on unknown tags.  i imagine 
i put in the error on the principle that misspellings should be flagged 
rather than silently ignored ...  but i wasn't considering the situation 
Aidan mentions.  maybe in its place a warning could be issued, at the 
same level as current warnings about unrecognized font/color/etc 
names.   (OTOH you can pretty easily define something like `msprinter' 
unilaterally in all circumstances, even on a non-MSWindows ...)

ben