I think we have nothing to do for that anymore, because it has to be done before where you modified and need to sort out against the given namelang. sorry for confusion. it looks similar but actually had to be done by different method.
On Tue, Jan 15, 2013 at 5:41 AM, Behdad Esfahbod <behdad@xxxxxxxxxx> wrote:
Ok, I've pushed this to master.
Akira, can you clean up the copy_lang now?
On 13-01-10 10:54 PM, Akira TAGOH wrote:
> Right. I've fixed partly this issue for namelang related stuff in 2.10.2. that
> would makes sense if user-defined values is copied from the pattern to the
> result. that said we may need to test it carefully if it has any side-effects
> in apps.
>
>
> On Fri, Jan 11, 2013 at 7:36 AM, Behdad Esfahbod <behdad@xxxxxxxxxx
> <mailto:behdad@xxxxxxxxxx>> wrote:> Fontconfig@xxxxxxxxxxxxxxxxxxxxx <mailto:Fontconfig@xxxxxxxxxxxxxxxxxxxxx>
>
> The SubstituteWithPat function copies the "best match" element value from the
> pattern to the font pattern for any element not in the font pattern. However,
> for custom elements that ends up just copying the first value. Which
> typically is NOT what you want. Typically you want all values to be copied.
> Right?
>
> For example, if I do this:
>
> <match target="pattern">
> <edit name="pangofontfeatures" mode="append">
> <string>smcp</string>
> <string>ss20</string>
> </edit>
> </match>
>
> only one of the strings makes it to the other side.
>
> Thoughts?
>
> --
> behdad
> http://behdad.org/
> _______________________________________________
> Fontconfig mailing list
--
behdad
http://behdad.org/
Akira TAGOH
_______________________________________________ Fontconfig mailing list Fontconfig@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/fontconfig
