[Fontconfig-bugs] [Bug 71287] size specific design selection support in OS/2 table version 5

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=71287

--- Comment #4 from Akira TAGOH <akira at tagoh.org> ---
(In reply to comment #3)
> - Why the renaming in FcConfigEvaluate()?  It takes a large chunk of the
> patch unnecessarily,

Because those vl and vr is destroyed at the end but FcValue returned by
FcConfigPromote() isn't malloc'd one. why it wasn't a problem ever is this
promotion won't be happened for FcLangSet in FcConfigEvaluate() but
FcConfigCompareValue().

> - In _FcValuePrintFile, perhaps better to print range using () instead of {}?

Aha. well, I don't have strong opinion on that format. just followed on
something we had a discussion for the range on the list recently.

> - I think we want to make the FC_SIZE to be able to accept both double and
> range.  I think you've got that already.  As such, please limit
> FcDefaultSubstitute changes to use double when a range is not necessary. 
> Ideally we shouldn't see any changes to that function whatsoever.  In fact,
> I think it would be fine to expect that the query-pattern has a double size,
> not a range size.  Think of this as FC_LANG.  The font patterns have a
> FcLangSet object for FC_LANG, but query-pattern has a string FC_LANG.

Sure.

> - I'd separate the cache version jump to a separate commit.

Okay.

> - FcCompareSizeRange() should return the minimum distance from one range to
> another.  If the two ranges overlap, that's zero.  Otherwise, return the
> minimum distance from one range's end to the other's.

Aha. sounds good.

> - Again, in FcNameParse, keep thee AddDouble as is.  Possibly add a way to
> specify a range, but that's not really important at least initially.

will fix that.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/fontconfig-bugs/attachments/20131210/576a1403/attachment.html>