On Fri, 28 Nov 2025 10:38:39 +0100 =?UTF-8?Q?S=C3=A9bastien_Noel?=
<[email protected]> wrote:
> I'll be happy to sponsor this font once htic clears the NEW queue.
thank you for the review and your proposal to sponsor the package :-)
Thanks for updating the packaging, Let's wait for htic then... :)
that is unfortunate/unexpected... I clearly miss the fact that the RFN
section mandate our build to be bit-for-bit identical to files
distributed by upstream... And they are not, for an still unknown reason
(upstream build is 5 years old, fontforge version is different, may that
be the cause ?)
A different version of fontforge could be the cause. Another possibility
may be the use of fontforge on another platform, say Windows, that leads
to tiny differences in the way floating point numbers are handled.
But, if I had to guess, I'd say that a changed timestamp somewhere is
responsible for the discrepancies.
Try using diffoscope (or diffoscope-minimal) to see if anything pops up.
Feel free to attach the log or the generated file, if you want another
pair of eyes to look over it.
Regards,
--
Gioele Barabucci
