Re: what remains before upload to unstable
Hi Frank, hi all,
On 03/19/2012 05:50 AM, Frank Küster wrote:
Norbert Preining<preining@logic.at> wrote:
So I see a few options, and it would be great if I hear opinions on that:
- extreme 1: ignore duplications,, and add lintian overrides
- extreme 2: ignore diffs, and always link
- tiring middle path: only link those that are cmp equal, keep the others
and add lintian overrides
I am personally for extreme 1.
I don't care much, but anyway we should send a note to the maintainer of
the font package. There's a chance they know a bit about the font
development, or at least care, and find out what the differences are
about faster than we could.
I am not sure ... I checked the libertine fonts, that are all different.
It turns out that the actual .otf fonts have the same version number,
but are generated with different versions of fontforge, at different
times.
I compared the versions of Debian libertine package, TeX Live, and what
I can download from www.linuxlibertine.com webpage or so. All three
versions are different.
BUT: I thought about it a bit and decided the following: Since ttf/otf
fonts normally ship their metrics within the font, and those engines
that use otf/ttf (xetex and luatex) are not using afm or tfm for these
fonts, I suggest that we simply DROP all fonts as is and add
depends and simple links to get it right. I.e., extreme 2 from above.
What do you and maybe someone else think?
BTW, here is a statistic:
all different are the fonts from the following packages:
fonts-oflb-asana-math (Asana-Math.otf)
otf-freefont (Free*)
fonts-linuxlibertine (Lin*)
fonts-comfortaa (Comfortaa*)
ttf-dejavu* (DejaVu*)
fonts-droid (Droid*)
ttf-freefont (Free*)
fonts-junicode (Junicode*)
fonts-hosny-amiri (amiri*)
all fonts are the same in the following packages:
fonts-gfs-* (GFS*)
fonts-inconsolata (Inconsolata)
fonts-sil-gentium* (Gen*)
fonts-ipa* (ipa*)
ttf-marvosym (marvosym.ttf)
Norbert
Reply to: