Hello, crowd. A question, that I don't think deserves a bugreport, illustrated with an example in reverse order.
After a fresh rsync on a sort-of fresh system:
livecd ~ # emerge -pvt gtk-engines
These are the packages that I would merge, in reverse order:
Calculating dependencies ...done!
[ebuild N ] x11-themes/gtk-engines-2.2.0 0 kB
[ebuild N ] media-libs/imlib-1.9.14-r3 0 kB
[ebuild NS ] x11-libs/gtk+-1.2.10-r11 -debug +nls 0 kB
However, I don't want to have gtk+-1.2 installed for no reason at all.
As you can see, it is not installed already, and yet, according to
the gtk-engines ebuild:
[ -n "${HAS_GTK1}" ] && DEPEND="${DEPEND} >=media-libs/imlib-1.8"
Well, I *do not* have GTK1, that HAS_GTK1 refers to, installed. I tried
to look at the gtk-engines2 eclass... but just touching it, worked as
well:
livecd ~ # touch /usr/portage/eclass/gtk-engines2.eclass
livecd ~ # emerge -pvt gtk-engines
These are the packages that I would merge, in reverse order:
Calculating dependencies QA Notice: has_version() in global scope: eclass
gtk-engines2
QA Notice: has_version() in global scope: eclass gtk-engines2
QA Notice: has_version() in global scope: eclass gtk-engines2
QA Notice: sed in global scope: x11-themes/gtk-engines-2.2.0
...done!
[ebuild N ] x11-themes/gtk-engines-2.2.0 0 kB
The gtk-engines2.eclass has a has_version and then sets the HAS_GTK1
variable. This has come up before many times, but is there something in
the works that solves this kind of problem? Is this a bug in the eclass,
since the message "QA Notice: has_version() in global scope:" sounds
like it is a Bad Thingâ and has to be avoided? More use flags maybe?
gtk2only, gtk1 or anything to stop this madness? Preventing metadata
from being generated for ebuilds that generate "QA Notice"s?
--
\/ Georgi Georgiev \/ Iowa State -- the high school after high \/
/\ [EMAIL PROTECTED] /\ school! -- Crow T. Robot /\
\/ +81(90)6266-1163 \/ \/
pgph1LA6v8qKf.pgp
Description: PGP signature
