1 |
Hi, |
2 |
|
3 |
Portage does not turn on USE flags of a package to satisfy a |
4 |
dependency, which is sometimes confusing. |
5 |
|
6 |
See https://bugs.gentoo.org/372513 |
7 |
|
8 |
mattst88@localhost ~/projects/gentoo-x86/sci-chemistry $ emerge cns -vp |
9 |
|
10 |
These are the packages that would be merged, in order: |
11 |
|
12 |
Calculating dependencies... done! |
13 |
[ebuild N ] sys-libs/libstdc++-v3-3.3.6 USE="(multilib) nls" 23,459 kB |
14 |
[ebuild N ] app-shells/tcsh-6.16 USE="perl -catalogs" 869 kB |
15 |
[ebuild N ] app-emulation/emul-linux-x86-compat-20100611 |
16 |
USE="(multilib)" 930 kB |
17 |
[ebuild N ] virtual/libstdc++-3.3 0 kB |
18 |
[ebuild N ] dev-lang/ifc-10.0.026-r1 40,378 kB |
19 |
[ebuild N ] virtual/fortran-0 USE="openmp" 0 kB |
20 |
[ebuild N F ] sci-chemistry/cns-1.2.1 USE="openmp" 31,981 kB |
21 |
|
22 |
Total: 7 packages (7 new), Size of downloads: 97,614 kB |
23 |
Fetch Restriction: 1 package (1 unsatisfied) |
24 |
mattst88@localhost ~/projects/gentoo-x86/sci-chemistry $ USE=fortran |
25 |
emerge cns -vp |
26 |
|
27 |
These are the packages that would be merged, in order: |
28 |
|
29 |
Calculating dependencies... done! |
30 |
[ebuild R ~] sys-devel/gcc-4.5.2 USE="fortran* graphite gtk lto |
31 |
mudflap (multilib) nls nptl openmp (-altivec) -bootstrap -build -doc |
32 |
(-fixed-point) -gcj (-hardened) (-libffi) -multislot -nocxx -nopie |
33 |
-nossp -objc -objc++ -objc-gc -test -vanilla (-n32%) (-n64%)" 0 kB |
34 |
[ebuild N ] app-shells/tcsh-6.16 USE="perl -catalogs" 869 kB |
35 |
[ebuild N ] virtual/fortran-0 USE="openmp" 0 kB |
36 |
[ebuild N F ] sci-chemistry/cns-1.2.1 USE="openmp" 31,981 kB |
37 |
|
38 |
Total: 4 packages (3 new, 1 reinstall), Size of downloads: 32,849 kB |
39 |
Fetch Restriction: 1 package (1 unsatisfied |
40 |
|
41 |
I'm sure there are good reasons for this behavior, but I'm not sure |
42 |
what they are. |
43 |
|
44 |
I wonder if there's some way we can manage this kind of situation? |
45 |
Perhaps portage could print alternative dependencies for virtuals, |
46 |
similar to the very helpful recent "The following keyword changes are |
47 |
necessary to proceed:" addition. |
48 |
|
49 |
Another similar situation is with ACCEPT_LICENSES. Attempting to |
50 |
emerge something like virtual/jre will suggest adding keywords for |
51 |
dev-java/ibm-jdk-bin, even though I'm on a mips system and this |
52 |
package will never run there. Suggesting other packages that fulfill |
53 |
virtual/jre would be much easier to understand. |
54 |
|
55 |
Thanks! |
56 |
Matt |