Google
  Web www.spinics.net

Re: Fixing broken deps - result of new dependency generator?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


On Sun, Jan 10, 2010 at 11:04:41AM -0600, Alan Dunn wrote:
[...]

I'd not realized that all the Fedora mailing lists had moved ...

> My first build of why in a while seems to have broken deps in Rawhide:
> 
> why has broken dependencies in the development tree:
> On x86_64:
>        why-2.23-1.fc13.x86_64 requires ocaml(Logic_decl) =
> 0:f63f0d96a92a779b9e74fc74cfc2423d
>        why-2.23-1.fc13.x86_64 requires ocaml(Ptree) =
> 0:dfb73bbb000b55e87b05c0cdc2c87af6
>        why-2.23-1.fc13.x86_64 requires ocaml(Logic) =
> 0:ddd0d46a3eaa2ce4a1a3d22195514e56
>        why-2.23-1.fc13.x86_64 requires ocaml(Cc) =
> 0:7017e5b498afc612807d5f6fdc2fc257
>        why-2.23-1.fc13.x86_64 requires ocaml(Types) =
> 0:c36604dc5d4202662cfca61be96c3303
>        why-2.23-1.fc13.x86_64 requires ocaml(Error) =
> 0:808ba08d6066ba67d6c769798e909ff8
>        why-2.23-1.fc13.x86_64 requires ocaml(Ast) =
> 0:2c019415b117907d730a302cdac2b248
> On i386:
>        why-2.23-1.fc13.i686 requires ocaml(Types) =
> 0:c36604dc5d4202662cfca61be96c3303
>        why-2.23-1.fc13.i686 requires ocaml(Logic_decl) =
> 0:f63f0d96a92a779b9e74fc74cfc2423d
>        why-2.23-1.fc13.i686 requires ocaml(Logic) =
> 0:ddd0d46a3eaa2ce4a1a3d22195514e56
>        why-2.23-1.fc13.i686 requires ocaml(Cc) =
> 0:7017e5b498afc612807d5f6fdc2fc257
>        why-2.23-1.fc13.i686 requires ocaml(Error) =
> 0:808ba08d6066ba67d6c769798e909ff8
>        why-2.23-1.fc13.i686 requires ocaml(Ptree) =
> 0:dfb73bbb000b55e87b05c0cdc2c87af6
>        why-2.23-1.fc13.i686 requires ocaml(Ast) =
> 0:2c019415b117907d730a302cdac2b248
> 
> I'm not sure what to do to fix this. Is this potentially a result of
> the new dependency generation mechanism? I don't think the spec file
> has any of the types of patterns that need to be changed as listed in
> https://www.redhat.com/archives/fedora-ocaml-list/2009-December/msg00011.html

So I think this is a bug in the dependency generator.  At the moment
it doesn't handle submodules, so for example if you have modules
called M.A, M.B, then it will generated dependency hashes for
ocaml(M), ocaml(A) and ocaml(B).  2 of the 3 are bogus.  I usually
hack round this by telling the dependency generator to ignore named
modules, which is what I did in this case:

http://cvs.fedoraproject.org/viewvc/devel/why/why.spec?r1=1.7&r2=1.8

> . Also, is there a way I might've detected this problem in advance?

If you try to use the rpm package (ie. rpm -Uvh) after building it
locally then you should be able to spot these quickly.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
libguestfs lets you edit virtual machines.  Supports shell scripting,
bindings from many languages.  http://et.redhat.com/~rjones/libguestfs/
See what it can do: http://et.redhat.com/~rjones/libguestfs/recipes.html
_______________________________________________
ocaml-devel mailing list
ocaml-devel@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/ocaml-devel

[Home]     [Fedora Users]     [Fedora Legacy List]     [Fedora Maintainers]     [Fedora Desktop]     [Red Hat 9 Bible]     [Fedora Bible]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [Yosemite Photos]     [KDE Users]     [Fedora Tools]

Powered by Linux

Google
  Web www.spinics.net

Google PageRank Checking tool