Re: Mass Bug Filing: Missing Build-Depends: graphviz
On Sun, Mar 20, 2016 at 06:51:23PM +0000, Bas Wijnen wrote:
> That also means that programs calling dot will need graphviz in their
> Build-Depends, no matter what the default is.
As is, a number of them do call dot without the build-dependency.
> > But this is inconsistent with having graphviz in the Suggests line for
> > doxygen.
>
> I agree that if it is the default, it should be Recommends, not Suggests. This
> doesn't change anything for the problem you're describing, however.
>
> > The disappoining moral for this is that nobody looks at their build logs...
>
> I don't think that's disappointing at all! It means we have built a system
> that will let us know when something is wrong.
That would be the case if doxygen propagated the error, which it does not.
> That means we don't need to poll for errors, because they will be pushed
> to us. (I think porters are still doing this manually at least some of
> the time; I think FTBFS bugs should be reported automatically.)
>
> In other words, my solution to this bug would be to make doxygen exit with an
> error code when calling dot fails. Then make will fail, it's an FTBFS, it gets
> fixed, and everyone is happy.
I've started a rebuild of all 552 packages in unstable that build-depend on
doxygen or anything that recursively pulls doxygen (eclipse-eclox doxyqml
doxygen-latex doxygen-gui doxygen-dbg python3-breathe python-breathe).
My build environment has a /usr/bin/dot that aborts the build when called,
which should detect this error even if doxygen output is redirected or
mangled. I'll let you know how widespread the FTBFS are when finished.
On a slow-ass machine, 57 builds done in 2.5 hours so far, so it'll take a
while.
--
A tit a day keeps the vet away.
Reply to: