Bug#675968: transition: biosig4c++

[Niels Thykier <niels@thykier.net>]

tags 675968 + pending
thanks

On 2012-06-04 18:03, Yaroslav Halchenko wrote:
> [...]
> 
> Please pardon me for apparently missing an important advisory for not using
> experimental first...


Hi,

I am not entirely sure if this was implied or not in your sentence, but
we would prefer getting a transition bug before the package is uploaded
to sid.  biosi4c++ is currently also involved in the octave transition,
though if openwalnut and sigviewer compiles without any issues there
should not be any issues.

> Transition is tiny -- just 2 dependent packages on libbiosig-dev
> packages (below). "Transition bug" against biosig4c++ source is
> #675967.
> 

The bts is acting up a bit, so I cannot actually read the contents of
#675967 right now.  Anyhow, it appears to be RC, which will stall the
transition so please close it or downgrade it.  For the purpose of
tracking the transition, this bug is sufficient.

> Outstanding FTBFS on ia64 was fixed  (closed now) and I expect it to
> build fine on sparc (which would resolve #633346).

Thanks for solving the ia64 issue so quickly - I see biosig4c++ has
already been rebuilt on all architectures (except sparc and armhf).  I
doubt armhf will cause any issues (given armel succeeded and that armhf
succeesfully built 1.3.0-1).
  As biosig4c++ has not been built on sparc previously, it is less of a
concern for the transition.  But it is nice to see another bug fixed.  :)

> The following source packages need to be rebuilt:
> 
> sigviewer     # this one might need a compatibility patch -- although seems to build nicely without
> openwalnut
> 

We already scheduled the binNMUs for these earlier today and indeed
sigviewer appears to be doing fine (still waiting for mipsel and armhf
though).  openwalut is still building, but I hope it will be as painless
as you expect.

> In terms of 'ben' lingo, my non-experienced guestimate that the transition
> should have the following parameters:
> 
> Affected: .build-depends ~ /libbiosig-dev/
> Good: .depends ~ /lbbiosig1/
> Bad: .depends ~ /lbbiosig0/
> 

There is a tracker at [1].  It uses a slightly different "affected" line
as I wrote the ben file before you reported this bug, but yours would
have been just as good (modulo the "lbbiosig" typo?).  You can ignore
the biosig4c++ being listed as bad (I was too lazy to have biosig4c++
filtered out).

[1] http://release.debian.org/transitions/html/biosig4c++.html

> What should be my course of action -- should I just close a transition
> bug (#675967) whenever transition slot is allocated?
> 
> [...] 
> 

As mentioned above - feel free to close it.  For now we will attempt to
complete the transition immediately as it does not appear to cause a lot
of issues.

~Niels