[Pkg-javascript-devel] node-buble_0.20.0+~0.19.2-1~bpo10+2_amd64.changes ACCEPTED into buster-backports

[Jonas Smedegaard]

Quoting Pirate Praveen (2021-01-14 15:33:43)
> 
> 
> On Thu, Jan 14, 2021 at 3:20 pm, Jonas Smedegaard <jonas at jones.dk> 
> wrote:
> > When I say that virtual package should be possible to prioritize 
> > through versioning, I don't mean nudging build environment to use 
> > different hinting to their various resolvers, but explicit 
> > contraints in declared package relations.
> > 
> > If the real package is resolved as a solution then it means the 
> > package permit that other version (and because real is treated as 
> > better than virtual, it wins).
> > 
> 
> That is not the case here, as node-acorn (>= 6~) cannot be satisfied 
> by the real package node-acorn 5.5.3

Sounds like you know with great confidence (so I will refrain from 
sharing my additional related theories here, to not annoy you further).


> > Try tighten (build-)dependencies to only possibly match the virtual 
> > package.
> > 
> 
> That is already done.

Interesting.  Then it sounds like a bug in the infrastructure.


> > That should lead to the virtual package getting installed.
> > 
> 
> No it does not, you can try it yourself if you don't believe me.

I believe you, and I am confident that you are doing the best that you 
can (and that "the best that you can" is quite good).

That said, I do not expect you to be an all-knowing deity (sorry), and 
am sharing what as I understand it is the expected behavior of Debian 
infrastructure - not directly to say "you are doing it wrong" but 
slightly more vaguely aid in moving this issue forward, e.g. through 
"oh, I didn't know that detail" or "sure, I know that and that can 
definitely be ruled out as a cause".

Hope that is acceptable clarification.


> > If it leads to no package being resolved or the unacceptably 
> > versioned real package getting resolved, then file a severe bug 
> > against the resoilver or build environment.
> 
> I don't know if that can be a severe bug. But reported against 
> aptitude and aspcud now.

Good.

If you share bug numbers here it may encourage some reading this to 
contribute to solving those bugs.  If that was the case already when you 
wrote the changelog entry, then for future sake you can consider a "(see 
bug#123456)" notice to such changelog antry for the same reason: steer 
potential help towards where they do most good instead of bothering busy 
developers with silly questions about things already well understood and 
ruled out.

Thanks for taking the time to share these details.


 - Jonas

-- 
 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-javascript-devel/attachments/20210114/6bc6afff/attachment.sig>