Hi Greg, thanks for your input. Regarding this:

> I find the choices that give rise to this approach to be problematic.
> Containers are a useful device for someone to choose, but if people feel
> they have to use a container, that's a clue that the situation is
> troubled.

> I suspect you are running into packages which require an exact
> dependency, rather than the proper approach of having a minimal version,
> trying hard to not to raise that without cause, and adapating to any
> released version >= the minimum.   Some upstreams, influenced by
> language-specific packaging tools, take the approach that their package
> is the most important thing in the world and therefore it's ok to insist
> on exact versions.

> So far, it seems that qgis does very well on not being unreasonable
> about dependencies (as long as gcc 12 turns out to be ok; see below).

If the QGIS repo is good at managing its dependencies in a way that limits its 
impact on the rest of my system - that's great. The problem is that any other 
repo could manage its dependencies differently. Over the past year I've 
probably worked across 50 different repos, many of which I do not maintain and 
cannot manage their dependencies. If I don't attempt to isolate each project 
within Conda, containers, or a VM then I am at risk of surprise breakages from 
what should be unrelated repos. In any repos that I do maintain I provide a 
containerised option whenever possible to ease the lives of other contributors.

> However, I think it's important for it to be reasonable to build without
> containers.

I agree - I'm not suggesting QGIS should _only_ build and run in a container, 
but I think the option of building and running in a container could be better 
supported and I'm happy to contribute to that effort if others also want it.

Tom
_______________________________________________
QGIS-Developer mailing list
[email protected]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Reply via email to