> > The serializer just needs to remove the _artifacts field and add an > _artifact field. Here's how I did it in docker, which is a total ripoff of > the file plugin. >
I know it's fairly simple to do manually, I just meant to do so automatically (unless we also need a serializer mixin like you said). On Mon, Jan 7, 2019 at 10:41 AM Austin Macdonald <amacd...@redhat.com> wrote: > The serializer just needs to remove the _artifacts field and add an > _artifact field. Here's how I did it in docker, which is a total ripoff of > the file plugin. > > https://github.com/pulp/pulp_docker/pull/291/ > > It might be worth making a serializer mixin also? (I can almost hear > jortel cringing about all these mixins) > > On Mon, Jan 7, 2019 at 10:32 AM Daniel Alley <dal...@redhat.com> wrote: > >> Given that single-artifact Content is likely to be a very common pattern >>> among plugins, maybe it would be best to add this as a mixin for pulpcore. >>> If that's the future of this field, we should definitely make it _artifact. >> >> >> +1 to this, I don't much like having to redefine this in every plugin. >> I'm curious about how to make it work with the serializers though. >> >> On Mon, Jan 7, 2019 at 10:13 AM Austin Macdonald <amacd...@redhat.com> >> wrote: >> >>> We have single-artifact Content in Docker as well. I've gone ahead and >>> named the field _artifact. >>> >>> Given that single-artifact Content is likely to be a very common pattern >>> among plugins, maybe it would be best to add this as a mixin for pulpcore. >>> If that's the future of this field, we should definitely make it _artifact. >>> >>> On Wed, Jan 2, 2019 at 12:24 PM David Davis <davidda...@redhat.com> >>> wrote: >>> >>>> In most plugins, Content only has a single artifact so we created a >>>> virtual field 'artifact' that we expose to end users. In a recent >>>> change[0], we prefixed the Content fields with underscores ('_') so we're >>>> considering renaming the field to '_artifact' to be consistent with other >>>> plugins that have '_artifacts' on Content. We could use some feedback by >>>> sprint planning (Jan 4) either here or on the issue: >>>> >>>> https://pulp.plan.io/issues/4282 >>>> >>>> [0] https://pulp.plan.io/issues/4206 >>>> >>>> David >>>> _______________________________________________ >>>> Pulp-dev mailing list >>>> Pulp-dev@redhat.com >>>> https://www.redhat.com/mailman/listinfo/pulp-dev >>>> >>> _______________________________________________ >>> Pulp-dev mailing list >>> Pulp-dev@redhat.com >>> https://www.redhat.com/mailman/listinfo/pulp-dev >>> >>
_______________________________________________ Pulp-dev mailing list Pulp-dev@redhat.com https://www.redhat.com/mailman/listinfo/pulp-dev
