On Oct 13, 2015 7:50 PM, "Robert Collins" <robe...@robertcollins.net> wrote: > > On 14 October 2015 at 12:39, Wes Turner <wes.tur...@gmail.com> wrote: > > > .... > > (.cfg | .yml) > JSONLD > > > > Should these build settings all be included in e.g. PEP 426 + JSONLD > > metadata.jsonld? > > - for backward compat, there would then be both metadata.json and > > metadata.jsonld (because of normative JSONLD form and PEP 426+) > > I'm sorry, I have literally no idea what you're talking about. I know > you have a thing about JSONLD, but I don't understand how or why it is > relevant to these design discussions, which are 99% fine tuning > backwards/forwards compat needs around the existing Python code thats > out there and will be interacting with what we produce. > > If you want to shift the needle here and get discussions talking about > things with JSONLD rather than PEP-426 (or some future leaner and less > enterprisy version) ... then for me at least, I need more clear > communication, less assumption that I have *any* overlap with where > you're coming from, and at least some connection to the problems we're > working on in your mails. > > > .cfg (ConfigParse): (section, key, value) > > > > .yml (YAML lib xyz): easily maps to JSONLD (which always maps to an RDF > > triple graph (of linked typed packages with attributes, generated from > > [...])) > > Still not following.
* egg-info and dist-info should generate JSONLD * sdist, bdist, and wheel should update **append to** the same JSONLD metadata document * if there are additional parameters which would be necessary to reproduce a package(s), those should be included in the package metadata ... the *provenance* of these build artifacts would then be represented in the package metadata document is this specific to sdist? nope, sorry, now is the time. > > -Rob > > -- > Robert Collins <rbtcoll...@hp.com> > Distinguished Technologist > HP Converged Cloud
_______________________________________________ Distutils-SIG maillist - Distutils-SIG@python.org https://mail.python.org/mailman/listinfo/distutils-sig
