Reference UGRID conventions in CF #459
Conversation
|
I can't say I carefully proofread it, but it looks good to me! Thanks for all the hard work. |
There was a problem hiding this comment.
This all looks sensible overall to me. I have raised a small number of questions and thoughts in-line, which relate more to the general underlying Issue #153 than the PR itself, but it feels useful to point to specific points from the new text suggested here for context.
Regarding the new data model:
- The update to
cfdm_cf_concepts.svg, extending with the two new variables, seems very natural to me. - As for the main CF Data Model schematic (here in
images/cfdm_field.svg), it took a while for me to get my head around the 'new way', but now I think that it makes sense conceptually.- Concerning specifically the change in the schematic, I thought it could be useful to have some summary of how the data model has changed, because the repositioning of nodes and rearrangement of relationship lines in the diagram could easily make it seem to someone who doesn't take a detailed look that it has changed quite drastically, but all that has really changed is that we have the new constructs 'DomainTopology' and 'CellConnectivity', right?
| A summary indicating how UGRID relates to other parts of the CF conventions, and which features of UGRID are excluded from CF, can be found in <<mesh-topology-variables>>. | ||
| To reduce the chance of ambiguities arising from their accidental re-use, all of the UGRID standardized attributes are specified in <<appendix-mesh-topology-attributes>> and <<attribute-appendix>>. | ||
|
|
||
| The UGRID conventions have their own conformance document, which should be used in conjunction with the CF conformance document when checking the validity of datasets. |
There was a problem hiding this comment.
On this note, I am wondering: how do we expect, if at all, that compliance checker tools such as the CF Checker can verify UGRID-related aspects? Will anything in the CF Conformance Document need updating or amending due to the coupling with UGRID and CF?
There was a problem hiding this comment.
The conformance document has been updated: https://github.com/cf-convention/cf-conventions/pull/459/files#diff-3b9c470edad8a09f463987db632803f1ecc22561199fa5771745ad472a62e0ee - which just devolves conformance to the UGRID conformance document.
There was a problem hiding this comment.
to have some summary of how the data model has changed,
I'm not in favour of writing into the conventions how they are different to a certain previous versions. That is the role of the issue discussion, I think.
There was a problem hiding this comment.
The conformance document has been updated
Aha, sorry I clearly completely missed that when working through the PR 🙂 Devolving sounds wise. All good!
There was a problem hiding this comment.
RE:
I'm not in favour of writing into the conventions how they are different to a certain previous versions. That is the role of the issue discussion, I think.
Fair enough, that seems sensible - but for the record and to check my understanding, since we are linked to the Issue here - is it indeed what I understand from looking in detail, that:
all that has really changed is that we have the new constructs 'DomainTopology' and 'CellConnectivity'
?
|
This is ready to go -- thanks all for the contributions! |
See issue #153 or discussion of these changes.
Note that #153 also depends on https://github.com/cf-convention/cf-convention.github.io/pull/210/files
This PR replaces #353
Release checklist
cf-conventions.adoc?cf-conventions.adocup to date? Versioning inspired by SemVer.history.adocup to date?For maintainers
After the merge remember to delete the source branch.
Tags are set at the conclusion of the annual meeting; until then
masteralways is a draft for the next version.