Closed cboettig closed 4 years ago
here is the (closed) issue: #31 I'm suggesting reversing that now that EML 2.2.0 is released
Would changing the NS from 2.1.1 to 2.2.0 represent a breaking change? The user-facing API looks version-independent. EML 2.1.1 -> 2.2.0 itself is a backwards compatible shift schema-wise. I'd say 0.3 is fine.
I'll PR this in a few.
Thanks. I think the only thing that would break is if you had been previously generating EML 2.2.0 using emld
-- I think you'd run into trouble parsing / validating etc. But I suppose that counts as a bugfix since the eml://
namespace is essentially invalid in 2.2.0, yes? In any event, since it wasn't the default behavior it is probably an edge case. So I'm happy just calling this 0.3.0 on account of the new default behavior to 2.2.0.
Thanks @amoeba ! @jeanetteclark 2.2.0 is also now the default version, https://github.com/ropensci/emld/pull/40/files#diff-66a2f207a5d0d3070029a1346bada357R14
Will send this up to CRAN
Thanks @cboettig and @jeanetteclark!
v0.3.0 is now on CRAN
This is just a follow-up on PR #40 to discuss issues for a new release.
[x] Need to think what the right version bump ought to be and update DESCRIPTION & NEWS.md (Maybe 0.2.1 or 0.3.0?, but really the namespace is a breaking change, and changing the default to 2.2.0 would be a significant shift as well...)
[x] Need to update NEWS.md to reflect PR #40 change in NAMESPACE
[x] @jeanetteclark notes: Before cutting the new release, we might want to switch back to setting the default EML version to 2.2.0 now that this is officially out. I created an issue (#31) requesting the switch to 2.1.1 until 2.2.0 got released.