Closed benjaminjb closed 1 year ago
OK, so v1.22 changes several API we want, then v1.25 did as well cite, so if we test 1.27, 1.24. and 1.21, we actually will hit all the changes.
I was also playing around with using GITHUB_ENV to set a matrix (thinking to randomize the midpoint test), but decided that I wanted this test to be more determinative in what it tested in the midrange.
As much as this satisfies the symmetricist in me, gonna check out some deprecations/changes in the different versions so our testing spread satisfies.