These are chat archives for WebDevStudios/CMB2

4th
Sep 2015
PG Lewis
@pglewis
Sep 04 2015 00:17
we were discussing add_group_field() earlier... that's actually one of the methods that simplifies down nicely due to the API changes:
because you can just call add_field on the parent object, and you can always get the parent object, it boils down to very very simple logic
PG Lewis
@pglewis
Sep 04 2015 00:33
backing up a few steps, when you do get the chance I'd like to verify you have our branch working for display the same way I do, when we left off we hadn't achieved that yet
Justin Sternberg
@jtsternberg
Sep 04 2015 04:10
Just tested, and is working as expected
as in, it displays the group, but not the group's values
PG Lewis
@pglewis
Sep 04 2015 04:11
right, that's as expected
Justin Sternberg
@jtsternberg
Sep 04 2015 04:11
so we're on the same page
PG Lewis
@pglewis
Sep 04 2015 04:11
it's safe to assume the remaining pieces /could/ be wired up
non grouped fields load save as normal, I only broke saving on the group fields when I swicthed them over to support nesting
I think the next step is decide if you want to adopt this approach and come up with a roadmap for working it into trunk
it's become far enough removed that merges aren't likely to work 1 to 1 and maintaining two branches is no fun
assuming we can use trunk as a beta testing branch, so we can also hit hit hard with code in the wild, that could be an option
I think a couple more days' work and it could be in shape enough to hit with further testing if we continued on
Justin Sternberg
@jtsternberg
Sep 04 2015 04:15
I'd say once saving works for grouped fields, we can look more seriously at merging to trunk
PG Lewis
@pglewis
Sep 04 2015 04:16
yeah, that's a bare minimum
and all tests passing, and all metabox example code I have so far working
Justin Sternberg
@jtsternberg
Sep 04 2015 04:16
as long as everything works as it should (to our best knowledge)
sounds good
PG Lewis
@pglewis
Sep 04 2015 04:16
totally agree there
this is already a third wave of step-wise refinement
first broke out the group field logic to its own class
second did away with meta_box
and this one did away with the field arrays
it's worked out fairly well so far as long as I get each pass as stable as possible before the next refactor
we can touch base again further next week, I'll have it whipped into shape the rest of the way by then
Justin Sternberg
@jtsternberg
Sep 04 2015 04:20
Sounds awesome. Again, i appreciate your hard work here
PG Lewis
@pglewis
Sep 04 2015 04:21
next phase after that: sell you on a CMB2_Field refactor
I cringe a little, it's going to be harder than the metabox object
Scott Kingsley Clark
@sc0ttkclark
Sep 04 2015 15:11
exciting, we're getting places :)