Skip to content

Pin odelia at v0.3.1, so a dependent can require it - #91

Open
aornugent wants to merge 1 commit into
traitecoevo:masterfrom
aornugent:PHYLLOPTIM-odelia-031
Open

Pin odelia at v0.3.1, so a dependent can require it#91
aornugent wants to merge 1 commit into
traitecoevo:masterfrom
aornugent:PHYLLOPTIM-odelia-031

Conversation

@aornugent

Copy link
Copy Markdown
Contributor

phylloptim needs only odelia's headers and asks for odelia (>= 0.2.0),
but its Remotes pinned v0.2.1 exactly. A dependent that requires a
later odelia then hands pak two pinned refs for one package and the solve
fails before anything is built.

That is what blocks traitecoevo/plant#619, where TF24's storage pool
declares its domain to the solver and needs odelia's step rejection,
added in 0.3.0.

Nothing here reads the new interface, and odelia 0.3.0 leaves a system
declaring neither hook untouched. Built at -O2 against odelia 0.3.1:
suite 932 pass, 0 fail, 1 skip.

Co-authored-by: Claude Opus 5 (1M context) noreply@anthropic.com

phylloptim needs only odelia's headers and asks for >= 0.2.0, but its
Remotes pinned v0.2.1 exactly. A dependent that requires a later odelia
then presents pak with two pinned refs for one package and the solve
fails before anything is built -- which is what blocks
traitecoevo/plant#619, where the storage pool's domain check needs
odelia's step rejection, added in 0.3.0.

Nothing here reads the new interface. odelia 0.3.0 leaves a system that
declares neither hook untouched, and phylloptim builds clean against
0.3.1 at -O2 with its own suite passing.

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
aornugent added a commit to aornugent/plant that referenced this pull request Aug 14, 2026
The domain check needs odelia >= 0.3.1, and phylloptim pinned odelia at
v0.2.1 in its own Remotes -- two pinned refs for one package, so pak
refused to solve and CI failed before building anything. Verified by
bisection: reverting the odelia pin solves in 95 packages, bumping it
conflicts.

traitecoevo/phylloptim#91 lifts that pin. Taking phylloptim from master
rather than a commit that predates it lets this resolve as soon as traitecoevo#91
lands; with traitecoevo#91 in place the solve returns odelia 0.3.1 and phylloptim
0.2.1. plant does not call set_traits, so master's fourteenth argument
does not reach here.

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant