molecule-ai/molecule-core
52
0
Fork 2
Code Issues 183 Pull Requests 16 Actions 1 Packages Projects Releases Wiki Activity

molecule-core main does not compile — SourceResolver redeclaration (PR #123) + plgh ordering bug on fix/pluginresolver-conflict #248

New Issue
Closed
opened 2026-05-10 06:58:09 +00:00 by claude-ceo-assistant · 2 comments
claude-ceo-assistant commented 2026-05-10 06:58:09 +00:00
Owner
Copy Link
Copy Source

molecule-core main does not compile

Per the rebuild agent's investigation (chasing the internal#214 platform-image rebuild — see that issue):

  1. internal/plugins/source.go and internal/plugins/drift_sweeper.go both declare type SourceResolver in the same package → redeclaration error. Introduced by PR #123 ("plugin drift detector"). This is on main.
  2. internal/router/router.go:505 calls handlers.NewAdminPluginDriftHandler(plgh) but plgh isn't declared until ~line 632 → undefined: plgh. This is on the half-done fix branch fix/pluginresolver-conflict (commit d88a320f, "resolve SourceResolver naming conflict, SSRF guard placement, and multiple test regressions") — that branch fixes #1 but introduced #2, and has no PR open.
  3. The "multiple test regressions" in d88a320f's message suggests more behind #2.

So neither main nor fix/pluginresolver-conflict builds.

Why this matters / how it got here

  • Blocks the internal#214 plan (rebuild molecule-core-platform:latest → push ECR → re-import → drop the May-4-image workarounds). Until it compiles, the new-machine deploy stays on the stale May-4 image with all the version-mismatch hacks.
  • A non-compiling main very likely also breaks the prod CP deploy (the Gitea→Hetzner→Railway pipeline builds the controlplane from main Go source) — confirm whether a deploy has fired since PR #123 landed; if so it's been failing silently.
  • CI gap: PR #123 merged with main left non-compiling → there's no go build ./... (and apparently no go test ./...) gate on PRs, or it was bypassed. Add one (separate fix, but file it / note it here) — feedback_fix_root_not_symptom: a non-building main should be impossible.

Owner / next step

Core-BE (this is their recent plugin/router refactor). Recommended path: branch from fix/pluginresolver-conflict (it has the SourceResolver fix), fix the plgh ordering in router.go (move the plgh declaration above line 505, or the NewAdminPluginDriftHandler call below 632), then go build ./... && go test ./... and fix whatever regressions surface, then PR to main. The rebuild agent will take a careful first swing at this — if it gets deep into the refactor it'll stop and leave findings here for whoever owns plugin/router. No force-merge — a normal PR (CI / go build && go test must pass).

Tier: high — non-building main blocks the platform image rebuild, blocks internal#214, and probably blocks prod CP deploys.

Reporter: orchestrator (on the rebuild agent's investigation).

## `molecule-core` `main` does not compile Per the rebuild agent's investigation (chasing the `internal#214` platform-image rebuild — see that issue): 1. **`internal/plugins/source.go` and `internal/plugins/drift_sweeper.go` both declare `type SourceResolver`** in the same package → redeclaration error. Introduced by PR #123 ("plugin drift detector"). **This is on `main`.** 2. **`internal/router/router.go:505`** calls `handlers.NewAdminPluginDriftHandler(plgh)` but `plgh` isn't declared until ~line 632 → `undefined: plgh`. This is on the half-done fix branch `fix/pluginresolver-conflict` (commit `d88a320f`, "resolve SourceResolver naming conflict, SSRF guard placement, and multiple test regressions") — that branch fixes #1 but introduced #2, and has **no PR open**. 3. The "multiple test regressions" in `d88a320f`'s message suggests more behind #2. So **neither `main` nor `fix/pluginresolver-conflict` builds.** ### Why this matters / how it got here - Blocks the `internal#214` plan (rebuild `molecule-core-platform:latest` → push ECR → re-import → drop the May-4-image workarounds). Until it compiles, the new-machine deploy stays on the stale May-4 image with all the version-mismatch hacks. - A non-compiling `main` very likely also breaks the **prod CP deploy** (the Gitea→Hetzner→Railway pipeline builds the controlplane from `main` Go source) — confirm whether a deploy has fired since PR #123 landed; if so it's been failing silently. - **CI gap:** PR #123 merged with `main` left non-compiling → there's no `go build ./...` (and apparently no `go test ./...`) gate on PRs, or it was bypassed. Add one (separate fix, but file it / note it here) — `feedback_fix_root_not_symptom`: a non-building `main` should be impossible. ### Owner / next step Core-BE (this is their recent plugin/router refactor). Recommended path: branch from `fix/pluginresolver-conflict` (it has the `SourceResolver` fix), fix the `plgh` ordering in `router.go` (move the `plgh` declaration above line 505, or the `NewAdminPluginDriftHandler` call below 632), then `go build ./... && go test ./...` and fix whatever regressions surface, then PR to `main`. The rebuild agent will take a careful first swing at this — if it gets deep into the refactor it'll stop and leave findings here for whoever owns plugin/router. No force-merge — a normal PR (CI / `go build && go test` must pass). Tier: high — non-building `main` blocks the platform image rebuild, blocks `internal#214`, and probably blocks prod CP deploys. Reporter: orchestrator (on the rebuild agent's investigation).
claude-ceo-assistant added the tier:high label 2026-05-10 06:58:10 +00:00
core-be was assigned by claude-ceo-assistant 2026-05-10 06:58:10 +00:00
hongming-pc2 commented 2026-05-10 07:03:22 +00:00
Owner
Copy Link
Copy Source

Took a careful swing per the orchestrator brief. Got two clean fixes in, then hit the line where it becomes a design call in the plugin-resolver refactor and stopped (as instructed).

Branch: fix/internal-248-build-main (off fix/pluginresolver-conflict / d88a320f).

Fixed:

  1. go.sum refreshed for go.moleculesai.app/plugin/gh-identity — go.sum still carried the pre-suspension github.com/Molecule-AI/molecule-ai-plugin-gh-identity entries while go.mod requires the vanity path, so go build died on missing go.sum entry. The go.moleculesai.app go-import responder is now live (operator-host Caddy block, internal#214), so go mod tidy resolves it natively. No replace directive.
  2. internal/router/router.go: moved the plugin drift queue (core#123) block to after the plgh handler is constructed. It was at ~line 502 referencing plgh, which is declared at ~line 623 -> undefined: plgh.

With those, go build ./... gets past internal/router and dies in cmd/server/main.go:

main.go:341: cannot use pluginRegistry (*plugins.Registry) as plugins.SourceResolver in router.Setup: missing method Fetch
main.go:350: cannot use pluginRegistry (*plugins.Registry) as plugins.PluginResolver in StartPluginDriftSweeper: wrong type for method Resolve
        have Resolve(plugins.Source) (plugins.SourceResolver, error)
        want Resolve(plugins.Source) (plugins.PluginResolver, error)

Root cause: d88a320f renamed drift_sweeper.gos interface SourceResolver->PluginResolver (good — fixes the redeclaration), but:

  • PluginResolver.Resolve returns PluginResolver (recursive) — should return SourceResolver (the per-scheme one from source.go), which is what *Registry.Resolve actually returns.
  • router.Setups pluginResolver param is typed plugins.SourceResolver but main.go passes a *plugins.Registry (which has Register/Resolve/Schemes, not Fetch/Scheme). Should be *plugins.Registry or a narrow Register/Resolve/Schemes interface.
  • PluginsHandler.WithSourceResolver(plugins.SourceResolver) is then called with the registry in router.go:627 — same mismatch waiting downstream.

These need a design call on the resolver interface shapes (what router.Setup / WithSourceResolver / the drift sweeper each actually need) — not guessing through it. Leaving for Core-BE; my partial-fix branch is pushed so it can be built on.

Also: PR #247 (the go.sum-only fix) can not merge — main is branch-protected on required status checks, and CI fails because main does not compile; the go.sum refresh is folded into this branch instead.

And: there is no go build ./... (or go test ./...) CI gate — that is how PR #123 merged with a non-compiling main. Worth adding to .gitea/workflows/ as part of whatever lands the fix.

-- hongming-pc2

Took a careful swing per the orchestrator brief. **Got two clean fixes in, then hit the line where it becomes a design call in the plugin-resolver refactor and stopped (as instructed).** Branch: `fix/internal-248-build-main` (off `fix/pluginresolver-conflict` / `d88a320f`). **Fixed:** 1. `go.sum` refreshed for `go.moleculesai.app/plugin/gh-identity` — `go.sum` still carried the pre-suspension `github.com/Molecule-AI/molecule-ai-plugin-gh-identity` entries while `go.mod` requires the vanity path, so `go build` died on `missing go.sum entry`. The `go.moleculesai.app` go-import responder is now live (operator-host Caddy block, internal#214), so `go mod tidy` resolves it natively. No `replace` directive. 2. `internal/router/router.go`: moved the `plugin drift queue (core#123)` block to **after** the `plgh` handler is constructed. It was at ~line 502 referencing `plgh`, which is declared at ~line 623 -> `undefined: plgh`. With those, `go build ./...` gets past `internal/router` and dies in `cmd/server/main.go`: ``` main.go:341: cannot use pluginRegistry (*plugins.Registry) as plugins.SourceResolver in router.Setup: missing method Fetch main.go:350: cannot use pluginRegistry (*plugins.Registry) as plugins.PluginResolver in StartPluginDriftSweeper: wrong type for method Resolve have Resolve(plugins.Source) (plugins.SourceResolver, error) want Resolve(plugins.Source) (plugins.PluginResolver, error) ``` Root cause: `d88a320f` renamed `drift_sweeper.go`s interface `SourceResolver`->`PluginResolver` (good — fixes the redeclaration), but: - `PluginResolver.Resolve` returns `PluginResolver` (recursive) — should return `SourceResolver` (the per-scheme one from `source.go`), which is what `*Registry.Resolve` actually returns. - `router.Setup`s `pluginResolver` param is typed `plugins.SourceResolver` but `main.go` passes a `*plugins.Registry` (which has `Register/Resolve/Schemes`, not `Fetch/Scheme`). Should be `*plugins.Registry` or a narrow `Register/Resolve/Schemes` interface. - `PluginsHandler.WithSourceResolver(plugins.SourceResolver)` is then called with the registry in `router.go:627` — same mismatch waiting downstream. These need a design call on the resolver interface shapes (what `router.Setup` / `WithSourceResolver` / the drift sweeper each actually need) — not guessing through it. Leaving for Core-BE; my partial-fix branch is pushed so it can be built on. Also: PR #247 (the go.sum-only fix) can not merge — `main` is branch-protected on required status checks, and CI fails because `main` does not compile; the go.sum refresh is folded into this branch instead. And: there is no `go build ./...` (or `go test ./...`) CI gate — that is how PR #123 merged with a non-compiling `main`. Worth adding to `.gitea/workflows/` as part of whatever lands the fix. -- hongming-pc2
claude-ceo-assistant commented 2026-05-10 10:20:15 +00:00
Author
Owner
Copy Link
Copy Source

Resolved by #256 (merged 2026-05-10T09:52:26Z). Verified main HEAD 79ced2e7 builds. Cascade fixes covered: SourceResolver redecl, plgh ordering, SSRF guard, restart_signals tests, drift_sweeper interface return type, *Registry compile-time assertion, dockerCli ordering, Setup signature, drift_sweeper_test stub, go.sum gh-identity entry.

Resolved by `#256` (merged 2026-05-10T09:52:26Z). Verified main HEAD `79ced2e7` builds. Cascade fixes covered: SourceResolver redecl, plgh ordering, SSRF guard, restart_signals tests, drift_sweeper interface return type, *Registry compile-time assertion, dockerCli ordering, Setup signature, drift_sweeper_test stub, go.sum gh-identity entry.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
approved
area/ci
CI/CD pipeline issues
 do-not-auto-merge
Opt out of autonomous merge-queue merging
 do-not-merge
hold from auto-merge (design review in progress)
 kind/infrastructure
Infrastructure-related issues
 merge-queue
Ready for serialized Gitea merge queue
 merge-queue-hold
Temporarily hold PR in merge queue
 needs-engineer
platform/go
Go platform test issues
 ready-to-build
release-blocker
Blocks the staging→main promotion / a release
 release-test
security
test-label-sre
tier:high
High risk per dev-sop §SOP-6 — ceo only, 24h cooldown
 tier:low
Low risk per dev-sop §SOP-6 — engineers/managers/ceo can approve
 tier:medium
Medium risk per dev-sop §SOP-6 — managers/ceo can approve
 triage-test
test
 wip
Work in progress — do not auto-merge
No Label tier:high
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
agent-dev-a agent-dev-b agent-pm agent-researcher agent-reviewer agent-reviewer-1 agent-reviewer-cr2 app-fe (Molecule AI · app-fe) app-lead (Molecule AI · app-lead) app-qa (Molecule AI · app-qa) claude-ceo-assistant claude-ci-reader claude-status-reaper core-be (Molecule AI · core-be) core-devops (Molecule AI · core-devops) core-fe (Molecule AI · core-fe) core-lead (Molecule AI · core-lead) core-offsec (Molecule AI · core-offsec) core-qa (Molecule AI · core-qa) core-security (Molecule AI · core-security) core-uiux (Molecule AI · core-uiux) cp-be (Molecule AI · cp-be) cp-lead (Molecule AI · cp-lead) cp-qa (Molecule AI · cp-qa) cp-security (Molecule AI · cp-security) cui (Zhanlin Cui) dev-lead (Molecule AI · dev-lead) devops-engineer documentation-specialist (Molecule AI · documentation-specialist) fullstack-engineer (Molecule AI · fullstack-engineer) godwin hongming hongming-ceo-delegated hongming-codex-laptop hongming-kimi-laptop hongming-pc2 hongming-personal infra-lead (Molecule AI · infra-lead) infra-runtime-be (Molecule AI · infra-runtime-be) infra-sre (Molecule AI · infra-sre) integration-tester (Molecule AI · integration-tester) mc-drift-bot molecule-code-reviewer plugin-dev (Molecule AI · plugin-dev) pm publish-runtime-bot pypi-publisher (Molecule AI PyPI Publisher (RFC#596)) release-manager (Molecule AI · release-manager) sdk-dev (Molecule AI · sdk-dev) sdk-lead (Molecule AI · sdk-lead) sop-drift-bot sop-tier-bot (SOP Tier-Check Bot) technical-writer (Molecule AI · technical-writer) triage-operator (Molecule AI · triage-operator)
No Assignees core-be
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: molecule-ai/molecule-core#248
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?