mesh-controller/cmd/mesh-controller/checks.go

package main

import (
	"context"
	"encoding/json"
	"fmt"
	"path"
	"slices"
	"sort"
	"strings"
	"time"

	"github.com/novox/mesh-controller/internal/catalogue"
	"github.com/novox/mesh-controller/internal/inventory"
	"github.com/novox/mesh-controller/internal/link"
)

// A pull request's merge check (novox/hq to-be 45 §9, ADR 0237 as amended 2026-10-06): the forge
// announces every pull request's new head, the controller decides what is checked, asks the build seat to
// check it, and says the verdict as `checked`, which the forge's holder sets as the pull request's
// statuses. **Before merge, never after**: every check the mesh had ran after a merge, on a machine.
//
// **The mesh's module graph decides, not the repository.** The controller holds the graph — every module,
// the repository and directory it is built from — and maps the pull request's changed paths onto it by
// the planner's own answer (reachOfMerge, touchedBy — the one place a changed file is mapped onto modules,
// for the merge handler, the release planner, the merge gate and this check; issue 280): **a changed file
// touches exactly the modules whose build reads it** — a module's own directory (the whole repository for
// one built from its root), or a repository its recipe packages. A file no build reads — a script at the
// root, a README — touches no module. A directory the change adds a module.json in, which the graph does
// not hold yet (said by the head, issue 278), is a new module and is checked too.
//
//   - touches a module: the build seat runs **the gate** — `mesh/merge-gate`, the touched manifests, every
//     machine composed with the change, the replays — and the repository's own merge-check.sh beside it;
//   - touches none, in a repository that is the mesh's (it sources a module on some branch, or shares the
//     core's owner): the gate is a pass that says so — a fact, not a missing check — and the repository's
//     own merge-check.sh runs as `mesh/repo-check`, a warning when it has none;
//   - touches none, anywhere else: the gate is a pass that says so, and nothing more is said.
//
// What is checked is decided here and run there (internal/builder/check.go).

// checkTimeout is how long one check may run on the build seat. Said here so the ask's watchdog (S6)
// and the builder agree on what late means.
const checkTimeout = 45 * time.Minute

// noModuleTouched is the gate's word for a change that touches nothing of the graph.
const noModuleTouched = "the change touches no module of the mesh's graph"

// noMergeCheck is the repository layer's word for a repository of the mesh with no merge-check.sh.
const noMergeCheck = "the repository declares no merge-check.sh: none of its own tests run before it merges"

// coreModules are the modules whose repositories are the mesh's core, by the directory a check finds
// each beside it — and whose owner is the mesh's own.
var coreModules = map[string]string{"mesh-controller": "mesh-controller", "mesh-host": "mesh-host",
	"node-tools": "mesh-tools", "nats": "mesh-catalog"}

// checkScope is what a pull request reaches of the mesh's graph, as the planner reckons it.
type checkScope struct {
	// Modules are the modules a merge of the change would move itself — built from the repository into
	// the pull request's base and reading a changed file, or packaging the repository's source — and
	// Dependents those the plan would build after them; New the directories it adds a module in.
	Modules    []string
	Dependents []string
	New        []string
	// Manifests are the moved modules' and the new ones' manifests in the change's tree.
	Manifests []string
	// Width is how many modules a merge would build, in how many tiers; Unread the changed files no
	// module's build reads.
	Width, Tiers int
	Unread       []string
	// Mesh says the repository is the mesh's: modules are built from it on some branch, its owner is the
	// core's, or the change adds a module to it.
	Mesh bool
	// Judge is who judges the gate (link.JudgeSelf, link.JudgeValidator, or the running controller).
	Judge string
	// From is a module built from the repository, for how the mesh clones it; nil when none is.
	From *inventory.Entry
	// Reach is the planner's whole answer, which the change plan is made from.
	Reach mergeReach
}

func (s checkScope) gated() bool { return len(s.Modules)+len(s.New) > 0 }

// pullScope is what a pull request reaches: **the planner's own answer** (reachOfMerge), asked as if the
// head were merged into the base — never a mapping of its own, so a change to what a merge touches
// changes what is checked with it (novox/hq ADR 0238).
func pullScope(p link.PullUpdated, entries []inventory.Entry, read map[string][]inventory.ReadRepository,
	edges []inventory.Edge) checkScope {
	m := link.SourceMoved{Owner: p.Owner, Repo: p.Repo, Base: p.Base, CloneURL: p.CloneURL, Commit: p.Commit,
		Paths: p.Paths, PathsTruncated: p.PathsTruncated, Removed: p.Removed, ModuleDirs: p.ModuleDirs,
		ModuleDirsSaid: p.ModuleDirsSaid}
	r := reachOfMerge(m, entries, read, edges)
	s := checkScope{Modules: r.Moved(), Dependents: r.Dependents(), New: r.Added, Unread: r.Unread,
		Width: len(r.Plan.Modules), Tiers: len(r.Plan.Tiers), Reach: r}
	for _, e := range append(append([]inventory.Entry{}, r.Touched...), r.Deleted...) {
		s.Manifests = append(s.Manifests, path.Join(strings.Trim(e.Source.Path, "/"), moduleManifestFile))
		switch e.Manifest.Module {
		case "mesh-controller":
			s.Judge = link.JudgeSelf
		case "mesh-host":
			if s.Judge == "" {
				s.Judge = link.JudgeValidator
			}
		}
	}
	for _, d := range r.Added {
		s.Manifests = append(s.Manifests, path.Join(d, moduleManifestFile))
	}
	sort.Strings(s.Manifests)
	s.Manifests = slices.Compact(s.Manifests)

	// Whose repository it is, for how it is cloned and whether its own check is the mesh's to run.
	owners := map[string]bool{}
	for i, e := range entries {
		if e.Provided {
			continue
		}
		if _, core := coreModules[e.Manifest.Module]; core {
			if owner := sourceOwner(e.Source.Repository); owner != "" {
				owners[owner] = true
			}
		}
		if sameRepository(e.Source.Repository, m) && s.From == nil {
			s.From = &entries[i]
		}
	}
	s.Mesh = s.From != nil || owners[strings.ToLower(p.Owner)] || s.gated()
	return s
}

// sourceOwner is the owner of a recorded repository, a path on the git seat or a URL: novox/mesh-host → novox.
func sourceOwner(repository string) string {
	parts := strings.Split(strings.Trim(strings.TrimSuffix(repository, ".git"), "/"), "/")
	if len(parts) < 2 {
		return ""
	}
	return strings.ToLower(parts[len(parts)-2])
}

// PullUpdated decides a pull request's merge check, and asks for it when there is something to run.
func (f following) PullUpdated(ctx context.Context, p link.PullUpdated) error {
	inv := f.open.inventory
	entries, err := inv.Catalogued(ctx)
	if err != nil {
		return err
	}
	read, err := inv.ReadRepositories(ctx)
	if err != nil {
		return err
	}
	edges, err := inv.Dependencies(ctx)
	if err != nil {
		return err
	}
	scope := pullScope(p, entries, read, edges)
	// The change plan of the commit at hand (ADR 0238): what a merge of it would build and send, posted
	// with the verdict whatever the verdict is.
	plan := changePlanOf(p.Owner+"/"+p.Repo, p.Base, p.Commit, scope.Reach, entries, func(module string) (inventory.Upgrade, bool) {
		u, err := inv.UpgradeOf(ctx, module)
		return u, err == nil
	})
	fmt.Print(planText(plan))
	direct := link.Checked{Owner: p.Owner, Repo: p.Repo, Number: p.Number, Commit: p.Commit,
		ID: link.NewBuildID(time.Now()), Verdict: "pass", Summary: noModuleTouched,
		Gate: &link.CheckLayer{Verdict: "pass", Summary: noModuleTouched}, Plan: &plan}
	switch {
	case !scope.gated() && !scope.Mesh:
		fmt.Printf("%s/%s#%d (%.8s): %s, and the repository is not the mesh's: said, nothing run\n",
			p.Owner, p.Repo, p.Number, p.Commit, noModuleTouched)
		sayChecked(ctx, direct)
		return nil
	case !scope.gated() && p.MergeCheckSaid && !p.MergeCheck:
		direct.RepoCheck = &link.CheckLayer{Verdict: "warning", Summary: noMergeCheck}
		fmt.Printf("%s/%s#%d (%.8s): %s; %s\n", p.Owner, p.Repo, p.Number, p.Commit, noModuleTouched, noMergeCheck)
		sayChecked(ctx, direct)
		return nil
	}
	request, err := checkRequestFor(ctx, f.open, p, scope, entries)
	if err != nil {
		return err
	}
	request.Check.Plan = &plan
	seat := buildSeatHeld(ctx)
	ask, err := askOverOn(seat)
	if err != nil {
		return err
	}
	defer ask.Close()
	if err := ask.Ask(ctx, request); err != nil {
		return err
	}
	what := "its own merge-check.sh alone: " + noModuleTouched
	if scope.gated() {
		what = fmt.Sprintf("the gate over %s (a merge would build %d module(s) in %d tier(s))",
			strings.Join(append(append([]string{}, scope.Modules...), prefixedAll("new:", scope.New)...), ", "),
			scope.Width, scope.Tiers)
	}
	fmt.Printf("%s/%s#%d (%.8s): asked %s to check it before it merges — %s — as %s\n", p.Owner, p.Repo, p.Number,
		p.Commit, seat, what, request.ID)
	return nil
}

// checkRequestFor is the ask for one pull request's head: the repository as the mesh clones it, the head,
// and what is read beside it.
func checkRequestFor(ctx context.Context, open *stores, p link.PullUpdated, scope checkScope,
	entries []inventory.Entry) (link.BuildRequest, error) {
	shelf := map[string]catalogue.Manifest{}
	for _, e := range entries {
		shelf[e.Manifest.Module] = e.Manifest
	}
	world, err := theRestOfTheMesh(ctx, open.inventory, shelf, "")
	if err != nil {
		return link.BuildRequest{}, err
	}
	clone := func(s inventory.Source) (string, error) {
		if s.Seat == "" {
			return s.Repository, nil
		}
		return clonedFromSeat(world, s.Seat, s.Repository)
	}
	// As the mesh clones a module built from it; a repository no module is built from, from the forge.
	source := inventory.Source{Seat: gitSeat, Repository: p.Owner + "/" + p.Repo}
	if scope.From != nil {
		source = scope.From.Source
	}
	repository, err := clone(source)
	if err != nil {
		return link.BuildRequest{}, err
	}
	current, err := open.inventory.CurrentBuilds(ctx)
	if err != nil {
		return link.BuildRequest{}, err
	}
	// Beside it, at what the mesh runs: each core repository by the module the mesh builds from it.
	beside := map[string]link.CheckedOut{}
	for _, e := range entries {
		dir, core := coreModules[e.Manifest.Module]
		if !core || e.Provided || e.Source.Repository == "" {
			continue
		}
		url, err := clone(e.Source)
		if err != nil {
			return link.BuildRequest{}, err
		}
		refs := besideRefs(dir, current[e.Manifest.Module].Commit)
		beside[dir] = link.CheckedOut{Repository: url, Ref: refs[dir]}
		if dir == "mesh-controller" {
			beside["mesh-controller-main"] = link.CheckedOut{Repository: url, Ref: refs["mesh-controller-main"]}
			if e.Source.Seat != "" {
				if lab, err := clone(inventory.Source{Seat: e.Source.Seat, Repository: siblingOf(e.Source.Repository,
					"mesh-lab")}); err == nil {
					beside["mesh-lab"] = link.CheckedOut{Repository: lab, Ref: refs["mesh-lab"]}
				}
			}
		}
	}
	return link.BuildRequest{
		ID:         link.NewBuildID(time.Now()),
		Repository: repository,
		Ref:        p.Commit,
		Held:       heldBy(ctx),
		Seats:      seatBases(ctx),
		Source:     sourceOnSeat(source),
		Check: &link.CheckRequest{Owner: p.Owner, Repo: p.Repo, Number: p.Number, Base: p.Base,
			Paths: p.Paths, Beside: beside, Modules: scope.Modules, Dependents: scope.Dependents, New: scope.New,
			Manifests: scope.Manifests, Judge: scope.Judge},
	}, nil
}

// besideRefs is the ref each repository is cloned at beside a check, by the directory it is found under:
// a core repository at the commit the mesh runs of the module built from it (`running`) — but the
// catalogue, whose checkout beside is what tests read its files from, at its main, what the next merge
// builds from; and beside the controller its main, for a judge the running controller predates, and the
// lab's main, whose replays every check runs. **One rule, read by the check the controller asks for and by
// the facts snapshot** (Facts.Beside), so a check run by hand clones what the build seat clones.
func besideRefs(dir, running string) map[string]string {
	switch dir {
	case "mesh-catalog":
		return map[string]string{dir: "main"}
	case "mesh-controller":
		return map[string]string{dir: running, "mesh-controller-main": "main", "mesh-lab": "main"}
	}
	return map[string]string{dir: running}
}

// siblingOf is another repository of the same owner: novox/mesh-controller → novox/mesh-lab.
func siblingOf(repository, name string) string {
	if cut := strings.LastIndex(repository, "/"); cut >= 0 {
		return repository[:cut+1] + name
	}
	return name
}

// sourceOnSeat is a source's seat form, nil for one on no seat.
func sourceOnSeat(s inventory.Source) *link.SourceOnSeat {
	if s.Seat == "" {
		return nil
	}
	return &link.SourceOnSeat{Seat: s.Seat, Repository: s.Repository}
}

// checkEvents is where the serving controller says a check's verdict; nil in a command.
var checkEvents link.Bus

// maxCheckReport is how much of a check's report travels in its verdict: enough for the failures and
// the machines, never a log.
const maxCheckReport = 60 << 10

// checked says a merge check's verdict as the controller's `checked`. Nothing is recorded or
// registered: a check builds nothing (issue 240's rule for a dry run, kept for a check).
func checked(ctx context.Context, result link.BuildResult) {
	sayChecked(ctx, checkedOf(result))
}

// checkedOf is what a check's outcome says: each layer, and an error — never a pass — for a check that
// could not run.
func checkedOf(result link.BuildResult) link.Checked {
	c := link.Checked{ID: result.ID, On: result.On, Commit: result.Ref}
	if result.Checked != nil {
		c.Owner, c.Repo, c.Number = result.Checked.Owner, result.Checked.Repo, result.Checked.Number
	}
	switch {
	case result.Check != nil:
		c.Verdict, c.Summary, c.Report = result.Check.Verdict, result.Check.Summary, result.Check.Report
		c.Gate, c.RepoCheck = result.Check.Gate, result.Check.RepoCheck
		if c.Gate == nil {
			// A build seat from before the layers: its verdict is the gate's.
			c.Gate = &link.CheckLayer{Verdict: c.Verdict, Summary: c.Summary}
		}
	case result.Failed != "":
		// The check could not run: an error, never read as a pass — on both layers it was asked for.
		c.Verdict, c.Summary = "error", "the check could not run: "+firstLine(result.Failed)
	default:
		c.Verdict, c.Summary = "error", "the build seat answered the check with no verdict"
	}
	if c.Verdict == "" {
		c.Verdict = "error"
	}
	if result.Check == nil {
		c.Gate = &link.CheckLayer{Verdict: c.Verdict, Summary: c.Summary}
		c.RepoCheck = &link.CheckLayer{Verdict: c.Verdict, Summary: c.Summary}
	}
	if result.Checked != nil && c.Gate != nil && len(c.Gate.Modules) == 0 {
		c.Gate.Modules = append(append([]string{}, result.Checked.Modules...), prefixedAll("new:", result.Checked.New)...)
	}
	if result.Checked != nil && c.Gate != nil && len(c.Gate.Dependents) == 0 {
		c.Gate.Dependents = result.Checked.Dependents
	}
	if result.Checked != nil {
		c.Plan = result.Checked.Plan
		// A delivery group's composed check (novox/hq ADR 0239): every head it judged, this one first.
		if g := result.Checked; g.Group != "" {
			c.Group = g.Group
			c.Members = append(c.Members, link.CheckedMember{Owner: g.Owner, Repo: g.Repo, Number: g.Number,
				Commit: result.Ref})
			for _, m := range g.Members {
				c.Members = append(c.Members, link.CheckedMember{Owner: m.Owner, Repo: m.Repo, Number: m.Number,
					Commit: m.Ref})
			}
		}
	}
	for _, l := range []*link.CheckLayer{c.Gate, c.RepoCheck} {
		if l != nil && l.Verdict == "" {
			l.Verdict = "error"
		}
	}
	if len(c.Report) > maxCheckReport {
		c.Report = "…" + c.Report[len(c.Report)-maxCheckReport:]
	}
	return c
}

func prefixedAll(prefix string, items []string) []string {
	out := make([]string, 0, len(items))
	for _, i := range items {
		out = append(out, prefix+i)
	}
	return out
}

// sayChecked says a merge check's verdict on the bus, where the forge's holder hears it.
func sayChecked(ctx context.Context, c link.Checked) {
	repo := "none"
	if c.RepoCheck != nil {
		repo = strings.ToUpper(c.RepoCheck.Verdict) + " — " + c.RepoCheck.Summary
	}
	fmt.Printf("%s: %s/%s#%d at %.8s checked on %s: gate %s — %s; repository %s\n", c.ID, c.Owner, c.Repo, c.Number,
		c.Commit, orSomewhere(c.On), strings.ToUpper(c.Verdict), c.Summary, repo)
	if checkEvents == nil {
		return
	}
	body, err := json.Marshal(c)
	if err != nil {
		return
	}
	stating, stop := context.WithTimeout(context.WithoutCancel(ctx), 10*time.Second)
	defer stop()
	if err := checkEvents.PublishSeatEvent(stating, link.MeshControllerSeat, link.KeyChecked, body); err != nil {
		fmt.Printf("%s: the verdict could not be said, so the pull request is not told it: %v\n", c.ID, err)
	}
}