route rpm virtual package paths to their member

e2e-docker/virtual_test.go

@@ -3,9 +3,17 @@
 package e2edocker
 
 import (
+	"bytes"
+	"compress/gzip"
+	"encoding/xml"
+	"fmt"
+	"io"
 	"net/http"
+	"os"
+	"os/exec"
 	"strings"
 	"testing"
+	"time"
 )
 
 // TestVirtualPyPIMerge uploads different packages to two pypi locals and
@@ -52,3 +60,79 @@ func TestVirtualHelmMerge(t *testing.T) {
 		t.Fatalf("merged helm index missing a member chart (want alpha and beta): %s", s)
 	}
 }
+
+// TestVirtualRPMMerge merges a remote rpm repo and a local rpm repo carrying the
+// same package: the first member wins the duplicate, its package download
+// routes through the virtual, and a real dnf can consume the merged repo.
+func TestVirtualRPMMerge(t *testing.T) {
+	createRepo(t, `{"name":"vrpm-remote","package_type":"rpm","repo_type":"remote","base_url":"`+mockUpstream()+`/rpm-mirror","stale_on_error":true}`)
+	createRepo(t, `{"name":"vrpm-local","package_type":"rpm","repo_type":"local"}`)
+	defer deleteRepo(t, "vrpm-remote")
+	defer deleteRepo(t, "vrpm-local")
+
+	pkg := fixtureBytes(t, "rpmrepo/Packages/e2e-testpkg-1.0-1.noarch.rpm")
+	uploadFile(t, "vrpm-local", "e2e-testpkg-1.0-1.noarch.rpm", pkg, "application/x-rpm")
+
+	createVirtual(t, `{"name":"vrpm","package_type":"rpm","members":["vrpm-remote","vrpm-local"]}`)
+	defer deleteVirtual(t, "vrpm")
+
+	resp, body := getEventually(t, api("/api/v1/virtual/vrpm/repodata/repomd.xml"), 15*time.Second)
+	if resp.StatusCode != http.StatusOK {
+		t.Fatalf("virtual repomd.xml: status %d: %s", resp.StatusCode, body)
+	}
+	var md struct {
+		Data []struct {
+			Type     string `xml:"type,attr"`
+			Location struct {
+				Href string `xml:"href,attr"`
+			} `xml:"location"`
+		} `xml:"data"`
+	}
+	if err := xml.Unmarshal(body, &md); err != nil {
+		t.Fatalf("parse repomd.xml: %v\n%s", err, body)
+	}
+	var primary []byte
+	for _, d := range md.Data {
+		if d.Type == "primary" {
+			resp, gz := doRequest(t, http.MethodGet, api("/api/v1/virtual/vrpm/"+d.Location.Href), nil, "")
+			if resp.StatusCode != http.StatusOK {
+				t.Fatalf("primary: status %d", resp.StatusCode)
+			}
+			r, err := gzip.NewReader(bytes.NewReader(gz))
+			if err != nil {
+				t.Fatal(err)
+			}
+			primary, _ = io.ReadAll(r)
+		}
+	}
+	href := "vrpm-remote/Packages/e2e-testpkg-1.0-1.noarch.rpm"
+	if strings.Count(string(primary), "<name>e2e-testpkg</name>") != 1 || !strings.Contains(string(primary), `href="`+href+`"`) {
+		t.Fatalf("merged primary should hold one e2e-testpkg owned by the first member:\n%s", primary)
+	}
+
+	resp, got := doRequest(t, http.MethodGet, api("/api/v1/virtual/vrpm/"+href), nil, "")
+	if resp.StatusCode != http.StatusOK || !bytes.Equal(got, pkg) {
+		t.Fatalf("package via virtual: status %d, %d bytes (want %d)", resp.StatusCode, len(got), len(pkg))
+	}
+
+	network := os.Getenv("COMPOSE_NETWORK")
+	internal := os.Getenv("ARTIFACTAPI_INTERNAL")
+	if network == "" || internal == "" {
+		t.Skip("COMPOSE_NETWORK/ARTIFACTAPI_INTERNAL not set; skipping real dnf")
+	}
+	if _, err := exec.LookPath("docker"); err != nil {
+		t.Skip("docker not available on the test host")
+	}
+	repoConf := fmt.Sprintf("[vrpm]\nname=vrpm\nbaseurl=%s/api/v1/virtual/vrpm/\nenabled=1\ngpgcheck=0\nmetadata_expire=0\n", strings.TrimRight(internal, "/"))
+	script := "set -euo pipefail; " +
+		"printf '%s' \"$REPO\" > /etc/yum.repos.d/vrpm.repo; " +
+		"dnf -y --disablerepo='*' --enablerepo=vrpm makecache; " +
+		"dnf -y --disablerepo='*' --enablerepo=vrpm list --available; " +
+		"dnf -y --disablerepo='*' --enablerepo=vrpm install e2e-testpkg; " +
+		"rpm -q e2e-testpkg"
+	out, err := exec.Command("docker", "run", "--rm", "--network", network, "-e", "REPO="+repoConf,
+		"rockylinux:9", "bash", "-c", script).CombinedOutput()
+	if err != nil {
+		t.Fatalf("real dnf against rpm virtual failed: %v\n%s", err, out)
+	}
+}

internal/api/v1/proxy.go

@@ -162,7 +162,16 @@ func (h *ProxyHandler) handleVirtual(w http.ResponseWriter, r *http.Request) {
 
 	proxyBaseURL := fmt.Sprintf("%s://%s", scheme(r), r.Host)
 
+	if loc, ok := h.virtualEngine.MemberRedirect(r.Context(), *virt, path, proxyBaseURL); ok {
+		http.Redirect(w, r, loc, http.StatusFound)
+		return
+	}
+
 	body, contentType, err := h.virtualEngine.Fetch(r.Context(), *virt, path, proxyBaseURL)
+	if errors.Is(err, virtual.ErrNotFound) {
+		http.Error(w, "not found", http.StatusNotFound)
+		return
+	}
 	if err != nil {
 		slog.Error("virtual fetch failed", "virtual", virtualName, "path", path, "error", err)
 		http.Error(w, "bad gateway", http.StatusBadGateway)

internal/virtual/engine.go

@@ -2,9 +2,14 @@ package virtual
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"io"
 	"log/slog"
+	"net/http"
+	"net/http/httptest"
+	"slices"
+	"strings"
 	"sync"
 
 	"git.unkin.net/unkin/artifactapi/internal/database"
@@ -23,6 +28,10 @@ func NewEngine(db *database.DB, proxyEngine *proxy.Engine) *Engine {
 }
 
 func (e *Engine) Fetch(ctx context.Context, virt models.Virtual, path string, proxyBaseURL string) ([]byte, string, error) {
+	if virt.PackageType == models.PackageRPM {
+		return e.fetchRPM(ctx, virt, path)
+	}
+
 	merger, err := GetMerger(virt.PackageType)
 	if err != nil {
 		return nil, "", fmt.Errorf("unsupported virtual type %q: %w", virt.PackageType, err)
@@ -133,3 +142,137 @@ func (e *Engine) fetchLocalIndex(ctx context.Context, remote models.Remote, path
 
 	return indexer.GenerateLocalIndex(ctx, e.db, remote.Name, path)
 }
+
+var ErrNotFound = errors.New("not found")
+
+// MemberRedirect maps an rpm virtual package path (<member>/<href>, as written
+// by MergeRPM) to the owning member's route. ok is false for any other path.
+func (e *Engine) MemberRedirect(ctx context.Context, virt models.Virtual, path, proxyBaseURL string) (string, bool) {
+	if virt.PackageType != models.PackageRPM || strings.HasPrefix(path, "repodata/") {
+		return "", false
+	}
+	name, rest, found := strings.Cut(path, "/")
+	if !found || !slices.Contains(virt.Members, name) {
+		return "", false
+	}
+	remote, err := e.db.GetRemote(ctx, name)
+	if err != nil {
+		return "", false
+	}
+	return fmt.Sprintf("%s/api/v1/%s/%s/%s", strings.TrimRight(proxyBaseURL, "/"), remote.RepoType, name, rest), true
+}
+
+func (e *Engine) fetchRPM(ctx context.Context, virt models.Virtual, path string) ([]byte, string, error) {
+	if !strings.HasPrefix(path, "repodata/") {
+		return nil, "", ErrNotFound
+	}
+
+	results := make([]*RPMMember, len(virt.Members))
+	var wg sync.WaitGroup
+	for i, name := range virt.Members {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			m, err := e.fetchRPMMember(ctx, name)
+			if err != nil {
+				slog.Warn("virtual member fetch failed", "virtual", virt.Name, "member", name, "error", err)
+				return
+			}
+			results[i] = m
+		}()
+	}
+	wg.Wait()
+
+	var members []RPMMember
+	for _, m := range results {
+		if m != nil {
+			members = append(members, *m)
+		}
+	}
+	if len(members) == 0 {
+		return nil, "", fmt.Errorf("no members reachable for virtual %q", virt.Name)
+	}
+
+	// ponytail: merges on every repodata request; cache keyed on member repomd checksums if member count/size makes this slow.
+	repo, err := MergeRPM(members)
+	if err != nil {
+		return nil, "", fmt.Errorf("merge rpm repodata: %w", err)
+	}
+	if path == "repodata/repomd.xml" {
+		return repo.Repomd, "application/xml", nil
+	}
+	if body, ok := repo.Files[path]; ok {
+		return body, "application/gzip", nil
+	}
+	return nil, "", ErrNotFound
+}
+
+func (e *Engine) fetchRPMMember(ctx context.Context, name string) (*RPMMember, error) {
+	remote, err := e.db.GetRemote(ctx, name)
+	if err != nil {
+		return nil, fmt.Errorf("remote %q: %w", name, err)
+	}
+	repomd, err := e.fetchMemberPath(ctx, *remote, "repodata/repomd.xml")
+	if err != nil {
+		return nil, err
+	}
+	locs, ts, err := parseRepomd(repomd)
+	if err != nil {
+		return nil, err
+	}
+	m := &RPMMember{RemoteName: name, Timestamp: ts, Data: map[string][]byte{}}
+	for t, href := range locs {
+		raw, err := e.fetchMemberPath(ctx, *remote, href)
+		if err != nil {
+			return nil, err
+		}
+		if m.Data[t], err = decompress(href, raw); err != nil {
+			return nil, fmt.Errorf("decompress %s: %w", href, err)
+		}
+	}
+	return m, nil
+}
+
+// fetchMemberPath reads one path from a member the same way its own route
+// would serve it: local index, synthesized remote (e.g. github_rpm), or proxy.
+func (e *Engine) fetchMemberPath(ctx context.Context, remote models.Remote, path string) ([]byte, error) {
+	prov, err := provider.Get(remote.PackageType)
+	if err != nil {
+		return nil, fmt.Errorf("provider %q: %w", remote.PackageType, err)
+	}
+
+	var serve func(w http.ResponseWriter, r *http.Request) bool
+	if remote.RepoType == models.RepoTypeLocal {
+		indexer, ok := prov.(provider.LocalIndexer)
+		if !ok {
+			return nil, fmt.Errorf("provider %q does not serve a local index", remote.PackageType)
+		}
+		serve = func(w http.ResponseWriter, r *http.Request) bool {
+			return indexer.ServeLocalIndex(w, r, e.db, remote.Name, path)
+		}
+	} else if rs, ok := prov.(provider.RemoteServer); ok {
+		serve = func(w http.ResponseWriter, r *http.Request) bool {
+			return rs.ServeRemote(w, r, remote, path, "", e.db)
+		}
+	}
+
+	if serve != nil {
+		rec := httptest.NewRecorder()
+		if serve(rec, httptest.NewRequestWithContext(ctx, http.MethodGet, "/"+path, nil)) {
+			if rec.Code != http.StatusOK {
+				return nil, fmt.Errorf("%s/%s: status %d", remote.Name, path, rec.Code)
+			}
+			return rec.Body.Bytes(), nil
+		}
+		if remote.RepoType == models.RepoTypeLocal {
+			return nil, fmt.Errorf("%s/%s: %w", remote.Name, path, ErrNotFound)
+		}
+	}
+
+	res, err := e.proxyEngine.Fetch(ctx, remote, path, prov)
+	if err != nil {
+		return nil, fmt.Errorf("fetch %s/%s: %w", remote.Name, path, err)
+	}
+	defer res.Reader.Close()
+	return io.ReadAll(res.Reader)
+}