Add local cargo registry type

- serve sparse index and config.json for local cargo repos from a cargo_index table
- add Web API publish, yank and unyank under the local repo api root
- drop index lines when the crate file is deleted

README.md

@@ -33,7 +33,6 @@ API: `http://localhost:8000` | Frontend: `http://localhost:5173`
 | `terraform` | `*/versions` | `*/download/*/*` |
 | `goproxy` | `@v/list`, `@latest` | `.info`, `.mod`, `.zip` |
 | `cargo` | sparse index files (`config.json` synthesized) | `crates/*/*.crate` |
-| `gems` | compact index (`versions`, `info/*`, `names`), `*specs.4.8.gz` | `gems/*.gem`, `quick/Marshal.4.8/*.gemspec.rz` |
 | `github_rpm` | `repodata/*` (synthesized) | `.rpm` (redirected) |
 
 Providers classify paths automatically. Users only configure what to proxy and TTLs.
@@ -280,6 +279,20 @@ content-addressable store (deduplicated by digest, reaped by GC once
 unreferenced); tags are mutable references and re-pushing a tag moves it. Blob
 uploads support both the monolithic and chunked (`POST`/`PATCH`/`PUT`) flows.
 
+### Local cargo registry
+
+A local `cargo` repo is a private crate registry: it serves the sparse index and
+the Web API publish, yank and unyank endpoints, so `cargo publish` works against
+it directly.
+
+```toml
+[registries.<repo>]
+index = "sparse+https://artifactapi.k8s.syd1.au.unkin.net/api/v1/local/<repo>/"
+```
+
+cargo requires a token for publish/yank (`CARGO_REGISTRIES_<REPO>_TOKEN`); like
+other local repos, artifactapi does not check it.
+
 ## Access Control
 
 | Field | Default | Behaviour |

e2e-docker/README.md

@@ -23,11 +23,12 @@ already-running stack.
 
 - **Repository lifecycle** — add / change / delete for remote, local and virtual repos.
 - **Caching** — one immutable artifact per remote package type (generic, docker,
-  helm, pypi, npm, rpm, alpine, puppet, terraform, goproxy, cargo, gems) proxied through the
+  helm, pypi, npm, rpm, alpine, puppet, terraform, goproxy, cargo) proxied through the
   mock upstream: first fetch `X-Artifact-Source: remote`, second `cache`, bytes
   verified against the origin fixture.
 - **Local uploads** — generic (upload/download), pypi (wheel + generated `simple/`
-  index), rpm (real package + **automatic repodata** generation).
+  index), rpm (real package + **automatic repodata** generation), cargo (publish,
+  sparse index, download, yank).
 - **Virtual repositories** — pypi simple-index merge and helm `index.yaml` merge
   across two members.
 - **Mirrorlist** — an rpm remote with a `mirrorlist` of extra upstream mirrors

e2e-docker/caching_test.go

@@ -30,7 +30,6 @@ func TestCachingPerProvider(t *testing.T) {
 		{"puppet", "puppet-releases/author-mod-1.0.0.tar.gz", "puppet-releases/author-mod-1.0.0.tar.gz"},
 		{"goproxy", "goproxy/example.com/mod/@v/v1.0.0.zip", "goproxy/example.com/mod/@v/v1.0.0.zip"},
 		{"cargo", "crates/mycrate/mycrate-1.0.0.crate", "crates/mycrate/mycrate-1.0.0.crate"},
-		{"gems", "gems/mygem-1.0.0.gem", "gems/mygem-1.0.0.gem"},
 		{"terraform", "hashicorp/aws/download/pkg.zip", "v1/providers/hashicorp/aws/download/pkg.zip"},
 		{"docker", "library/testimg/blobs/blobdata", "v2/library/testimg/blobs/blobdata"},
 	}

e2e-docker/cargo_local_test.go

@@ -0,0 +1,56 @@
+//go:build dockere2e
+
+package e2edocker
+
+import (
+	"bytes"
+	"encoding/binary"
+	"net/http"
+	"strings"
+	"testing"
+)
+
+// TestLocalCargoRegistry publishes through the cargo Web API, then reads the
+// sparse index, downloads the crate and yanks it.
+func TestLocalCargoRegistry(t *testing.T) {
+	createRepo(t, `{"name":"local-cargo","package_type":"cargo","repo_type":"local"}`)
+	defer deleteRepo(t, "local-cargo")
+	base := api("/api/v1/local/local-cargo")
+
+	meta := `{"name":"e2e-crate","vers":"0.1.0","deps":[],"features":{}}`
+	crate := []byte("e2e crate bytes")
+	var body []byte
+	body = binary.LittleEndian.AppendUint32(body, uint32(len(meta)))
+	body = append(body, meta...)
+	body = binary.LittleEndian.AppendUint32(body, uint32(len(crate)))
+	body = append(body, crate...)
+
+	if resp, b := doRequest(t, http.MethodPut, base+"/api/v1/crates/new", body, "application/octet-stream"); resp.StatusCode != http.StatusOK {
+		t.Fatalf("publish: status %d: %s", resp.StatusCode, b)
+	}
+	if resp, _ := doRequest(t, http.MethodPut, base+"/api/v1/crates/new", body, "application/octet-stream"); resp.StatusCode != http.StatusConflict {
+		t.Fatalf("republish: status %d, want 409", resp.StatusCode)
+	}
+
+	resp, b := doRequest(t, http.MethodGet, base+"/config.json", nil, "")
+	if resp.StatusCode != http.StatusOK || !strings.Contains(string(b), `"api":"`) {
+		t.Fatalf("config.json: status %d: %s", resp.StatusCode, b)
+	}
+
+	resp, b = doRequest(t, http.MethodGet, base+"/e2/e-/e2e-crate", nil, "")
+	if resp.StatusCode != http.StatusOK || !strings.Contains(string(b), `"cksum":"`) || !strings.Contains(string(b), `"yanked":false`) {
+		t.Fatalf("index: status %d: %s", resp.StatusCode, b)
+	}
+
+	resp, b = doRequest(t, http.MethodGet, base+"/crates/e2e-crate/e2e-crate-0.1.0.crate", nil, "")
+	if resp.StatusCode != http.StatusOK || !bytes.Equal(b, crate) {
+		t.Fatalf("download: status %d", resp.StatusCode)
+	}
+
+	if resp, b := doRequest(t, http.MethodDelete, base+"/api/v1/crates/e2e-crate/0.1.0/yank", nil, ""); resp.StatusCode != http.StatusOK {
+		t.Fatalf("yank: status %d: %s", resp.StatusCode, b)
+	}
+	if _, b := doRequest(t, http.MethodGet, base+"/e2/e-/e2e-crate", nil, ""); !strings.Contains(string(b), `"yanked":true`) {
+		t.Fatalf("index after yank: %s", b)
+	}
+}

e2e-docker/fixtures/gems/mygem-1.0.0.gem

@@ -1 +0,0 @@
-mygem-1.0.0 e2e fixture

internal/api/v1/cargo.go

@@ -0,0 +1,148 @@
+package v1
+
+import (
+	"bytes"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"log/slog"
+	"net/http"
+
+	"github.com/go-chi/chi/v5"
+
+	"git.unkin.net/unkin/artifactapi/internal/database"
+	"git.unkin.net/unkin/artifactapi/internal/provider"
+	"git.unkin.net/unkin/artifactapi/internal/provider/cargo"
+	"git.unkin.net/unkin/artifactapi/pkg/models"
+)
+
+// This file implements the cargo registry Web API (publish, yank, unyank) for
+// local cargo repositories. config.json points api at /api/v1/local/<repo>, so
+// cargo appends its own /api/v1/crates/... below that.
+
+const maxCratePublish = 64 << 20
+
+func (h *ProxyHandler) localCargoRemote(r *http.Request) (*models.Remote, bool) {
+	remote, err := h.db.GetRemote(r.Context(), chi.URLParam(r, "localName"))
+	if err != nil {
+		return nil, false
+	}
+	return remote, remote.RepoType == models.RepoTypeLocal && remote.PackageType == models.PackageCargo
+}
+
+func cargoError(w http.ResponseWriter, status int, format string, args ...any) {
+	w.Header().Set("Content-Type", "application/json")
+	w.WriteHeader(status)
+	_ = json.NewEncoder(w).Encode(map[string]any{
+		"errors": []map[string]string{{"detail": fmt.Sprintf(format, args...)}},
+	})
+}
+
+func cargoOK(w http.ResponseWriter, v any) {
+	w.Header().Set("Content-Type", "application/json")
+	w.WriteHeader(http.StatusOK)
+	_ = json.NewEncoder(w).Encode(v)
+}
+
+func (h *ProxyHandler) cargoPublish(w http.ResponseWriter, r *http.Request) {
+	remote, ok := h.localCargoRemote(r)
+	if !ok {
+		cargoError(w, http.StatusNotFound, "no local cargo registry %q", chi.URLParam(r, "localName"))
+		return
+	}
+	ctx := r.Context()
+
+	body, err := io.ReadAll(io.LimitReader(r.Body, maxCratePublish+1))
+	if err != nil {
+		cargoError(w, http.StatusBadRequest, "read body: %v", err)
+		return
+	}
+	if len(body) > maxCratePublish {
+		cargoError(w, http.StatusRequestEntityTooLarge, "publish body exceeds %d bytes", maxCratePublish)
+		return
+	}
+	meta, crate, err := cargo.ParsePublish(body)
+	if err != nil {
+		cargoError(w, http.StatusBadRequest, "%v", err)
+		return
+	}
+	line, err := cargo.NewIndexLine(meta, crate)
+	if err != nil {
+		cargoError(w, http.StatusBadRequest, "%v", err)
+		return
+	}
+
+	existing, err := h.db.ListCargoIndexEntries(ctx, remote.Name, meta.Name)
+	if err != nil {
+		cargoError(w, http.StatusInternalServerError, "%v", err)
+		return
+	}
+	for _, e := range existing {
+		if e.Name != meta.Name {
+			cargoError(w, http.StatusBadRequest, "crate %q already exists as %q", meta.Name, e.Name)
+			return
+		}
+		if e.Version == meta.Vers {
+			cargoError(w, http.StatusConflict, "crate version `%s@%s` already exists", meta.Name, meta.Vers)
+			return
+		}
+	}
+
+	lineJSON, err := json.Marshal(line)
+	if err != nil {
+		cargoError(w, http.StatusInternalServerError, "%v", err)
+		return
+	}
+
+	path := cargo.CratePath(meta.Name, meta.Vers)
+	result, err := h.cas.Store(ctx, bytes.NewReader(crate), "application/gzip")
+	if err != nil {
+		cargoError(w, http.StatusInternalServerError, "store failed: %v", err)
+		return
+	}
+	if err := h.db.UpsertBlob(ctx, result.ContentHash, result.S3Key, result.SizeBytes, "application/gzip"); err != nil {
+		cargoError(w, http.StatusInternalServerError, "record blob: %v", err)
+		return
+	}
+	if err := h.db.CreateLocalFile(ctx, remote.Name, path, result.ContentHash); err != nil {
+		if errors.Is(err, database.ErrAlreadyExists) {
+			cargoError(w, http.StatusConflict, "crate version `%s@%s` already exists", meta.Name, meta.Vers)
+			return
+		}
+		cargoError(w, http.StatusInternalServerError, "record file: %v", err)
+		return
+	}
+	entry := &provider.CargoIndexEntry{RepoName: remote.Name, FilePath: path, Name: meta.Name, Version: meta.Vers, Line: string(lineJSON)}
+	if err := h.db.InsertCargoIndexEntry(ctx, entry); err != nil {
+		// Drop the file so a retry is not rejected as a duplicate.
+		_ = h.db.DeleteLocalFile(ctx, remote.Name, path)
+		cargoError(w, http.StatusInternalServerError, "record index: %v", err)
+		return
+	}
+
+	slog.Info("cargo: published", "repo", remote.Name, "crate", meta.Name, "version", meta.Vers)
+	cargoOK(w, map[string]any{"warnings": map[string][]string{"invalid_categories": {}, "invalid_badges": {}, "other": {}}})
+}
+
+func (h *ProxyHandler) cargoYank(yanked bool) http.HandlerFunc {
+	return func(w http.ResponseWriter, r *http.Request) {
+		remote, ok := h.localCargoRemote(r)
+		if !ok {
+			cargoError(w, http.StatusNotFound, "no local cargo registry %q", chi.URLParam(r, "localName"))
+			return
+		}
+		name, version := chi.URLParam(r, "crate"), chi.URLParam(r, "version")
+		found, err := h.db.SetCargoYanked(r.Context(), remote.Name, name, version, yanked)
+		if err != nil {
+			cargoError(w, http.StatusInternalServerError, "%v", err)
+			return
+		}
+		if !found {
+			cargoError(w, http.StatusNotFound, "crate version `%s@%s` not found", name, version)
+			return
+		}
+		slog.Info("cargo: yank updated", "repo", remote.Name, "crate", name, "version", version, "yanked", yanked)
+		cargoOK(w, map[string]bool{"ok": true})
+	}
+}

internal/api/v1/proxy.go

@@ -40,8 +40,10 @@ func NewProxyHandler(engine *proxy.Engine, virtualEngine *virtual.Engine, db *da
 func (h *ProxyHandler) Routes() chi.Router {
 	r := chi.NewRouter()
 	r.Get("/remote/{remoteName}/*", h.handleProxy)
-	r.Head("/remote/{remoteName}/*", h.handleProxyHead)
 	r.Get("/local/{localName}/*", h.handleLocal)
+	r.Put("/local/{localName}/api/v1/crates/new", h.cargoPublish)
+	r.Delete("/local/{localName}/api/v1/crates/{crate}/{version}/yank", h.cargoYank(true))
+	r.Put("/local/{localName}/api/v1/crates/{crate}/{version}/unyank", h.cargoYank(false))
 	r.Get("/virtual/{virtualName}/*", h.handleVirtual)
 	return r
 }

internal/api/v1/routes_test.go

@@ -1,17 +0,0 @@
-package v1
-
-import (
-	"net/http"
-	"testing"
-
-	"github.com/go-chi/chi/v5"
-)
-
-func TestRoutesRemoteHead(t *testing.T) {
-	r := (&ProxyHandler{}).Routes()
-	for _, m := range []string{http.MethodGet, http.MethodHead} {
-		if !r.Match(chi.NewRouteContext(), m, "/remote/rubygems/versions") {
-			t.Errorf("%s /remote/rubygems/versions not routed", m)
-		}
-	}
-}

internal/database/cargo_index.go

@@ -0,0 +1,56 @@
+package database
+
+import (
+	"context"
+	"strings"
+
+	"git.unkin.net/unkin/artifactapi/internal/provider"
+)
+
+func (db *DB) InsertCargoIndexEntry(ctx context.Context, e *provider.CargoIndexEntry) error {
+	_, err := db.Pool.Exec(ctx, `
+		INSERT INTO cargo_index (repo_name, file_path, name, version, line, yanked)
+		VALUES ($1, $2, $3, $4, $5, $6)
+	`, e.RepoName, e.FilePath, e.Name, e.Version, e.Line, e.Yanked)
+	return err
+}
+
+func (db *DB) ListCargoIndexEntries(ctx context.Context, repoName, name string) ([]provider.CargoIndexEntry, error) {
+	rows, err := db.Pool.Query(ctx, `
+		SELECT repo_name, file_path, name, version, line, yanked
+		FROM cargo_index
+		WHERE repo_name = $1 AND lower(name) = $2
+		ORDER BY id
+	`, repoName, strings.ToLower(name))
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+
+	var result []provider.CargoIndexEntry
+	for rows.Next() {
+		var e provider.CargoIndexEntry
+		if err := rows.Scan(&e.RepoName, &e.FilePath, &e.Name, &e.Version, &e.Line, &e.Yanked); err != nil {
+			return nil, err
+		}
+		result = append(result, e)
+	}
+	return result, rows.Err()
+}
+
+// SetCargoYanked reports false when no such crate version exists.
+func (db *DB) SetCargoYanked(ctx context.Context, repoName, name, version string, yanked bool) (bool, error) {
+	tag, err := db.Pool.Exec(ctx, `
+		UPDATE cargo_index SET yanked = $4
+		WHERE repo_name = $1 AND lower(name) = $2 AND version = $3
+	`, repoName, strings.ToLower(name), version, yanked)
+	if err != nil {
+		return false, err
+	}
+	return tag.RowsAffected() > 0, nil
+}
+
+func (db *DB) DeleteCargoIndexEntry(ctx context.Context, repoName, filePath string) error {
+	_, err := db.Pool.Exec(ctx, `DELETE FROM cargo_index WHERE repo_name = $1 AND file_path = $2`, repoName, filePath)
+	return err
+}

internal/database/database_test.go

@@ -425,3 +425,38 @@ func TestRPMMetadata(t *testing.T) {
 		t.Fatalf("list rpm rows: %v %v", len(rows), err)
 	}
 }
+
+func TestCargoIndex(t *testing.T) {
+	requireDB(t)
+	for _, v := range []string{"0.1.0", "0.2.0"} {
+		if err := testDB.InsertCargoIndexEntry(ctx(), &provider.CargoIndexEntry{
+			RepoName: "cargo-db", FilePath: "crates/Unkin-Hello/Unkin-Hello-" + v + ".crate",
+			Name: "Unkin-Hello", Version: v, Line: `{"vers":"` + v + `"}`,
+		}); err != nil {
+			t.Fatal(err)
+		}
+	}
+
+	found, err := testDB.SetCargoYanked(ctx(), "cargo-db", "unkin-hello", "0.1.0", true)
+	if err != nil || !found {
+		t.Fatalf("yank: found=%v err=%v", found, err)
+	}
+	if found, _ := testDB.SetCargoYanked(ctx(), "cargo-db", "unkin-hello", "9.9.9", true); found {
+		t.Error("yank of a missing version reported found")
+	}
+
+	entries, err := testDB.ListCargoIndexEntries(ctx(), "cargo-db", "unkin-hello")
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(entries) != 2 || entries[0].Version != "0.1.0" || !entries[0].Yanked || entries[1].Yanked {
+		t.Fatalf("entries = %+v", entries)
+	}
+
+	if err := testDB.DeleteCargoIndexEntry(ctx(), "cargo-db", entries[0].FilePath); err != nil {
+		t.Fatal(err)
+	}
+	if entries, _ = testDB.ListCargoIndexEntries(ctx(), "cargo-db", "UNKIN-HELLO"); len(entries) != 1 {
+		t.Fatalf("after delete: %+v", entries)
+	}
+}

internal/provider/cargo/local.go

@@ -0,0 +1,303 @@
+package cargo
+
+import (
+	"context"
+	"crypto/sha256"
+	"encoding/binary"
+	"encoding/hex"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"log/slog"
+	"net/http"
+	"strings"
+
+	"git.unkin.net/unkin/artifactapi/internal/provider"
+)
+
+// publishDep is a dependency as cargo sends it in the publish metadata.
+type publishDep struct {
+	Name               string   `json:"name"`
+	VersionReq         string   `json:"version_req"`
+	Features           []string `json:"features"`
+	Optional           bool     `json:"optional"`
+	DefaultFeatures    bool     `json:"default_features"`
+	Target             *string  `json:"target"`
+	Kind               string   `json:"kind"`
+	Registry           *string  `json:"registry"`
+	ExplicitNameInToml *string  `json:"explicit_name_in_toml"`
+}
+
+// PublishMeta is the JSON half of a `PUT /api/v1/crates/new` body; fields the
+// index does not carry are ignored.
+type PublishMeta struct {
+	Name        string              `json:"name"`
+	Vers        string              `json:"vers"`
+	Deps        []publishDep        `json:"deps"`
+	Features    map[string][]string `json:"features"`
+	Links       *string             `json:"links"`
+	RustVersion *string             `json:"rust_version"`
+}
+
+type indexDep struct {
+	Name            string   `json:"name"`
+	Req             string   `json:"req"`
+	Features        []string `json:"features"`
+	Optional        bool     `json:"optional"`
+	DefaultFeatures bool     `json:"default_features"`
+	Target          *string  `json:"target"`
+	Kind            string   `json:"kind"`
+	Registry        *string  `json:"registry,omitempty"`
+	Package         *string  `json:"package,omitempty"`
+}
+
+// IndexLine is one line of a sparse index file.
+type IndexLine struct {
+	Name        string              `json:"name"`
+	Vers        string              `json:"vers"`
+	Deps        []indexDep          `json:"deps"`
+	Cksum       string              `json:"cksum"`
+	Features    map[string][]string `json:"features"`
+	Features2   map[string][]string `json:"features2,omitempty"`
+	Yanked      bool                `json:"yanked"`
+	Links       *string             `json:"links,omitempty"`
+	V           int                 `json:"v,omitempty"`
+	RustVersion *string             `json:"rust_version,omitempty"`
+}
+
+// ParsePublish splits a publish body: u32le JSON length, JSON, u32le crate length, crate.
+func ParsePublish(body []byte) (PublishMeta, []byte, error) {
+	var meta PublishMeta
+	jsonBytes, rest, err := lengthPrefixed(body)
+	if err != nil {
+		return meta, nil, fmt.Errorf("metadata: %w", err)
+	}
+	crate, rest, err := lengthPrefixed(rest)
+	if err != nil {
+		return meta, nil, fmt.Errorf("crate: %w", err)
+	}
+	if len(rest) != 0 {
+		return meta, nil, errors.New("trailing bytes after crate")
+	}
+	if err := json.Unmarshal(jsonBytes, &meta); err != nil {
+		return meta, nil, fmt.Errorf("metadata: %w", err)
+	}
+	return meta, crate, nil
+}
+
+func lengthPrefixed(b []byte) ([]byte, []byte, error) {
+	if len(b) < 4 {
+		return nil, nil, errors.New("truncated length")
+	}
+	n := binary.LittleEndian.Uint32(b)
+	b = b[4:]
+	if uint64(n) > uint64(len(b)) {
+		return nil, nil, errors.New("truncated body")
+	}
+	return b[:n], b[n:], nil
+}
+
+// NewIndexLine validates the metadata and translates it into index form.
+func NewIndexLine(meta PublishMeta, crate []byte) (IndexLine, error) {
+	if !validName(meta.Name) {
+		return IndexLine{}, fmt.Errorf("invalid crate name %q", meta.Name)
+	}
+	if !validVersion(meta.Vers) {
+		return IndexLine{}, fmt.Errorf("invalid version %q", meta.Vers)
+	}
+	sum := sha256.Sum256(crate)
+	line := IndexLine{
+		Name:        meta.Name,
+		Vers:        meta.Vers,
+		Deps:        make([]indexDep, 0, len(meta.Deps)),
+		Cksum:       hex.EncodeToString(sum[:]),
+		Features:    map[string][]string{},
+		Links:       meta.Links,
+		RustVersion: meta.RustVersion,
+	}
+	for _, d := range meta.Deps {
+		dep := indexDep{
+			Name:            d.Name,
+			Req:             d.VersionReq,
+			Features:        d.Features,
+			Optional:        d.Optional,
+			DefaultFeatures: d.DefaultFeatures,
+			Target:          d.Target,
+			Kind:            d.Kind,
+			Registry:        d.Registry,
+		}
+		if dep.Features == nil {
+			dep.Features = []string{}
+		}
+		if dep.Kind == "" {
+			dep.Kind = "normal"
+		}
+		if d.ExplicitNameInToml != nil && *d.ExplicitNameInToml != "" {
+			pkg := d.Name
+			dep.Name, dep.Package = *d.ExplicitNameInToml, &pkg
+		}
+		line.Deps = append(line.Deps, dep)
+	}
+	// Older cargo cannot parse dep:/?/ feature syntax, so those features live in
+	// features2 behind v:2, as on crates.io.
+	for name, vals := range meta.Features {
+		if vals == nil {
+			vals = []string{}
+		}
+		if usesNewFeatureSyntax(vals) {
+			if line.Features2 == nil {
+				line.Features2 = map[string][]string{}
+			}
+			line.Features2[name] = vals
+			continue
+		}
+		line.Features[name] = vals
+	}
+	if line.Features2 != nil {
+		line.V = 2
+	}
+	return line, nil
+}
+
+func usesNewFeatureSyntax(vals []string) bool {
+	for _, v := range vals {
+		if strings.HasPrefix(v, "dep:") || strings.Contains(v, "?/") {
+			return true
+		}
+	}
+	return false
+}
+
+func validName(name string) bool {
+	if name == "" || len(name) > 64 {
+		return false
+	}
+	for i, c := range name {
+		alpha := (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
+		if i == 0 && !alpha {
+			return false
+		}
+		if !alpha && (c < '0' || c > '9') && c != '-' && c != '_' {
+			return false
+		}
+	}
+	return true
+}
+
+func validVersion(v string) bool {
+	if v == "" || len(v) > 128 || !strings.ContainsAny(v[:1], "0123456789") {
+		return false
+	}
+	return strings.Trim(v, "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789.-+") == ""
+}
+
+// IndexPath is the sparse-index file path for a crate name.
+func IndexPath(name string) string {
+	n := strings.ToLower(name)
+	switch len(n) {
+	case 0:
+		return ""
+	case 1:
+		return "1/" + n
+	case 2:
+		return "2/" + n
+	case 3:
+		return "3/" + n[:1] + "/" + n
+	}
+	return n[:2] + "/" + n[2:4] + "/" + n
+}
+
+// CratePath is where a published .crate is stored, matching config.json's dl.
+func CratePath(name, version string) string {
+	return crateDir + name + "/" + name + "-" + version + ".crate"
+}
+
+// RenderIndex joins stored lines into an index file with each yanked flag applied.
+func RenderIndex(entries []provider.CargoIndexEntry) ([]byte, error) {
+	var b strings.Builder
+	for _, e := range entries {
+		var line IndexLine
+		if err := json.Unmarshal([]byte(e.Line), &line); err != nil {
+			return nil, fmt.Errorf("%s %s: %w", e.Name, e.Version, err)
+		}
+		line.Yanked = e.Yanked
+		raw, err := json.Marshal(line)
+		if err != nil {
+			return nil, err
+		}
+		b.Write(raw)
+		b.WriteByte('\n')
+	}
+	return []byte(b.String()), nil
+}
+
+// LocalConfig is config.json for a local registry rooted at base.
+func LocalConfig(base string) map[string]string {
+	return map[string]string{
+		"dl":  base + "/" + crateDir + "{crate}/{crate}-{version}.crate",
+		"api": base,
+	}
+}
+
+func (p *Provider) ServeLocalIndex(w http.ResponseWriter, r *http.Request, files provider.FileStore, repoName, path string) bool {
+	path = strings.TrimLeft(path, "/")
+	if path == "config.json" {
+		w.Header().Set("Content-Type", "application/json")
+		_ = json.NewEncoder(w).Encode(LocalConfig(requestBase(r) + "/api/v1/local/" + repoName))
+		return true
+	}
+	name := path[strings.LastIndex(path, "/")+1:]
+	if name == "" || IndexPath(name) != path {
+		return false
+	}
+
+	reader, ok := files.(provider.CargoIndexReader)
+	if !ok {
+		http.Error(w, "cargo index not available", http.StatusInternalServerError)
+		return true
+	}
+	entries, err := reader.ListCargoIndexEntries(r.Context(), repoName, name)
+	if err != nil {
+		http.Error(w, err.Error(), http.StatusInternalServerError)
+		return true
+	}
+	if len(entries) == 0 {
+		http.Error(w, "not found", http.StatusNotFound)
+		return true
+	}
+	body, err := RenderIndex(entries)
+	if err != nil {
+		http.Error(w, err.Error(), http.StatusInternalServerError)
+		return true
+	}
+	w.Header().Set("Content-Type", "text/plain")
+	w.WriteHeader(http.StatusOK)
+	_, _ = w.Write(body)
+	return true
+}
+
+func (p *Provider) GenerateLocalIndex(ctx context.Context, files provider.FileStore, repoName, path string) ([]byte, error) {
+	return nil, fmt.Errorf("cargo local index generation for virtual repos not supported")
+}
+
+func (p *Provider) AfterDelete(ctx context.Context, repoName, storagePath string, db provider.MetadataDeleter) error {
+	deleter, ok := db.(provider.CargoIndexDeleter)
+	if !ok {
+		return nil
+	}
+	if err := deleter.DeleteCargoIndexEntry(ctx, repoName, storagePath); err != nil {
+		slog.Error("cargo index: delete failed", "repo", repoName, "path", storagePath, "error", err)
+		return err
+	}
+	return nil
+}
+
+func requestBase(r *http.Request) string {
+	scheme := "http"
+	if r.TLS != nil {
+		scheme = "https"
+	} else if fwd := r.Header.Get("X-Forwarded-Proto"); fwd != "" {
+		scheme = fwd
+	}
+	return scheme + "://" + r.Host
+}

internal/provider/cargo/local_test.go

@@ -0,0 +1,218 @@
+package cargo
+
+import (
+	"context"
+	"crypto/sha256"
+	"encoding/binary"
+	"encoding/hex"
+	"encoding/json"
+	"net/http/httptest"
+	"strings"
+	"testing"
+
+	"git.unkin.net/unkin/artifactapi/internal/provider"
+)
+
+func publishBody(meta string, crate []byte) []byte {
+	var b []byte
+	b = binary.LittleEndian.AppendUint32(b, uint32(len(meta)))
+	b = append(b, meta...)
+	b = binary.LittleEndian.AppendUint32(b, uint32(len(crate)))
+	return append(b, crate...)
+}
+
+const sampleMeta = `{
+  "name": "unkin-hello", "vers": "0.2.0",
+  "deps": [
+    {"name": "serde", "version_req": "^1", "features": ["derive"], "optional": false,
+     "default_features": true, "target": null, "kind": "normal",
+     "registry": "https://github.com/rust-lang/crates.io-index"},
+    {"name": "rand_core", "version_req": "^0.6", "features": null, "optional": true,
+     "default_features": false, "target": "cfg(unix)", "kind": "dev",
+     "explicit_name_in_toml": "rc"}
+  ],
+  "features": {"default": ["std"], "std": [], "extra": ["dep:rc"], "weak": ["serde?/std"]},
+  "links": null, "rust_version": "1.70", "description": "ignored"
+}`
+
+func TestParsePublishAndIndexLine(t *testing.T) {
+	crate := []byte("fake crate bytes")
+	meta, gotCrate, err := ParsePublish(publishBody(sampleMeta, crate))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if string(gotCrate) != string(crate) {
+		t.Fatalf("crate = %q", gotCrate)
+	}
+	line, err := NewIndexLine(meta, crate)
+	if err != nil {
+		t.Fatal(err)
+	}
+	sum := sha256.Sum256(crate)
+	if line.Cksum != hex.EncodeToString(sum[:]) {
+		t.Errorf("cksum = %s", line.Cksum)
+	}
+
+	raw, _ := json.Marshal(line)
+	var got map[string]any
+	_ = json.Unmarshal(raw, &got)
+	if _, ok := got["version_req"]; ok {
+		t.Error("publish-only field leaked into index line")
+	}
+	if got["v"] != float64(2) {
+		t.Errorf("v = %v, want 2 with features2 present", got["v"])
+	}
+	if got["rust_version"] != "1.70" {
+		t.Errorf("rust_version = %v", got["rust_version"])
+	}
+	if _, ok := line.Features["extra"]; ok {
+		t.Error("dep: feature must move to features2")
+	}
+	if len(line.Features2) != 2 || len(line.Features) != 2 {
+		t.Errorf("features = %v features2 = %v", line.Features, line.Features2)
+	}
+
+	deps := got["deps"].([]any)
+	serde := deps[0].(map[string]any)
+	if serde["req"] != "^1" || serde["registry"] != "https://github.com/rust-lang/crates.io-index" || serde["target"] != nil {
+		t.Errorf("serde dep = %v", serde)
+	}
+	if _, ok := serde["package"]; ok {
+		t.Error("package set without a rename")
+	}
+	renamed := deps[1].(map[string]any)
+	if renamed["name"] != "rc" || renamed["package"] != "rand_core" || renamed["kind"] != "dev" {
+		t.Errorf("renamed dep = %v", renamed)
+	}
+	if f, ok := renamed["features"].([]any); !ok || len(f) != 0 {
+		t.Errorf("null features must render as [], got %v", renamed["features"])
+	}
+}
+
+func TestIndexLineNoNewFeaturesOmitsV(t *testing.T) {
+	line, err := NewIndexLine(PublishMeta{Name: "a", Vers: "1.0.0", Features: map[string][]string{"x": nil}}, nil)
+	if err != nil {
+		t.Fatal(err)
+	}
+	raw, _ := json.Marshal(line)
+	s := string(raw)
+	for _, bad := range []string{`"v":`, `"features2"`, `"links"`} {
+		if strings.Contains(s, bad) {
+			t.Errorf("%s present in %s", bad, s)
+		}
+	}
+	if !strings.Contains(s, `"deps":[]`) || !strings.Contains(s, `"features":{"x":[]}`) {
+		t.Errorf("deps/features must be non-null: %s", s)
+	}
+}
+
+func TestParsePublishRejectsMalformed(t *testing.T) {
+	good := publishBody(`{"name":"a","vers":"1.0.0"}`, []byte("c"))
+	for name, body := range map[string][]byte{
+		"empty":     nil,
+		"short":     good[:len(good)-1],
+		"trailing":  append(append([]byte{}, good...), 'x'),
+		"bad json":  publishBody(`{`, []byte("c")),
+		"no length": good[:6],
+	} {
+		if _, _, err := ParsePublish(body); err == nil {
+			t.Errorf("%s: expected error", name)
+		}
+	}
+}
+
+func TestNewIndexLineValidation(t *testing.T) {
+	for _, m := range []PublishMeta{
+		{Name: "", Vers: "1.0.0"},
+		{Name: "1abc", Vers: "1.0.0"},
+		{Name: "../x", Vers: "1.0.0"},
+		{Name: "ok", Vers: "1.0.0/../../x"},
+		{Name: "ok", Vers: ""},
+		{Name: strings.Repeat("a", 65), Vers: "1.0.0"},
+	} {
+		if _, err := NewIndexLine(m, nil); err == nil {
+			t.Errorf("%q@%q accepted", m.Name, m.Vers)
+		}
+	}
+	if _, err := NewIndexLine(PublishMeta{Name: "Foo_bar-2", Vers: "1.0.0-rc.1+build.5"}, nil); err != nil {
+		t.Error(err)
+	}
+}
+
+func TestIndexPath(t *testing.T) {
+	for name, want := range map[string]string{
+		"a": "1/a", "ab": "2/ab", "abc": "3/a/abc", "Serde": "se/rd/serde", "unkin-hello": "un/ki/unkin-hello",
+	} {
+		if got := IndexPath(name); got != want {
+			t.Errorf("IndexPath(%q) = %q, want %q", name, got, want)
+		}
+	}
+}
+
+type fakeIndex struct {
+	entries []provider.CargoIndexEntry
+	asked   string
+}
+
+func (f *fakeIndex) ListCargoIndexEntries(_ context.Context, _, name string) ([]provider.CargoIndexEntry, error) {
+	f.asked = name
+	return f.entries, nil
+}
+func (f *fakeIndex) ListFilesByPrefix(context.Context, string, string) ([]provider.FileEntry, error) {
+	return nil, nil
+}
+func (f *fakeIndex) ListPackages(context.Context, string) ([]string, error) { return nil, nil }
+
+func TestServeLocalIndex(t *testing.T) {
+	line, _ := NewIndexLine(PublishMeta{Name: "unkin-hello", Vers: "0.1.0"}, []byte("x"))
+	raw, _ := json.Marshal(line)
+	files := &fakeIndex{entries: []provider.CargoIndexEntry{
+		{Name: "unkin-hello", Version: "0.1.0", Line: string(raw), Yanked: true},
+		{Name: "unkin-hello", Version: "0.2.0", Line: strings.Replace(string(raw), "0.1.0", "0.2.0", 1)},
+	}}
+	p := &Provider{}
+
+	rec := httptest.NewRecorder()
+	if !p.ServeLocalIndex(rec, httptest.NewRequest("GET", "http://aa.example/x", nil), files, "crates-local", "un/ki/unkin-hello") {
+		t.Fatal("index path not served")
+	}
+	lines := strings.Split(strings.TrimSpace(rec.Body.String()), "\n")
+	if len(lines) != 2 || !strings.Contains(lines[0], `"yanked":true`) || !strings.Contains(lines[1], `"yanked":false`) {
+		t.Errorf("index body = %q", rec.Body.String())
+	}
+	if files.asked != "unkin-hello" {
+		t.Errorf("looked up %q", files.asked)
+	}
+
+	rec = httptest.NewRecorder()
+	req := httptest.NewRequest("GET", "http://aa.example/config.json", nil)
+	req.Header.Set("X-Forwarded-Proto", "https")
+	p.ServeLocalIndex(rec, req, files, "crates-local", "config.json")
+	var cfg map[string]string
+	_ = json.Unmarshal(rec.Body.Bytes(), &cfg)
+	if cfg["api"] != "https://aa.example/api/v1/local/crates-local" ||
+		cfg["dl"] != "https://aa.example/api/v1/local/crates-local/crates/{crate}/{crate}-{version}.crate" {
+		t.Errorf("config = %v", cfg)
+	}
+
+	for _, path := range []string{"crates/unkin-hello/unkin-hello-0.1.0.crate", "xx/ki/unkin-hello", ""} {
+		if p.ServeLocalIndex(httptest.NewRecorder(), httptest.NewRequest("GET", "/", nil), files, "r", path) {
+			t.Errorf("%q should fall through", path)
+		}
+	}
+
+	files.entries = nil
+	rec = httptest.NewRecorder()
+	p.ServeLocalIndex(rec, httptest.NewRequest("GET", "/", nil), files, "r", "3/a/abc")
+	if rec.Code != 404 {
+		t.Errorf("unknown crate code = %d", rec.Code)
+	}
+}
+
+func TestCratePathMatchesDL(t *testing.T) {
+	dl := LocalConfig("http://h/api/v1/local/r")["dl"]
+	want := strings.NewReplacer("{crate}", "unkin-hello", "{version}", "0.1.0").Replace(dl)
+	if got := "http://h/api/v1/local/r/" + CratePath("unkin-hello", "0.1.0"); got != want {
+		t.Errorf("CratePath = %q, dl expands to %q", got, want)
+	}
+}

internal/provider/cargo_index.go

@@ -0,0 +1,24 @@
+package provider
+
+import "context"
+
+// CargoIndexEntry is one published crate version in a local cargo registry.
+// Line is the sparse-index JSON line; Yanked is applied when it is served.
+type CargoIndexEntry struct {
+	RepoName string
+	FilePath string
+	Name     string
+	Version  string
+	Line     string
+	Yanked   bool
+}
+
+// CargoIndexReader lists every version of one crate (matched case-insensitively).
+// *database.DB satisfies it and CargoIndexDeleter.
+type CargoIndexReader interface {
+	ListCargoIndexEntries(ctx context.Context, repoName, name string) ([]CargoIndexEntry, error)
+}
+
+type CargoIndexDeleter interface {
+	DeleteCargoIndexEntry(ctx context.Context, repoName, filePath string) error
+}

internal/provider/gems/gems.go

@@ -1,59 +0,0 @@
-// Package gems proxies a RubyGems source: mutable compact index (versions,
-// info/*, names) and legacy specs indexes, plus immutable .gem and
-// quick/Marshal.4.8 gemspec downloads.
-package gems
-
-import (
-	"context"
-	"net/http"
-	"strings"
-
-	"git.unkin.net/unkin/artifactapi/internal/auth"
-	"git.unkin.net/unkin/artifactapi/internal/provider"
-	"git.unkin.net/unkin/artifactapi/pkg/models"
-)
-
-func init() {
-	provider.Register(&Provider{})
-}
-
-type Provider struct{}
-
-func (p *Provider) Type() models.PackageType { return models.PackageGems }
-
-func (p *Provider) Classify(path string) provider.Mutability {
-	if isGem(path) || isQuickSpec(path) {
-		return provider.Immutable
-	}
-	return provider.Mutable
-}
-
-func (p *Provider) ContentType(path string) string {
-	path = strings.TrimLeft(path, "/")
-	if path == "versions" || path == "names" || strings.HasPrefix(path, "info/") {
-		return "text/plain; charset=utf-8"
-	}
-	return "application/octet-stream"
-}
-
-func (p *Provider) UpstreamURL(remote models.Remote, path string) string {
-	return strings.TrimRight(remote.BaseURL, "/") + "/" + strings.TrimLeft(path, "/")
-}
-
-func (p *Provider) RewriteResponse(_ []byte, _ models.Remote, _ string) ([]byte, error) {
-	return nil, nil
-}
-
-func (p *Provider) AuthHeaders(_ context.Context, remote models.Remote) (http.Header, error) {
-	return auth.BasicHeaders(remote), nil
-}
-
-func isGem(path string) bool {
-	path = strings.TrimLeft(path, "/")
-	return strings.HasPrefix(path, "gems/") && strings.HasSuffix(path, ".gem")
-}
-
-func isQuickSpec(path string) bool {
-	path = strings.TrimLeft(path, "/")
-	return strings.HasPrefix(path, "quick/Marshal.4.8/") && strings.HasSuffix(path, ".gemspec.rz")
-}

internal/provider/gems/gems_test.go

@@ -1,58 +0,0 @@
-package gems
-
-import (
-	"testing"
-
-	"git.unkin.net/unkin/artifactapi/internal/provider"
-	"git.unkin.net/unkin/artifactapi/pkg/models"
-)
-
-func TestClassify(t *testing.T) {
-	p := &Provider{}
-	tests := []struct {
-		path string
-		want provider.Mutability
-	}{
-		{"gems/rack-3.1.8.gem", provider.Immutable},
-		{"/gems/nokogiri-1.16.7-x86_64-linux.gem", provider.Immutable},
-		{"quick/Marshal.4.8/rack-3.1.8.gemspec.rz", provider.Immutable},
-		{"versions", provider.Mutable},
-		{"names", provider.Mutable},
-		{"info/rack", provider.Mutable},
-		{"info/gems", provider.Mutable},
-		{"specs.4.8.gz", provider.Mutable},
-		{"latest_specs.4.8.gz", provider.Mutable},
-		{"prerelease_specs.4.8.gz", provider.Mutable},
-		{"api/v1/dependencies", provider.Mutable},
-		{"quick/Marshal.4.8/", provider.Mutable},
-	}
-	for _, tt := range tests {
-		if got := p.Classify(tt.path); got != tt.want {
-			t.Errorf("Classify(%q) = %v, want %v", tt.path, got, tt.want)
-		}
-	}
-}
-
-func TestContentType(t *testing.T) {
-	p := &Provider{}
-	tests := map[string]string{
-		"versions":            "text/plain; charset=utf-8",
-		"info/rack":           "text/plain; charset=utf-8",
-		"names":               "text/plain; charset=utf-8",
-		"gems/rack-3.1.8.gem": "application/octet-stream",
-		"specs.4.8.gz":        "application/octet-stream",
-	}
-	for path, want := range tests {
-		if got := p.ContentType(path); got != want {
-			t.Errorf("ContentType(%q) = %q, want %q", path, got, want)
-		}
-	}
-}
-
-func TestUpstreamURL(t *testing.T) {
-	p := &Provider{}
-	r := models.Remote{BaseURL: "https://rubygems.org/"}
-	if got := p.UpstreamURL(r, "/info/rack"); got != "https://rubygems.org/info/rack" {
-		t.Errorf("UpstreamURL = %q", got)
-	}
-}

internal/server/server.go

@@ -25,7 +25,6 @@ import (
 	_ "git.unkin.net/unkin/artifactapi/internal/provider/cargo"
 	"git.unkin.net/unkin/artifactapi/internal/provider/deb"
 	_ "git.unkin.net/unkin/artifactapi/internal/provider/docker"
-	_ "git.unkin.net/unkin/artifactapi/internal/provider/gems"
 	_ "git.unkin.net/unkin/artifactapi/internal/provider/generic"
 	_ "git.unkin.net/unkin/artifactapi/internal/provider/goproxy"
 	_ "git.unkin.net/unkin/artifactapi/internal/provider/helm"

migrations/0003_cargo_index.sql

@@ -0,0 +1,15 @@
+-- Local cargo registries: one sparse-index line per published crate version.
+-- line holds the index JSON minus yanked, which flips on yank/unyank.
+CREATE TABLE IF NOT EXISTS cargo_index (
+	id           BIGSERIAL PRIMARY KEY,
+	repo_name    TEXT NOT NULL,
+	file_path    TEXT NOT NULL,
+	name         TEXT NOT NULL,
+	version      TEXT NOT NULL,
+	line         TEXT NOT NULL,
+	yanked       BOOLEAN NOT NULL DEFAULT FALSE,
+	created_at   TIMESTAMPTZ DEFAULT NOW(),
+	UNIQUE(repo_name, file_path)
+);
+
+CREATE INDEX IF NOT EXISTS idx_cargo_index_repo_name ON cargo_index(repo_name, lower(name));

pkg/models/package_type.go

@@ -17,7 +17,6 @@ const (
 	PackageTerraform    PackageType = "terraform"
 	PackageGoProxy      PackageType = "goproxy"
 	PackageCargo        PackageType = "cargo"
-	PackageGems         PackageType = "gems"
 	PackageGitHubRPM    PackageType = "github_rpm"
 	PackageGitHubDeb    PackageType = "github_deb"
 	PackageGitHubAlpine PackageType = "github_alpine"
@@ -36,7 +35,6 @@ var validPackageTypes = map[PackageType]bool{
 	PackageTerraform:    true,
 	PackageGoProxy:      true,
 	PackageCargo:        true,
-	PackageGems:         true,
 	PackageGitHubRPM:    true,
 	PackageGitHubDeb:    true,
 	PackageGitHubAlpine: true,

pkg/models/package_type_test.go

@@ -19,7 +19,6 @@ func TestPackageTypeValid(t *testing.T) {
 		models.PackageTerraform,
 		models.PackageGoProxy,
 		models.PackageCargo,
-		models.PackageGems,
 		models.PackageGitHubRPM,
 		models.PackageGitHubDeb,
 		models.PackageGitHubAlpine,

ui/src/components/UsageInstructions.tsx

@@ -229,6 +229,25 @@ go mod download`,
       ];
 
     case 'cargo':
+      if (isLocal) {
+        return [
+          {
+            title: 'Add the private registry (real sparse registry)',
+            language: 'toml',
+            code: `# $CARGO_HOME/config.toml
+[registries.${name}]
+index = "sparse+${url}/api/v1/local/${name}/"`,
+            note: 'cargo requires a token for publish/yank: set CARGO_REGISTRIES_' + name.toUpperCase().replace(/-/g, '_') + '_TOKEN to any value.',
+          },
+          {
+            title: 'Publish, depend on and yank crates',
+            language: 'bash',
+            code: `cargo publish --registry ${name}
+cargo add <crate> --registry ${name}
+cargo yank --version <version> --registry ${name}`,
+          },
+        ];
+      }
       return [
         {
           title: 'Point crates.io at this sparse registry',
@@ -243,16 +262,6 @@ index = "sparse+${proxy}/"`,
         },
       ];
 
-    case 'gems':
-      return [
-        {
-          title: 'Use this remote as the RubyGems source',
-          language: 'bash',
-          code: `gem sources --clear-all --add ${proxy}/
-bundle config mirror.https://rubygems.org ${proxy}`,
-        },
-      ];
-
     case 'puppet':
       return [
         {

ui/src/pages/Locals.tsx

@@ -18,7 +18,6 @@ const typeColors: Record<string, 'blue' | 'green' | 'yellow' | 'red' | 'default'
   terraform: 'blue',
   goproxy: 'green',
   cargo: 'yellow',
-  gems: 'red',
 };
 
 export function Locals() {

ui/src/pages/Remotes.tsx

@@ -18,7 +18,6 @@ const typeColors: Record<string, 'blue' | 'green' | 'yellow' | 'red' | 'default'
   terraform: 'blue',
   goproxy: 'green',
   cargo: 'yellow',
-  gems: 'red',
 };
 
 export function Remotes() {