rdev/internal/adapter/zot/client.go

// Package zot provides a client for interacting with the zot container registry.
//
// TODO: Deploy recommended Zot config with gcInterval, retention policies, and
// deduplication. Current live config has no periodic GC — old tags accumulate
// until disk fills. Add Zot manifests to deployments/k8s/base/zot/ for version
// control. See .claude/guides/ops/zot-registry.md for the recommended config.
package zot

import (
	"context"
	"encoding/json"
	"fmt"
	"io"
	"log/slog"
	"net/http"
	"strings"
	"time"

	"github.com/orchard9/rdev/internal/domain"
	"github.com/orchard9/rdev/internal/port"
)

// Client interacts with the zot container registry.
type Client struct {
	url        string
	httpClient *http.Client
	logger     *slog.Logger
}

// Ensure Client implements port.RegistryProvider at compile time.
var _ port.RegistryProvider = (*Client)(nil)

// NewClient creates a new zot client.
// The URL should be the registry base URL (e.g., "https://registry.threesix.ai").
func NewClient(url string) *Client {
	return &Client{
		url: url,
		httpClient: &http.Client{
			Timeout: 30 * time.Second,
		},
		logger: slog.Default(),
	}
}

// WithLogger sets the logger for the client.
func (c *Client) WithLogger(logger *slog.Logger) *Client {
	c.logger = logger
	return c
}

// maxResponseBodySize is the maximum response body size (10MB) to prevent OOM on large responses.
const maxResponseBodySize = 10 * 1024 * 1024

// doWithRetry executes an HTTP request with up to 3 attempts and exponential backoff.
// It retries on network errors and 5xx status codes, but NOT on 4xx client errors.
//
// NOTE: This assumes the request has no body (GET, HEAD, DELETE) since the body
// cannot be re-read on retry. If POST/PUT support is needed, the caller must
// provide a body factory or buffer the body for re-use.
func (c *Client) doWithRetry(req *http.Request) (*http.Response, error) {
	var lastErr error
	for attempt := 0; attempt < 3; attempt++ {
		if attempt > 0 {
			backoff := time.Duration(1<<uint(attempt-1)) * time.Second
			c.logger.Warn("retrying registry request",
				"attempt", attempt+1,
				"backoff", backoff,
				"method", req.Method,
				"url", req.URL.String(),
			)
			time.Sleep(backoff)
		}

		resp, err := c.httpClient.Do(req)
		if err != nil {
			lastErr = err
			continue
		}
		if resp.StatusCode >= 500 {
			body, _ := io.ReadAll(io.LimitReader(resp.Body, maxResponseBodySize))
			resp.Body.Close()
			lastErr = fmt.Errorf("registry returned %d: %s", resp.StatusCode, string(body))
			continue
		}
		return resp, nil
	}
	return nil, fmt.Errorf("registry request failed after 3 attempts: %w", lastErr)
}

// Check returns the health status of the registry.
// A 200 or 401 response indicates the registry is healthy (401 means auth required but registry is up).
func (c *Client) Check(ctx context.Context) domain.RegistryStatus {
	start := time.Now()

	req, err := http.NewRequestWithContext(ctx, http.MethodGet, c.url+"/v2/", nil)
	if err != nil {
		return domain.RegistryStatus{
			Healthy:     false,
			URL:         c.url,
			Error:       fmt.Sprintf("failed to create request: %v", err),
			LastChecked: time.Now().UTC(),
		}
	}

	resp, err := c.doWithRetry(req)
	latency := time.Since(start)

	if err != nil {
		return domain.RegistryStatus{
			Healthy:     false,
			URL:         c.url,
			Latency:     latency.String(),
			Error:       fmt.Sprintf("connection error: %v", err),
			LastChecked: time.Now().UTC(),
		}
	}
	defer func() { _ = resp.Body.Close() }()

	// 200 = healthy, 401 = healthy but requires auth
	healthy := resp.StatusCode == http.StatusOK || resp.StatusCode == http.StatusUnauthorized

	status := domain.RegistryStatus{
		Healthy:     healthy,
		URL:         c.url,
		Latency:     latency.String(),
		LastChecked: time.Now().UTC(),
	}

	if !healthy {
		status.Error = fmt.Sprintf("unexpected status code: %d", resp.StatusCode)
	}

	return status
}

// catalogResponse is the OCI catalog API response.
type catalogResponse struct {
	Repositories []string `json:"repositories"`
}

// tagsResponse is the OCI tags list API response.
type tagsResponse struct {
	Name string   `json:"name"`
	Tags []string `json:"tags"`
}

// ListRepositories returns all repositories in the registry.
func (c *Client) ListRepositories(ctx context.Context) ([]string, error) {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, c.url+"/v2/_catalog", nil)
	if err != nil {
		return nil, fmt.Errorf("create request: %w", err)
	}

	resp, err := c.doWithRetry(req)
	if err != nil {
		return nil, fmt.Errorf("request failed: %w", err)
	}
	defer func() { _ = resp.Body.Close() }()

	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("unexpected status: %d", resp.StatusCode)
	}

	body, err := io.ReadAll(io.LimitReader(resp.Body, maxResponseBodySize))
	if err != nil {
		return nil, fmt.Errorf("read body: %w", err)
	}

	var catalog catalogResponse
	if err := json.Unmarshal(body, &catalog); err != nil {
		return nil, fmt.Errorf("parse catalog: %w", err)
	}

	return catalog.Repositories, nil
}

// ListProjectRepositories returns all repositories for a specific project.
// This includes both the main repo and sub-repos like cache.
func (c *Client) ListProjectRepositories(ctx context.Context, projectID string) ([]string, error) {
	allRepos, err := c.ListRepositories(ctx)
	if err != nil {
		return nil, err
	}

	var projectRepos []string
	prefix := projectID + "/"
	for _, repo := range allRepos {
		if repo == projectID || strings.HasPrefix(repo, prefix) {
			projectRepos = append(projectRepos, repo)
		}
	}
	return projectRepos, nil
}

// DeleteRepository deletes all tags and manifests for a repository.
// This triggers Zot's garbage collection to reclaim storage.
func (c *Client) DeleteRepository(ctx context.Context, repo string) error {
	// List all tags for the repository
	tags, err := c.listTags(ctx, repo)
	if err != nil {
		return fmt.Errorf("list tags for %s: %w", repo, err)
	}

	// Delete each tag's manifest
	for _, tag := range tags {
		if err := c.deleteManifest(ctx, repo, tag); err != nil {
			c.logger.Warn("failed to delete manifest", "repo", repo, "tag", tag, "error", err)
			// Continue with other tags
		}
	}

	c.logger.Info("deleted repository", "repo", repo, "tags_deleted", len(tags))
	return nil
}

// DeleteProjectRepositories deletes all repositories for a project.
func (c *Client) DeleteProjectRepositories(ctx context.Context, projectID string) error {
	repos, err := c.ListProjectRepositories(ctx, projectID)
	if err != nil {
		return fmt.Errorf("list project repos: %w", err)
	}

	for _, repo := range repos {
		if err := c.DeleteRepository(ctx, repo); err != nil {
			c.logger.Warn("failed to delete repo", "repo", repo, "error", err)
			// Continue with other repos
		}
	}

	c.logger.Info("deleted project repositories", "project", projectID, "repos_deleted", len(repos))
	return nil
}

// listTags returns all tags for a repository.
func (c *Client) listTags(ctx context.Context, repo string) ([]string, error) {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, c.url+"/v2/"+repo+"/tags/list", nil)
	if err != nil {
		return nil, fmt.Errorf("create request: %w", err)
	}

	resp, err := c.doWithRetry(req)
	if err != nil {
		return nil, fmt.Errorf("request failed: %w", err)
	}
	defer func() { _ = resp.Body.Close() }()

	if resp.StatusCode == http.StatusNotFound {
		return nil, nil // Repo doesn't exist, nothing to delete
	}
	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("unexpected status: %d", resp.StatusCode)
	}

	body, err := io.ReadAll(io.LimitReader(resp.Body, maxResponseBodySize))
	if err != nil {
		return nil, fmt.Errorf("read body: %w", err)
	}

	var tags tagsResponse
	if err := json.Unmarshal(body, &tags); err != nil {
		return nil, fmt.Errorf("parse tags: %w", err)
	}

	return tags.Tags, nil
}

// deleteManifest deletes a specific manifest by tag.
func (c *Client) deleteManifest(ctx context.Context, repo, tag string) error {
	// First, get the manifest digest
	headReq, err := http.NewRequestWithContext(ctx, http.MethodHead, c.url+"/v2/"+repo+"/manifests/"+tag, nil)
	if err != nil {
		return fmt.Errorf("create head request: %w", err)
	}
	headReq.Header.Set("Accept", "application/vnd.oci.image.manifest.v1+json, application/vnd.docker.distribution.manifest.v2+json")

	headResp, err := c.doWithRetry(headReq)
	if err != nil {
		return fmt.Errorf("head request failed: %w", err)
	}
	defer func() { _ = headResp.Body.Close() }()

	if headResp.StatusCode == http.StatusNotFound {
		return nil // Already deleted
	}
	if headResp.StatusCode != http.StatusOK {
		return fmt.Errorf("head unexpected status: %d", headResp.StatusCode)
	}

	digest := headResp.Header.Get("Docker-Content-Digest")
	if digest == "" {
		return fmt.Errorf("no digest in response")
	}

	// Delete by digest
	delReq, err := http.NewRequestWithContext(ctx, http.MethodDelete, c.url+"/v2/"+repo+"/manifests/"+digest, nil)
	if err != nil {
		return fmt.Errorf("create delete request: %w", err)
	}

	delResp, err := c.doWithRetry(delReq)
	if err != nil {
		return fmt.Errorf("delete request failed: %w", err)
	}
	defer func() { _ = delResp.Body.Close() }()

	if delResp.StatusCode != http.StatusAccepted && delResp.StatusCode != http.StatusOK && delResp.StatusCode != http.StatusNotFound {
		return fmt.Errorf("delete unexpected status: %d", delResp.StatusCode)
	}

	return nil
}