clonepack/internal/clone/rpm/cloner.go

package rpm

import (
	"compress/gzip"
	"context"
	"encoding/xml"
	"fmt"
	"io"
	"net/http"
	"os"
	"path/filepath"
	"strings"
	"time"
)

type Progress struct {
	File      string
	BytesDone int64
}

type ProgressFunc func(p Progress)

type Cloner struct {
	SourceURL  string
	DestDir    string
	HTTPClient *http.Client
	OnProgress ProgressFunc
}

func New(sourceURL, destDir string) *Cloner {
	return &Cloner{
		SourceURL:  strings.TrimRight(sourceURL, "/"),
		DestDir:    destDir,
		HTTPClient: &http.Client{Timeout: 30 * time.Minute},
	}
}

func (c *Cloner) Clone(ctx context.Context) error {
	if err := os.MkdirAll(filepath.Join(c.DestDir, "repodata"), 0o755); err != nil {
		return fmt.Errorf("create repodata dir: %w", err)
	}
	if err := os.MkdirAll(filepath.Join(c.DestDir, "Packages"), 0o755); err != nil {
		return fmt.Errorf("create Packages dir: %w", err)
	}

	repomd, err := c.fetchRepoMD(ctx)
	if err != nil {
		return fmt.Errorf("fetch repomd.xml: %w", err)
	}

	var primaryEntry *RepoMDEntry
	for i, entry := range repomd.Data {
		if entry.Type == "primary" {
			primaryEntry = &repomd.Data[i]
			continue
		}
		if err := c.downloadMetadataFile(ctx, entry); err != nil {
			return fmt.Errorf("download metadata %s: %w", entry.Type, err)
		}
	}

	if primaryEntry == nil {
		return fmt.Errorf("no primary metadata found in repomd.xml")
	}

	packages, err := c.fetchPrimary(ctx, *primaryEntry)
	if err != nil {
		return fmt.Errorf("fetch primary.xml: %w", err)
	}

	for _, pkg := range packages {
		if err := ctx.Err(); err != nil {
			return err
		}
		if err := c.downloadPackage(ctx, pkg); err != nil {
			return fmt.Errorf("download package %s: %w", pkg.Name, err)
		}
	}

	return nil
}

func (c *Cloner) fetchRepoMD(ctx context.Context) (*RepoMD, error) {
	url := c.SourceURL + "/repodata/repomd.xml"
	data, err := c.fetchBytes(ctx, url)
	if err != nil {
		return nil, err
	}

	dest := filepath.Join(c.DestDir, "repodata", "repomd.xml")
	if err := writeFile(dest, data); err != nil {
		return nil, err
	}

	var repomd RepoMD
	if err := xml.Unmarshal(data, &repomd); err != nil {
		return nil, fmt.Errorf("parse repomd.xml: %w", err)
	}
	return &repomd, nil
}

func (c *Cloner) fetchPrimary(ctx context.Context, entry RepoMDEntry) ([]Package, error) {
	url := c.SourceURL + "/" + entry.Location.Href
	data, err := c.fetchBytes(ctx, url)
	if err != nil {
		return nil, err
	}

	if entry.Checksum.Type == "sha256" {
		if err := verifyChecksum(data, entry.Checksum.Value); err != nil {
			return nil, fmt.Errorf("primary.xml.gz: %w", err)
		}
	}

	dest := filepath.Join(c.DestDir, entry.Location.Href)
	if err := os.MkdirAll(filepath.Dir(dest), 0o755); err != nil {
		return nil, err
	}
	if err := writeFile(dest, data); err != nil {
		return nil, err
	}

	gz, err := gzip.NewReader(strings.NewReader(string(data)))
	if err != nil {
		return nil, fmt.Errorf("open gzip: %w", err)
	}
	defer gz.Close()

	var primary PrimaryMetadata
	if err := xml.NewDecoder(gz).Decode(&primary); err != nil {
		return nil, fmt.Errorf("parse primary.xml: %w", err)
	}
	return primary.Packages, nil
}

func (c *Cloner) downloadMetadataFile(ctx context.Context, entry RepoMDEntry) error {
	url := c.SourceURL + "/" + entry.Location.Href
	dest := filepath.Join(c.DestDir, entry.Location.Href)

	n, err := c.downloadAndVerify(ctx, url, dest, entry.Checksum.Type, entry.Checksum.Value)
	if err != nil {
		return err
	}
	if c.OnProgress != nil {
		c.OnProgress(Progress{File: entry.Location.Href, BytesDone: n})
	}
	return nil
}

func (c *Cloner) downloadPackage(ctx context.Context, pkg Package) error {
	url := c.SourceURL + "/" + pkg.Location.Href
	dest := filepath.Join(c.DestDir, pkg.Location.Href)

	if err := os.MkdirAll(filepath.Dir(dest), 0o755); err != nil {
		return err
	}

	n, err := c.downloadAndVerify(ctx, url, dest, pkg.Checksum.Type, pkg.Checksum.Value)
	if err != nil {
		return err
	}
	if c.OnProgress != nil {
		c.OnProgress(Progress{File: pkg.Location.Href, BytesDone: n})
	}
	return nil
}

func (c *Cloner) downloadAndVerify(ctx context.Context, url, destPath, checksumType, expectedChecksum string) (int64, error) {
	return DownloadAndVerify(ctx, c.HTTPClient, url, destPath, checksumType, expectedChecksum)
}

func (c *Cloner) fetchBytes(ctx context.Context, url string) ([]byte, error) {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, url, nil)
	if err != nil {
		return nil, err
	}
	resp, err := c.HTTPClient.Do(req)
	if err != nil {
		return nil, err
	}
	defer resp.Body.Close()

	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("HTTP %d for %s", resp.StatusCode, url)
	}
	return io.ReadAll(resp.Body)
}

func writeFile(path string, data []byte) error {
	if err := os.MkdirAll(filepath.Dir(path), 0o755); err != nil {
		return err
	}
	return os.WriteFile(path, data, 0o644)
}