1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301
|
// Copyright (c) 2018-2023, Sylabs Inc. All rights reserved.
// This software is licensed under a 3-clause BSD license. Please consult the
// LICENSE.md file distributed with the sources of this project regarding your
// rights to use or distribute this software.
package client
import (
"context"
"errors"
"fmt"
"io"
"net/http"
"net/url"
"os"
"strconv"
"strings"
jsonresp "github.com/sylabs/json-resp"
)
var errRequestedImageNotFound = fmt.Errorf("requested image was not found in the library")
// DownloadImage will retrieve an image from the Container Library, saving it
// into the specified io.Writer. The timeout value for this operation is set
// within the context. It is recommended to use a large value (ie. 1800 seconds)
// to prevent timeout when downloading large images.
func (c *Client) DownloadImage(ctx context.Context, w io.Writer, arch, path, tag string, callback func(int64, io.Reader, io.Writer) error) error {
if arch != "" && !c.apiAtLeast(ctx, APIVersionV2ArchTags) {
c.Logger.Log("This library does not support architecture specific tags")
c.Logger.Log("The image returned may not be the requested architecture")
}
if strings.Contains(path, ":") {
return fmt.Errorf("%w: malformed image path: %s", errBadRequest, path)
}
if tag == "" {
tag = "latest"
}
apiPath := fmt.Sprintf("v1/imagefile/%s:%s", strings.TrimPrefix(path, "/"), tag)
q := url.Values{}
q.Add("arch", arch)
c.Logger.Logf("Pulling from URL: %s", apiPath)
req, err := c.newRequest(ctx, http.MethodGet, apiPath, q.Encode(), nil)
if err != nil {
return err
}
res, err := c.HTTPClient.Do(req)
if err != nil {
return err
}
defer res.Body.Close()
if res.StatusCode == http.StatusNotFound {
return errRequestedImageNotFound
}
if res.StatusCode != http.StatusOK {
err := jsonresp.ReadError(res.Body)
if err != nil {
return fmt.Errorf("download did not succeed: %w", err)
}
if res.StatusCode == http.StatusUnauthorized {
return ErrUnauthorized
}
return fmt.Errorf("%w: unexpected http status code: %d", errHTTP, res.StatusCode)
}
c.Logger.Logf("OK response received, beginning body download")
if callback != nil {
err = callback(res.ContentLength, res.Body, w)
} else {
_, err = io.Copy(w, res.Body)
}
if err != nil {
return err
}
c.Logger.Logf("Download complete")
return nil
}
// Downloader defines concurrency (# of requests) and part size for download operation.
type Downloader struct {
// Concurrency defines concurrency for multi-part downloads.
Concurrency uint
// PartSize specifies size of part for multi-part downloads. Default is 5 MiB.
PartSize int64
// BufferSize specifies buffer size used for multi-part downloader routine.
// Default is 32 KiB.
// Deprecated: this value will be ignored. It is retained for backwards compatibility.
BufferSize int64
}
// NoopProgressBar implements ProgressBarInterface to allow disabling the progress bar
type NoopProgressBar struct{}
// Init is a no-op
func (*NoopProgressBar) Init(int64) {}
// ProxyReader is a no-op
func (*NoopProgressBar) ProxyReader(r io.Reader) io.ReadCloser { return io.NopCloser(r) }
// IncrBy is a no-op
func (*NoopProgressBar) IncrBy(int) {}
// Abort is a no-op
func (*NoopProgressBar) Abort(bool) {}
// Wait is a no-op
func (*NoopProgressBar) Wait() {}
// ProgressBar provides a minimal interface for interacting with a progress bar.
// Init is called prior to concurrent download operation.
type ProgressBar interface {
// Initialize progress bar. Argument is size of file to set progress bar limit.
Init(int64)
// ProxyReader wraps r with metrics required for progress tracking. Only useful for
// single stream downloads.
ProxyReader(io.Reader) io.ReadCloser
// IncrBy increments the progress bar. It is called after each concurrent
// buffer transfer.
IncrBy(int)
// Abort terminates the progress bar.
Abort(bool)
// Wait waits for the progress bar to complete.
Wait()
}
// ConcurrentDownloadImage implements a multi-part (concurrent) downloader for
// Cloud Library images. spec is used to define transfer parameters. pb is an
// optional progress bar interface. If pb is nil, NoopProgressBar is used.
//
// The downloader will handle source files of all sizes and is not limited to
// only files larger than Downloader.PartSize. It will automatically adjust the
// concurrency for source files that do not meet minimum size for multi-part
// downloads.
func (c *Client) ConcurrentDownloadImage(ctx context.Context, dst *os.File, arch, path, tag string, spec *Downloader, pb ProgressBar) error {
if pb == nil {
pb = &NoopProgressBar{}
}
if strings.Contains(path, ":") {
return fmt.Errorf("%w: malformed image path: %s", errBadRequest, path)
}
name := strings.TrimPrefix(path, "/")
if tag == "" {
tag = "latest"
}
// Check for direct OCI registry access
if err := c.ociDownloadImage(ctx, arch, name, tag, dst, spec, pb); err != nil {
if !errors.Is(err, errOCIDownloadNotSupported) {
// Return OCI download error or fallback to legacy download
return err
}
c.Logger.Log("Fallback to (legacy) library download")
return c.legacyDownloadImage(ctx, arch, name, tag, dst, spec, pb)
}
return nil
}
func (c *Client) legacyDownloadImage(ctx context.Context, arch, name, tag string, dst io.WriterAt, spec *Downloader, pb ProgressBar) error {
if arch != "" && !c.apiAtLeast(ctx, APIVersionV2ArchTags) {
c.Logger.Log("This library does not support architecture specific tags")
c.Logger.Log("The image returned may not be the requested architecture")
}
apiPath := fmt.Sprintf("v1/imagefile/%v:%v", name, tag)
q := url.Values{}
q.Add("arch", arch)
c.Logger.Logf("Pulling from URL: %s", apiPath)
customHTTPClient := &http.Client{
Transport: c.HTTPClient.Transport,
CheckRedirect: func(req *http.Request, via []*http.Request) error {
if req.Response.StatusCode == http.StatusSeeOther {
return http.ErrUseLastResponse
}
maxRedir := 10
if len(via) >= maxRedir {
return fmt.Errorf("%w: stopped after %d redirects", errHTTP, maxRedir)
}
return nil
},
Jar: c.HTTPClient.Jar,
Timeout: c.HTTPClient.Timeout,
}
req, err := c.newRequest(ctx, http.MethodGet, apiPath, q.Encode(), nil)
if err != nil {
return err
}
res, err := customHTTPClient.Do(req)
if err != nil {
return err
}
defer res.Body.Close()
if res.StatusCode == http.StatusNotFound {
return errRequestedImageNotFound
}
if res.StatusCode == http.StatusOK {
// Library endpoint does not provide HTTP redirection response, treat as single stream download
c.Logger.Log("Library endpoint does not support concurrent downloads; reverting to single stream")
size, err := parseContentLengthHeader(res.Header.Get("Content-Length"))
if err != nil {
return err
}
return c.download(ctx, dst, res.Body, size, pb)
}
if res.StatusCode != http.StatusSeeOther {
if res.StatusCode == http.StatusUnauthorized {
return ErrUnauthorized
}
return fmt.Errorf("%w: unexpected http status %d", errHTTP, res.StatusCode)
}
// Get image metadata to determine image size
img, err := c.GetImage(ctx, arch, fmt.Sprintf("%v:%v", name, tag))
if err != nil {
return err
}
redirectURL, err := url.Parse(res.Header.Get("Location"))
if err != nil {
return err
}
var creds credentials
if c.AuthToken != "" && samehost(c.BaseURL, redirectURL) {
// Only include credentials if redirected to same host as base URL
creds = bearerTokenCredentials{authToken: c.AuthToken}
}
// Use redirect URL to download artifact
return c.multipartDownload(ctx, redirectURL.String(), creds, dst, img.Size, spec, pb)
}
// samehost returns true if host1 and host2 are, in fact, the same host by
// comparing scheme (https == https) and host, including port.
//
// Hosts will be treated as dissimilar if one host includes domain suffix
// and the other does not, even if the host names match.
func samehost(host1, host2 *url.URL) bool {
return strings.EqualFold(host1.Scheme, host2.Scheme) && strings.EqualFold(host1.Host, host2.Host)
}
func parseContentLengthHeader(val string) (int64, error) {
if val == "" {
return int64(-1), nil
}
size, err := strconv.ParseInt(val, 10, 64)
if err != nil {
return -1, fmt.Errorf("parsing Content-Length header %v: %w", val, err)
}
return size, nil
}
// download implements a simple, single stream downloader
func (c *Client) download(_ context.Context, w io.WriterAt, r io.Reader, size int64, pb ProgressBar) error {
pb.Init(size)
defer pb.Wait()
proxyReader := pb.ProxyReader(r)
defer proxyReader.Close()
written, err := io.Copy(&filePartDescriptor{start: 0, end: size - 1, w: w}, proxyReader)
if err != nil {
pb.Abort(true)
return err
}
c.Logger.Logf("Downloaded %v byte(s)", written)
return nil
}
|