-
Notifications
You must be signed in to change notification settings - Fork 2
/
gofetch.go
406 lines (338 loc) · 10.2 KB
/
gofetch.go
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
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
// This Source Code Form is subject to the terms of the Mozilla Public
// License, version 2.0. If a copy of the MPL was not distributed with this
// file, You can obtain one at http://mozilla.org/MPL/2.0/.
package gofetch
import (
"crypto/md5"
"crypto/sha1"
"crypto/sha256"
"crypto/sha512"
"fmt"
"hash"
"io"
"net/http"
"os"
"path"
"path/filepath"
"strconv"
"strings"
"sync"
"github.com/hooklift/httpclient"
"github.com/mitchellh/go-homedir"
"github.com/pkg/errors"
)
// ProgressReport represents the current download progress of a given file
type ProgressReport struct {
// Total length in bytes of the file being downloaded
Total int64
// Written bytes to disk on a write by write basis. It does not accumulate.
WrittenBytes int64
}
// Fetcher represents an instance of gofetch, holding global configuration options.
type Fetcher struct {
destDir string
etag bool
concurrency int
algorithm string
checksum string
httpClient *http.Client
}
// Option as explained in http://commandcenter.blogspot.com/2014/01/self-referential-functions-and-design.html
type Option func(*Fetcher)
// WithDestDir allows you to set the destination directory for the downloaded files.
// By default it is set to: ./
func WithDestDir(dir string) Option {
return func(f *Fetcher) {
f.destDir = dir
}
}
// WithConcurrency allows you to set the number of goroutines used to download a specific
// file. By default it is set to 1.
func WithConcurrency(c int) Option {
return func(f *Fetcher) {
f.concurrency = c
}
}
// WithETag enables ETag support, meaning that if an already downloaded file is currently on disk and matches the ETag value returned by the server,
// it will not be downloaded again. By default it is set to false. Be aware that different servers, serving the same file,
// are likely to return different ETag values, causing the file to be re-downloaded, even though it might already exist on disk.
func WithETag() Option {
return func(f *Fetcher) {
f.etag = true
}
}
// WithHTTPClient allows to provide a custom HTTP client. By default a HTTP client with support for read/write timeouts is used.
func WithHTTPClient(c *http.Client) Option {
return func(f *Fetcher) {
f.httpClient = c
}
}
// WithChecksum verifies the file once it is fully downloaded using the provided hash and expected value.
func WithChecksum(alg, value string) Option {
return func(f *Fetcher) {
f.algorithm = alg
f.checksum = value
}
}
var workDir string
func init() {
dir, err := homedir.Dir()
if err != nil {
fmt.Printf(`Unable to get user home directory err=%s\n`, err)
}
workDir = filepath.Join(dir, ".gofetch")
}
// New creates a new instance of goFetch with the given options.
func New(opts ...Option) *Fetcher {
// Creates instance and assigns defaults.
gofetch := &Fetcher{
concurrency: 1,
destDir: "./",
httpClient: httpclient.Default(),
}
for _, opt := range opts {
opt(gofetch)
}
return gofetch
}
// Fetch downloads content from the provided URL. It supports resuming and
// parallelizing downloads while being very memory efficient.
func (gf *Fetcher) Fetch(url string, progressCh chan<- ProgressReport) (*os.File, error) {
if url == "" {
return nil, errors.New("URL is required")
}
// We need to make a preflight request to get the size of the content and check if the server
// supports requesting byte ranges.
res, err := http.Head(url)
if err != nil {
return nil, err
}
if !strings.HasPrefix(res.Status, "2") {
return nil, fmt.Errorf("HTTP requests returned a non 2xx status code: %s", res.Status)
}
if res.Header.Get("Accept-Ranges") != "bytes" {
// Server does not support sending byte ranges, setting concurrency to 1
gf.concurrency = 1
}
fileName := path.Base(url)
destFilePath := filepath.Join(gf.destDir, fileName)
var etag string
if gf.etag {
// Go's stdlib returns header value enclosed in double quotes.
etag = strings.Trim(res.Header.Get("ETag"), `"`)
if etag == "" {
goto FETCH
}
// Create directory if it doesn't exist, we ignore errors if it already exists.
cacheDir := filepath.Join(workDir, fileName)
etagPath := filepath.Join(cacheDir, etag)
os.MkdirAll(cacheDir, 0700)
if _, err := os.Stat(etagPath); err == nil {
// Our file has been already fully downloaded, return a file
// descriptor to it and skip fetching altogether.
fi, err := os.Stat(destFilePath)
if err == nil && fi.Size() == res.ContentLength {
if progressCh != nil {
close(progressCh)
}
return os.Open(destFilePath)
}
} else {
f, err := os.Create(etagPath)
if err != nil {
return nil, err
}
f.Close()
}
}
FETCH:
f, err := gf.parallelFetch(url, destFilePath, res.ContentLength, progressCh)
if err != nil {
return nil, err
}
if gf.algorithm != "" {
if err := gf.verify(f, gf.algorithm, gf.checksum); err != nil {
return nil, errors.Wrap(err, "failed veryfing file integrity")
}
// We need to make sure we return the file descriptor ready to be read by the user again
f.Seek(0, 0)
}
return f, nil
}
func (gf *Fetcher) verify(f *os.File, algorithm string, checksum string) error {
var hasher hash.Hash
switch algorithm {
case "md5":
hasher = md5.New()
case "sha1":
hasher = sha1.New()
case "sha256":
hasher = sha256.New()
case "sha512":
hasher = sha512.New()
default:
return fmt.Errorf("unsupported hashing algorithm: %s", algorithm)
}
// Makes sure file cursor is positioned at the beginning
_, err := f.Seek(0, 0)
if err != nil {
return err
}
_, err = io.Copy(hasher, f)
if err != nil {
return err
}
result := fmt.Sprintf("%x", hasher.Sum(nil))
if result != checksum {
return fmt.Errorf("checksum does not match\n found: %s\n expected: %s", result, checksum)
}
return nil
}
// parallelFetch fetches using multiple goroutines, each piece is streamed down
// to disk which makes it very efficient in terms of memory usage.
func (gf *Fetcher) parallelFetch(url, destFilePath string, length int64, progressCh chan<- ProgressReport) (*os.File, error) {
if progressCh != nil {
defer close(progressCh)
}
var wg sync.WaitGroup
report := ProgressReport{Total: length}
concurrency := int64(gf.concurrency)
chunkSize := length / concurrency
remainingSize := length % concurrency
chunksDir := filepath.Join(gf.destDir, path.Base(url)+".chunks")
if err := os.MkdirAll(chunksDir, 0760); err != nil {
return nil, err
}
var errs []error
for i := int64(0); i < concurrency; i++ {
min := chunkSize * i
max := chunkSize * (i + 1)
if i == (concurrency - 1) {
// Add the remaining bytes in the last request
max += remainingSize
}
wg.Add(1)
go func(min, max int64, chunkNumber int) {
defer wg.Done()
chunkFile := filepath.Join(chunksDir, strconv.Itoa(chunkNumber))
if err := gf.fetch(url, chunkFile, min, max, report, progressCh); err != nil {
fmt.Printf("gofetch: error %#v\n", err)
errs = append(errs, err)
}
}(min, max, int(i))
}
wg.Wait()
if len(errs) > 0 {
return nil, fmt.Errorf("errors: \n %s", errs)
}
file, err := gf.assembleChunks(destFilePath, chunksDir)
if err != nil {
return nil, err
}
os.RemoveAll(chunksDir)
// Makes sure to return the file on the correct offset so it can be
// consumed by users.
_, err = file.Seek(0, 0)
if err != nil {
return nil, err
}
return file, err
}
// assembleChunks join all the data pieces together
func (gf *Fetcher) assembleChunks(destFile, chunksDir string) (*os.File, error) {
file, err := os.Create(destFile)
if err != nil {
return nil, err
}
for i := 0; i < gf.concurrency; i++ {
chunkFile, err := os.Open(filepath.Join(chunksDir, strconv.Itoa(i)))
if err != nil {
return nil, err
}
// Deferring within a loop is not ideal but we expect not too many chunks to exists.
defer chunkFile.Close()
if _, err := io.Copy(file, chunkFile); err != nil {
return nil, err
}
}
return file, nil
}
// fetch downloads files using one unbuffered HTTP connection and supports
// resuming downloads if interrupted.
func (gf *Fetcher) fetch(url, destFile string, min, max int64,
report ProgressReport, progressCh chan<- ProgressReport) error {
req, err := http.NewRequest("GET", url, nil)
if err != nil {
return err
}
// In order to resume previous interrupted downloads we need to open the file
// in append mode.
file, err := os.OpenFile(destFile, os.O_RDWR|os.O_CREATE|os.O_APPEND, 0660)
if err != nil {
return err
}
defer file.Close()
fi, err := file.Stat()
if err != nil {
return err
}
currFileSize := fi.Size()
currChunkSize := (max - min)
// Report bytes written already into the file
if progressCh != nil {
report.WrittenBytes = currFileSize
progressCh <- report
}
// There is nothing to do if chunk data file was fully downloaded.
if currFileSize > 0 && currFileSize == currChunkSize {
return nil
}
// Adjusts min to resume file download from where it was left off.
if currFileSize > 0 {
min = min + currFileSize
}
// Prepares writer to report download progress.
writer := fetchWriter{
Writer: file,
progressCh: progressCh,
progressReport: report,
}
brange := fmt.Sprintf("bytes=%d-%d", min, max-1)
if max == -1 {
brange = fmt.Sprintf("bytes=%d-", min)
}
req.Header.Add("Range", brange)
//fmt.Printf("range %s\n", brange)
res, err := gf.httpClient.Do(req)
if err != nil {
return err
}
defer res.Body.Close()
if !strings.HasPrefix(res.Status, "2") {
return fmt.Errorf("HTTP requests returned a non 2xx status code: %s", res.Status)
}
reader := res.Body.(io.Reader)
if max > 0 {
// Known content-length, so we only read from body the amount of bytes of requested chunk.
reader = io.LimitReader(res.Body, currChunkSize)
}
_, err = io.Copy(&writer, reader)
return err
}
// fetchWriter implements a custom io.Writer so we can send granular
// progress reports when streaming down content.
type fetchWriter struct {
io.Writer
//progressCh is the channel sent by the user to get download updates.
progressCh chan<- ProgressReport
// report is the structure sent through the progress channel.
progressReport ProgressReport
}
func (fw *fetchWriter) Write(b []byte) (int, error) {
n, err := fw.Writer.Write(b)
if fw.progressCh != nil {
fw.progressReport.WrittenBytes = int64(n)
fw.progressCh <- fw.progressReport
}
return n, err
}