Created
November 14, 2013 22:53
-
-
Save howeyc/7475785 to your computer and use it in GitHub Desktop.
Download an s3 bucket.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
// Download a public s3 bucket. | |
// | |
// Relies on the directory listing being public. | |
// | |
// Example usage: | |
// s3dl -bucket <bucketName> | |
// | |
// -> Will download all files in the "<bucketName>" s3 bucket | |
// to "<bucketName>" folder. | |
package main | |
import ( | |
"crypto/md5" | |
"encoding/xml" | |
"flag" | |
"fmt" | |
"io" | |
"log" | |
"net/http" | |
"net/url" | |
"os" | |
"path/filepath" | |
"strconv" | |
"strings" | |
) | |
type Content struct { | |
Key string | |
Md5Sum string `xml:"ETag"` | |
Size int64 | |
} | |
type Result struct { | |
Contents []Content | |
IsTruncated bool | |
} | |
func main() { | |
var bucketName string | |
flag.StringVar(&bucketName, "bucket", "", "bucket to download") | |
flag.Parse() | |
bucketUrl := url.URL{Scheme: "https", Host: bucketName + ".s3.amazonaws.com"} | |
// Request for list of files | |
// (Possibly requiring multiple requests if bucket has lots of items) | |
var result Result | |
for { | |
reqUrl := bucketUrl | |
// If we already recieved a response, | |
// continue from the end of list already received. | |
if len(result.Contents) > 0 { | |
reqValues := reqUrl.Query() | |
reqValues.Set("marker", result.Contents[len(result.Contents)-1].Key) | |
reqUrl.RawQuery = reqValues.Encode() | |
} | |
res, err := http.Get(reqUrl.String()) | |
if err != nil { | |
log.Fatal(err) | |
} | |
xmlDecode := xml.NewDecoder(res.Body) | |
xmlDecode.Decode(&result) | |
res.Body.Close() | |
// Received last of the list | |
if !result.IsTruncated { | |
break | |
} | |
} | |
// Download all the files to an output directory with bucket name | |
downloadContents(bucketUrl, bucketName, result.Contents) | |
} | |
func downloadContents(bucketUrl url.URL, bucketName string, contents []Content) { | |
sum := md5.New() | |
for _, content := range contents { | |
fmt.Print("Downloading ", content.Key, "... ") | |
dir, file := filepath.Split(content.Key) | |
outDir := filepath.Join(bucketName, dir) | |
os.MkdirAll(outDir, 0777) | |
if file != "" { | |
fileUrl := bucketUrl | |
fileUrl.Path = "/" + content.Key | |
resFile, resErr := http.Get(fileUrl.String()) | |
if resErr == nil { | |
lfile, ecreate := os.Create(filepath.Join(outDir, file)) | |
if ecreate != nil { | |
fmt.Println(ecreate) | |
continue | |
} | |
sum.Reset() | |
ofile := io.MultiWriter(lfile, sum) | |
io.Copy(ofile, resFile.Body) | |
resFile.Body.Close() | |
lfile.Close() | |
// Unquote ETag field | |
content.Md5Sum, _ = strconv.Unquote(content.Md5Sum) // Remove Quotes | |
// Look for (and convert) block-size type of Md5 | |
// Assumes only one block if finds this type of md5. | |
if dashIdx := strings.LastIndex(content.Md5Sum, "-"); dashIdx > 0 { | |
md5sum := sum.Sum(nil) | |
sum.Reset() | |
sum.Write(md5sum) | |
content.Md5Sum = content.Md5Sum[:dashIdx] | |
} | |
// Check Size and Md5 | |
if content.Size != resFile.ContentLength { | |
fmt.Print("Invalid Size!") | |
} else if content.Md5Sum != fmt.Sprintf("%x", sum.Sum(nil)) { | |
fmt.Print("Invalid Md5Sum!") | |
} else { | |
fmt.Print("Complete.") | |
} | |
fmt.Println() | |
} else { | |
fmt.Print(resErr) | |
} | |
} | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment