Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/c4milo/gofetch
Go library for downloading content from internet, in parallel.
https://github.com/c4milo/gofetch
downloader etags fetcher gofetch golang golang-library http parallel
Last synced: about 1 month ago
JSON representation
Go library for downloading content from internet, in parallel.
- Host: GitHub
- URL: https://github.com/c4milo/gofetch
- Owner: c4milo
- License: mpl-2.0
- Created: 2015-10-02T14:09:45.000Z (about 9 years ago)
- Default Branch: master
- Last Pushed: 2017-07-04T18:05:41.000Z (over 7 years ago)
- Last Synced: 2024-06-20T17:35:00.632Z (5 months ago)
- Topics: downloader, etags, fetcher, gofetch, golang, golang-library, http, parallel
- Language: Go
- Homepage:
- Size: 10 MB
- Stars: 27
- Watchers: 4
- Forks: 2
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# Gofetch
[](https://travis-ci.org/c4milo/gofetch)
[](https://godoc.org/github.com/c4milo/gofetch)
Go library to download files from the internerds using Go 1.7 or greater.
## Features
* Resumes downloads if interrupted.
* Allows parallel downloading of a single file by requesting multiple data chunks at once over HTTP.
* Reports download progress through a Go channel if indicated to do so.
* Supports file integrity verification if a checksum is provided.
* Supports ETags, skipping downloading a file if it hasn't changed on the server.
* Can be combined with https://github.com/cenkalti/backoff to support retrying with exponential back-off
## Gotchas
When downloading file chunks concurrently, you may encounter some issues:
* Servers may limit the number of concurrent connections you have open against them
* Servers may accept the connections but will not send anything, in this case the default HTTP client will timeout for you.
If you provide your own client, make sure it has proper timeouts or your connection will block for several seconds, depending
on your operating system network timeouts.
## Example
```go
package main
import (
"fmt"
"io"
"os"
"github.com/c4milo/gofetch"
)
func main() {
gf := gofetch.New(
gofetch.WithDestDir(os.TempDir()),
gofetch.WithConcurrency(10),
gofetch.WithETag(),
)
progressCh := make(chan gofetch.ProgressReport)
var myFile *os.File
go func() {
var err error
myFile, err = gf.Fetch(
"http://releases.ubuntu.com/15.10/ubuntu-15.10-server-amd64.iso",
progressCh)
if err != nil {
panic(err)
}
}()
// pogressCh is close by gofetch once a download finishes
var totalWritten int64
for p := range progressCh {
// p.WrittenBytes does not accumulate, it represents the chunk size written
// in the current operation.
totalWritten += p.WrittenBytes
fmt.Printf("%d of %d\n", totalWritten, p.Total)
}
destFile, err := os.Create("/tmp/ubuntu-15.10-server-amd64.iso")
if err != nil {
panic(err)
}
defer func() {
if err := destFile.Close(); err != nil {
panic(err)
}
}()
if _, err := io.Copy(destFile, myFile); err != nil {
panic(err)
}
}
```