How do I pipe a downloaded file to standard output in bash?

Question

How is it possible to pipe out wget's downloaded file? If not what alternatives should I use?

score 54 · Accepted Answer · answered Jun 15 '09 at 11:01

54

wget -O - -o /dev/null  http://google.com

answered Jun 15 '09 at 11:01

pQd

30,179
6
66
110

13

The -o /dev/null is only necessary if you truly don't care about errors, since without that errors will be written to stderr (while the file is written to stdout). – Greg Hewgill Jun 15 '09 at 11:03
right. or you can check error codes... as long as you pipe output to a file not another command. – pQd Jun 15 '09 at 11:23
AFAIK, no need to -o /dev/null because that goes to stderr anyway. – Milan Babuškov Jun 15 '09 at 11:27
12

You can use wget -qO- $URL to simplify things. – Oli Jul 05 '11 at 12:04
1

wget -qO- $URL works if you're using Wget on Windows – Chris S Jan 27 '12 at 12:29

score 38 · Answer 2 · answered Jun 15 '09 at 11:13

38

Or use curl, where it's the default behaviour.

curl http://www.google.com/

http://curl.haxx.se/

answered Jun 15 '09 at 11:13

GodEater

560

4

-s won't hurt to make it quiet, particularly with a small file. – Asclepius Oct 06 '14 at 22:54

score 6 · Answer 3 · answered Jun 15 '09 at 11:54

6

There are other methods you can use instead of wget and curl:

You can use lynx:

# lynx -source http://www.google.com

w3m:

# w3m -dump_source http://www.google.com

and libwww-perl comes with a handy program called GET (as well has HEAD and POST, which do what you think they do)

# GET http://www.google.com

answered Jun 15 '09 at 11:54

David Pashley

23,637

1

Offtopic, but I've used lynx in some of my scripts to parse html for me automatically whenever I've needed the content of a page and didn't care about the markup. It's great for that – Matt Simmons Jun 15 '09 at 12:46
indeed, both lynx and w3m have a -dump option. I prefer w3m for its table and frame support. – David Pashley Jun 15 '09 at 14:00

score 1 · Answer 4 · answered Aug 20 '11 at 14:41

1

This is how I did it:

URL='http://wordpress.org/extend/plugins/akismet/'
curl -s "$URL" | egrep -o "http://downloads.wordpress.org/plugin/[^']+" | xargs wget -qO-

answered Aug 20 '11 at 14:41

Roger

473

score 0 · Answer 5 · answered Jun 15 '09 at 14:10

0

Just to add another option: I often use lwp-request, from libwww-perl, for this. It outputs to STDOUT by default and is more likely than curl to be installed on the systems I use (your situation my vary).

answered Jun 15 '09 at 14:10

Jeff Tang

151

1

FWIW, Mac OS has curl (not wget) and I believe other BSD's do as well. As do a lot of embedded *nix systems I've used. Not sure if Perl is more common than curl. – Wyatt Ward Feb 02 '16 at 21:18

score 0 · Answer 6 · answered Jan 28 '16 at 15:37

0

I suggest to use Aria2. It's powerful downloader.

aria2c http://google.com.tw

answered Jan 28 '16 at 15:37

Phil Huang

11

Does it output to standard output (the crux of the question)? – Peter Mortensen Aug 14 '21 at 18:13

How do I pipe a downloaded file to standard output in bash?

6 Answers6