Using wildcards in wget or curl query

2019-01-10 19:44发布

Is it possible to use wildcards in wget queries when downloading from directories? Basically, I have a site, say, www.download.example.com/dir/version/package.rpm. Howevery, the version directory changes all the time and sometimes contains multiple RPM packages. Is there a single wget query that would get me only and all the packages in the version directory without knowing what that version is? In my ideal world, the query would be something like:

wget www.download.example.com/dir/*/*.rpm

If there is a way to do this with curl, that would work as well.

3条回答
2楼-- · 2019-01-10 19:50

If you are able to found a pattern in your query, you can use the bash brace expansion to do this task.

For example, in your case, you may use something like:

wget www.download.example.com/dir/{version,old}/package{00..99}.rpm

Also, you may combine this with the -A and -R parameters to filter your results.

查看更多
Lonely孤独者°
3楼-- · 2019-01-10 19:54

You can't use wildcards in wget but the -A flag should work. From the wget manpage:

You want to download all the gifs from a directory on an http server. You tried wget http://www.server.com/dir/*.gif, but that didn't work because http retrieval does not support globbing. In that case, use: wget -r -l1 --no-parent -A.gif http://www.server.com/dir/

Edit: found a related question

Regarding directories:

There's a utility called LFTP, which has some support for globbing. Take a look at the manpage. There's another question on Linux & Unix that covers its usage in a scenario similar to yours.

查看更多
闹够了就滚
4楼-- · 2019-01-10 20:01

Although the above solution kind of works, it fails when you just want to download certain directories, but not all. For example if you have:

http://site.io/like/
http://site.io/like2/
http://site.io/nolike/

Instead put the directory names you want in a text file, e.g.: dirs.txt:

like/
like2/

Then use wget with the following command options -i dirs.txt -B <base-URL> like so:

wget -nH -nc -np -r -e robots=off -R "index.html*" -i dirs.txt -B http://site.io/

Since, I don't think you can use directories in the -A and -R lists. (?)

查看更多
登录 后发表回答