I've been trying to come up with a good way of pushing virus updates to others on my network. We have Linux servers, and Windows workstations, and they only talk to each other via an intermediary server. The Windows workstations are otherwise not connected to the internet.

My goal is to say "wget http://my.antiviruscorp.com/updates/tables*.zip"

I know wget won't support wildcards on HTTP and my AV corp doesn't have an FTP site.

I've been told that curl/libcurl could do what I want, but I'm having the damnest time figuring out how the hell to get it to grab the file I want. I've tried curl'ing the direct link, but that downloads it and posts the binary to my screen, and not a file.

Can anyone give me a heads up on where to begin? The man pages don't seem to illustrate this particular use, nor to the FAQ's on the main curl site.

Otherwise, is there a way to grep a webpage so I can just find out what the full name of the tables file I need to DL is and then just push that name on the end of the wget script as a variable. This would actually be my preferred route because this way I can hack the script to only DL exactly what I want.

Thanks in advance,
Sam

Sam:

Do they have a filename named 'current' that is a link the to current file to download?

You can do a 'wget -O - http://cool.list.com/ | wget --force-html -i - '

Sadly, no. It has a 3 letter prefix which stays constant, and then a numeric suffix which changes as they release new tables. I want to be able to write a script which downloads any files beginning with xxx*.zip. I can't seem to make your script download any of the files, just the http content. Is there something else that needs to be there? Some other variable besides the web URL?

Thanks in advance,
Sam

This should do it, although it will download all files of type zip.

You want to download all the zips from a directory on an HTTP
server. You tried wget http://www.server.com/dir/*.zip, but that
didn't work because HTTP retrieval does not support globbing. In
that case, use:

wget -r -l1 --no-parent -A.zip http://www.server.com/dir/

More verbose, but the effect is the same. -r -l1 means to retrieve
recursively, with maximum depth of 1. --no-parent means that ref-
erences to the parent directory are ignored, and -A.zip means to
download only the GIF files. -A "*.zip" would have worked too.

The best solution would be:

wget -r -l1 --no-parent -A "XXX*.zip" http://www.server.com/dir/

The results would be in a local directory on your box.

Tried your snippet, here's what I get:

Now if I manually go to the directory with the update zip (/ftp/products/pattern/, it's empty, yet when I fill in the explicit zip filename it downloads. There is no index.html. It appears to be an empty folder. Despite the /ftp/ in the address, I cannot seem to get it to talk to the site using an ftp protocol, and their support staff insists that there is no ftp.

I'd appreciate any further insight.

Thanks in advance,
Sam

Sam I can't resolve the address you provided to debug. I tried the IP's listed and they don't work either. If you modified the IP's or name you can contact me directly. At this point, I don't know what else to try. If the site requires you to login then you may need to enter that info into your wget options.

I was doing a little work on the system and needed to refer back to the thread here. Realized I never posted the working solution. This will manually DL antivirus updates from Trend Micro.

Hope it is helpful to someone else.

Sam

works great thanks dehuszar.

shann