On Thu, 10 May 2007 16:04:41 -0500 (CDT)
Steven M. Schweda wrote:
> From: R Kimber
>
> > Yes there's a web page. I usually know what I want.
>
> There's a difference between knowing what you want and being able
> to describe what you want so that it makes sense to someone who does
> not know what you want.
Well I was wondering if wget had a way of allowing me to specify it.
> > But won't a recursive get get more than just those files? Indeed,
> > won't it get everything at that level? The accept/reject options
> > seem to assume you know what's there and can list them to exclude
> > them. I only know what I want. [...]
>
> Are you trying to say that you have a list of URLs, and would like
> to use one wget command for all instead of one wget command per URL?
> Around here:
>
> ALP $ wget -h
> GNU Wget 1.10.2c, a non-interactive network retriever.
> Usage: alp$dka0:[utility]wget.exe;13 [OPTION]... [URL]...
> [...]
>
> That "[URL]..." was supposed to suggest that you can supply more than
> one URL on the command line. Subject to possible command-line length
> limitations, this should allow any number of URLs to be specified at
> once.
>
> There's also "-1" ("--input-file=FILE"). No bets, but it looks as
> if you can specify "-" for FILE, and it'll read the URLs from stdin,
> so you could pipe them in from anything.
Thanks, but my point is I don't know the full URL, just the pattern.
What I'm trying to download is what I might express as:
http://www.stirling.gov.uk/*.pdf
but I guess that's not possible. I just wondered if it was possible
for wget to filter out everything except *.pdf - i.e. wget would look
at a site, or a directory on a site, and just accept those files that
match a pattern.
- Richard
--
Richard Kimber
http://www.psr.keele.ac.uk/
