On Wed, 24 Aug 2005 14:30:13 -0400, "Lars Nilsson"
<chamaeleon_at_gmail.com> said:
> On 8/24/05, leegold <leegold_at_fastmail.fm> wrote:
> > Tried posting in the PHP list but no answer - please forgive my cross
> > posting.
> > Would appreciate a link or brief explaination for example showing the
> > idea to do the following:
> >
> > Wondered if I could get a simple newbie example for the code to do the
> > following, I'll be using PHP:
> >
> > I want to go to a website, and then, on that site click a link. The
> > link will take me to a new page which shows a jpeg graphic. I then want
> > to to download (via http protocol) this jpeg to a folder on my local
> > hard drive.
> >
> > If you give me a general case of the code I'll put in the specifics.
> > There's much more, but this would be a start.
> >
> > Lee G.
>
> 1. Get first page
> 2. Parse HTML for first page and extract url for second page
> 3. Get second page based on extracted url
> 4. Parse HTML for second page and extract url for jpeg
> 5. Get jpeg
>
> Now you can just fill in the specifics. ;) libcurl only provides you
> with the tools to do step 1, 3 and 5. It is your responsibility to do
> the things that do not involve the transfer of data from the server to
> your program (which includes figuring out what to do with the HTML
> code and extract any relevant information, etc).

This helps!!!
So steps 2,4 involve regex and text processing - OK excellent.
1,3 are seem well documented on the haxx site.
But 5 - what would be the curl command? I see upload cmds in the man but
didn't see a download cmd...I now see another utility called wget, but
would prefer curl if there's way?

Thanks so much,
Lee G.

The assumption for
> the steps is of course that it is required to retrieve the two pages
> before attempting the jpeg url (due to athentication, cookies, or lack
> of a priori knowledge of the jpeg url, or for any other reason that
> you have not mentioned).
>
> Lars Nilsson
>
Received on 2005-08-24