If you had a page that followed the W3C standards, couldn't you grab the page with LWP::Simple, store that to a file, get the size, and use HTML::TokeParser to parse out the image sizes?

Update: After thinking about this on a long drive today ( and checking the W3C website to make sure ) I realized that all you're going to be able to parse are the alloted pixel sizes of a web page. I beleive that other monks in the thread have pointed out ways to get image sizes.

You might be able to determine a rough size without having to download all the images. If you download the HTML page, you could then get a HEAD for each image it has, and add up all the reported sizes. Of course, this relies on the web server reporting a proper size, and I don't know how reliable this is, but it seems like it should be pretty good. (Doesn't HTTP use the Content-Length to determine how much to download? If so, then the Content-Length would probably be pretty accurate.)

HTH

Update: after re-reading, I'm not sure if I really answered your questions at all, but this is a thought I had. Perhaps it will at least give you some ideas...

Using wget might be much easier in this situation. It has a bunch of switches to download all the graphics related to the page. Curl and ht-track also do the same thing, but I am not familiar with them.

---

See node 362257 for a more in-depth view on this topic and a couple of good hints on the possible complexities involved.

The code may be a bit old and crufty but I think my column on calculating download time may be right up your alley.

I wrote HTTP::Size to do just this. From the docs:

get_sizes( URL, BASE_URL )

The get_sizes function is like get_size, although for HTML pages it also fetches all of the images then sums the sizes of the original page and image sizes. It returns a total download size. In list context it returns the total download size and a hash reference whose keys are the URLs of the images found in the HTML and whose values are hash references with these keys: