Make an arrays of links contained in one big string - image

Question

I have a big string ( a html code from web page).
Now the problem is how to parse the links to images.
I want to make an array of all the links to images in that web page.
I know how to do this i java but I do not know how to do the parse strings and do a string manipulations in shell. I know there are many tricks and I guess this can be easy done.
in the end I want to get something like this
#!/bin/bash
read BIG_STRING <<< $(curl some_web_page_with_links_to_images.com)
#parse the big string and fill the LINKS variable
# fill this with the links to image somewhow (.jpg and .png only)
#after the parsing the LINKS should look like this
LINKS=("www.asd.com/asd1.jpg" "www.asd.com/asd.jpg" "www.asd.com/asd2123.jpg")
#I need the parsing and to fill the LINKS variable with the links from the web page
# get length of an array
tLen=${#LINKS[#]}
for (( i=0; i<${tLen}; i++ ));
do
echo ${LINKS[$i]}
done
Thanks, for the responses you saved me days of frustrations

Answer 1

Why not start with the right tool? Parsing HTML is hard, especially with sed. If you have the mojo tool from the Mojolicious project you can do this:
mojo get http://example.com a attr href
And then just check whether each line ends with jpg, png, or whatever.

Answer 2

It's hard to offer more than approximations. Let's assume all interesting links are href="" attributes, and there's at most one href attribute per line (And the links are also one line only, actually I'm not sure if newlines are allowed inside URLs.
Let's assume your sourcefile is called test.html.
The following should print all links under these assumptions:
sed -n 's/.*\<href="\([^"]*\)".*/\1/p' test.html
To understand how this works, you should know what regular expressions are and have read up a tutorial on sed (particularly how the s ubstitute command works)