How to parse a complex HTML page in Zorro?

It just looks very difficult in c-lite, what would be my best option?
I bacsically have a file downloaded from a website (with http_transfer) and would like to find values in those file which I would use in indicators.

The usual way for parsing numbers out of a website is identifying the text in the HTM source immediately before the number you want. Then do a text search with strstr, and get the subsequent number with atoi or atof.