If there is ANY way to do this other than scraping a web site, you should use it. Many sites will provide APIs that expose the data in a more stable, manageable way, and if one is available, that will be a decidedly superior way to go about it.

The big issue with scraping a web page is that it will stop working sooner or later, and usually sooner. Web mavens can't resist changing around the designs of their pages, and whenever they make a change, your code will break. If the site is quite active, those changes could happen almost daily, though they will likely be less frequent than that.