Have done a few demos with this. As there is more than one available I hope I used the best one (masa haru 9 dec 2010). It would be useful for those not familiar with the IDE to have a good way of choosing a library as essential as this.

First big problem is the use of fixed length buffers. If you are scraping something out of a large web page you need huge buffers. In this environment I would something that is more economical with this scarce resource. It would also be better if you did not need to know the size of the page you are trying to access. The kind of interface you have in higher level languages (e.g. get a line at a time) would be useful.

The second is apparent lack of a destructor. I would like to make sure the whole thing is closed down and resources released. One of my tests does continuous polling and currently dies after a few iterations. This is probably my code but it would be nice to know that repeated calls do not produce any kind of leak. I will do some more tests on this.

Having said all the above, the facility to see code and doc for libraries at least makes it possible to address the above if you really need to.

If you want to get large web pages, you can alway use th HTTPStream. And, AFAIK, the HTTPClient automatically closes its resources.