Short while back I published a tutorial on how to easily find and extract text blocks within HTML. Time has come to evolve. We need to dig deeper and get the sentences out of the text chunks.
Had greatly improved speed based on (PR) Papa Rage's comment in last performance bottleneck post but had to go further as DNS resolving functions in Windows are as slow as it gets.
See what happened, in terms of speed, after rewriting the DNS resolving functions with a twist.
