I have done plenty of regex parsing of xml with Perl. It has been very useful. Over time I have also used things like the index function to eck out some addit

I have done plenty of regex parsing of xml with Perl. It has been very useful. Over time I have also used things like the index function to eck out some additional performance. Recently I replaced this with a xml tokenizer I wrote in Go that can deal with invalid or corrupt xml. On top of this I have used a state machine to make it possible to handle different situations.