This is probably the simplest way of doing your specific task, but there are of course other ways. The tool Xtract has a parser for XPath-like expressions, including more complex predicates, which get translated directly to combinator-style accessors. Unfortunately, the set of combinators used in Xtract is slightly different from those in the rest of HaXml, but they bear a close relationship.
Before i start, I apologise for the simplicity of these questions, here goes: I am just trying out a few things with the Xtract tool. I have a simple XML document: <greetings> <greeting>hello</greeting> <greeting>world</greeting> </greetings> I am using the Xtract query /greetings/greeting[1]/- hoping to get the text "world" e.g. mkElem "h1" [applypred (keep) (parseXtract "/greetings/greeting[1]/-")] but this just returns the entire xml document. Have i got something fundamentally wrong here? Thanks Tom Spencer