On 12/12/11 02:42, Malcolm Wallace wrote:
The extra parameter "i" is for "information" attached to each node of the tree. As you have correctly guessed, the parser fills in this field with positional information relating to the original source document, which is useful for instance if you are validating or checking the original document. When building new parts of a document, it is perfectly fine to attach "noPos".
You can alternatively replace all of the informational items in the tree, with for instance "fmap (const ())" if you don't care about them.
The information fields are useful for other purposes though, e.g. to hold the relevant xmlns namespace for subtrees; or to distinguish added/removed/modified subtrees in a diff-like viewer.
Regards, Malcolm
Thanks, I understand what's going on now. I like the noPos solution because the other ones make the reader think there's something important going on; with noPos, I can just comment "gotta give it a Posn, and this is the only one I know how to make."