>So for a small XML document, it probably makes more sense just to code something by hand to read the XML doc yourself, no? I was thinking that referencing v4 of the MSXML parser would be faster than what is in VFP7.

Doubtful. VFP's internal XML parsing is blazing fast. I'm not sure if they're using MSXML or not internally, but if they are they are low level early binding to it, which will be much faster than anything we can do with the late bound MSXML objects.

Speed is not an issue you need to worry about - it's whether the data can be retrieved in an acceptable format. It may even be faster to use an imperfect CURSORTOXML import and then post process it than doing it manually.


+++ Rick ---

>
>
>>
>>Importing on the other hand must deal with a number of different paths through the XML data and check for all the possibilities of types attributes and document structure in addition to just reading through the document structure. Furthermore VFP's XMLToCursor() often requires two passes through the data to determine the full structure of the table.