[Ann] MTML Pull-Parser Data Table

[Mark Brownell]

Concerning something that Wouter found:

On Tuesday, March 23, 2004, at 06:54  AM, Wouter wrote:

> Hi,
>
> I hope you don't mind the intrusion as I didn't want to put this on  
> the list.
> At:
>
> http://www.mail-archive.com/use-revolution@lists.runrev.com/ 
> msg27362.html
>
> you gave these very nice handlers in a reply on Thomas McGrath III.
> They seem to be some of your basic handlers, copied and pasted into  
> this mail to the list.
> That  is why I wanted to point you to:
>
> function PNLPgetAttribute tAttribute, strngToSearch
>
> because that one contains a couple of errors.
> But if you corrected this already (or you did put them intentionally  
> in :-) then consider this mail as nonexistent.
> Any way thank you very much for these otherwise "very" nice handlers.
>
> Greetings,
> Wouter

It would not surprise me if I pasted a broken version of this into an  
e-mail/list. More than likely its problems stem from using the tZap  
property in the PNLPgetAttribute example:

-- put PNLPgetAttribute("name", tZap) into theAttribute
function PNLPgetAttribute tAttribute, strngToSearch

strngToSearch should be parsed from an element tag set ("<bob",  
"</bob>", tZap) first in order to parse that element's attributes.  
(Note: "<bob" is missing the ">" on the start tag. This is to allow for  
the attributes to follow.) In other words it's only a pull-parser for  
attributes after it is separated out from its element as the new  
strngToSearch. This is probably another case of bad explanations or  
comments in my code. On the other hand it could be broken. I just put  
up an example using the element version:
-- put PNLPgetElement("<record>", "</record>", tZap) into theElement
function PNLPgetElement tStTag, tEdTag, stngToSch
to make an XML table. I should have explained that parsing attributes  
is a two step process.

Thanks for showing me this Wouter.

Mark