Hi all,

I use very simple XML (plain tags and CDATAs, no attributes or so) to pass 
structured text to a search engine. Using the XML enables me to search for 
words in specified fields, defined by the XML tags. The search engine calls 
back on every word it comes across during the indexing process, including the 
tags. I would like to know that a word that is passed to me is a field name, 
but the callback does not provide this information. I was thinking about 
using some characters that are not used in the indexed text. For example, the 
character '^' would never be in the indexed text (within the CDATA). So, if I 
could use a tag named <^fld>, I could check the first letter of each word 
passed by the callback, and if it has the '^' at the beginning it will 
certainly be a field name.

The problem is that I couldn't find any suitable non-alphabetical character 
other that underscore that I can use at the beginning of the tag, and 
underscore is definitively a valid value character (the indexed text is only 
letter, numbers, '_' and '@'). What I'm looking for are a few characters that 
are not letters, and can be used in a tag. The characters can be in Unicode.

(BTW, I've tried looking at the standard, but I'm not familiar with the 
terms as I hardly work with XML, and I've figured any XML expert could easily 
answer my question)

TIA,
Kal