Worked like a charm!  Decided to go with ^ as my delimeter character since
it is fairly uncommonly used in the real world.  Thanks!

-----Original Message-----
From: Bill Moseley [mailto:moseley@hank.org]
Sent: Thursday, April 24, 2003 5:48 PM
To: Multiple recipients of list
Subject: [SWISH-E] Re: how to handle multiple occurrences of an xml
field?


On Thu, 24 Apr 2003, Steven Moy wrote:

> 
> 
> If I have documents with multiple fields of the same name, e.g. the field
> named <client> in an example like this:
> <salesperson>Mr. Foo</salesperson>
> <company>XYZ Corp.</company>
> <client>John Doe</client>
> <client>Jane Doe</client>
> <client>John Smith</client>
> 
> If I then choose to display the field contents in the search results ( -x
> "<salesperson>|<client>|<company>|") it displays the clients all
> concatenated together:
> 
> Mr.Foo|John Doe Jane Doe John Smith|XYZ Corp.
> 
> 
> Is there any way to get at the clients individually?  Or maybe at least
get
> a separator between the clients?  For example "John Doe#JaneDoe#John
Smith"
> would be great.  Then I could just parse that string.

Do you mind recompiling?

in src/docprop.c search for:

    if ( add_a_space )
        p->propValue[p->propLen++] = ' ';



-- 
Bill Moseley moseley@hank.org