Date:         Thu, 19 Jan 2006 04:45:50 -0800
Reply-To:     mjessup@YAHOO.COM
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         mjessup@YAHOO.COM
Organization: http://groups.google.com
Subject:      ODS HTML split character and non-breaking spaces
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset="iso-8859-1"

I have been working with tagsets and recently come across an odd
behavior. If a split character appears as the last character for a
column header text and is the only split character the ODS engine seems
to replace all of the spaces with non-breaking spaces (&nbsp;). For
example in the following code:

proc report data=mydata NOWINDOWS SPLIT="^";;
 column (
        "TEST" cfld
        ("Group 1"
                ("cfld1" cfld1)
                ("cfld2" cfld2)
        )
        ("Upper^Group"
                ("Group^2^"
                        ("nfld1" nfld1)
                )
                ("Group 3^"
                        ("cfld3 ^" cfld3)
                        ("nfld2" nfld2)
                )
        )
);

The text will be translated as follows:
Upper^Group = Upper<br>Group
Group^2^ = Group<br>2
Group 3^" = Group&nbsp;3
cfld3 ^ = cfld3&nbsp;

Has anyone else encountered this problem, and does anyone know how to
stop this automatic substitution. I have been able to reproduce this
both in a custom tagset based on the CHTML tagset and with the default
HTML. I am using 9.1.3 and have tested both on a windows and solaris
machine.