Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!caen!news
From: amber@engin.umich.edu (Lee Liming)
Newsgroups: comp.windows.x.motif
Subject: OSF 1.1.2 XmStringLtoRCreate()
Message-ID: <6JC-!J=@engin.umich.edu>
Date: 14 Jun 91 03:04:27 GMT
Organization: University of Michigan (CAEN)
Lines: 29

We've recently upgraded our OSF/Motif from 1.1.1 to 1.1.2.  One of the bug 
fixes listed for this version is "pir 2652", which supposedly fixes a problem 
with "extra" segments with zero-length text ("") being added to compound 
strings created using the XmStringLtoRCreate() function.

After recompiling a home-brew application with the new version, I discovered 
that the new XmStringLtoRCreate() will consider ANY zero-length text segment to
be "extra", including those embedded between non-zero-length segments in the 
argument passed to it.  It consequently throws them all away.

Our revision 1.1 OSF/Motif Programmer's Guide says nothing about this function
dropping empty text segments from the compound strings it creates.  The ability
to have zero-length segments was useful for creating blank lines in the text
displayed in XmLabel widgets, for example, perhaps in cases where you don't
know at compile-time what text will be displayed.

Was this really the intention behind this bug fix, or was it overkill?  Is 
there something obvious (or a "greater purpose") that I'm overlooking?  Maybe
compound strings weren't meant to be used this way?

For now, I'm bypassing this "feature" by simply replacing zero-length segments
with segments containing a space.  It seems like this is a hack, though, and 
shouldn't be necessary.  What's wrong with zero-length segments in compound
strings?
--
-------------------------------------------------------------------------------
                                                                     Lee Liming
                                             Computer Aided Engineering Network
                                                     The University of Michigan