[omniORB] Putting a Sequence into an Any and marshalling it

[Gary D. Duzan]

In Message <35BC3A42.255E9A6C@eps.agfa.be> ,
   "Gert Steenssens" <gsteenss@eps.agfa.be> wrote:

=>Gary D. Duzan wrote:
=>> 
=>> In Message <35B8A385.75FC9306@eps.agfa.be> ,
=>>    "Gert Steenssens" <gsteenss@eps.agfa.be> wrote:
=>> 
=>> =>I've encountered a little problem while trying to put an 
=>> Unbounded Sequence into an Any 
=>
=>> This sounds like the problem I reported earlier. Check out bug/patch
=>> #12 on the OmniORB News Page, or
=>> < http://www.orl.co.uk/omniORB/bugs/12.html >. Hopefully that will
=>> clear up the problem.
=>
=>
=>Well actually, I'm not sure it is...
=>I looked at the suggested patch, and it seems they want me to replace 
=>'this->length()' by 'member_tc->length()' in typeCode.cc, but the strange thing
=>is; line 663 already has 'member_tc->length()' in it. (or maybe the patch is the
=>otherway around ?)

   Yes, it should be the other way around. The code in question is
trying to get rid of aliases for sequences. 'member_tc->length()'
doesn't make sense since the the sequence member type may or may not
have a length, What it really needed is the length of the sequence,
which is 'this->length()' here.

=>I must say I downloaded the 2.5.0 ORB only last week, so probably it was already
=>patched for this bug ?
=>As far as I can tell either way the same thing happens with or without patch
=>applied...

   If you are using shared libraries, did you make sure the old ones
were flushed out of memory before trying the new ones? That has bitten
me a few times.
   Even if this patch doesn't address the root of your problem, it is
a good idea to apply it anyway as it will likely cause problems later.
   Good luck...

					Gary Duzan
					GTE Laboratories