Re: record copy function

In article <3s7eh3$b1g@cssun.mathcs.emory.edu> johnl@informix.com (Jonathan Leffler) writes: >>From: "Watje, Paul" <watjep@emspo01.hasting.com> >>Date: Tue, 20 Jun 95 14:45:00 PDT >>X-Informix-List-Id: <list.6659> >> >>Who ever posted the source for the record comparision >>function, THANKS, it really came in handy. I know i4gl, >>has that function in a library(unsupported :^(, but >>with the source I could also use it under RDS. > >It does? I don't think it does, and, with all due respect, I think I >know the I4GL soruce code better than you do. That's why the C-code is >necessary. There is no built-in record comparison available in I4GL >because the language does not support the operation. Incidentally, what >does the C code you are using do if the records differ in size by an even >number of elements? > Hi Jonathan, It could very well be the one I wrote and posted some time ago. You raise a good point about records differing in size by an even number of elements and of course the function has no way of detecting that it was called that way. It was designed strictly for use with records defined exactly the same, either via similar declarations or as elements within the same array. However, one good point leads to another. Why doesn't the language support this concept directly so I would not have had to write such a beast? (I do think the language support should allow an option for whether nulls in paired-elements are ignored during the comparison.) One of your many fans, Scott