Re: Phantom error in 4GL

I've seen something similar with a load statement, code ran OK without the load but core dumped with it in. Never got to the bottom of it, just split the function into separate modules and it worked. Weird Daniel Wright <dw420@airmail.net> wrote in article <A85E35FB64985517.C3A4F9EB6C12FEFB.D74E83E5474622A5@library-proxy.airnews.ne t>... > The first time I heard of this error, I didn't believe it. > > A 4GL compiles fine, and is syntactically correct. However, weird > errors, such as "incorrect number of host variables supplied to an OPEN > or Syntax error (-201) appear as a result of a PREPARE or OPEN. > > In fact, the same code that compiled and ran fine before being checked > into our source code control system (PVCS, in this case), failed upon > being recompiled and executed (The only difference being that a few > comments were added to the top of the file.) > > As I said, I didn't believe it at first, but I have seen at least 2 > examples of this behavior in the past month, and am told by those who > know, that it is not that uncommon. Supposedly Informix will not admit > to the bug, although Senior Tech Supporticians (Is that a word?) will > agree that it exists. > And the fix? Simply move a function within the source code....it > doesn't really matter, just rearrange it a bit, and it will work. > I've seen this happen in version 5.??, and am told it can happen in > version 7.?? as well. > > Has anyone else experienced this? > Is it true Informix will not admit to a bug? > If so, I am posting this to "hold their feet to the fire" to admit a bug > and fix it. > This kind of crap scares me, and makes me wonder if many of our > customer's complaints are a result of an Informix bug. > [Yeah, it bothers me, because, I am likely to be awoken at 4:00 AM to > fix the 'holy grail' of Informix bugs, which naturally, I assume is > ours, since Informix really "QA's their shit"]. > If not, either tell me that me and my co-workers who have experienced > this are crazy or at the very least hallucinating. > > I've never personally encountered this bug, but I write that off to > mostly working w/ ESQL/C, and I have only heard it reported w/ 4GL, > since I have seen another demonstrate it to me on several separate > occasions w/ c4gl. > > > > -- > - Danny Wright > danwright@bigfoot.com > #include <stddisclaimers.h> > Idiot, n.: > A member of a large and powerful tribe whose influence in human > affairs has always been dominant and controlling. > - Ambrose Bierce, "The Devil's Dictionary" > > >