home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!europa.asd.contel.com!emory!salmon.demon.co.uk
- From: neil@salmon.demon.co.uk ("Neil S. Briscoe")
- Newsgroups: comp.databases.informix
- Subject: Redeclaration of local variables (?)
- Message-ID: <9371@emory.mathcs.emory.edu>
- Date: 21 Aug 92 12:08:05 GMT
- Sender: walt@mathcs.emory.edu
- Reply-To: neil@salmon.demon.co.uk ("Neil S. Briscoe")
- Lines: 63
- X-Informix-List-ID: <list.1387>
-
- Jonathon,
-
- You wrote:
-
- >>I had the message:
- >>ut_merge.c Line 4351 redeclaration of prepare_sel_
- >
- >This is therefore probably a C compiler error message. If it came
- >from 4GL, the file referenced would be ut_merge.4gl; if it came from
- >ESQL, the file referenced would be ut_merge.ec. Also, the message
- >format is more like that used by the C compiler...
-
- Sorry. Well of course you're right and I new the message was not from
- c4gl itself by then, I merely mentioned this so that folks would know
- I was working from make and not from the programmers environment.
-
- >In fact, I would expect an error message to refer to the .ec file even
- >if the compiler was compiling the C produced from the .ec file...
-
- I would expect that. But both this version and 4.10.... on the Sparc
- always state blah.c. Only lowly 1.10.00B states blah.ec.
-
- >When you checked the source, did you look at the I4GL code or the generated
- >C code? I would expect your observation to be valid for both, which would
- >then indicate a broken C compiler -- however, you don't get many broken
- >C compilers on Unix boxes, so it needs more careful scrutiny.
-
- I must admit that so far I've only checked the 4GL. I fixed this
- particular variable by making it global. The reason I did this was
- because it, amongst other variables, were listed in ld's list of
- undefined symbols. I've grepped for all of these variables in the 4gl
- code and the objects are in the library.
-
- >Have you tracked down the declarations of prepare_sel_ in the C code?
- >Are these internal to the function or external to them? If internal and
- >the compiler is wittering, then you have a seriously defective compiler.
- >If external, then we have a question of what is being done on your other
- >machines where it does work. Where are the variables declared on those
- >other machines?
-
- Thanks for the hint. As you might guess, I don't often program in C.
-
- >None of this is a definitive answer, but to give one would need much more
- >information than should be sent over the net...
-
- Accepted. Again, thanks very much.
-
- Regards
- Neil
-
-
- Neil S. Briscoe Telephone: +44 252 376737
- System Administrator Fax: +44 252 376644
- Email: neil@salmon.demon.co.uk
- nbriscoe@cix.compulink.co.uk
- Bartec Medical Systems Ltd.
- Impression House
- Invincible Road "!##@!** Listen who swears
- Farnborough Christopher Robin has fallen downstairs"
- Hants GU14 7NP -- The Goodies Book of Criminal Records
- England
- "A man is incomplete until he's married -- then he's finished!"
- -- seen on a desktop calendar
-