Informix: Dumb decision guys!

We've just discovered the following, and think it's a wee bit of a dumb idea. Does anybody agree? ------ Informix have reported that in future run time errors:- -1226, -1213, -1214, -1202 will NO LONGER be reported at run time. -1226 Decimal or money value exceeds maximum precision. -1213 A character to numeric conversion process failed. -1214 Value too large to fit in a SMALLINT. -1202 An attempt was made to divide by zero. Their reason is what they call a NIS (Not in spec) feature. The feature has been available but has now been withdrawn. A workaround (as mentioned by Informix) is to compile the programs with the -anyerr parameter, and to some extent this works, *except* for things involved with SELECT statements. The following wee program illustrates that it's now perfectly reasonable to select CHAR columns into SMALLINT fields without an error: main define i smallint -- Point at a database (You'll want to change this line) database stores6 -- Program should die if we have an error whenever any error stop -- Clear out, and startup, the errorlog run "rm -f dumb.log" CALL startlog("dumb.log") -- Select a char column into a smallint select tabname into i from systables where tabname = "systables" -- Look ma -- no problems! display "nothing wrong, status = ", status -- But check out the errorlog... display "but what's this?" run "cat dumb.log" end main The above *might* be a bug, or it might be expected behaviour. Waddaya reckon? (This with 6.02.UC1 RDS) Regards, Kerry S ------------------------------------,------------------------------------------ Kerry Sainsbury, kerry@kcbbs.gen.nz | THE INFORMIX FAQ v2.4 Oct 95 Quanta Systems, Auckland | http://www.iiug.org/techinfo/faq/ New Zealand. Work: +64 9 377-4473 | ftp://ftp.iiug.org/pub/informix/faq Home: +64 9 279-3571 | ftp://kcbbs.gen.nz:/pub/informix/