Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!newstop!exodus!exodus-bb!khb From: khb@chiba.Eng.Sun.COM (chiba) Newsgroups: comp.lang.fortran Subject: Re: Is this kosher? Message-ID: Date: 31 Jan 91 00:04:26 GMT References: <20773@hydra.gatech.EDU> <2796@dftsrv.gsfc.nasa.gov> <20785@hydra.gatech.EDU> <2812@dftsrv.gsfc.nasa.gov> <91030.132436KENCB@SLACVM.SLAC.STANFORD.EDU> Sender: news@exodus.Eng.Sun.COM Organization: Sun MegaSystems Lines: 22 In-reply-to: KENCB@SLACVM.SLAC.STANFORD.EDU's message of 30 Jan 91 21:24:36 GMT In article <91030.132436KENCB@SLACVM.SLAC.STANFORD.EDU> KENCB@SLACVM.SLAC.STANFORD.EDU writes: ... but you must not transfer control into a block. Thus, you must not transfer control from one statement block to another." The second sentence is the relavent one. However, MUCH to my SURPRISE, John's test program fragment compiled just FINE under VAX Fortran! Sheez! I think it's time for an SPR on this... but then I don't know the history (PDP-11 compatibility, etc.). In general, a FORTRAN processor (x3.9-1978 compliant), may do anything it wants to with a non-standard compliant program. Similarly with vendor documents, the constraint is on the user to not write illegal (viz. code which violates the vendor's document) code. If you do, it MAY work. That isn't a bug. -- ---------------------------------------------------------------- Keith H. Bierman kbierman@Eng.Sun.COM | khb@chiba.Eng.Sun.COM SMI 2550 Garcia 12-33 | (415 336 2648) Mountain View, CA 94043