ags@h.cc.purdue.edu (Dave Seaman) (09/21/88)
In article <3821@lanl.gov> jlg@lanl.gov (Jim Giles) writes: >You are the one who keeps >introducing into the discussion things which weren't relevant to the >original issue. On the contrary, you are the one who keeps changing the subject and talking about the general case. To see clearly what is happening, take a look at the subject line above. I have changed it. I am now talking ONLY about the special case. I expect you to do likewise, unless you use a different subject line. Do we have agreement on the ground rules? Good. Now, I will ask my question one last time: why is my code example illegal? No, the answer is not "because side effects are not allowed in Fortran." That is not answering the question. That is changing the subject. Get the picture? Let me help. The answer is, "because section _____ of the standard says _________________". You fill in the blanks. No paraphrasing, please. I insist on an exact quote. How about it? ----------------------------------------------------------------------- Fair warning: I intend to post again, with THE GENERAL CASE as my subject. Since you were the one who introduced that discussion and kept returning to it every time I tried to discuss MY subject, I thought you were entitled to know. Of course, you don't have to respond, if you find it too difficult to stick to logic and can no longer accuse me of changing the subject. -- Dave Seaman ags@j.cc.purdue.edu
jlg@lanl.gov (Jim Giles) (09/22/88)
From article <3994@h.cc.purdue.edu>, by ags@h.cc.purdue.edu (Dave Seaman): > Now, I will ask my question one last time: why is my code example > illegal? > The example you gave appears to be legal. The optimization I gave for your example also appears to be legal. It appears that the semantics of Fortran may be ambiguous in this case. I claim that any case in which the semantics of a programming language are ambiguous is a case which the programmer should regard as illegal. I regard multiple side effects in C expressions as illegal - because the semantics of such things are ambiguous YES!!! I ADMIT IT!!!! You have found a case that I thought was illegal and now it appears to be _merely_ ambiguous in meaning. Now, are you still going to maintain that it should be used by programmers (as you have been doing). Or, do you think it should have the same status as the 'i=f(i++)' type of operation in C? That is, it should be illegal but the committee hasn't said so yet. J. Giles Los Alamos