bobal@microsoft.UUCP (Bob Allison) (06/14/89)
Not that this has anything to do with FORTRAN, but for those of us who love autoindent, but don't like tabs, the appropriate vi setting is: :set ai ts=90 This is great since it will autoindent, but it won't use a tab unless you're beyond the 90th column (I rarely write a line longer than 80 characters), so it just blank-fills. Anyway, back to FORTRAN: I have to say that my biggest problem with pointers in the current draft is this whole issue of pointing to sections of arrays. Boy, am I getting tired of dope vectors. My second biggest problem is that it is an attribute. Bill Leonard, in email to members of the committee complained that it was awfully difficult to use this form of pointers to insert items in a linked list (at least, I think that was the example), a fairly common usage of pointers. As I recall, someone came back with a solution which was deemed "elegant" and was implemented using recursion. Ugh! I just want to walk around lists and stuff. Bill, could you post the problem again and see if anyone else can come up with a solution which doesn't require retraining everyone who has ever used pointers in any other language? Oh yes, a small warning on the public review: while there have been some requests for another four month public review, ANSI procedures typically call for subsequent public reviews to be two months in length. So, be prepared once the call goes out to get moving, since the time it took some people to get drafts from Global Engineering during the first public review exceeded two months. As soon as X3 sets the date of the public review, I'm sure it will be posted on the net. Frankly, I doubt there will be enough time to get the information into any journals for publication before the review is over. Bob Allison uunet!microsoft!bobal
brainerd@unmvax.unm.edu (Walt Brainerd) (06/15/89)
In article <6028@microsoft.UUCP>, bobal@microsoft.UUCP (Bob Allison) writes: > . . . Bill Leonard, in > email to members of the committee complained that it was awfully difficult > to use this form of pointers to insert items in a linked list (at least, > I think that was the example), a fairly common usage of pointers. As I > recall, someone came back with a solution which was deemed "elegant" and > was implemented using recursion. Ugh! I just want to walk around > lists and stuff. Bill, could you post the problem again and see if > anyone else can come up with a solution which doesn't require retraining > everyone who has ever used pointers in any other language? > The responses showed that Bill was all wet and that it was very easy to do linked lists, etc. It is only a little more difficult to create inelegant programs to do the same things by explicitly diddling with the pointers. If anyone else requests it, I will do an inelegant list insertion program, but I may not be able to write as bad a program as Bob, so maybe he should try, too, after he sees how it is done. Gee, the members of X3J3 are going to have to learn how to write Fortran 8x programs one of these days.