the g0 edit descriptor
in [Fortran]
|
Prev: Foray Build Tool Project Started
Next: write matrix
From: James Giles on 3 Jul 2008 19:21 If anyone here can answer, I won't send the following as an F2008 public review response: I can't really tell in the document whether g0 is allowed to be g0.d (that is, a digits part). If it can, then it's likely to become one of my favorite features. -- J. Giles "I conclude that there are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies." -- C. A. R. Hoare "Simplicity is prerequisite for reliability" -- E. W. Dijkstra
From: Dan Nagle on 3 Jul 2008 20:09 Hi, On 2008-07-03 19:21:40 -0400, "James Giles" <jamesgiles(a)worldnet.att.net> said: > If anyone here can answer, I won't send the following as an > F2008 public review response: > > I can't really tell in the document whether g0 is allowed to > be g0.d (that is, a digits part). If it can, then it's > likely to become one of my favorite features. I think so. I don't see any prohibition (on output). (You can't on input because the processor doesn't know how many characters to read.) On input (10.7.2.3.2) allows w=0. However, please submit a comment, unless a definitive response to the contrary appears. The g0 form is intended for output, and we may have overlooked a possibility within the input case. Good catch. -- Cheers! Dan Nagle
From: James Giles on 3 Jul 2008 20:17 Dan Nagle wrote: .... >> I can't really tell in the document whether g0 is allowed to >> be g0.d (that is, a digits part). If it can, then it's >> likely to become one of my favorite features. > > I think so. I don't see any prohibition (on output). Well, that's where I want it: output. I want to be able to demand at least enough digits that subsequent input will yield the same internal value as was written. For IEEE single that's 9. If I have to take the same default d that list directed produces that's of little value to me. -- J. Giles "I conclude that there are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies." -- C. A. R. Hoare "Simplicity is prerequisite for reliability" -- E. W. Dijkstra
From: Tobias Burnus on 4 Jul 2008 02:08 On 4 Jul., 01:21, "James Giles" <jamesgi...(a)worldnet.att.net> wrote: > If anyone here can answer, I won't send the following as an > F2008 public review response: > I can't really tell in the document whether g0 is allowed to > be g0.d (that is, a digits part). If it can, then it's > likely to become one of my favorite features. It is not allowed: G w [ . d [ E e ] ] "C1007 (R1006) For the G edit descriptor, d shall be specified if and only if w is not zero." Tobias
From: James Giles on 4 Jul 2008 02:46
Tobias Burnus wrote: > On 4 Jul., 01:21, "James Giles" <jamesgi...(a)worldnet.att.net> wrote: >> If anyone here can answer, I won't send the following as an >> F2008 public review response: >> I can't really tell in the document whether g0 is allowed to >> be g0.d (that is, a digits part). If it can, then it's >> likely to become one of my favorite features. > > It is not allowed: > > G w [ . d [ E e ] ] > > "C1007 (R1006) For the G edit descriptor, d shall be specified if and > only if w is not zero." Ok. That means my comment will be that the constraint removes any value I saw in the feature. -- J. Giles "I conclude that there are two ways of constructing a software design: One way is to make it so simple that there are obviously no deficiencies and the other way is to make it so complicated that there are no obvious deficiencies." -- C. A. R. Hoare "Simplicity is prerequisite for reliability" -- E. W. Dijkstra |