- VLFloat, numbers view and mathematicl problems - 4 Updates
- That's while I love C++ over C - 2 Updates
- I think references should have been const by default - 1 Update
| wij <wyniijj@gmail.com>: Nov 03 02:45AM -0700 I have recently released a C++ library update. https://sourceforge.net/projects/cscall/files/latest/download The main change is adding a class VLFloat for variable length floating-point number. Although a work of a bit more than two months, not extensively tested, yet many mathematical questions can be practically answered and verified. (thanks to C++'s expressiveness, if one use it properly) What is more significant, IMO, is about the number view. This is the beginning https://sourceforge.net/projects/cscall/files/MisFiles/Infinity-en.txt/download Note that I am not particularly interested in the "0.999...=1" such a little piece of cake. To the most interested, I guess, would be: 1. lim(x->∞) 1/x=0 is illogical as a foundation of a 'logical' theory. 2. Euler's identity is very questionable. https://sourceforge.net/projects/cscall/files/MisFiles/NumberView-en.txt/download I showed that Calculus is also questionable for transcendental functions. As Engineering math., this is OK, because small errors can often be ignored not to say infinitesimal errors. In all, mathematicians should re-think the relevant topics, I am just a programmer. |
| Paavo Helde <myfirstname@osa.pri.ee>: Nov 03 10:53PM +0200 03.11.2021 11:45 wij kirjutas: > In all, mathematicians should re-think the relevant topics, I am just a programmer. In related news, da Vinci should clearly re-think his oil paintings because I am troubled to map them to RGB values. |
| red floyd <no.spam.here@its.invalid>: Nov 03 02:12PM -0700 On 11/3/2021 2:45 AM, wij wrote: > 1. lim(x->∞) 1/x=0 is illogical as a foundation of a 'logical' theory. Perhaps you should have done your calculus homework in school. > 2. Euler's identity is very questionable. Again, perhaps you should have done your calculus homeowork. |
| red floyd <no.spam.here@its.invalid>: Nov 03 02:14PM -0700 On 11/3/2021 2:45 AM, wij wrote: > 1. lim(x->∞) 1/x=0 is illogical as a foundation of a 'logical' theory. > 2. Euler's identity is very questionable. You should contact the people responsible for the Fields medal. They will be very interested in your mechanism for overthrowing all of current advanced mathematics. |
| Bonita Montero <Bonita.Montero@gmail.com>: Nov 03 08:39PM +0100 (x == y ? a : b) = c; |
| Bart <bc@freeuk.com>: Nov 03 08:28PM On 03/11/2021 19:39, Bonita Montero wrote: > (x == y ? a : b) = c; Hey, I can do that: (x = y | a | b) := z and also: (n | a, b, c | d) := z (Assign to one of n lvalues.) Or even: if x then a elsif y then b else c fi := z Lots of such things in an expression-based language. But this is in a language not much higher than C in level; it doesn't require the vast complexity of C++. If C doesn't support this, that's a deliberate choice. It means someone having to instead write: *(x == y ? &a : &b) = c; It's not a big enough reason to switch. |
| "Alf P. Steinbach" <alf.p.steinbach@gmail.com>: Nov 03 12:38AM +0100 On 2 Nov 2021 22:49, Paavo Helde wrote: >> stop at first error, like I believe (think I remember) old Turbo C++ did. > Hmm, something like this? > g++ -fmax-errors=1 At least some years ago, telling g++ to stop on error did not speed up things (e.g. it still prepares itself for re-synchronizing with the source and continuing spewing out diagnostics), and when the error message has supporting notes those subsequent lines are not shown. About same story with Visual C++, but I don't remember any details. The whole batch compilation idea is 1950-ish, as I see it. - Alf |
| You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page. To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com. |
No comments:
Post a Comment