Hi Derek,
Thanks! I'll try it out later.
Best regards,
Luis
On Fri, Mar 06, 2015 at 08:43:02AM -0700, Derek Lamb wrote:
> Hi Luis,
>
> PDL-2.007_12 was just announced and has fixes for longlong operands, and a
> whole bunch more modulus tests. Give it a whirl and let us know if I missed
> anything.
>
> best,
> Derek
>
> > On Feb 17, 2015, at 11:07 PM, Derek Lamb <[email protected]> wrote:
> >
> > Hi Luis,
> >
> > You are correct, longlong(600851475143)%longlong(71) still fails. Sorry
> > for not checking. I have a good idea on solving that… stand by.
> >
> > Just wanted to make sure you don't spend time chasing a dead-end with this
> > parentheses bit:
> >
> > pdl> p (600851475143)%(71) #WRONG and DIFFERENT from above
> > 600851475143
> >
> > Notice that the returned value is the first argument. What you wrote
> > translated into roughly:
> >
> > pdl> print(600851475143)%(71)
> >
> > In general, if you're going to follow 'p', 'print' or any function with a
> > (, you need a closing paren to match it. Apparently "p" in the debugger
> > behaves differently.
> >
> > Derek
> >
> >> On Feb 16, 2015, at 6:54 PM, Luis Mochan <[email protected]> wrote:
> >>
> >> Dear Derek,
> >> I downloaded and compiled the git version and I made the following
> >> tests from the pdl-code directory
> >>
> >> mochan@em:~/Downloads/pdl-code$ ./pdl
> >> ...
> >> Loaded PDL v2.007 (supports bad values) #MY OLD PDL
> >> ...
> >> pdl> p longlong(10)%longlong(5) #WRONG result
> >> 53870592
> >> pdl> q
> >>
> >>
> >> mochan@em:~/Downloads/pdl-code$ ./pdl -Mblib
> >> ...
> >> Loaded PDL v2.00711 (supports bad values) #NEW VERSION
> >> ...
> >> pdl> p longlong(10)%longlong(5) #CORRECT result!
> >> 0
> >> pdl> p 600851475143%71 #CORRECT result
> >> 0
> >> pdl> p longlong(600851475143)%longlong(71) #still WRONG result
> >> 609885356032
> >> pdl> p (600851475143)%(71) #WRONG and DIFFERENT from above
> >> 600851475143
> >>
> >> So it seems that the problem with longs is solved when using small
> >> numbers, but it remains when using large numbers. Furthermore, I found
> >> (accidentally) another situation which yields the wrong result: just
> >> putting the numbers between parenthesis yields an erroneous result
> >> which is different.
> >>
> >> In order to test if the problem is the loss of precission when reading
> >> the large number, as suggested by Chris
> >> (http://sourceforge.net/p/pdl/bugs/343/)
> >> I built it by parts:
> >>
> >> pdl> p 600851475143%71
> >> 0
> >> pdl> p +(longlong(6008514)*100000+75143)
> >> 600851475143
> >> pdl> p +(longlong(6008514)*100000+75143)%longlong(71)
> >> 609885356032
> >> pdl> p +(longlong(6008514)*100000+75143)%71
> >> 609885356032
> >>
> >>
> >> The problem with the parenthesis seemed to be unrelated to pdl's, so I
> >> tried
> >> it in the perl debugger:
> >>
> >> mochan@em:~/Downloads/pdl-code$ perl -d -e 'print Hi'
> >> ...
> >> main::(-e:1): print Hi
> >> DB<1> p 600851475143%71
> >> 0
> >> DB<2> p (600851475143)%(71)
> >> 0
> >> DB<3> q
> >>
> >> It seems the culprit is not my perl, as it yields the correct result with
> >> or without parenthesis.
> >>
> >> In summary, part of the problem was corrected with your commit, but it
> >> seems that other problems remain, and apparently they are not (only)
> >> related to the conversion from floating point.
> >>
> >> Thanks and regards,
> >> Luis
> >>
> >>
> >> --
> >>
> >> o
> >> W. Luis Mochán, | tel:(52)(777)329-1734 /<(*)
> >> Instituto de Ciencias Físicas, UNAM | fax:(52)(777)317-5388 `>/ /\
> >> Apdo. Postal 48-3, 62251 | (*)/\/ \
> >> Cuernavaca, Morelos, México | [email protected] /\_/\__/
> >> GPG: 791EB9EB, C949 3F81 6D9B 1191 9A16 C2DF 5F0A C52B 791E B9EB
> >>
> >>
> >>
> >> _______________________________________________
> >> Perldl mailing list
> >> [email protected]
> >> http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
> >>
> >
> >
> > _______________________________________________
> > Perldl mailing list
> > [email protected]
> > http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
> >
>
>
--
o
W. Luis Mochán, | tel:(52)(777)329-1734 /<(*)
Instituto de Ciencias Físicas, UNAM | fax:(52)(777)317-5388 `>/ /\
Apdo. Postal 48-3, 62251 | (*)/\/ \
Cuernavaca, Morelos, México | [email protected] /\_/\__/
GPG: 791EB9EB, C949 3F81 6D9B 1191 9A16 C2DF 5F0A C52B 791E B9EB
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
pdl-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/pdl-general
