[erlang-questions] : Bug? Pretty-printing floating point values
Mon Dec 10 09:40:18 CET 2007
On 11/22/07, Pierpaolo Bernardi <olopierpa@REDACTED> wrote:
> On Nov 22, 2007 9:05 AM, Raimo Niskanen
> <raimo+erlang-questions@REDACTED> wrote:
> > I guess float_to_list/1 produces 20 digits because previously the
> > external term format for floats was that kind of printout -
> > 20 decimal digits with exponent. Nowdays it is the IEEE format.
> > And I guess it was chosen so you kind of could guarantee to be able to
> > recreated exactly the right number after a roundtrip to external format.
> > Therefore significantly more than 16 digits was needed.
> > I do not know if 20 digits can be proved to be always enough, though...
> The problem of printing and reading floating point numbers with the
> minimal number of digits
> to guarantee round trip invariance has been investigated and solved.
> See the papers and the public domain code already posted by Doug Currie.
> These algorithms are used in all serious common lisp and scheme implementations.
> It would be very smart for the Erlang implementors to borrow this
> public domain code.
I spent some time today writing up an Erlang implementation of Bob
Burger's "Printing Floating-Point Numbers Quickly and Accurately"
algorithm. It's in the mochiweb repository here:
It's probably not as quick as the scheme implementation because it
doesn't use a look-up table for powers of 10, but it's probably fast
enough for most uses.
More information about the erlang-questions