5 ms·

It is worth noting that most scripting languages today are still restricted by 64-bit integers and IEEE floating points limitations. Python and Ruby are probab

by mmoez 7y ago

It is worth noting that most scripting languages today are still restricted by 64-bit integers and IEEE floating points limitations. Python and Ruby are probably the only popular ones that ship by default with unlimited precision integers. For arbitrary precision reals in Python, there is mpmath http://mpmath.org/ http://mpmath.org/ . There is also the standard decimal module which "provides support for fast correctly-rounded decimal floating point arithmetic" : https://docs.python.org/3/library/decimal.html https://docs.python.org/3/library/decimal.html

m4r35n357 7y ago
For MP work in Python, gmpy2 - https://pypi.org/project/gmpy2/ https://pypi.org/project/gmpy2/ comes with a much better selection of MP functions, backed by MPFR, MPC. Also, the buit-in Decimal class is horribly slow by comparison to gmpy2.
- fdej 7y ago
  Also (shameless self-promotion) python-flint! http://fredrikj.net/python-flint/ http://fredrikj.net/python-flint/
- baruchel 7y ago
  Purpose of the Decimal module is not really arbitrary precision computation but rather exact computation on decimal numbers wherever it matters (financial arithmetic, etc.). This module should never be used for scientific purposes.
292355744930110 7y ago
In Go "Numeric constants represent exact values of arbitrary precision" https://golang.org/ref/spec#Constants https://golang.org/ref/spec#Constants
- divs1210 7y ago
  I don't think Go qualifies as a 'scripting language', even though the term is very loosely defined.
  - 292355744930110 7y ago
    I missed the part about scripting languages. Is LISP a scripting language?
    - kazinator 7y ago
      Yes. 1. Most Lisp dialects, including Scheme and ANSI Lisp, have interactive modes for evaluating expressions. 2. Most Lisp dialects have a mode of processing files in the same manner: reading individual forms and evaluating them immediately. I.e. it is not necessary to translate files into objects that have to be linked to produce an executable program.
- munificent 7y ago
  This is only true for values calculated at compile time. Once the result is stored in a variable it is converted to a fixed-size representation. All runtime calculations and results are fixed size.
threwawasy1228 7y ago
I think Haskell allows for arbitrary-precision with the Integer type. It is limited only by the available memory in the machine.
- divs1210 7y ago
  Even though the term 'scripting language' is very loosely defined, I don't think Haskell qualifies as one.
  - ky3 7y ago
    Haskell has shell hash-bang scripting capabilities just like Python and Ruby do. See https://stackoverflow.com/questions/8676331/how-to-run-a-haskell-file-in-interpreted-mode/8676377#8676377 https://stackoverflow.com/questions/8676331/how-to-run-a-has...
    - kazinator 7y ago
      Almost anything can be hash-bang, including C. https://rosettacode.org/wiki/Native_shebang https://rosettacode.org/wiki/Native_shebang https://rosettacode.org/wiki/Multiline_shebang https://rosettacode.org/wiki/Multiline_shebang
      - ky3 7y ago
        True, but the difference between C and Haskell hash-bangs is that GHC Haskell provides an interpreter. GCC hash-bang is still an edit-compile-link-debug cycle. Just like with Python and Ruby, Haskell hash-bang accelerates the cycle by jettisoning the compile and linking steps.
        kazinator 7y ago
        The Rosetta examples show a solution that only compiles the C file if it has not been previously compiled, or if it has a newer modification timestamp than the executable.
mpolun 7y ago
BigInts are coming to Javascript -- they're supported in most browsers and in node now, but they haven't officially been added to the standard.
nikofeyn 7y ago
> Python and Ruby are probably the only popular ones that ship by default with unlimited precision integers. racket has exact numbers of unlimited size. https://docs.racket-lang.org/reference/numbers.html https://docs.racket-lang.org/reference/numbers.html
jxy 7y ago
In addition to Scheme and CL, Perl: https://perldoc.perl.org/bignum.html https://perldoc.perl.org/bignum.html Perl6: https://docs.perl6.org/language/math#Arithmetic https://docs.perl6.org/language/math#Arithmetic J: https://code.jsoftware.com/wiki/Vocabulary/xco https://code.jsoftware.com/wiki/Vocabulary/xco ocaml: https://opam.ocaml.org/packages/zarith https://opam.ocaml.org/packages/zarith
ddragon 7y ago
Julia (as a "scripting" language for scientific purposes) also supports them natively, but defaults to normal int/float for performance reasons, which might create some gotchas for people who use the language but are not programmers. https://docs.julialang.org/en/v1/manual/integers-and-floating-point-numbers/#Arbitrary-Precision-Arithmetic-1 https://docs.julialang.org/en/v1/manual/integers-and-floatin... If I remember correctly, Elixir uses arbitrary precision integers by default, though I'm not sure about float.
hajile 7y ago
BigInt works in Firefox, Chrome, Opera, and upcoming Edge at the moment. Safari preview also has at least partial support.

by mmoez 7y ago

m4r35n357 7y ago

For MP work in Python, gmpy2 - https://pypi.org/project/gmpy2/ https://pypi.org/project/gmpy2/ comes with a much better selection of MP functions, backed by MPFR, MPC. Also, the buit-in Decimal class is horribly slow by comparison to gmpy2.

fdej 7y ago
Also (shameless self-promotion) python-flint! http://fredrikj.net/python-flint/ http://fredrikj.net/python-flint/
baruchel 7y ago
Purpose of the Decimal module is not really arbitrary precision computation but rather exact computation on decimal numbers wherever it matters (financial arithmetic, etc.). This module should never be used for scientific purposes.

292355744930110 7y ago

In Go "Numeric constants represent exact values of arbitrary precision" https://golang.org/ref/spec#Constants https://golang.org/ref/spec#Constants

divs1210 7y ago
I don't think Go qualifies as a 'scripting language', even though the term is very loosely defined.
- 292355744930110 7y ago
  I missed the part about scripting languages. Is LISP a scripting language?
  - kazinator 7y ago
    Yes. 1. Most Lisp dialects, including Scheme and ANSI Lisp, have interactive modes for evaluating expressions. 2. Most Lisp dialects have a mode of processing files in the same manner: reading individual forms and evaluating them immediately. I.e. it is not necessary to translate files into objects that have to be linked to produce an executable program.
munificent 7y ago
This is only true for values calculated at compile time. Once the result is stored in a variable it is converted to a fixed-size representation. All runtime calculations and results are fixed size.

threwawasy1228 7y ago

I think Haskell allows for arbitrary-precision with the Integer type. It is limited only by the available memory in the machine.

divs1210 7y ago
Even though the term 'scripting language' is very loosely defined, I don't think Haskell qualifies as one.
- ky3 7y ago
  Haskell has shell hash-bang scripting capabilities just like Python and Ruby do. See https://stackoverflow.com/questions/8676331/how-to-run-a-haskell-file-in-interpreted-mode/8676377#8676377 https://stackoverflow.com/questions/8676331/how-to-run-a-has...
  - kazinator 7y ago
    Almost anything can be hash-bang, including C. https://rosettacode.org/wiki/Native_shebang https://rosettacode.org/wiki/Native_shebang https://rosettacode.org/wiki/Multiline_shebang https://rosettacode.org/wiki/Multiline_shebang
    - ky3 7y ago
      True, but the difference between C and Haskell hash-bangs is that GHC Haskell provides an interpreter. GCC hash-bang is still an edit-compile-link-debug cycle. Just like with Python and Ruby, Haskell hash-bang accelerates the cycle by jettisoning the compile and linking steps.
      - kazinator 7y ago
        The Rosetta examples show a solution that only compiles the C file if it has not been previously compiled, or if it has a newer modification timestamp than the executable.

mpolun 7y ago

BigInts are coming to Javascript -- they're supported in most browsers and in node now, but they haven't officially been added to the standard.

nikofeyn 7y ago

> Python and Ruby are probably the only popular ones that ship by default with unlimited precision integers. racket has exact numbers of unlimited size. https://docs.racket-lang.org/reference/numbers.html https://docs.racket-lang.org/reference/numbers.html

jxy 7y ago

In addition to Scheme and CL, Perl: https://perldoc.perl.org/bignum.html https://perldoc.perl.org/bignum.html Perl6: https://docs.perl6.org/language/math#Arithmetic https://docs.perl6.org/language/math#Arithmetic J: https://code.jsoftware.com/wiki/Vocabulary/xco https://code.jsoftware.com/wiki/Vocabulary/xco ocaml: https://opam.ocaml.org/packages/zarith https://opam.ocaml.org/packages/zarith

ddragon 7y ago

Julia (as a "scripting" language for scientific purposes) also supports them natively, but defaults to normal int/float for performance reasons, which might create some gotchas for people who use the language but are not programmers. https://docs.julialang.org/en/v1/manual/integers-and-floating-point-numbers/#Arbitrary-Precision-Arithmetic-1 https://docs.julialang.org/en/v1/manual/integers-and-floatin... If I remember correctly, Elixir uses arbitrary precision integers by default, though I'm not sure about float.

hajile 7y ago

BigInt works in Firefox, Chrome, Opera, and upcoming Edge at the moment. Safari preview also has at least partial support.