Beefy Boxes and Bandwidth Generously Provided by pair Networks
We don't bite newbies here... much
 
PerlMonks  

TANGENT = %.17g formatting (Re^8:int() function)

by pryrt (Abbot)
on Oct 26, 2020 at 13:39 UTC ( [id://11123182]=note: print w/replies, xml ) Need Help??


in reply to Re^7: int() function
in thread int() function

"%.17g" formatting

I was surprised to see 16 digits after the decimal point in a "%.17" notation. Per perl's sprintf docs:

%g    a floating-point number, in %e or %f notation

But that's not what actually happens in Perl:

C:>perl -MConfig -le "print +5.3130431399669280e-34; printf qq($_\n), ++5.3130431399669280e-34 for qw/%.17f %.17e %.17g/; print qq($^V\t|\t$ +Config{myuname})" 5.31304313996693e-34 0.00000000000000000 5.31304313996692843e-34 5.3130431399669284e-34 v5.30.0 | Win32 strawberry-perl 5.30.0.1 #1 Thu May 23 12:20:46 +2019 x64

... In this instance, "%.17g" is neither equivalent to "%.17f" (17 digits after the fixed decimal point) nor to "%.17e" (17 digits after the floating decimal point).

I tried on an ancient linux machine with v5.8.5, with similar results:

% perl -MConfig -le 'print +5.3130431399669280e-34; printf qq($_\n), + +5.3130431399669280e-34 for qw/%.17f %.17e %.17g/; print qq($]\t|\t$Co +nfig{myuname})' 5.31304313996693e-34 0.00000000000000000 5.31304313996692843e-34 5.3130431399669284e-34 5.008005 | linux builder4.centos.org 2.6.9-55.0.9.elsmp # +1 smp thu sep 27 18:27:41 edt 2007 i686 i686 i386 gnulinux

The sprintf docs have used that phrasing at least as far back as v5.005 . Am I misunderstanding what I'm seeing? or has perl documentation about %g been wrong for that long?

Replies are listed 'Best First'.
Re: TANGENT = %.17g formatting (Re^8:int() function)
by syphilis (Archbishop) on Oct 26, 2020 at 23:57 UTC
    %g a floating-point number, in %e or %f notation

    The full piece of documentation is:
    %e a floating-point number, in scientific notation %f a floating-point number, in fixed decimal notation %g a floating-point number, in %e or %f notation
    I think the last of those 3 is just shorthand for:
    %g a floating-point number, in scientific notation or fixed decima +l notation
    Note that "%.*g" will give you scientific notation and "%*g" will give you fixed decimal notation (ie fixed to 6 digits of precision).
    I think it's just meant to provide a quick indication of what "%g" formatting does, for anyone who is unfamiliar with it.
    Those are the only excuses I can come up with ;-)

    Cheers,
    Rob

Log In?
Username:
Password:

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: note [id://11123182]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this?Last hourOther CB clients
Other Users?
Others cooling their heels in the Monastery: (2)
As of 2024-04-25 21:09 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    No recent polls found