[Tutor] String formatting expression "g" conversion type case.

[Barnaby Scott]

On 24/01/2013 13:29, Krupkina Lesya Olegovna wrote:
> Hello!
> I’m newcomer to Python and I’m on documentation reading stage and trying some of examples.
> I’m using Win7 x64 OS and Python 2.7.3 (default, Apr 10 2012, 23:24:47) [MSC v.1500 64 bit (AMD64)].
> I try to understand how string format expression (%)works. Everything is almost clear but except one case: using ofg(G) conversion type and # flag.
> Let’s take a look at documentation here:
> http://docs.python.org/release/2.7.3/library/stdtypes.html#sequence-types-str-unicode-list-tuple-bytearray-buffer-xrange
> Document declares for g(G) conversion type in case of using # flag (4th note):
> “The precision determines the number of significant digits before and after the decimal point and defaults to 6”.
>
> I have noticed behavior that does not meet documentation declaration and looks like a bug in case when using g(G) conversion type with # flag
> with  omitted  precision  and  zero integer part of the decimal. Could
> someone, please comment the case it is a bug or right use case result? If it is correct, please explain why.
>
> Steps to reproduce the case:
>
> 1.Start python interactive mode
> 2.Enter string with g(G) conversion type and using #flag like this: "%#g"%0.3 – precision parameter is omitted and integer part of the decimal is zero.
> 3.Watch the output results
>
> Actual result:
>
> Python outputs decimal as declared as but with more significant digits than default value of 6 - if integer part of the decimal is equal to zero.
>>>> "%#g"%0.3
> '0.300000'
>>>> "%#G"%0.3
> '0.300000'
>>>> "%#G"%0.004
> '0.00400000'
>>>>
>
> Expected results:
> As declared in documentation – there will be 6 significant digits before and after decimal point by default.
>
> Thanks,
> Regards, Lesya.
>
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> To unsubscribe or change subscription options:
> http://mail.python.org/mailman/listinfo/tutor
>


My experience is that things that look odd at first sight are VERY 
rarely bugs - especially in something as mature as Python 2.7.

Maybe the documentation could be slightly clearer, but I see it as 
saying the following:

'#g' and '#G' both convert to exponential format (with some exceptions) 
with a default precision of 6 *significant figures*. It is certainly 
talking about significant figures, not decimal places. So your final 
example may look odd, but it really is doing what it says - there are 6 
s.f. shown, and it has not converted to exponential form because the 
exponent would not be less than -4.

With significant digits before the decimal point, it appears to convert 
to exponential form only if the number of them exceeds the precision.

Best

Barnaby