How to make Python format floats with certain amount of significant digits?

Question

I want my Python (2.4.3) output numbers to have a certain format. Specifically, if the number is a terminating decimal with <= 6 significant digits, show it all. However, if it has > 6 significant digits, then output only 6 significant digits.

"A" shows how Python is writing the floats. "B" shows how I want them written. How can I make Python format my numbers in that way?

A:
10188469102.605597
5.5657188485
3.539
22.1522612479
0
15.9638450858
0.284024
7.58096703786
24.3469152383

B:
1.01885e+10
5.56572
3.539
22.1523
0
15.9638
0.284024
7.58097
24.3469

Factitious

Floats lack the precision required to make this possible. – Coop

"Format decimal" - I think you mean "format float", because Decimal is a library in Python. – Compel

@BurhanKhalid yes, you're correct. – Factitious

possible duplicate of Nicely representing a floating-point number in python – Sunil

You really need to accept joachim's answer, not mine. – Compel

Just a tip, I'd advise using the latest version of python 2 (2.7) 2.4.x is very old now and in terms of compatability and use to others it'd be worth updating to the latest version – Pabulum

Answer 1

You'll want the g modifier for format that drops insignificant zeroes;

>>> "{0:.6g}".format(5.5657188485)
'5.56572'
>>> "{0:.6g}".format(3.539)
'3.539'

---

Sorry, my update also includes the fact that I am restricted to using Python 2.4.3, which does not have format() function.

The format specifiers work even without the .format() function:

>>> for i in a:
...    print '%.6g' % (i,)
...
1.01885e+10
5.56572
3.539
22.1523
0
15.9638
0.284024
7.58097
24.3469

Selfsupporting

Updated my question with a large number in example as well – Factitious

@Factitious As far as I can see, the format gives your exact B output. – Selfsupporting

Sorry, my update also includes the fact that I am restricted to using Python 2.4.3, which does not have format() function. – Factitious

Answer 2

There is a way to retain trailing zeros so that it consistently shows the number of significant digits. Not exactly what OP wanted, but probably useful to many.

a = [10188469102.605597,5.5657188485,3.539,22.1522612479,0,15.9638450858,0.284024,7.58096703786,24.3469152383]

for i in a:
    print("{:#.6g}".format(i))

Output

1.01885e+10
5.56572
3.53900
22.1523
0.00000
15.9638
0.284024
7.58097
24.3469

Note that this will only work with the format function and not with % operator.

According to the docs:

The '#' option causes the “alternate form” to be used for the conversion. The alternate form is defined differently for different types. This option is only valid for integer, float, complex and Decimal types.

'g': General format ... insignificant trailing zeros are removed from the significand, and the decimal point is also removed if there are no remaining digits following it, unless the '#' option is used.

Angloamerican

thanks; according to the docs, the '#' goes btw sign and the optional '0': docs.python.org/3.4/library/string.html – Burgrave

Answer 3

try this way

a=[10188469102.605597,5.5657188485,3.539,22.1522612479,0,15.9638450858,0.284024,7.58096703786,24.3469152383]

 for i in a:
    if i >100:
        print '{:.6e}'.format(i)
    else:
        print '{:.6f}'.format(i)

for lower version of python

for i in a:
    if i >100:
        print '%6e'%i
    else:
        print '%6f'%i

output

1.018847e+10
5.565719
3.539000
22.152261
0.000000
15.963845
0.284024
7.580967
24.346915

Exertion

I cannot use format() function. That is not available for Python 2.4.3. – Factitious