I want to replace the old string formatting behaviour with the new python string formatting syntax in my scripts, but how to avoid rounding when I deal with floats?
The old version
print ('%02d:%02d:%02d' % (0.0,0.9,67.5))
yields 00:00:67
while my (obviously wrong) translation into the new syntax
print ('{0:0>2.0f}:{1:0>2.0f}:{2:0>2.0f}'.format(0.0,0.9,67.5))
yields 00:01:68.
How to avoid rounding here and get the old output with the new format syntax?
Explicitly convert the arguments to ints:
>>> '{:02}:{:02}:{:02}'.format(int(0.0), int(0.9), int(67.5))
'00:00:67'
BTW, you don't need to specify argument index ({0}, {1}, ..) if you use Python 2.7+, Python 3.1+ (autonumbering).
int :( I thought that format has a way to truncate the decimal part.The "rules" are simple:
'%d' % 7.7 # truncates to 7
'%.0f' % 7.7 # rounds to 8
format(7.7, 'd') # refuses to convert
format(7.7, '.0f') # rounds to 9
To have complete control over the presentation, you can pre-convert the float to an integer. There are several ways to do that depending on your needs:
>>> math.trunc(f) # ignore the fractional part
67
>>> math.floor(f) # round down
67
>>> math.ceil(f) # round up
68
>>> round(f) # round nearest
68
format(7.7, '.0f') rounds to 8 and not 7