我个人使用humanize库:

>>> import datetime
>>> humanize.naturalday(datetime.datetime.now())
'today'
>>> humanize.naturalday(datetime.datetime.now() - datetime.timedelta(days=1))
'yesterday'
>>> humanize.naturalday(datetime.date(2007, 6, 5))
'Jun 05'
>>> humanize.naturaldate(datetime.date(2007, 6, 5))
'Jun 05 2007'
>>> humanize.naturaltime(datetime.datetime.now() - datetime.timedelta(seconds=1))
'a second ago'
>>> humanize.naturaltime(datetime.datetime.now() - datetime.timedelta(seconds=3600))
'an hour ago'

当然，它并没有给您所寻找的确切答案(实际上是str(timeA - timeB)，但我发现，一旦超过几个小时，显示很快就变得不可读了。Humanize支持更大的值，这些值是人类可读的，而且本地化得很好。

它的灵感来自于Django的贡献。humanize模块，显然，既然你在使用Django，你应该使用它。

下面是一个通用函数，用于将timedelta对象或常规数字(以秒或分钟等形式)转换为格式化良好的字符串。我采用了mpounsett对一个重复问题的出色回答，使其更加灵活，提高了可读性，并添加了文档。

你会发现这是迄今为止最灵活的答案，因为它允许你:

动态自定义字符串格式，而不是硬编码。 省略特定的时间间隔没有问题(参见下面的例子)。

功能:

from string import Formatter
from datetime import timedelta

def strfdelta(tdelta, fmt='{D:02}d {H:02}h {M:02}m {S:02}s', inputtype='timedelta'):
    """Convert a datetime.timedelta object or a regular number to a custom-
    formatted string, just like the stftime() method does for datetime.datetime
    objects.

    The fmt argument allows custom formatting to be specified.  Fields can 
    include seconds, minutes, hours, days, and weeks.  Each field is optional.

    Some examples:
        '{D:02}d {H:02}h {M:02}m {S:02}s' --> '05d 08h 04m 02s' (default)
        '{W}w {D}d {H}:{M:02}:{S:02}'     --> '4w 5d 8:04:02'
        '{D:2}d {H:2}:{M:02}:{S:02}'      --> ' 5d  8:04:02'
        '{H}h {S}s'                       --> '72h 800s'

    The inputtype argument allows tdelta to be a regular number instead of the  
    default, which is a datetime.timedelta object.  Valid inputtype strings: 
        's', 'seconds', 
        'm', 'minutes', 
        'h', 'hours', 
        'd', 'days', 
        'w', 'weeks'
    """

    # Convert tdelta to integer seconds.
    if inputtype == 'timedelta':
        remainder = int(tdelta.total_seconds())
    elif inputtype in ['s', 'seconds']:
        remainder = int(tdelta)
    elif inputtype in ['m', 'minutes']:
        remainder = int(tdelta)*60
    elif inputtype in ['h', 'hours']:
        remainder = int(tdelta)*3600
    elif inputtype in ['d', 'days']:
        remainder = int(tdelta)*86400
    elif inputtype in ['w', 'weeks']:
        remainder = int(tdelta)*604800

    f = Formatter()
    desired_fields = [field_tuple[1] for field_tuple in f.parse(fmt)]
    possible_fields = ('W', 'D', 'H', 'M', 'S')
    constants = {'W': 604800, 'D': 86400, 'H': 3600, 'M': 60, 'S': 1}
    values = {}
    for field in possible_fields:
        if field in desired_fields and field in constants:
            values[field], remainder = divmod(remainder, constants[field])
    return f.format(fmt, **values)

演示:

>>> td = timedelta(days=2, hours=3, minutes=5, seconds=8, microseconds=340)

>>> print strfdelta(td)
02d 03h 05m 08s

>>> print strfdelta(td, '{D}d {H}:{M:02}:{S:02}')
2d 3:05:08

>>> print strfdelta(td, '{D:2}d {H:2}:{M:02}:{S:02}')
 2d  3:05:08

>>> print strfdelta(td, '{H}h {S}s')
51h 308s

>>> print strfdelta(12304, inputtype='s')
00d 03h 25m 04s

>>> print strfdelta(620, '{H}:{M:02}', 'm')
10:20

>>> print strfdelta(49, '{D}d {H}h', 'h')
2d 1h

如果您的包中碰巧有IPython(您应该)，那么它(到目前为止)有一个非常好的持续时间格式器(以浮点秒为单位)。它被用于许多地方，例如%%时间单元格魔法。我喜欢它在短时间内产生的格式:

>>> from IPython.core.magics.execution import _format_time
>>> 
>>> for v in range(-9, 10, 2):
...     dt = 1.25 * 10**v
...     print(_format_time(dt))

1.25 ns
125 ns
12.5 µs
1.25 ms
125 ms
12.5 s
20min 50s
1d 10h 43min 20s
144d 16h 13min 20s
14467d 14h 13min 20s

我想这样做，所以写了一个简单的函数。它对我来说非常有用，而且非常通用(支持年到微秒，以及任何粒度级别，例如，你可以选择“2天4小时48分钟”和“2天4小时”和“2天4.8小时”等。

def pretty_print_timedelta(t, max_components=None, max_decimal_places=2):
''' 
Print a pretty string for a timedelta. 
For example datetime.timedelta(days=2, seconds=17280) will be printed as '2 days, 4 hours, 48 minutes'. Setting max_components to e.g. 1 will change this to '2.2 days', where the 
number of decimal points can also be set. 
'''
time_scales = [timedelta(days=365), timedelta(days=1), timedelta(hours=1), timedelta(minutes=1), timedelta(seconds=1), timedelta(microseconds=1000), timedelta(microseconds=1)]
time_scale_names_dict = {timedelta(days=365): 'year',  
                         timedelta(days=1): 'day', 
                         timedelta(hours=1): 'hour', 
                         timedelta(minutes=1): 'minute', 
                         timedelta(seconds=1): 'second', 
                         timedelta(microseconds=1000): 'millisecond', 
                         timedelta(microseconds=1): 'microsecond'}
count = 0
txt = ''
first = True
for scale in time_scales:
    if t >= scale: 
        count += 1
        if count == max_components:
            n = t / scale
        else:
            n = int(t / scale)
            
        t -= n*scale
        
        n_txt = str(round(n, max_decimal_places))
        if n_txt[-2:]=='.0': n_txt = n_txt[:-2]
        txt += '{}{} {}{}'.format('' if first else ', ', n_txt, time_scale_names_dict[scale], 's' if n>1 else '', )
        if first:
            first = False
        
        
if len(txt) == 0: 
    txt = 'none'
return txt

他已经有一个timedelta对象，所以为什么不使用其内置方法total_seconds()将其转换为秒，然后使用divmod()获得小时和分钟?

hours, remainder = divmod(myTimeDelta.total_seconds(), 3600)
minutes, seconds = divmod(remainder, 60)

# Formatted only for hours and minutes as requested
print '%s:%s' % (hours, minutes)

无论时间增量是天还是年，这都是有效的。

>>> str(datetime.timedelta(hours=10.56))
10:33:36

>>> td = datetime.timedelta(hours=10.505) # any timedelta object
>>> ':'.join(str(td).split(':')[:2])
10:30

将timedelta对象传递给str()函数调用的格式化代码与输入print td相同。因为你不想要秒，我们可以用冒号分隔字符串(3部分)，然后只用前2部分把它重新组合在一起。