python字符串格式化操作

Question

Faulty code:

pos_1 = 234
pos_n = 12890
min_width = len(str(pos_n)) # is there a better way for this?

# How can I use min_width as the minimal width of the two conversion specifiers?
# I don't understand the Python documentation on this :(
raw_str = '... from %(pos1)0*d to %(posn)0*d ...' % {'pos1':pos_1, 'posn': pos_n}

Required output:

... from 00234 to 12890 ...

           ______________________EDIT______________________

New code:

# I changed my code according the second answer
pos_1 = 10234 # can be any value between 1 and pos_n
pos_n = 12890
min_width = len(str(pos_n))

raw_str = '... from % *d to % *d ...' % (min_width, pos_1, min_width, pos_n)

New Problem:

There is one extra whitespace (I marked it _ ) in front of the integer values, for intigers with min_width digits:

print raw_str
... from _10234 to _12890 ...

Also, I wonder if there is a way to add Mapping keys?

Answer 1

pos_1 = 234
pos_n = 12890
min_width = len(str(pos_n))

raw_str = '... from %0*d to %0*d ...' % (min_width, pos_1, min_width, pos_n)

Answer 2

"1234".rjust(13,"0")

Should do what you need

addition:

a = ["123", "12"]    
max_width = sorted([len(i) for i in a])[-1]

put max_width instead of 13 above and put all your strings in a single array a (which seems to me much more usable than having a stack of variables).

additional nastyness: (Using array of numbers to get closer to your question.)

a = [123, 33, 0 ,223]
[str(x).rjust(sorted([len(str(i)) for i in a])[-1],"0") for x in a]

Who said Perl is the only language to easily produce braindumps in? If regexps are the godfather of complex code, then list comprehension is the godmother.

(I am relatively new to python and rather convinced that there must be a max-function on arrays somewhere, which would reduce above complexity. .... OK, checked, there is. Pity, have to reduce the example.)

[str(x).rjust(max([len(str(i) for i in a]),"0") for x in a]

And please observe below comments on "not putting calculation of an invariant (the max value) inside the outer list comprehension".

Answer 3

Concerning using a mapping type as second argument to '%':

I presume you mean something like that '%(mykey)d' % {'mykey': 3} , right?! I think you cannot use this if you use the "%*d" syntax, since there is no way to provide the necessary width arguments with a dict.

But why don't you generate your format string dynamically:

fmt = '... from %%%dd to %%%dd ...' % (min_width, min_width)
# assuming min_width is e.g. 7 fmt would be: '... from %7d to %7d ...'
raw_string = fmt % pos_values_as_tuple_or_dict

This way you decouple the width issue from the formatting of the actual values, and you can use a tuple or a dict for the latter, as it suits you.