str vs repr

[Skaperen]

besides the difference in how they handle strings, is there anything in which str() will handle differently than repr()?  i have the idea of simulating the print() function in my columnizer class, and applying str() to each argument in the col.print() method call (so that col.print(foo) formats like print(foo)).

[Ofnuts]

If the intention is readability, use str(). If it is unambiguousness use repr(). See https://stackoverflow.com/questions/1436...on#2626364

[Skaperen]

the intention is to make the .print() method in class columnize behave in a manner similar to how print() behaves, in terms of how it forms the line to be printed.  for example print(9) outputs '9' plus the line end character to go to the next line, so i want columnize.print(9) to "output" '9' to the current line in the current column, and go to the next line.  i am thinking that applying the str() function to each argument and doing sep_option.join() with a list of those would achieve the same effect.  the sep= and end= options would supported and the file= and flush= options would make no sense and raise an exception.

[wavic]

In [1]: def say(text):
   ...:     print(text, end='')
   ...:     

In [2]: print('text')
text
                       # new line
In [3]: say('text')
text                  # no additional new line 
In [4]: 

[Skaperen]

In [1]: def say(text):
   ...:     print(text, end='')
   ...:     

In [2]: print('text')
text
                       # new line
In [3]: say('text')
text                  # no additional new line 
In [4]: 

which means i need to implement end= and handle '\n' the same way, to do it up right.

[wavic]

Yes. If you need the line not to end with '\n' it could end with nothing or a dollar sign for example.
This is why one get a different result in this situation:

$ echo Tom | md5sum

vs

$ echo -n Tom | md5sum

The first one add a new line after Tom and md5sum is calculating 'Tom\n'
But you know echo

[Skaperen]

as long as i make the defaults be sep=' ' and end='\n' i think it should be OK.

[Skaperen]

str(chr(256)) != repr(chr(256))  

---

it looks like i want to use repr() in python2 and use ascii() in python3.  so i think i can do:

if sys.version_info.major < 3:
   ascii = repr

and just use ascii() to get the escape sequence conversion i want.

[Ofnuts]

Better:

if sys.version_info.major < 3:
   readable = repr
else:
   readable= ascii

[Skaperen]

Better:

if sys.version_info.major < 3:
   readable = repr
else:
   readable= ascii

i don't think i like that name. anyone have ideas for another? it should be understandable by py3 coders. if they see it in py3 code they should be able to understand what it does, in the context of actual use.