Question

I’m running Python 2.7.2 (64 Bit) on Windows 7. I’m a little confused about ‘universal newline mode’ documented here: http://docs.python.org/library/functions.html#open

From the documentation it seems ‘universal newline mode’ should not be in effect unless a ‘U’ is specified in the mode parameter of open(). However I see that as the default behavior! So is the documentation indeed misleading or am I missing something?

f = open("c:/Temp/test.txt", "wb")
f.write("One\r\nTwo\r\nThree\r\nFour"); f.close()

f = open("c:/Temp/test.txt", "rb")
f.read(); f.close()
'One\r\nTwo\r\nThree\r\nFour'

f = open("c:/Temp/test.txt", "r")
f.read(); f.close()
'One\nTwo\nThree\nFour'

f = open("c:/Temp/test.txt", "rt")
f.read(); f.close()
'One\nTwo\nThree\nFour'

f = open("c:/Temp/test.txt", "rU")
f.read(); f.close()
'One\nTwo\nThree\nFour'

Seems like “r”, “rt”, “rU” all have the same behavior?

Answer 1

You’re observing this because \r\n is the line terminator on Windows, so the t mode converts it to \n. On Unix (MacOS here), t doesn’t affect \r\n and there’s no conversion. The difference between t and U is that U converts \r\n and \r to \n on every platform, while t is platform dependent and only converts LT for the given platform.

Replace your test string to "One\r\nTwo\nThree\rFour" to see the effect of U.