解析文件名中反斜杠和正斜杠的混合

Question

I am getting filename from an api in this format containing mix of / and \\ .

infilename = 'c:/mydir1/mydir2\\mydir3\\mydir4\\123xyz.csv'

When I try to parse the directory structure, \\ followed by a character is converted into single character.

Is there a way around to get each component correctly?

What I already tried:

path.normpath didn't help. 

infilename = 'c:/mydir1/mydir2\mydir3\mydir4\123xyz.csv'
os.path.normpath(infilename)

out:
'c:\\mydir1\\mydir2\\mydir3\\mydir4Sxyz.csv'

Answer 1

that's not visible in your example but writing this:

infilename = 'c:/mydir1/mydir2\mydir3\mydir4\123xyz.csv'

isn't a good idea because some of the lowercase (and a few uppercase) letters are interpreted as escape sequences if following an antislash. Notorious examples are \\t , \\b , there are others. For instance:

infilename = 'c:/mydir1/mydir2\thedir3\bigdir4\123xyz.csv'

doubly fails because 2 chars are interpreted as "tab" and "backspace".

When dealing with literal Windows-style path (or regexes), you have to use the raw prefix, and better, normalize your path to get rid of the slashes.

infilename = os.path.normpath(r'c:/mydir1/mydir2\mydir3\mydir4\123xyz.csv')

However, the raw prefix only applies to literals . If the returned string appears, when printing repr(string) , as 'the\\terrible\\\\dir' , then tab chars have already been put in the string, and there's nothing you can do except a lousy post-processing.

Answer 2

use r before the string to process it as a raw string (ie no string formatting).

eg

infilename = r'C:/blah/blah/blah.csv'

More details here: https://docs.python.org/3.6/reference/lexical_analysis.html#string-and-bytes-literals

Answer 3

Instead of parsing by \\ try parsing by \\\\ . You usually have to escape by \\ so the \\ character is actually \\\\ .