使用字典和转义字符时奇怪的 Python 行为

Question

I'm new to Python, and I'm trying to perform simple tasks the way I used to, but I've faced an interesting... feature?

The code below works just how I want it to:

def cleanLDAP(search):
  escChars = {'(':r'\28', ')':r'\29' }
  for ch, val in escChars.items():
    if ch in search:
      search = search.replace(ch, val)
  return search

cleanLDAP('(123)')

The output is '\\\\28123\\\\29' as I expect, but when I change escChars as follows:

escChars = {'(':r'\28', ')':r'\29', '\\': '\5c' }

the output become a bit weird: '\\x05c28123\\x05c29'

I understand that I might miss some implicit encoding changes, but still I want to know the reason why does this happening? Thank you in advance!

Answer 1

5c in utf-8 is \\ .

When you try save a string as \\5c Python returns the utf-8 hex for 5c since you prefixed the string with \\ this causes your value to become: \\x05c .

'\5c'
#'\x05c'

'5c'
#'5c'

escChars
#{'(': '\\28', ')': '\\29', '\\': '\x05c'}

When you iterate over your keys, the ch it tests against in your iteration is \\ , because you did not save your key with the raw format string r .

for ch, value in escChars.items(): 
    print(ch, value)

#( \28
#) \29
#\ c

Finally, since you are modifying your everytime you find a match during iteration, you're checking if the \\ exists after you added it in via replace()

This leads you to do your first replacement, then replaces the \\ you inserted into the string with the utf-8 symbol for \\ .

The simple fix here is to save your key with the r to ensure the code will only match against \\\\ and not \\ , and save your value with the same to ensure it does not get converted to hex.

def cleanLDAP(search):
    escChars = {'(':r'\28', ')':r'\29', r'\\': r'\5c' }
    for ch, val in escChars.items():
        if ch in search:
            search = search.replace(ch, val)
    return search

>>> cleanLDAP('(123)')

#'\\28123\\29'

Answer 2

Change to -

escChars = {'(':r'\28', ')':r'\29', '\\': r'\5c' }

You missed adding r'\\5c' and just did '\\5c' . This makes it hexadecimal.

To understand with an example -

a='\5'
a
ord(a)

Returns '\\x05' and 5 respectively