Error :cannot use a string pattern on a bytes-like object

Question

Hy am using Python RegEx to show all internet wirless profiles connected to a computer.There is error (TypeError: cannot use a string pattern on a bytes-like object) in my Second last line pls anyone help to identifi my mistake.Thanks

My Program

import subprocess,re
command = "netsh wlan show profile"
output = subprocess.check_output(command, shell=True)  
network_names = re.search("(Profile\s*:\s)(.*)", output)  
print(network_names.group(0))

.....................................................

ERROR

line 8, in <module>


 return _compile(pattern, flags).search(string)


TypeError: cannot use a string pattern on a bytes-like object

You could try str(output) in your re.search or output.decode('utf-8') maybe? — Chris
– Chris, Commented May 11, 2020 at 0:03
output = output.decode() ? subprocess return bytes and you have to manually convert to string/unicode (using default 'utf-8' or other encoding - ie. decode('latin1') - if system uses different encoding then utf-8) — furas
– furas, Commented May 11, 2020 at 0:03

Jiří Baum · Accepted Answer · 2020-05-11 00:15:49Z

23

Python 3 distinguishes "bytes" and "string" types; this is especially important for Unicode strings, where each character may be more than one byte, depending on the character and the encoding.

Regular expressions can work on either, but it has to be consistent — searching for bytes within bytes, or strings within strings.

Depending on what you need, there are two solutions:

Decode the output variable before searching in it; for instance, with: output_text = output.decode('utf-8')

This depends on the encoding that you are using; UTF-8 is the most common these days.

The matched group will be a string.
Search with bytes by adding a b prefix to the regular expression. A regular expression should also use the r prefix, so it becomes: re.search(br"(Profile\s*:\s)(.*)", output)

The matched group will be a bytes object.

answered May 11, 2020 at 0:15

Jiří Baum

6,9882 gold badges19 silver badges19 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

0xsteve Over a year ago

Decoding the output variable before searching worked for me..

drootang · Accepted Answer · 2020-05-11 00:10:21Z

3

From the documentation for Popen.stdout:

If the stdout argument was PIPE, this attribute is a readable stream object as returned by open(). Reading from the stream provides output from the child process. If the encoding or errors arguments were specified or the universal_newlines argument was True, the stream is a text stream, otherwise it is a byte stream. If the stdout argument was not PIPE, this attribute is None.

So without setting these options you get a byte stream.

subprocess.check_output supports an encoding keyword argument. Set this to 'utf8' and you will get a text stream:

output = subprocess.check_output(command, shell=True, encoding='utf8')

answered May 11, 2020 at 0:10

drootang

2,5001 gold badge23 silver badges33 bronze badges

Comments

Zhd Zilin · Accepted Answer · 2020-05-11 00:07:35Z

0

I tried the same code on my computer with python 2.7. Works perfect.

Output is a str object on my side.

I think you can add a line after this code "output = subprocess.check_output(command, shell=True)", the line is print(type(output)).

You may see the real data type, if it's not str, try to use output = str(output) to convert it to str

answered May 11, 2020 at 0:07

Zhd Zilin

1337 bronze badges

3 Comments

furas Over a year ago

Python 2 treats bytes as string but Python 3 doesn't treast bytes as string

Zhd Zilin Over a year ago

So I said use str method to convert to str

Jiří Baum Over a year ago

The downside of using output = str(output) is that (a) it'll add b' and ' marks around the text, and (b) it won't work well for accented characters, emoji, etc. For instance, instead of café it'll print out b'caf\xc3\xa9' Using the .decode() method will treat all these characters correctly.

L Tyrone · Accepted Answer · 2024-04-05 02:48:24Z

I recently had a similar issue. I was trying to convert an input from a .csv table into a number, removing the '£' prefix.

e.g. £25.68 need to be 25.68

The .csv file was imported with latin1 encoding due to '£' not being readable by pd.pandas. This meant values were bytes, these just needed to be converted to strings.

OutgoingArray = ['£3.13', '£11.50', '£5.90', '£4.72']
    iteration=0
    temp = []
    for O in OutgoingArray:  # For loop cycles through 'Amount' column and removes '£' from number.
        #print(O) # Prints the current value from statement being processed
        temp = (re.findall('\d+.+\d+', str(O) )) # Identifies the number in the string 
        # "str(O)" is the bit that fixed my code
        #print(temp[0])
        OutgoingArray[iteration] = float(temp[0]) # Replaces string with prefix '£' with a float value.
        iteration += 1

Output:

OutgoingArray >> [3.13, 11.5, 5.9, 4.72]

Collectives™ on Stack Overflow

Error :cannot use a string pattern on a bytes-like object

4 Answers 4

1 Comment

Comments

3 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

1 Comment

Comments

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related