fix line 71 IndexError at Windows

[sunrabbit123]

I used this app
but I get this bug

Traceback (most recent call last):
  File "C:\Users\user\AppData\Roaming\Python\Python39\Scripts\wifi-password-script.py", line 33, in <module>
    sys.exit(load_entry_point('wifi-password==1.0.7', 'console_scripts', 'wifi-password')())
  File "C:\Users\user\AppData\Roaming\Python\Python39\site-packages\wifi_password\wifi_password.py", line 116, in main
    password = get_password(args.ssid)
  File "C:\Users\user\AppData\Roaming\Python\Python39\site-packages\wifi_password\wifi_password.py", line 76, in get_password
    password = re.findall(r"Key Content\s+:\s(.*)", password)[0]
IndexError: list index out of range

That's why I added this.

def get_password(ssid):
    if ssid == "":
        print_error("SSID is not defined")

    if sys.platform == "darwin":
        password = run_command(f"security find-generic-password -l \"{ssid}\" -D 'AirPort network password' -w")
        password = password.replace("\n", "")

    elif sys.platform == "linux":
        # Check if the user is running with super user privilages
        if os.geteuid() != 0:
            password = run_command(f"sudo nmcli -s -g 802-11-wireless-security.psk connection show '{ssid}'")
        else:
            password = run_command(f"nmcli -s -g 802-11-wireless-security.psk connection show '{ssid}'")

        password = password.replace("\n", "")

    elif sys.platform == "win32":
        password = run_command(f"netsh wlan show profile name=\"{ssid}\" key=clear | findstr Key").replace("\r", "")
        try:
            password = re.findall(r"Key Content\s+:\s(.*)", password)[0]
        except IndexError:
            password = ""
    if password == "":
        print_error("Could not find password")

    return password

[sdushantha]

Thank you for bringing this up.

Do you know why this error is happening? Is it because the WiFi does not have a password?

[sunrabbit123]

In my case, the OS language is not English.

I'm from korea, So I use Korean.

Engilsh case:

Key Content : !@#!$!@#

Korean Case:

키 콘텐츠 : !@#!@#

Different languages display differently.
It looks good to get OS language and classify it automatically.
Korean example is here

[sdushantha]

This is interesting. Does #25 happen to fix the issue you are facing?

[sunrabbit123]

Probably.

However, it is difficult for me to understand the solution of #25.

Can I ask you to explain the code of #25?
What does LANG=C cause?

[sdushantha]

@sunrabbit123 The 'LANG' environment variable indicates the language/locale and encoding, where "C" is the default language setting, English.

[sunrabbit123]

I understand. Thank you.
Let's think about LANG=C.

But I think the solution is to make a big dictionary to get keywords that fit your language.

Examples are below.

keyword_dict = {
    "en_US" : ["Key", "Key Contents"],
    "ko_KR" : ["키" : "키 콘텐츠"]
    .
    .
    .
}
key = keyword_dict["en_US"]

[sdushantha]

Thank you for your suggestion.

Could you try replacing the current run_function function with the one below and let me know if that worked for you?

def run_command(command):
    env = os.environ.copy()
    env["LANG"] = "C"
    output, _ = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL, shell=True).communicate()
    return output.decode("utf-8").rstrip('\r\n')

I think setting the locale would be a much better approach at fixing this issue as less code is needed to do so.

[sunrabbit123]

LANG=C didn't work for me.

The ploblem is netsh wlan show profiles name=SSID key=clear | findstr Key

netsh wlan show interfaces | findstr SSID
    SSID                   : SK_WiFiGIGA42C0
    BSSID                  : 00:23:aa:dc:42:c3

netsh wlan show profiles name="SK_WiFiGIGA42C0" key=clear | findstr 키
    보안 키           : 있음
    키 콘텐츠            : password

but

netsh wlan show profiles name="SK_WiFiGIGA42C0" key=clear | findstr Key

ERROR: Could not find password

and

We also found additional problems.
When decoding with utf-8, the following error appears.

Traceback (most recent call last):
  File "C:\Users\user\AppData\Roaming\Python\Python39\Scripts\wifi-password-script.py", line 33, in <module>
    sys.exit(load_entry_point('wifi-password==1.0.7', 'console_scripts', 'wifi-password')())
  File "C:\Users\user\AppData\Roaming\Python\Python39\site-packages\wifi_password\wifi_password.py", line 124, in main
    password = get_password(args.ssid)
  File "C:\Users\user\AppData\Roaming\Python\Python39\site-packages\wifi_password\wifi_password.py", line 78, in get_password
    password = run_command(f"netsh wlan show profiles name=\"{ssid}\" key=clear | findstr 키").replace("\r", "")
  File "C:\Users\user\AppData\Roaming\Python\Python39\site-packages\wifi_password\wifi_password.py", line 23, in run_command
    return output.decode("utf-8").rstrip('\r\n')
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xba in position 4: invalid start byte

But the solution is simple.
You can decode it as follows:

return output.decode("euc-kr")

[sdushantha]

This means we would have to have a translation of the word "Key" in every language and then decode the output based on the system's language, right?

I'll have to think about this because having a dictionary of the translation of all languages seems to be a waste of lines of code in my opinion. But if there is no other option, we'll have to stick with that then.

[sunrabbit123]

I think so.

However, rather than supporting all languages, I think it is better to provide basic support for English, but to provide opportunities for contributors to participate.

You can know your own language with a simple Python script.

The code is as follows:

import locale
lang, _ = locale.getdefaultlocale()
print(lang)

Windows supports approximately 106 languages.
I don't think it's a big file even if it's symmetrical with the key.

I think it would be good to use json.

Thank you for reviewing my opinion.
But I wish I had a better way.

Because my method is not very efficient.

[sdushantha]

Could you try this instead?

def run_command(command):
    env = os.environ.copy()
    env["LC_ALL"] = "C"
    output, _ = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL, shell=True).communicate()
    return output.decode("utf-8").rstrip('\r\n')

Does this work? The change I made is env["LC_ALL"] = "C". According to this AskUbuntu post, it should set the output of the commands to English.

[sunrabbit123]

Unfortunately, it doesn't work on Windows.

However, I will find a way to change the output of the command to English

[sdushantha]

Most of the command line tools are translated to local languages, but from time to time user may want the output of command line tools to be in English. For Windows 7 system on languages such as East Asian (Japanese, Korean, Chinese Simplify and Traditional), Russian and Greek there is a trick to do that.

1. Start console window on localized version of Windows 7.

2. Change code page in console window to English by typing chcp 437 and press Enter.

3. Run your application and you should see output is in English.

Source: https://channel9.msdn.com/Blogs/witastre/How-to-see-command-line-tools-output-in-English-on-East-Asian-Russian-and-Greek-language-of-Windows-

[sunrabbit123]

run_command("chcp 437")

It works very sexy.
However, it is a problem that the page is updated.
This 'chcp' command cannot be modified because it has only one language code.

I think you can put this code at the beginning of the main function.

def main():
    run_command("chcp 437")
    .
    .
    .
    .

[celianvdb]

Could you try this instead?

def run_command(command):
    env = os.environ.copy()
    env["LC_ALL"] = "C"
    output, _ = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL, shell=True).communicate()
    return output.decode("utf-8").rstrip('\r\n')

Does this work? The change I made is env["LC_ALL"] = "C". According to this AskUbuntu post, it should set the output of the commands to English.

After created the env variable you need to apply it for your subprocess. To do that you have to add env as parameters in subprocess.Popen() methods by adding env=env. So it becomes :

 def run_command(command):
     env = os.environ.copy()
     env["LC_ALL"] = "C"
     output, _ = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL, shell=True, env=env).communicate()
     return output.decode("utf-8").rstrip('\r\n')

[celianvdb]

run_command("chcp 437")

It works very sexy.
However, it is a problem that the page is updated.
This 'chcp' command cannot be modified because it has only one language code.

I think you can put this code at the beginning of the main function.

def main():
    run_command("chcp 437")
    .
    .
    .
    .

I don't think that's a good idea because chcp doesn't exist on linux, by doing that, wifi-password will crash/throw an error on linux.

[sdushantha]

@celianvdb

I don't think that's a good idea because chcp doesn't exist on linux, by doing that, wifi-password will crash/throw an error on Linux.

I agree with you, it is better to put it in the areas where Windows commands get executed. The only reason why I said to put it under the main function was for testing purposes.

[sdushantha]

The language issue has been fixed in #45