Get error code cmd

• Overview
• Part 1 – Getting Started
• Part 2 – Variables
• Part 3 – Return Codes
• Part 4 – stdin, stdout, stderr
• Part 5 – If/Then Conditionals
• Part 6 – Loops
• Part 7 – Functions
• Part 8 – Parsing Input
• Part 9 – Logging
• Part 10 – Advanced Tricks

Today we’ll cover return codes as the right way to communicate the outcome of your script’s execution to the world. Sadly, even
skilled Windows programmers overlook the importance of return codes.

Return Code Conventions

By convention, command line execution should return zero when execution succeeds and non-zero when execution fails. Warning messages
typically don’t effect the return code. What matters is did the script work or not?

Checking Return Codes In Your Script Commands

The environmental variable %ERRORLEVEL% contains the return code of the last executed program or script. A very helpful feature is
the built-in DOS commands like ECHO, IF, and SET will preserve the existing value of %ERRORLEVEL%.

The conventional technique to check for a non-zero return code using the NEQ (Not-Equal-To) operator of the IF command:

IF %ERRORLEVEL% NEQ 0 (
  REM do something here to address the error
)

Another common technique is:

IF ERRORLEVEL 1 (
  REM do something here to address the error
)

The ERRORLEVEL 1 statement is true when the return code is any number equal to or greater than 1. However, I don’t use this technique because
programs can return negative numbers as well as positive numbers. Most programs rarely document every possible return code, so I’d rather explicity
check for non-zero with the NEQ 0 style than assuming return codes will be 1 or greater on error.

You may also want to check for specific error codes. For example, you can test that an executable program or script is in your PATH by simply
calling the program and checking for return code 9009.

SomeFile.exe
IF %ERRORLEVEL% EQU 9009 (
  ECHO error - SomeFile.exe not found in your PATH
)

It’s hard to know this stuff upfront – I generally just use trial and error to figure out the best way to check the return code of the program or
script I’m calling. Remember, this is duct tape programming. It isn’t always pretty, but, it gets the job done.

Conditional Execution Using the Return Code

There’s a super cool shorthand you can use to execute a second command based on the success or failure of a command. The first program/script must
conform to the convention of returning 0 on success and non-0 on failure for this to work.

To execute a follow-on command after sucess, we use the && operator:

SomeCommand.exe && ECHO SomeCommand.exe succeeded!

To execute a follow-on command after failure, we use the || operator:

SomeCommand.exe || ECHO SomeCommand.exe failed with return code %ERRORLEVEL%

I use this technique heavily to halt a script when any error is encountered. By default, the command processor will continue executing
when an error is raised. You have to code for halting on error.

A very simple way to halt on error is to use the EXIT command with the /B switch (to exit the current batch script context, and not the command
prompt process). We also pass a specific non-zero return code from the failed command to inform the caller of our script about the failure.

SomeCommand.exe || EXIT /B 1

A simliar technique uses the implicit GOTO label called :EOF (End-Of-File). Jumping to EOF in this way will exit your current script with
the return code of 1.

SomeCommand.exe || GOTO :EOF

Tips and Tricks for Return Codes

I recommend sticking to zero for success and return codes that are positive values for DOS batch files. The positive values are a good idea
because other callers may use the IF ERRORLEVEL 1 syntax to check your script.

I also recommend documenting your possible return codes with easy to read SET statements at the top of your script file, like this:

SET /A ERROR_HELP_SCREEN=1
SET /A ERROR_FILE_NOT_FOUND=2

Note that I break my own convention here and use uppercase variable names – I do this to denote that the variable is constant and should not
be modified elsewhere. Too bad DOS doesn’t support constant values like Unix/Linux shells.

Some Final Polish

One small piece of polish I like is using return codes that are a power of 2.

SET /A ERROR_HELP_SCREEN=1
SET /A ERROR_FILE_NOT_FOUND=2
SET /A ERROR_FILE_READ_ONLY=4
SET /A ERROR_UNKNOWN=8

This gives me the flexibility to bitwise OR multiple error numbers together if I want to record numerous problems in one error code.
This is rare for scripts intended for interactive use, but, it can be super helpful when writing scripts you support but you don’t
have access to the target systems.

@ECHO OFF
SETLOCAL ENABLEEXTENSIONS

SET /A errno=0
SET /A ERROR_HELP_SCREEN=1
SET /A ERROR_SOMECOMMAND_NOT_FOUND=2
SET /A ERROR_OTHERCOMMAND_FAILED=4

SomeCommand.exe
IF %ERRORLEVEL% NEQ 0 SET /A errno^|=%ERROR_SOMECOMMAND_NOT_FOUND%

OtherCommand.exe
IF %ERRORLEVEL% NEQ 0 (
    SET /A errno^|=%ERROR_OTHERCOMMAND_FAILED%
)

EXIT /B %errno%

If both SomeCommand.exe and OtherCommand.exe fail, the return code will be the bitwise combination of 0x1 and 0x2, or decimal 3. This return code tells
me that both errors were raised. Even better, I can repeatedly call the bitwise OR with the same error code and still interpret which errors were
raised.

<< Part 2 – Variables


Part 4 – stdin, stdout, stderr >>

By default when a command line execution is completed it should either return zero when execution succeeds or non-zero when execution fails. When a batch script returns a non-zero value after the execution fails, the non-zero value will indicate what is the error number. We will then use the error number to determine what the error is about and resolve it accordingly.

Following are the common exit code and their description.

Error Level

The environmental variable %ERRORLEVEL% contains the return code of the last executed program or script.

By default, the way to check for the ERRORLEVEL is via the following code.

Syntax

IF %ERRORLEVEL% NEQ 0 ( 
   DO_Something 
)

It is common to use the command EXIT /B %ERRORLEVEL% at the end of the batch file to return the error codes from the batch file.

EXIT /B at the end of the batch file will stop execution of a batch file.

Use EXIT /B < exitcodes > at the end of the batch file to return custom return codes.

Environment variable %ERRORLEVEL% contains the latest errorlevel in the batch file, which is the latest error codes from the last command executed. In the batch file, it is always a good practice to use environment variables instead of constant values, since the same variable get expanded to different values on different computers.

Let’s look at a quick example on how to check for error codes from a batch file.

Example

Let’s assume we have a batch file called Find.cmd which has the following code. In the code, we have clearly mentioned that we if don’t find the file called lists.txt then we should set the errorlevel to 7. Similarly, if we see that the variable userprofile is not defined then we should set the errorlevel code to 9.

if not exist c:lists.txt exit 7 
if not defined userprofile exit 9 
exit 0

Let’s assume we have another file called App.cmd that calls Find.cmd first. Now, if the Find.cmd returns an error wherein it sets the errorlevel to greater than 0 then it would exit the program. In the following batch file, after calling the Find.cnd find, it actually checks to see if the errorlevel is greater than 0.

Call Find.cmd

if errorlevel gtr 0 exit 
echo “Successful completion”

Output

In the above program, we can have the following scenarios as the output −

• If the file c:lists.txt does not exist, then nothing will be displayed in the console output.

• If the variable userprofile does not exist, then nothing will be displayed in the console output.

• If both of the above condition passes then the string “Successful completion” will be displayed in the command prompt.

Loops

In the decision making chapter, we have seen statements which have been executed one after the other in a sequential manner. Additionally, implementations can also be done in Batch Script to alter the flow of control in a program’s logic. They are then classified into flow of control statements.

Looping through Command Line Arguments

The ‘for’ statement can also be used for checking command line arguments. The following example shows how the ‘for’ statement can be used to loop through the command line arguments.

Example

@ECHO OFF 
:Loop 

IF "%1"=="" GOTO completed 
FOR %%F IN (%1) DO echo %%F 
SHIFT 
GOTO Loop 
:completed

Output

Let’s assume that our above code is stored in a file called Test.bat. The above command will produce the following output if the batch file passes the command line arguments of 1,2 and 3 as Test.bat 1 2 3.

1 
2 
3

У командный процессора cmd.exe есть такая вещь — уровень ошибки (error level). Это код выхода (exit code) программы, которую вы запускали последней. Проверить уровень ошибки можно при помощи команды IF ERRORLEVEL.

IF ERRORLEVEL 1 ECHO error level is 1 or more

<sidebar>
Проверка IF ERROR LEVEL n срабатывает, если уровень ошибки n или выше. Это, вероятно, потому, что многие программы выражают разную степень ошибки все большими и большими кодах выхода. К примеру, программа diff имеет 3 кода выхода: «0» означает, что файлы одинаковые, «1» — разные, «2» — случилось что-то страшное. Некоторые программы используют код выхода «0» для успеха и все остальное для ошибки.
</sidebar>

Вдобавок к этому внутреннему состоянию, вы, если хотите, можете создать переменную окружения с именем ERRORLEVEL, так же, как вы можете создать переменную с именем FRED. Но, как и FRED, эта переменная не повлияет на уровень ошибки.

rem this next command sets the error level to zero
CMD /C EXIT 0
set ERRORLEVEL=1
if ERRORLEVEL 1 echo Does this print?


Сообщение не будет отображено, поскольку переменная ERRORLEVEL не имеет никакого влияния на уровень ошибки. Это просто переменная, имя которой совпадает с концепцией командного процессора.

set BANKBALANCE=$1 000 000,00

«Эй, когда я пытаюсь снять денег, у меня ошибка — „недостаточно денег на счету“».

Однако, есть вариант, когда включено расширенный режим командного процессора, и вы используете %ERRORLEVEL%.
В этом случае командный процессор ищет переменную с таким именем и, если не находит, заменяет %ERRORLEVEL% на текущее значение внутреннего уровня ошибки. Это запасной вариант — как указать адрес соседа запасным адресом доставки товара, на случай, если вас нет дома. Однако это не повлияет на посылки, доставляемые соседу.

То же поведение и у %CD%: если вы не установили переменную с таким именем, подставляется текущий каталог командного процессора. Но изменить каталог при помощи set CD=C:Windows нельзя.

Вероятно, есть несколько причин для такого поведения:
— Чтобы можно было вывести уровень ошибки в лог:
ECHO error level is %ERRORLEVEL%>logfile
— Чтобы можно было выполнять другие сравнения с уровнем ошибки — например, чтобы проверять равенство:
IF %ERRORLEVEL% EQU 1 echo Different!

Но я отклонился от темы. На сегодня мой тезис такой: уровень ошибки — это не то же самое, что переменная %ERRORLEVEL%.

Провел небольшой тест.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

@echo off
Setlocal EnableDelayedExpansion
set flash=f:
set videofile=Cougar.Town.S01E01.rus.LostFilm.TV.avi
 
Echo Тест успешного копирования
copy nul nul
echo errorlevel %errorlevel%
Echo Тест копирования в защищенную правами папку
cmd /c copy c:test.log c:Windowskmsem1.log
echo errorlevel %errorlevel%
Echo Тест копирования несуществующего файла
cmd /c copy abracadabra L:*.*
echo errorlevel %errorlevel%
verify off
if exist %flash%%videofile% (
  del %flash%%videofile%& ping 127.1>nul
)
Echo Тест копирования через XCOPY прав доступа ACL без административных привилегий
XCOPY /O i:%videofile% %flash%
echo errorlevel %errorlevel%
echo.
Echo Тест копирования видеофайла (268 MB) на флеш
Set STime=%time%
copy i:%videofile% %flash%
echo errorlevel %errorlevel%
Set ETime=%time%
Call :TimeElapsed "%STime%" "%ETime%" ret
Echo Прошло: %ret% с.
del %flash%%videofile%
ping 4 127.1>nul
Set STime=%time%
Echo Тест копирования видеофайла (268 MB) на флеш с верификацией записи
copy /v i:%videofile% %flash%
echo errorlevel %errorlevel%
Set ETime=%time%
Call :TimeElapsed "%STime%" "%ETime%" ret
Echo Прошло: %ret% с.
pause
goto :eof
 
 
:TimeElapsed %1-StartTime %2-EndTime %3-var_result
Call :TimeToMSec "%~1" TimeS_ms
Call :TimeToMSec "%~2" TimeE_ms
Set /A diff=TimeE_ms-TimeS_ms
Set /A diffSS=diff/100
Set /A diffms=%diff% %% 100
Set %3=%diffSS%,%diffms%
Exit /B
 
:TimeToMSec %1-Time 2-var_mSec
For /F "Tokens=1-4 Delims=,:" %%A in ("%~1") do (
  Set /A HH=%%A
  Set MM=1%%B& Set /A MM=!MM!-100
  Set SS=1%%C& Set /A SS=!SS!-100
  Set mS=1%%D& Set /A mS=!mS!-100
)
Set /A %~2=(HH*60*60+MM*60+SS)*100+mS
Exit /B

Один раз получил странную ошибку. Copy скопировала файл, после чего написала, что «я такого имени не знаю…» =)) Error 9009

По итогам +Errorlevel:

xcopy

errorlevel = 0, даже, если пользователь ответил «Не заменять файл».

Также видим, что в случаях с защищенной правами папкой или когда файл не существует
мы получаем Errorlevel 1 (а не ожидаемые дефолтовые 5 и 2 соответственно).

copy

errorlevel = 0, даже если пользователь ответил «Не заменять файл»
Чтобы консоль задала этот вопрос:
1) для одиночного файла нужно указать ключ /-y

2) для маски файлов работает по-умолчанию. Обратная операция — «принудительная замена»: ключ /y

Сравнение copy с verify off
1) без ключа
2) с ключем /v
показало, что в большинстве случаев с верификацией для копирования файла в ~250 МБ затрачивается на ~1,5 сек. больше времени. Не знаю, что именно оно проверяет, но уж файл точно целиком не считывается.

7 ответов

echo Exit Code is %errorlevel%

if errorlevel

@echo off
my_nify_exe.exe
if errorlevel 1 (
   echo Failure Reason Given is %errorlevel%
   exit /b %errorlevel%
)

start /wait something.exe
echo %errorlevel%

@echo off
my_nify_exe.exe
if %ERRORLEVEL% EQU 0 (
   echo Success
) else (
   echo Failure Reason Given is %errorlevel%
   exit /b %errorlevel%
)

#include "stdafx.h"
#include "windows.h"
#include "stdio.h"
#include "tchar.h"
#include "stdio.h"
#include "shellapi.h"

int _tmain( int argc, TCHAR *argv[] )
{

    CString cmdline(GetCommandLineW());
    cmdline.TrimLeft('"');
    CString self(argv[0]);
    self.Trim('"');
    CString args = cmdline.Mid(self.GetLength()+1);
    args.TrimLeft(_T("" "));
    printf("Arguments passed: '%ws'n",args);
    STARTUPINFO si;
    PROCESS_INFORMATION pi;

    ZeroMemory( &si, sizeof(si) );
    si.cb = sizeof(si);
    ZeroMemory( &pi, sizeof(pi) );

    if( argc < 2 )
    {
        printf("Usage: %s arg1,arg2....n", argv[0]);
        return -1;
    }

    CString strCmd(args);
    // Start the child process. 
    if( !CreateProcess( NULL,   // No module name (use command line)
        (LPTSTR)(strCmd.GetString()),        // Command line
        NULL,           // Process handle not inheritable
        NULL,           // Thread handle not inheritable
        FALSE,          // Set handle inheritance to FALSE
        0,              // No creation flags
        NULL,           // Use parent environment block
        NULL,           // Use parent starting directory 
        &si,            // Pointer to STARTUPINFO structure
        &pi )           // Pointer to PROCESS_INFORMATION structure
    ) 
    {
        printf( "CreateProcess failed (%d)n", GetLastError() );
        return GetLastError();
    }
    else
        printf( "Waiting for "%ws" to exit.....n", strCmd );

    // Wait until child process exits.
    WaitForSingleObject( pi.hProcess, INFINITE );
    int result = -1;
    if(!GetExitCodeProcess(pi.hProcess,(LPDWORD)&result))
    { 
        printf("GetExitCodeProcess() failed (%d)n", GetLastError() );
    }
    else
        printf("The exit code for '%ws' is %dn",(LPTSTR)(strCmd.GetString()), result );
    // Close process and thread handles. 
    CloseHandle( pi.hProcess );
    CloseHandle( pi.hThread );
    return result;
}

usage: logit.sh [-h] [-p] [-i=n] [-s] <description>
example: logit.sh -p error -i 501 -s myscript.sh "failed to run the mount command"

LGT_TEMP_FILE="$(mktemp --suffix .cmd)"
cat<<EOF>$LGT_TEMP_FILE
    @echo off
    set LGT_EXITCODE="$LGT_ID"
    exit /b %LGT_ID%
EOF
unix2dos "$LGT_TEMP_FILE"

__create_event () {
    local cmd="eventcreate /ID $LGT_ID /L Application /SO $LGT_SOURCE /T $LGT_PRIORITY /D "
    if [[ "$1" == *';'* ]]; then
        local IFS=';'
        for i in "$1"; do
            $cmd "$i" &>/dev/null
        done
    else
        $cmd "$LGT_DESC" &>/dev/null
    fi
}

cmd /c "$(cygpath -wa "$LGT_TEMP_FILE")"
__create_event