Bat обработка ошибок

Every scripting and programming language contains an error handler like Java contains try-catch for error handling. In a Batch script, there is no direct way to do this, but we can create an error handler in the Batch script using a built-in variable of the Batch script name %ERRORLEVEL%.

This article will show how we can create a Batch script to handle errors and failures. Also, we are going to some examples that make the topic easier.

Error Handling in Batch Script

When a command successfully executes, it always returns an EXIT CODE that indicates whether the command successfully executed or failed to execute. So, to create an error handling in a Batch file, we can use that EXIT CODE in our program.

You can follow below general format to create an error handler:

@Echo off
SomeCommand && (
  ECHO Message for Success
) || (
  ECHO Message for Failure or Error
)

We can also do that by checking the variable named %ERRORLEVEL%. If the variable contains a value not equal to 0, then there might be a problem or error when executing the command. To test the %ERRORLEVEL% variable, you can follow the below example codes:

@ECHO off
Some Command Here !!!
IF %ERRORLEVEL% NEQ 0 (Echo Error found when running the command &Exit /b 1)

You must note that the keyword NEQ means Not Equal. And the variable %ERRORLEVEL% will only contain a non-zero value if there is a problem or error in the code.

An Example That Contains Errors

Below, we shared an example. We will run a Batch file named Your_file.bat from a location.

We intentionally removed that file from the directory. So it’s an error command.

The code for our example will be:

@echo off
ECHO Running a Batch file
CD G:BATCH
CALL Your_file.bat
IF  errorlevel 1 GOTO ERROR
ECHO The file run successfully.
GOTO EOF

:ERROR
ECHO The file didn't run successfully.
CMD /k
EXIT /b 1

:EOF

Now, as the file doesn’t exist in the directory, it will show an error, and you will get the below output when you run the code shared above.

Output:

Running a Batch file
The system cannot find the path specified.
'Your_file.bat' is not recognized as an internal or external command,
operable program or batch file.
The file didn't run successfully.

An Error-Free Code Example That Runs Successfully

In the example above, we made a mistake on the code intentionally to understand how the code works. If we correct it like below:

@echo off
ECHO Running a Batch file
CALL "G:BATCHYourfile.bat"
IF  errorlevel 1 GOTO ERROR
ECHO The file runs successfully.
GOTO EOF

:ERROR
ECHO The file didn't run successfully.
CMD /k
EXIT /b 1

:EOF

Then we will get an output like this:

Running a Batch file
This is from the first file
The file runs successfully.

Remember, all commands we discussed here are only for the Windows Command Prompt or CMD environment.

An equivalent for the exit code $? in batch is the errorlevel accessible via the variable %ERRORLEVEL%. In batch scripting, there exists a special form of the IF-clause that uses the errorlevel: IF ERRORLEVEL n is equivalent to: «if the errorlevel is greater or equal to n«. As in bash, a value of 0 means the previous command executed without errors (not always, some commands don’t set the errorlevel). So you have two choices:

1. IF %ERRORLEVEL% NEQ 0 which is the literal translation of your if-statement in your bash script. Be aware you’ll have to use the delayed expansion version IF !ERRORLEVEL! NEQ 0 inside code blocks delimited with (...) (check this question that explains why)
2. IF ERRORLEVEL 1 which is IMHO more suitable (depends on your preferences though). Also works inside (...) code blocks but assumes commands never set negative errorlevels (never encountered one but is worth mentionning)

Normally setting a variable with the set command doesn’t change the errorlevel so you can choose either one of the two options. But as I don’t have a windows machine to test it right now, I’ll play it safe and give you the first option. I’ll edit my answer once I’ve tested the second option.

set exit_code=%ERRORLEVEL%
if %exit_code% NEQ 0 (  
    echo Error calling sqlplus, return code = %exit_code%
    exit /b %exit_code%
)

Some last advice: use exit /b
instead of exit. exit will also exit the command line in batch

EDIT I’ve tested the second version and it works as well. So the code above can be replaced with:

set exit_code=%ERRORLEVEL%
IF ERRORLEVEL 1 (  
    echo Error calling sqlplus, return code = %exit_code%
    exit /b %exit_code%
)

This blog article explains quite well why this version should be used instead of the of the first one using the %ERRORLEVEL% variable.

Another option @aschipfl kindly reminded me of in his comment is the use of the conditional executions operators && and ||:

• command1 && command2 will execute command2 only if command1 succeeded
• command1 || command2 will execute command2 only if command1 failed.

Both operators use the exit code (in most cases the same as errorlevel but not always) to determine if a command failed or succeeded.

I’ve also seen you’ve posted your batch-file. There is a little something I’d like to point out though: goto @ERROR_HANDLING_SQL will throw an error because there is no label @ERROR_HANDLING_SQL. There exists a label ERROR_HANDLING_SQL at the end of your script though. If you were trying to jump to that label you should replace all occurences of goto @ERROR_HANDLING_SQL with goto ERROR_HANDLING_SQL

An equivalent for the exit code $? in batch is the errorlevel accessible via the variable %ERRORLEVEL%. In batch scripting, there exists a special form of the IF-clause that uses the errorlevel: IF ERRORLEVEL n is equivalent to: «if the errorlevel is greater or equal to n«. As in bash, a value of 0 means the previous command executed without errors (not always, some commands don’t set the errorlevel). So you have two choices:

1. IF %ERRORLEVEL% NEQ 0 which is the literal translation of your if-statement in your bash script. Be aware you’ll have to use the delayed expansion version IF !ERRORLEVEL! NEQ 0 inside code blocks delimited with (...) (check this question that explains why)
2. IF ERRORLEVEL 1 which is IMHO more suitable (depends on your preferences though). Also works inside (...) code blocks but assumes commands never set negative errorlevels (never encountered one but is worth mentionning)

Normally setting a variable with the set command doesn’t change the errorlevel so you can choose either one of the two options. But as I don’t have a windows machine to test it right now, I’ll play it safe and give you the first option. I’ll edit my answer once I’ve tested the second option.

set exit_code=%ERRORLEVEL%
if %exit_code% NEQ 0 (  
    echo Error calling sqlplus, return code = %exit_code%
    exit /b %exit_code%
)

Some last advice: use exit /b
instead of exit. exit will also exit the command line in batch

EDIT I’ve tested the second version and it works as well. So the code above can be replaced with:

set exit_code=%ERRORLEVEL%
IF ERRORLEVEL 1 (  
    echo Error calling sqlplus, return code = %exit_code%
    exit /b %exit_code%
)

This blog article explains quite well why this version should be used instead of the of the first one using the %ERRORLEVEL% variable.

Another option @aschipfl kindly reminded me of in his comment is the use of the conditional executions operators && and ||:

• command1 && command2 will execute command2 only if command1 succeeded
• command1 || command2 will execute command2 only if command1 failed.

Both operators use the exit code (in most cases the same as errorlevel but not always) to determine if a command failed or succeeded.

I’ve also seen you’ve posted your batch-file. There is a little something I’d like to point out though: goto @ERROR_HANDLING_SQL will throw an error because there is no label @ERROR_HANDLING_SQL. There exists a label ERROR_HANDLING_SQL at the end of your script though. If you were trying to jump to that label you should replace all occurences of goto @ERROR_HANDLING_SQL with goto ERROR_HANDLING_SQL

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