A try-catch block is a language construct used in programming that allows a programmer to handle errors or exceptions in their code. The try block contains code that may throw an exception, and the catch block catches and handles that exception if it is thrown.
try
do_something_risky()
catch
print "The risky thing just failed."
Many modern programming languages, including JS, Python, and C++, have built-in support for try-catch blocks. However, the C programming language does not provide this functionality out of the box.
Why should you use try-catch?
The ability to handle exceptions in code is extremely useful for a number of reasons. First and foremost, it allows a program to gracefully handle errors or exceptional situations that may arise during execution. This can prevent the program from crashing or behaving unpredictably. In the case of embedded code, try-catch statements are particularly useful for writing your unit tests.
In embedded code, you probably use an assert function to handle errors, allowing you to check some conditions and manage critical situations. Most of the time, those assert functions display a log or traces and stop the program.
Let’s take a simple function as a reference:
void assert(bool statement)
{
if (!statement)
{
printf("This is bad\n");
while(1);
}
}
int freq(int period){
assert(period != 0);
return 1/period;
}
int main() {
freq(0); // divide by zero to assert
}
In this example, the main function will properly assert in the freq function if the value is 0.
When you create your unit test to validate the freq functions, you will have to check all possible values of period, but you don’t want your test to stop its execution by testings the limits. This is where you need a try-catch block.
How to create a try-catch bloc in C?
C does not have built-in support for try-catch blocks, but it is possible to create a similar construct using macros and the standards setjmp.h lib. Here's an example of how to create a simple try-catch macro in C following our previous example:
#include <setjmp.h>
jmp_buf ex_buf;
#define TRY if (setjmp(ex_buf) == 0)
#define CATCH else
#define THROW longjmp(ex_buf, 1)
void assert(bool statement)
{
if (!statement)
{
printf("This is bad\n");
THROW;
while(1);
}
}
int freq(int period){
assert(period!=0);
return 1/period;
}
int main() {
TRY {
// Code that may throw an exception
freq(0); // Divide by zero to throw an exception
}
CATCH {
// Handle the exception
printf("An exception occurred, the test just failed\n");
}
return 0;
}
This macro defines a TRY block, using a jmp_buf variable to check if an exception has been thrown and to save the current context. If no exception has been thrown, the code in the TRY block is executed.
If the THROW macro is called, the code will jump back to the setjmp function of the TRY block and restore the saved context. Then the setjmp function will return the integer passed to the longjmp(in this case, 1). So if an exception is thrown, the if statement of the second execution of the TRY macro will be false and the CATCH block will be executed instead.
In the example above, the code in the TRY block attempts to divide by zero, which will throw an exception. This will jump back on the TRY macro and make the if statement fail. The CATCH block catches the exception and prints a message into the console.
While this macro is a simplified version of a try-catch block, it demonstrates how a similar construct can be implemented in C.
Please note that the following try-catch code is not suitable for use in a production environment. While it was designed for a specific, well-controlled unit test environment, it may cause issues with nested calls or multi-process situations. Therefore, we cannot guarantee its safety or reliability.
Please use this code at your own risk.
If you want a practical example of try catch usage, you can check out the Luos unit tests and try-catch macro.
In conclusion, try-catch blocks are a powerful tool for handling errors and exceptions in code. While the C programming language does not provide built-in support for this functionality, it is possible to create a close behavior using macros. By using try-catch blocks, programmers can make their unit-test simpler.
We advise that you thoroughly consider the risks associated with using this before proceeding.
Get started with Luos
