NAME
Feature::Compat::Try - make try/catch syntax available
SYNOPSIS
use Feature::Compat::Try;
sub foo
{
try {
attempt_a_thing();
return "success";
}
catch ($e) {
warn "It failed - $e";
return "failure";
}
}
DESCRIPTION
This module makes syntax support for try/catch control flow easily available.
Perl added such syntax at version 5.34.0, and extended it to support optional finally blocks at 5.35.9, which is enabled by
use feature 'try';
On that version of perl or later, this module simply enables the core feature equivalent to using it directly. On such perls, this module will install with no non-core dependencies, and requires no C compiler.
On older versions of perl before such syntax is available, it is currently provided instead using the Syntax::Keyword::Try module, imported with a special set of options to configure it to recognise exactly and only the same syntax as the core perl feature, thus ensuring that any code using it will still continue to function on that newer perl.
KEYWORDS
try
try {
STATEMENTS...
}
...
A try statement provides the main body of code that will be invoked, and must be followed by a catch statement. It may optionally be followed by a finally statement.
Execution of the try statement itself begins from the block given to the statement and continues until either it throws an exception, or completes successfully by reaching the end of the block.
The body of a try {} block may contain a return expression. If executed, such an expression will cause the entire containing function to return with the value provided. This is different from a plain eval {} block, in which circumstance only the eval itself would return, not the entire function.
The body of a try {} block may contain loop control expressions (redo, next, last) which will have their usual effect on any loops that the try {} block is contained by.
The parsing rules for the set of statements (the try block and its associated catch) are such that they are parsed as a self-contained statement. Because of this, there is no need to end with a terminating semicolon.
Even though it parses as a statement and not an expression, a try block can still yield a value if it appears as the final statement in its containing sub or do block. For example:
my $result = do {
try { attempt_func() }
catch ($e) { "Fallback Value" }
};
catch
...
catch ($var) {
STATEMENTS...
}
A catch statement provides a block of code to the preceding try statement that will be invoked in the case that the main block of code throws an exception. A new lexical variable is created to store the exception in.
Presence of this catch statement causes any exception thrown by the preceding try block to be non-fatal to the surrounding code. If the catch block wishes to optionally handle some exceptions but not others, it can re-raise it (or another exception) by calling die in the usual manner.
As with try, the body of a catch {} block may also contain a return expression, which as before, has its usual meaning, causing the entire containing function to return with the given value. The body may also contain loop control expressions (redo, next or last) which also have their usual effect.
finally
...
finally {
STATEMENTS...
}
A finally statement provides an optional block of code to the preceding try/catch pair which is executed afterwards, both in the case of a normal execution or a thrown exception. This code block may be used to provide whatever clean-up operations might be required by preceding code.
Because it is executed during a stack cleanup operation, a finally {} block may not cause the containing function to return, or to alter the return value of it. It also cannot see the containing function's @_ arguments array (though as it is block scoped within the function, it will continue to share any normal lexical variables declared up until that point). It is protected from disturbing the value of $@. If the finally {} block code throws an exception, this will be printed as a warning and discarded, leaving $@ containing the original exception, if one existed.
COMPATIBILITY NOTES
This module may use either Syntax::Keyword::Try or the perl core try feature to implement its syntax. While the two behave very similarly, and both conform to the description given above, the following differences should be noted.
Visibility to
caller()The
Syntax::Keyword::Trymodule implementstryblocks by usingevalframes. As a result, they are visible to thecaller()function and hence to things likeCarp::longmesswhen viewed as stack traces.By comparison, core's
feature 'try'creates a new kind of context stack entry that is ignored bycaller()and hence these blocks do not show up in stack traces.This should not matter to most use-cases - e.g. even
Carp::croakwill be fine here. But if you are usingcaller()with calculated indexes to inspect the state of callers to your code and there may betryframes in the way, you will need to somehow account for the difference in stack height.B::DeparseThe core
feature 'try'is implemented by emitting real opcodes that represent its behaviour, which is recognised by the version of B::Deparse that ships with core perl. As a result, any code using this implementation will deparse currently with tools likeperl -MO=Deparse ..., or others related to it such as coverage checkers.By comparison, since
Syntax::Keyword::TryusesOP_CUSTOMit is not recognised byB::Deparseand so attempts to deparse this will result in error messages likeunexpected OP_CUSTOM (catch) at ...This is rather unavoidable due to the way that
B::Deparseis implemented and does not easily support custom operators.
AUTHOR
Paul Evans <leonerd@leonerd.org.uk>