include() failed...Or did it?

[Itshim]

Doing some testing, I have found that when and include statement fails to open a stream it returns boolean FALSE (along with generating warnings), but an included file could also return FALSE. I would like to know if there is a way to determine (programmatically) if the include failed or it was successful and the file is returning FALSE.


-- -- -- -- -- -- -- --

If you give someone a program, you will frustrate them for a day
but if you teach them how to program, you will frustrate them for a lifetime.

 

[jpadie]

if ((include $filename) === FALSE){

//do something
}

 

[Itshim]

That is what I normally do, but it occurred to me this morning that an include file could have the statement:

return FALSE;

In which case the code:

if ((include $filename) === FALSE){

//do something
}

would not be able to determine if the file was included or not.

I am not actually having a problem seeing as none of my includes actually return anything, I just thought of the situation and was wondering if there was a way to detect what was happening.

I thought of adding something to a custom error handler but that seemed messy. You could check using class_exists() or function_exists() (if the file defines a class or a function). Actually that just gave me an idea, you could use the function get_included_files() and see if the file exists in the returned array... I guess I just answered my own question.

--Sorry for the train of thought post.


-- -- -- -- -- -- -- --

If you give someone a program, you will frustrate them for a day
but if you teach them how to program, you will frustrate them for a lifetime.

 

[jpadie]

you're right. this, I think, is a bug in the include handling of php, the include itself worked so should not report false.

so i think you are left with two options:

1. use require instead
2. use the FALSE test PLUS test for an E_WARNING in error_get_last() (obviously you need to test for the particular line so as not to pick up a previous error).

 

[Itshim]

Excellent, error_get_last() looks like the right tool for the job.

I have a tendency to stay away from require() calls, because if it fails it issues an fatal error (E_ERROR), problem with that being that there is no way to trap a fatal error.

I was hoping that require() would be changed to issue a E_RECOVERABLE_ERROR, which I feel better represents the state of the engine at fail time.

E_RECOVERABLE_ERROR - Catchable fatal error. It indicates that a probably dangerous error occured, but did not leave the Engine in an unstable state. If the error is not caught by a user defined handle (see also set_error_handler()), the application aborts as it was an E_ERROR.

I know I'm probably alone on this because the actual definition of the word 'require' is: to have need of; need.

Thanks for the help.