Wrong, it's an unreliable hack.
That's exactly one of the reasons why you shouldn't use overloading. 
Undesired side-effects, and yet more hacks to work around stuff being 
broken by the hack called overloading.

The mbstring extension allows working with multi-byte string 
encodings. Pragmatically, using mbstring.func_overloading is the best 
way to develop a UTF-8 application in PHP.



When mbstring.func_overloading is enabled, the string functions treat 
strings a sequence of characters, not bytes. Usually, this is desired, 
as you wouldn't want to truncate a string part way through a 
multi-byte character.



When this is not desired is when you are working with binary files or 
streams. Usually in this case you want to work with bytes and not 
characters. The attached patch explicitly uses bytes (the '8bit' 
encoding) for string calculations in a binary context if mbstring 
function overloading is enabled.

That doesn't make any sense. If you use function overloading, the 
string functions are already replaced by the mbstring variants. That's 
the whole purpose of that functionality, so there is no need to do 
this explicitly in wrapper methods.

Beside that, mbstring function overloading is evil and should be 
removed from the face of this earth completely.

Patch to make File_PDF work when mbstring.func_overload is enabled.

When mbstring.func_overload is enabled, the strlen, substr and strpos 
functions get overloaded with multi-byte character versions. Since 
these functions are used to do byte calculations in File_PDF, using 
mbstring.func_overload breaks File_PDF. The attached patch makes 
File_PDF work even with mbstring.func_overload is used.