utf8::is_utf8 valid introspection?

[McA]

Hi all,

there is much stuff out there for utf8::is_utf8. My question: Is this a valid/accepted/reliable function to introspect a perl string? Is it valid to rely on the upgrading semantics when I concatenate a utf8 flagged string with an unflagged one?

Best regards
McA

[dave_the_m]

Code that needs to use utf8::is_utf8 (apart from for debugging purposes) is, in general, likely to be buggy. Most of the time your code shouldn't need to care what internal format perl's currently using to store strings.

Except of course for "the Unicode bug", where the state of the utf8 flag on strings effects things like regexes for chars in the range 0x80..0xff. This has been reduced in more recent perls by the addition of things like the //a match modifier.

Dave.

[ikegami]

If you need to work around a bug, just use

utf8::upgrade($var);

or

utf8::downgrade($var);

to get the the string in the expected storage format (regardless of the current storage format).

The only use I can think of for utf8::is_utf8 is for debugging, but I use Devel::Peek's Dump when I want to peek at a scalar's internals.