Switching on internal UTF-8 flaq on DBI result from database

[liz]

After having spent the better part of a day figuring out how perfectly good UTF-8 data out of a MySQL database was being mangled by Perl, I found that this was caused by Perl upgrading the string which was already UTF-8 (but didn't have the internal UTF-8 flag set). So I needed a way to set that flag, I found the "_utf8_on" of Encode through graff's answer Re: restore unicode data from database?.

However, doing this "manually" on everything you fetch out of a database, becomes tedious quickly. Since I frequently use the selectall_arrayref fetching methods, I created this little sub that could can "wrap" around such a call.

It expects the arrayref of arrayrefs as input, and also returns it. So you can "inline" the call to "_utf8_on_all_arrayref".

require Encode; # needs to be done only once in the beginning
sub _utf8_on_all_arrayref {
 
# For all the records specified
#  Switch on the UTF-8 flag for all values
# Return the original reference

    foreach (@{$_[0]}) {
        Encode::_utf8_on( $_ ) foreach @{$_};
    }
    $_[0];
} #_utf8_on_all_arrayref

[Aristotle]

require is a no-op if called multiple times on the same module, so you can just move it into the function. You can also shorten the code by collapsing the outer for into a map:

sub _utf8_on_all_arrayref {
    require Encode;
    Encode::_utf8_on($_) for map @$_, @{$_[0]};
    $_[0];
}

sub _utf8_on_all_arrayref {
    require Encode;
    Encode::_utf8_on($$_) for map \(@$_), @{$_[0]};
    $_[0];
}

[liz]

Well, I think if you can avoid calling require over and over again, you should. Any require is at least a lookup in %INC each time you execute it.

use Benchmark qw(:hireswallclock timethese);
timethese( 1000000,{
  one => sub { require Benchmark },
  two => sub { require Benchmark; require Benchmark },
});
__END__
$ perl 1
Benchmark: timing 1000000 iterations of one, two...
 one: 2.43888 wallclock secs ( 1.70 usr +  0.00 sys =  1.70 CPU) @ 588
+235.29/s (n=1000000)
 two: 5.11982 wallclock secs ( 2.77 usr +  0.00 sys =  2.77 CPU) @ 361
+010.83/s (n=1000000)

Also, for some reason your list flattening with map() does not work. Not sure whether this is a bug in Perl, or a conceptual problem with using map() and $_. Observe:

use Encode qw(_utf8_on);
$a = [['']];
foreach (@$a) { _utf8_on($_) foreach @$_ } # my way
print utf8::is_utf8( $a->[0][0] ),$/;
$a = [['']];
_utf8_on( $_ ) for map @$_, @{$a}; # Aristotle's way
print utf8::is_utf8( $a->[0][0] ),$/;
__END__
1

which should show two 1's instead of 1.

But also from a performance point of view, the extra list flattening with map() is not very efficient:

use Encode qw(_utf8_on);
use Benchmark qw(:hireswallclock timethese);
push @$a,[(0) x 10] foreach 1..10;
timethese( 10000,{
  liz => sub {
    foreach (@{$a}) {
        _utf8_on( $_ ) foreach @{$_};
    }
},
  Aristotle => sub {
    _utf8_on( $_ ) for map @$_, @{$a};
},
});
__END__
Benchmark: timing 10000 iterations of Aristotle, liz...
 Aristotle: 4.37344 wallclock secs ( 3.73 usr + 0.00 sys = 3.73 CPU) @
+ 2680.97/s (n=10000)
 liz: 4.06957 wallclock secs ( 2.73 usr + 0.00 sys = 2.73 CPU) @ 3663.
+00/s (n=10000)

Liz

Update:
The way Aristotle proposed doesn't work because map() creates a copy of the elements, on which the UTF-8 flag is set and then discarded. See $_ and list flattening with map() for more info.

[Aristotle]

Doh. Of course, map aliases $_ during the expression, but the expression's result is completely separate. I updated my previous reply with a version that works. Returning a new value rather than just operating on an alias is the reason for the performance hit of course; unfortunately, I can only think of only two other remotely relevant aliasing constructs in Perl, neither of which are any help here: grep returns aliases, and the @_ in a function call contains aliases rather than copies. I can't see how to use either to achieve something less awkward than a nested for loop though.