Front page | perl.perl5.porters |
Postings from July 2001
On Tue, Jul 03, 2001 at 06:16:37PM +0000, Ton Hospel wrote:
> In article <5.1.0.14.0.20010702140058.01b6c9c0@exchi01>,
> "Craig A. Berry" <craigberry@mac.com> writes:
> > The existing logic in openunicode() bails out if it fails to open the first
> > filespec it tries. I believe the intent is to iterate through @INC until it
> > successfully finds and opens the file or exhausts the list.
> >
> > --- lib/Unicode/UCD.pm;-0 Sun Jul 1 01:56:14 2001
> > +++ lib/Unicode/UCD.pm Mon Jul 2 14:49:25 2001
> > @@ -44,6 +44,7 @@
> > for my $d (@INC) {
> > use File::Spec;
> > $f = File::Spec->catfile($d, "unicode", @path);
> > + next unless -f $f;
> > if (open($$rfh, $f)) {
> > last;
> > } else {
>
> Strictly speaking that introduces a race. Normally it's better to
> check the open result against ENOENT, but the cost would be to
> have to load one of the Errno modules. Another way may be to do a
> next in the else branch always, but remember $! and report the last one
> on failure.
Hmmm. Let's croak only if nothing worked.
unless (defined $$rfh) {
for my $d (@INC) {
use File::Spec;
$f = File::Spec->catfile($d, "unicode", @path);
last if open($$rfh, $f);
}
croak __PACKAGE__, ": failed to find ",join("/",@path)," in @INC\n"
unless defined $rfh;
}
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
Thread Previous
|
Thread Next