This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
C<not> should be C<!> in utf8decode.t, to fix a precedence bug.
authorNicholas Clark <nick@ccl4.org>
Fri, 11 Mar 2011 23:18:45 +0000 (23:18 +0000)
committerNicholas Clark <nick@ccl4.org>
Fri, 11 Mar 2011 23:18:45 +0000 (23:18 +0000)
commit8567041cf58773c5e66c77659b993b81515bd622
tree55a7218ed204fb82f772d5401fe086609aed68f6
parentd066ad0f7923df77c196bc64360843fa02f2aef1
C<not> should be C<!> in utf8decode.t, to fix a precedence bug.

Fixing the bug opens a whole can of worms. The test hasn't actually been
testing failure since 35bcd33832d74e56. Indeed, 35bcd33832d74e56 increases
the verbosity of warnings from unpack, meaning that the test needs to aggregate
lines in its $SIG{__WARN__} handler. At which point it passes again.

Until 097fb8e2acde8522, which adds ", immediately" to the UTF-8 diagnostics.
Then 872c91ae155f6880 makes most every form of invalid Unicode warning free,
and also overlong UTF-8 warning free. 618432450ee258c0 makes overlong UTF-8
warn once more. Finally, 08ca2aa38a29585f means that 'U0U*' needs to be
replaced with 'C0U*'.

So now it passes, and (I believe) actually *tests* what it intends to test.
Long may that state of affairs endure.
t/op/utf8decode.t