Bug #101046 for Encode: Test suite of 2.67 and 2.68 fails with perl 5.21.7 and 5.21.8

Sat Dec 20 16:50:12 2014 SREZIC [...] cpan.org - Ticket created

Subject:

Test suite of 2.67 fails with perl 5.21.7

t/encoding.t ............... Failed 2/31 subtests ... Test Summary Report ------------------- t/encoding.t (Wstat: 0 Tests: 31 Failed: 2) Failed tests: 12, 14 Fail happens on a Debian/wheezy and FreeBSD 9.2 system. Interestingly "make test" of perl 5.21.7 went fine without errors on these systems, and the bundled Encode version seems to be the same... Regards, Slaven

Sat Dec 20 16:52:17 2014 SREZIC [...] cpan.org - Correspondence added

On 2014-12-20 16:50:12, SREZIC wrote: Show quoted text

> t/encoding.t ............... > Failed 2/31 subtests > > ... > > Test Summary Report > ------------------- > t/encoding.t (Wstat: 0 Tests: 31 Failed: 2) > Failed tests: 12, 14 > > > Fail happens on a Debian/wheezy and FreeBSD 9.2 system. > > Interestingly "make test" of perl 5.21.7 went fine without errors on > these systems, and the bundled Encode version seems to be the same... >

See also http://matrix.cpantesters.org/?dist=Encode;perl=5.21.7;reports=1 for available fail reports.

Thu Jan 22 12:08:48 2015 SREZIC [...] cpan.org - Subject changed from 'Test suite of 2.67 fails with perl 5.21.7' to 'Test suite of 2.67 fails with perl 5.21.7 and 5.21.8'

Thu Jan 22 12:20:01 2015 SREZIC [...] cpan.org - Subject changed from 'Test suite of 2.67 fails with perl 5.21.7 and 5.21.8' to 'Test suite of 2.67 and 2.68 fails with perl 5.21.7 and 5.21.8'

Fri Jan 23 14:31:06 2015 ether [...] cpan.org - Requestor ETHER added

Thu Feb 05 06:43:42 2015 KENTNL [...] cpan.org - Requestor KENTNL added

Thu Feb 05 07:22:48 2015 KENTNL [...] cpan.org - Correspondence added

Curious. Seems to be something in XS stuff.

Simply untarring and running `prove -blvr t/encoding.t` *without* doing a make run *passes* tests ( on blead ) ( and without warning )

However, doing `perl Makefile.PL && make` and then doing `prove -blvr t/encoding.t` makes it fail.

Where its triggering those warnings is the strangest thing:

PERL5OPT="-d:Confess" prove -blvr t/encoding.t

But either way, I don't think the warnings are directly related to the problem, just a failure to include those warnings only when deprecations are warned about. ( I could be very wrong about that of course )

Attached however is the reduced and instrumented form of the failing parts of the test, and the output emitted:

t/encoding.t .. Setting ${^ENCODING} is deprecated at /home/kent/.cpanm/work/1423137170.272187/Encode-2.70/blib/lib/encoding.pm line 134.
        encoding::import("encoding", "greek") called at t/encoding.t line 5
        main::BEGIN() called at t/encoding.t line 5
        eval {...} called at t/encoding.t line 5

1..2
not ok 1
# expected: 223
# got: 943
not ok 2
# expected: 223
# got: 943
Failed 2/2 subtests

Removing the `use encoding 'greek' ` from the file makes these 2 tests pass.... but of course, make the rest of the file fail =)

Doing some diving shows this reduced test fails all the way back with encode 2.00 in the exact same way with the same warnings... so its somehow a "blead breaks cpan without breaking itself with modules that are shipped on CPAN".

Wut.

Subject:

encoding_reduce.t

print "1..2\n"; no warnings "deprecated"; use encoding "greek"; # iso 8859-7 (no "latin" alias, surprise...) # charnames must still work use charnames ':full'; { my $expected = 0xdf; my $got = ord("\N{LATIN SMALL LETTER SHARP S}"); my $state = $got == $expected; print "not " unless $state; print "ok 1\n"; if ( not $state ) { print "# expected: $expected\n# got: $got\n"; } } # combine { my $c = "\xDF\N{LATIN SMALL LETTER SHARP S}" . chr(0xdf); my $expected = 0xdf; my $got = ord(substr($c, 1, 1)); my $state = $got == $expected; print "not " unless $state; print "ok 2\n"; if ( not $state ) { print "# expected: $expected\n# got: $got\n"; } }

Subject:

encoding_tap.txt

Setting ${^ENCODING} is deprecated at /home/kent/.cpanm/work/1423136354.224896/Encode-2.70/blib/lib/encoding.pm line 134. encoding::import("encoding", "latin1") called at t/encoding.t line 20 main::BEGIN() called at t/encoding.t line 20 eval {...} called at t/encoding.t line 20 Setting ${^ENCODING} is deprecated at /home/kent/.cpanm/work/1423136354.224896/Encode-2.70/blib/lib/encoding.pm line 134. encoding::import("encoding", "greek") called at t/encoding.t line 21 main::BEGIN() called at t/encoding.t line 21 eval {...} called at t/encoding.t line 21 t/encoding.t .. 1..31 ok 1 ok 2 ok 3 ok 4 ok 5 ok 6 ok 7 ok 8 ok 9 ok 10 ok 11 not ok 12 ok 13 not ok 14 ok 15 ok 16 ok 17 ok 18 ok 19 ok 20 ok 21 ok 22 ok 23 ok 24 ok 25 ok 26 ok 27 ok 28 ok 29 ok 30 ok 31 Failed 2/31 subtests Test Summary Report ------------------- t/encoding.t (Wstat: 0 Tests: 31 Failed: 2) Failed tests: 12, 14 Files=1, Tests=31, 1 wallclock secs ( 0.05 usr 0.00 sys + 0.09 cusr 0.01 csys = 0.15 CPU) Result: FAIL

Thu Feb 05 07:23:13 2015 KENTNL [...] cpan.org - Broken in 2.66 added

Thu Feb 05 07:23:13 2015 KENTNL [...] cpan.org - Broken in 2.68 added

Wed Feb 11 23:18:34 2015 ANDK [...] cpan.org - Correspondence added

The relevant commits in bleadperl: commit c0659f7 Author: Karl Williamson <khw@cpan.org> AuthorDate: Wed Nov 19 17:02:58 2014 -0700 Commit: Karl Williamson <khw@cpan.org> CommitDate: Thu Nov 20 21:45:18 2014 -0700 perldelta for deprecating ${^ENCODING} commit a3ee04b Author: Karl Williamson <khw@cpan.org> AuthorDate: Wed Nov 19 16:49:53 2014 -0700 Commit: Karl Williamson <khw@cpan.org> CommitDate: Thu Nov 20 21:45:18 2014 -0700 perlvar: More text about ${^ENCODING}; refer by perldiag commit d5b4785 Author: Dagfinn Ilmari Mannsåker <ilmari@ilmari.org> AuthorDate: Mon Nov 10 23:54:46 2014 +0000 Commit: Karl Williamson <khw@cpan.org> CommitDate: Thu Nov 20 21:45:18 2014 -0700 Deprecate setting ${^ENCODING} The commiter added a no warnings in t/op/leaky-magic.t, and made other minor changes because of rebasing issues. commit 3e66930 Author: Karl Williamson <khw@cpan.org> AuthorDate: Tue Nov 18 17:03:03 2014 -0700 Commit: Karl Williamson <khw@cpan.org> CommitDate: Thu Nov 20 21:45:18 2014 -0700 Make encoding pragma lexical in scope The encoding pragma is deprecated, but in the meantime it causes spooky action at a distance with other modules that it may be combined with. In these modules, operations such as chr(), ord(), and utf8::upgrade() will suddenly start doing the wrong thing. The documentation for 'encoding' has said to call it after loading other modules, but this may be impractical. This is especially bad with anything that auto-loads at first use, like \N{} does now for charnames. There is an issue with combining this with setting the variable ${^ENCODING} directly. The potential for conflicts has always been there, and remains. This commit introduces a shadow hidden variable, subservient to ${^ENCODING} (to preserve backwards compatibility) that has lexical scope validity. The pod for 'encoding' has been revamped to be more concise, clear, use more idiomatic English, and to speak from a modern perspective.

Wed Feb 11 23:18:36 2015 The RT System itself - Status changed from 'new' to 'open'

Thu Mar 12 19:27:48 2015 SREZIC [...] cpan.org - Correspondence added

On 2014-12-20 16:50:12, SREZIC wrote: Show quoted text

> t/encoding.t ............... > Failed 2/31 subtests > > ... > > Test Summary Report > ------------------- > t/encoding.t (Wstat: 0 Tests: 31 Failed: 2) > Failed tests: 12, 14 > > > Fail happens on a Debian/wheezy and FreeBSD 9.2 system. > > Interestingly "make test" of perl 5.21.7 went fine without errors on > these systems, and the bundled Encode version seems to be the same...

Seems to be fixed with 2.71.

Thu Mar 12 19:27:54 2015 SREZIC [...] cpan.org - Status changed from 'open' to 'resolved'

Thu Mar 12 19:27:54 2015 SREZIC [...] cpan.org - Fixed in 2.71 added