Bug #118885 for Mail-Builder: Fails with Encode 2.87

Mon Nov 21 15:02:53 2016 SREZIC [...] cpan.org - Ticket created

CC:	DANKOGAI [...] cpan.org
Subject:	Fails with Encode 2.87

I see the following test failure on my smoker systems: ... # Failed test 'Cc header encoding ok' # at t/07_builder_integration.t line 165. # got: '=?UTF-8?B?dmVyeSBsb25nIG5hbWUgdGhhdCBleGNlZWRzIHRoZSA3NSBjaGFyYWN0ZXIg?= # =?UTF-8?B?bGltaXQgb2YgYW4gZW5jb2RlZCB3b3JkIMODwrxmdC04IG7Dg8KkbcODwqk=?= # <recipient5@test.com> # ' # expected: '=?UTF-8?B?dmVyeSBsb25nIG5hbWUgdGhhdCBleGNlZWRzIHRoZSA3NSBjaGFyYWN0ZXIgbGk=?= =?UTF-8?B?bWl0IG9mIGFuIGVuY29kZWQgd29yZCDDg8K8ZnQtOCBuw4PCpG3Dg8Kp?= <recipient5@test.com> # ' # Looks like you failed 1 test of 68. t/07_builder_integration.t .. Dubious, test returned 1 (wstat 256, 0x100) Failed 1/68 subtests ... This happens only with Encode 2.87, it seems. Statistical analysis (negative theta is "bad"): **************************************************************** Regression 'mod:Encode' **************************************************************** Name Theta StdErr T-stat [0='const'] 1.0000 0.0000 18898845728483576.00 [1='eq_2.60'] 0.0000 0.0000 1.28 [2='eq_2.72'] 0.0000 0.0000 0.91 [3='eq_2.72_01'] -0.0000 0.0000 -1.36 [4='eq_2.80'] -0.0000 0.0000 -1.11 [5='eq_2.80_01'] -0.0000 0.0000 -1.36 [6='eq_2.85'] -0.0000 0.0000 -2.27 [7='eq_2.86'] -0.0000 0.0000 -2.07 [8='eq_2.87'] -1.0000 0.0000 -17818002628278870.00 R^2= 1.000, N= 55, K= 9 ****************************************************************

Mon Mar 13 15:19:46 2017 MAROS [...] cpan.org - Correspondence added

Thank you for your bug report. This issue should be fixed in 2.12

Mon Mar 13 15:19:47 2017 The RT System itself - Status changed from 'new' to 'open'

Mon Mar 13 15:19:53 2017 MAROS [...] cpan.org - Status changed from 'open' to 'resolved'

Mon Mar 13 15:36:01 2017 SREZIC [...] cpan.org - Correspondence added

On 2017-03-13 15:19:46, MAROS wrote: Show quoted text

> Thank you for your bug report. This issue should be fixed in 2.12

Well, now it's the other way around: it passes for newer Encode versions and fails for older ones: ... # Failed test 'Cc header encoding ok' # at t/07_builder_integration.t line 166. # '=?UTF-8?B?dmVyeSBsb25nIG5hbWUgdGhhdCBleGNlZWRzIHRoZSA3NSBjaGFyYWN0ZXIgbGk=?= =?UTF-8?B?bWl0IG9mIGFuIGVuY29kZWQgd29yZCDDvGZ0LTggbsOkbcOp?= <recipient5@test.com> # ' # doesn't match '(?^:^=\?UTF-8\?B\?dmVyeSBsb25nIG5hbWUgdGhhdCBleGNlZWRzIHRoZSA3NSBjaGFyYWN0ZXIg\?=\s*=\?UTF-8\?B\?bGltaXQgb2YgYW4gZW5jb2RlZCB3b3JkIMO8ZnQtOCBuw6Rtw6k=\?=\s*<recipient5\@test\.com>\s*$)' # Looks like you failed 1 test of 68. t/07_builder_integration.t .. Dubious, test returned 1 (wstat 256, 0x100) Failed 1/68 subtests ... Statistical analysis now (theta=1 is "good", theta=0 is "bad"): **************************************************************** Regression 'mod:Encode' **************************************************************** Name Theta StdErr T-stat [0='const'] -0.0000 0.0000 -2.18 [1='eq_2.72_01'] 0.0000 0.0000 1.54 [2='eq_2.80'] 0.0000 0.0000 1.78 [3='eq_2.80_01'] -0.0000 0.0000 -7.57 [4='eq_2.86'] -0.0000 0.0000 -4.63 [5='eq_2.87'] 1.0000 0.0000 18403081521694608.00 [6='eq_2.88'] 1.0000 0.0000 19241279394544768.00 R^2= 1.000, N= 37, K= 7 **************************************************************** Maybe you should accept both variants in the test? Or increase the prereq version for Encode?

Mon Mar 13 15:36:02 2017 SREZIC [...] cpan.org - Status changed from 'resolved' to 'open'

Wed May 31 03:17:37 2017 SREZIC [...] cpan.org - Correspondence added

On 2017-03-13 15:36:01, SREZIC wrote: Show quoted text

> On 2017-03-13 15:19:46, MAROS wrote:

> > Thank you for your bug report. This issue should be fixed in 2.12

> > Well, now it's the other way around: it passes for newer Encode > versions and fails for older ones: > > ... > # Failed test 'Cc header encoding ok' > # at t/07_builder_integration.t line 166. > # '=?UTF- > 8?B?dmVyeSBsb25nIG5hbWUgdGhhdCBleGNlZWRzIHRoZSA3NSBjaGFyYWN0ZXIgbGk=?= > =?UTF-8?B?bWl0IG9mIGFuIGVuY29kZWQgd29yZCDDvGZ0LTggbsOkbcOp?= > <recipient5@test.com> > # ' > # doesn't match '(?^:^=\?UTF- > 8\?B\?dmVyeSBsb25nIG5hbWUgdGhhdCBleGNlZWRzIHRoZSA3NSBjaGFyYWN0ZXIg\?=\s*=\?UTF- > 8\?B\?bGltaXQgb2YgYW4gZW5jb2RlZCB3b3JkIMO8ZnQtOCBuw6Rtw6k=\?=\s*<recipient5\@test\.com>\s*$)' > # Looks like you failed 1 test of 68. > t/07_builder_integration.t .. > Dubious, test returned 1 (wstat 256, 0x100) > Failed 1/68 subtests > ... > > Statistical analysis now (theta=1 is "good", theta=0 is "bad"): > > **************************************************************** > Regression 'mod:Encode' > **************************************************************** > Name Theta StdErr T-stat > [0='const'] -0.0000 0.0000 -2.18 > [1='eq_2.72_01'] 0.0000 0.0000 1.54 > [2='eq_2.80'] 0.0000 0.0000 1.78 > [3='eq_2.80_01'] -0.0000 0.0000 -7.57 > [4='eq_2.86'] -0.0000 0.0000 -4.63 > [5='eq_2.87'] 1.0000 0.0000 18403081521694608.00 > [6='eq_2.88'] 1.0000 0.0000 19241279394544768.00 > > R^2= 1.000, N= 37, K= 7 > **************************************************************** > > Maybe you should accept both variants in the test? Or increase the > prereq version for Encode?

Still a problem with 2.13.

Sat Jun 03 04:33:24 2017 KENTNL [...] cpan.org - Requestor KENTNL added