Bug #53636 for Test-LongString: Added Diagnostics seems to break depending modules

Tue Jan 12 23:00:17 2010 jjnapiork [...] cpan.org - Ticket created

Subject:

Added Diagnostics seems to break depending modules

Hi, You may not consider this a bug, but the recently changed diagnostic messages, while helpful, seems to break some automated tests in depending modules which are expecting your output to continue to conform to a particular pattern. Currently I find "Test-WWW-Mechanize" to be busted in this manner, it's possible there might be others. Right now that module depends on your old style diagnostics. I will contact the author of that module to see if I can get a suitable patch accepted but would you consider some means of supporting backward compatible error messages? Thank you. I'm be happy to help if I can. Sincerely, John Napiorkowski (jjnapiork@cpan.org)

Fri Jan 22 10:19:15 2010 RGARCIA [...] cpan.org - Taken

Sat Jan 23 03:54:55 2010 RGARCIA [...] cpan.org - Correspondence added

I'm considering releasing a new version with this patch : http://consttype.org/cgi/gitweb.cgi?p=Test-LongString.git;a=commitdiff;h=6b219e6042e03f03695c11d171d046f406fc3de9 that turns off LCSS reporting. Would it be sufficient ?

Sat Jan 23 03:54:56 2010 The RT System itself - Status changed from 'new' to 'open'

Wed Jan 27 15:07:18 2010 jjnapiork [...] cpan.org - Correspondence added

On Sat Jan 23 03:54:55 2010, RGARCIA wrote: Show quoted text

> I'm considering releasing a new version with this patch : > > http://consttype.org/cgi/gitweb.cgi?p=Test- > LongString.git;a=commitdiff;h=6b219e6042e03f03695c11d171d046f406fc3de9 > > that turns off LCSS reporting. Would it be sufficient ?

SOrry for the delayed response. If you release a developer candidate I'd be happy to test it against the CPAN modules that rely on the older behavior. For your knowledge it seems the Test::WWW::Mechanize people are aware of the issue and is possible they are also prepping a patch. Thanks!

Sat Jan 30 08:40:23 2010 RGARCIA [...] cpan.org - Correspondence added

Version 0.13 allows the ability to turn off LCSS output (although this is not the default) with use Test::LongString lcss => 0; This syntax will also work (but do nothing) in older versions of T::LS.

Sat Jan 30 08:40:25 2010 RGARCIA [...] cpan.org - Status changed from 'open' to 'resolved'

Sun Jan 31 21:29:49 2010 jjnapiork [...] cpan.org - Correspondence added

On Sat Jan 30 08:40:23 2010, RGARCIA wrote: Show quoted text

> Version 0.13 allows the ability to turn off LCSS output (although this > is not the default) with > > use Test::LongString lcss => 0; > > This syntax will also work (but do nothing) in older versions of T::LS.

Okay I'll see if this helps, thanks for checking in! Sincerely, John Napiorkowski

Sun Jan 31 21:29:52 2010 The RT System itself - Status changed from 'resolved' to 'open'

Wed Feb 10 17:19:40 2010 jjnapiork [...] cpan.org - Correspondence added

On Sun Jan 31 21:29:49 2010, JJNAPIORK wrote: Show quoted text

> On Sat Jan 30 08:40:23 2010, RGARCIA wrote:

> > Version 0.13 allows the ability to turn off LCSS output (although this > > is not the default) with > > > > use Test::LongString lcss => 0; > > > > This syntax will also work (but do nothing) in older versions of T::LS.

> > Okay I'll see if this helps, thanks for checking in! > > Sincerely, > John Napiorkowski

Hey, Just FYI, that change solve the problems I was having. using ... lcss=>0 fixes it for now. Later on we might consider supporting the extended diagnostics format. Thanks for being so responsive! John Napiorkowski

Wed Feb 10 17:19:42 2010 jjnapiork [...] cpan.org - Status changed from 'open' to 'resolved'