Skip Menu |

This queue is for tickets about the MarpaX-ESLIF CPAN distribution.

Report information
The Basics
Id: 120784
Status: resolved
Priority: 0/
Queue: MarpaX-ESLIF

People
Owner: Nobody in particular
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: Normal
Broken in:
  • 1.0.48
  • 1.0.49
Fixed in: 1.0.50

Attachments
fail.MarpaX-ESLIF-1.0.49.amd64-freebsd.10.1-release.1490827733.17969.rpt



Subject: Undefined symbol "libiconv"
t/00-compile.t fails on non-Linux systems. On FreeBSD systems I see: ... # Failed test 'MarpaX/ESLIF.pm loaded ok' # at t/00-compile.t line 49. # got: '65280' # expected: '0' Can't load '/home/cpansand/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-9/blib/arch/auto/MarpaX/ESLIF/ESLIF.so' for module MarpaX::ESLIF: /home/cpansand/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-9/blib/arch/auto/MarpaX/ESLIF/ESLIF.so: Undefined symbol "libiconv" at /usr/perl5.20.2p/lib/5.20.2/amd64-freebsd/DynaLoader.pm line 193. at /home/cpansand/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-9/blib/lib/MarpaX/ESLIF.pm line 18. BEGIN failed--compilation aborted at /home/cpansand/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-9/blib/lib/MarpaX/ESLIF.pm line 18. Compilation failed in require at -e line 1. # Looks like you failed 1 test of 4. t/00-compile.t ......... Dubious, test returned 1 (wstat 256, 0x100) Failed 1/4 subtests ... On Mac OS X it looks like this: ... # Failed test 'MarpaX/ESLIF.pm loaded ok' # at t/00-compile.t line 49. # got: '65280' # expected: '0' Can't load '/Users/eserte/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-0/blib/arch/auto/MarpaX/ESLIF/ESLIF.bundle' for module MarpaX::ESLIF: dlopen(/Users/eserte/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-0/blib/arch/auto/MarpaX/ESLIF/ESLIF.bundle, 2): Symbol not found: _iconv Referenced from: /Users/eserte/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-0/blib/arch/auto/MarpaX/ESLIF/ESLIF.bundle Expected in: flat namespace in /Users/eserte/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-0/blib/arch/auto/MarpaX/ESLIF/ESLIF.bundle at /opt/perl-5.25.11/lib/5.25.11/darwin-2level/DynaLoader.pm line 197. BEGIN failed--compilation aborted at /Users/eserte/.cpan/build/2017032906/MarpaX-ESLIF-1.0.48-0/blib/lib/MarpaX/ESLIF.pm line 18. Compilation failed in require at -e line 1. # Looks like you failed 1 test of 4. t/00-compile.t ......... Dubious, test returned 1 (wstat 256, 0x100) Failed 1/4 subtests ...
Many thanks for this. Indeed, having no such box, I was wondering what was happening. Hopefully fixed with version 1.0.49. To be confirmed.
Still failed. Sorry about that. I don't see any logfile yet for MarpaX::ESLIF on cpantesters which makes things more difficult. Regards. On Wed Mar 29 15:45:02 2017, JDDPAUSE wrote: Show quoted text
> Many thanks for this. Indeed, having no such box, I was wondering what > was happening. Hopefully fixed with version 1.0.49. To be confirmed.
Can you please send the 1.0.49 log excerpts ? And may I ask to cpantesters if there is a pb... does not sound normal to me that http://www.cpantesters.org/show/MarpaX-ESLIF.json is absent, c.f. http://matrix.cpantesters.org/?dist=MarpaX-ESLIF+1.0.49 . Many thanks.
On 2017-03-30 08:57:10, JDDPAUSE wrote: Show quoted text
> Can you please send the 1.0.49 log excerpts ? > And may I ask to cpantesters if there is a pb... does not sound normal > to me that http://www.cpantesters.org/show/MarpaX-ESLIF.json is > absent, c.f. http://matrix.cpantesters.org/?dist=MarpaX-ESLIF+1.0.49 . > > > Many thanks.
Attached a report generated on a freebsd 10 system. And looking at the munin graphs it seems that the synchronization between metabase and cpantesters.org is lagging, about seven days.
Subject: fail.MarpaX-ESLIF-1.0.49.amd64-freebsd.10.1-release.1490827733.17969.rpt

Message body not shown because it is not plain text.

On 2017-03-30 16:06:05, SREZIC wrote: Show quoted text
> On 2017-03-30 08:57:10, JDDPAUSE wrote:
> > Can you please send the 1.0.49 log excerpts ? > > And may I ask to cpantesters if there is a pb... does not sound > > normal > > to me that http://www.cpantesters.org/show/MarpaX-ESLIF.json is > > absent, c.f. http://matrix.cpantesters.org/?dist=MarpaX-ESLIF+1.0.49 > > . > > > > > > Many thanks.
> > Attached a report generated on a freebsd 10 system.
It seems that $iconv_lib is set, but except for a diagnostic message never used. The following fix seems to help: diff --git a/Makefile.PL b/Makefile.PL index 87b394b..84de10e 100644 --- a/Makefile.PL +++ b/Makefile.PL @@ -230,7 +230,7 @@ my %WriteMakefileArgs = ( %WriteMakefileArgs, OBJECT => 'ESLIF$(OBJ_EXT)', LDFROM => join(' ', '$(OBJECT)', @ldfrom), - dynamic_lib => { OTHERLDFLAGS => $cpp_libs }, + dynamic_lib => { OTHERLDFLAGS => "$cpp_libs $iconv_lib" }, INC => join(' ', map { "-I$_ " } (File::Spec->catdir('src', 'output', 'include'), File::Spec->catdir('src', 'include'), File::Spec->catdir('src', 'output', '3rdparty', 'genericLogger', 'output', 'include'), File::Spec->catdir('src', 'output', '3rdparty', 'genericLogger', 'include'), File::Spec->catdir('src', 'output', '3rdparty', 'genericStack', 'include'), File::Spec->catdir('src', 'output', '3rdparty', 'genericHash', 'include') ) ), );
On Thu Mar 30 16:15:37 2017, SREZIC wrote: Show quoted text
> It seems that $iconv_lib is set, but except for a diagnostic message > never used. The following fix seems to help: > > diff --git a/Makefile.PL b/Makefile.PL > index 87b394b..84de10e 100644 > --- a/Makefile.PL > +++ b/Makefile.PL > @@ -230,7 +230,7 @@ my %WriteMakefileArgs = ( > %WriteMakefileArgs, > OBJECT => 'ESLIF$(OBJ_EXT)', > LDFROM => join(' ', '$(OBJECT)', @ldfrom), > - dynamic_lib => { OTHERLDFLAGS => $cpp_libs }, > + dynamic_lib => { OTHERLDFLAGS => "$cpp_libs $iconv_lib" }, > INC => join(' ', map { "-I$_ " } (File::Spec->catdir('src', > 'output', 'include'), File::Spec->catdir('src', 'include'), > File::Spec->catdir('src', 'output', '3rdparty', 'genericLogger', > 'output', 'include'), File::Spec->catdir('src', 'output', '3rdparty', > 'genericLogger', 'include'), File::Spec->catdir('src', 'output', > '3rdparty', 'genericStack', 'include'), File::Spec->catdir('src', > 'output', '3rdparty', 'genericHash', 'include') ) ), > );
Oups, indeed! Many thanks for your help. Cheers.