Bug #122469 for POE-Component-Client-MPD: libpoe-component-client-mpd-perl: FTBFS: test suite failures

Thu Jul 13 16:07:43 2017 gregoa [...] cpan.org - Ticket created

From:	gregoa [...] cpan.org
Subject:	libpoe-component-client-mpd-perl: FTBFS: test suite failures

We have the following bug reported to the Debian package of POE-Component-Client-MPD (https://bugs.debian.org/868253): It doesn't seem to be a bug in the packaging, so you may want to take a look. Thanks! The "new mpd" mentioned below is version 0.20.9. ------8<-----------8<-----------8<-----------8<-----------8<----- Package: libpoe-component-client-mpd-perl Version: 2.001-1 Severity: serious This package fails to build from source on current sid due to testsuite failures. The build also hangs for me consistently, though tests.reproducible-builds.org seems to get past that every now and then. This probably regressed with the new mpd around 2017-06-19. Log excerpt (ending with a SIGINT after hanging for ten minutes or so): t/21-conn-non_mpd.t ...... 1..1 ok 1 - wrong server ok Can't locate object method "duration" via package "Audio::MPD::Common::Item::Song" at /<<PKGBUILDDIR>>/blib/lib/POE/Component/Client/MPD/Connection.pm line 228. # Looks like you planned 34 tests but ran 9. t/23-conn-dialog.t ....... 1..34 ok 1 - got a mpd_error event ok 2 - unknown command ok 3 - got a mpd_error event ok 4 - bad password ok 5 - got a mpd_data event ok 6 - no error message ok 7 - got a mpd_data event ok 8 - no error message ok 9 - commands return stuff Dubious, test returned 255 (wstat 65280, 0xff00) Failed 25/34 subtests [...] t/43-cmds-settings.t ..... 1..30 ok 1 - command 'repeat' returned an ok status ok 2 - command 'status' returned an ok status ok 3 - repeat is on ok 4 - command 'repeat' returned an ok status ok 5 - command 'status' returned an ok status ok 6 - repeat is off ok 7 - command 'repeat' returned an ok status ok 8 - command 'status' returned an ok status ok 9 - repeat is on ok 10 - command 'repeat' returned an ok status ok 11 - command 'status' returned an ok status ok 12 - repeat is off ok 13 - command 'fade' returned an ok status ok 14 - command 'status' returned an ok status ok 15 - enabling fading ok 16 - command 'fade' returned an ok status ok 17 - command 'status' returned an ok status ok 18 - disabling fading by default ok 19 - command 'random' returned an ok status ok 20 - command 'status' returned an ok status ok 21 - random is on ok 22 - command 'random' returned an ok status ok 23 - command 'status' returned an ok status ok 24 - random is off ok 25 - command 'random' returned an ok status ok 26 - command 'status' returned an ok status ok 27 - random is on ok 28 - command 'random' returned an ok status ok 29 - command 'status' returned an ok status ok 30 - random is off ok E: ABORT: Received INT signal (requesting cleanup and shutdown) -- Niko Tyni ntyni@debian.org ------8<-----------8<-----------8<-----------8<-----------8<----- Thanks for considering, gregor herrmann, Debian Perl Group

Sat Aug 18 07:56:50 2018 dam [...] cpan.org - Correspondence added

На 13 юли 2017, чт 23:07:43, GREGOA написа: Show quoted text

> We have the following bug reported to the Debian package of > POE-Component-Client-MPD (https://bugs.debian.org/868253): > > It doesn't seem to be a bug in the packaging, so you may want to take > a look. Thanks! > > The "new mpd" mentioned below is version 0.20.9.

It appears song duration is now reported as 'duration' and not as 'time' as it was reported before. Here's a patch that accounts for that change, handling any 'duration' keys as if they were 'time', keeping compatibility with Audio::MPD::Common::Item::Song

Subject:

time-not-duration.patch

Description: adapt to a change in MPD protocol (time -> duration) It appears MPD reports song duration as 'diration', not 'time' as before. . This patch accounts for that change by replacing 'duration' key with 'time' key, keeping compatibility with Audio::MPD::Common::Item::Song Bug-Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=868253 Bug: https://rt.cpan.org/Ticket/Display.html?id=122469 Author: Damyan Ivanov <dmn@debian.org> --- a/lib/POE/Component/Client/MPD/Connection.pm +++ b/lib/POE/Component/Client/MPD/Connection.pm @@ -217,6 +217,10 @@ sub _got_data { my ($k,$v) = split /:\s+/, $input, 2; $k = lc $k; $k =~ s/-/_/; + if ( $k eq 'duration' ) { + $k = 'time'; + $v = int($v + 0.5); + } if ( $k eq 'file' || $k eq 'directory' || $k eq 'playlist' ) { # build a new amc-item

Sat Aug 18 07:56:51 2018 The RT System itself - Status changed from 'new' to 'open'

Sat Aug 18 08:31:51 2018 jquelin [...] cpan.org - Correspondence added

On Sat Aug 18 07:56:50 2018, DAM wrote: Show quoted text

> It appears song duration is now reported as 'duration' and not as > 'time' as it was reported before. > > Here's a patch that accounts for that change, handling any 'duration' > keys as if they were 'time', keeping compatibility with > Audio::MPD::Common::Item::Song

Thanks. Unfortunately I don't have that much free time those days. Do you want comaint on the dist?

Sat Jan 19 09:37:26 2019 fsfs [...] debian.org - Correspondence added

Show quoted text

> On Sat Aug 18 07:56:50 2018, DAM wrote:

> > It appears song duration is now reported as 'duration' and not as > > 'time' as it was reported before.

I believe a more correct fix for this is extending the list of possible song attributes, see https://github.com/jquelin/audio-mpd-common/pull/4

Sat Jan 19 14:23:32 2019 jquelin [...] cpan.org - Correspondence added

On Sat Jan 19 09:37:26 2019, fsfs@debian.org wrote: Show quoted text

> > On Sat Aug 18 07:56:50 2018, DAM wrote:

> > > It appears song duration is now reported as 'duration' and not as > > > 'time' as it was reported before.

> > I believe a more correct fix for this is extending the list of > possible song attributes, see https://github.com/jquelin/audio-mpd- > common/pull/4

Unfortunately, I don't have the time needed to work on my modules currently. I am open to grant comaint to whoever wants to help.