cplay package dont install any player, leaving the frontend useless

Bug #139459 reported by QUASAR FREAK
10
Affects Status Importance Assigned to Milestone
cplay (Ubuntu)
Triaged
Medium
Unassigned

Bug Description

Binary package hint: cplay

the package only depends on python and without require a player the program is useless.

Now:

# aptitude show cplay
Package: cplay
State: installed
Version: 1.49-10
Priority: optional
Section: universe/sound
Maintainer: Peter Samuelson <email address hidden>
Uncompressed Size: 238k
Depends: python
Recommends: vorbis-tools, python-pyvorbis
Suggests: mpg321 | madplay | splay | mpg123, speex, mikmod, xmp, sox, python-id3 (>= 1.2-1)
Conflicts: madplay (< 0.15)

Maybe:
Depends: python, mpg321 | madplay | splay | mpg123

Revision history for this message
Barry deFreese (bddebian) wrote :

This package hasn't seen an update in a couple of years. Unfortunately I'm not real familiar with it. Do you have a suggestion for front-ends that are known to work? Thanks for the bug report!

Changed in cplay:
status: New → Incomplete
Revision history for this message
QUASAR FREAK (quasarfreak) wrote :

this frontend work excellent when you install any of the suggested players. =)

Revision history for this message
Barry deFreese (bddebian) wrote :

Aye sorry I meant players. How are we supposed to depend on a player if there are multiple players that work? Who decides which is the "best" player? Thanks.

Revision history for this message
QUASAR FREAK (quasarfreak) wrote :

maybe we can let the user decide asking what of the four players install but we must install one if none of these are present.

Revision history for this message
Stephen Paulger (stephen-paulger) wrote :

Presumably ogg support is wanted and only ogg123 can do that. So I guess moving vorbis-tools to depends from reccomends.

If you look in the cplay source (search for "PLAYERS") you can see it makes a distinction between 3 different types of player FrameOffsetPlayer, TimeOffsetPlayer and NoOffsetPlayer.

NoOffsetPlayer means that you can't rewind or fast-forward using that player.

Offset capable players that support mp3 are splay, mpg123, mpg321 and madplay. When a track is played in cplay it goes through the list in that order to find one that is installed. So if you have all 4 installed it will use splay. So you might say that there is an preference within the program for splay.

If you put "python, vorbis-tools, mpg321 | madplay | splay | mpg123" into depends does it ask the user which one to install if none are installed?

Revision history for this message
Connor Imes (ckimes) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in it recently. We were wondering is this still an issue for you? Can you try with latest Ubuntu release? Thanks in advance.

Revision history for this message
Stephen Paulger (stephen-paulger) wrote :

The dependencies for this program have not changed since this bug was created.

Revision history for this message
Connor Imes (ckimes) wrote :

This application doesn't seem to be in development anymore. I believe this is supposed to be its homepage - http://www.tf.hut.fi/~flu/hacks/cplay/ which has an error message (I think). The link posted takes you elsewhere, which also seems to have an error message.
As for dependencies, having it ask you for a program that uses it is not how the system works. There are certainly many, many packages out there that are useless without programs that use them, but we can't expect the package managers to ask you each time you install such packages (though you could try to file a bug report to request that if you really thought it was necessary or useful w/out being annoying). If you want to see what available packages depend on a specific package, use
 apt-cache rdepends <package>
If you were to run
 apt-cache rdepends python | grep cplay
you would see that cplay depends on python.

I feel that, because of the above, this bug can be closed with status Won't Fix - please let me know what you think. Thank you.

Revision history for this message
QUASAR FREAK (quasarfreak) wrote :

I don't understand why is important that this soft isn't in develop or maintained anymore, the program runs fine once a player is installed, this isn't a cplay bug.

If showing to the user the option to choose any of the compatible players is annoying just install one player or like gentoo, arch and many others distro do show a little message that inform to the user that he must install a compatible player for make this program usefull.

I really don't see the point to the "There are certainly many, many packages out there that are useless without programs that use them" if in other packages are dependency problems must be corrected there too, good packages and dependency resolv is why ppl choose a Debian based distribution instead a RH one.

In my opinion this bug must be fixed here in Ubuntu or Debian and has nothing to do with this program not being developed anymore.

I remark, if you install any of the compatible players Cplay works like a charm.

/* Sorry for my english */

Revision history for this message
Martin Brook (martin.brook) wrote :

There seem to be two problems here:

1. the packaging system isn't flexible enough to include a rule requiring 'at least one of a set of packages' to be installed.

2. the program itself doesn't alert the user to the need to install a compatible player, instead it just fails silently when you attempt to play a track.

I'd suggest the best user experience would be achieved by having a post-installation script (or a first-run script) which checks for the availability of audio players, and (if none are installed) prompt the user to install one via the package manager. I'm sure I've seen this done for other packages before.

I agree with QUASAR_FREAK that the fact that this is no longer under development isn't really relevant.

Revision history for this message
William Grant (wgrant) wrote :

The package should Recommend (rather than Suggest) the disjunction of the supported players. We should find a preferred player and have that as the first disjunct.

Changed in cplay:
importance: Undecided → Medium
status: Incomplete → Triaged
Revision history for this message
William Grant (wgrant) wrote :

> 1. the packaging system isn't flexible enough to include a rule requiring 'at least one of a set of packages' to be installed.

That's not true. We can represent disjunctions fine, as is done in the Suggests line in cplay. The Suggests should be moved to a Recommends or Depends. I'm not quite sure which is better, but perhaps the latter now I think about it.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.