SoK Project Query

vedant agarwala vedant.kota at gmail.com
Mon Nov 24 14:36:23 UTC 2014


On Mon, Nov 24, 2014 at 5:56 PM, Nitul Datt <nitul1991 at gmail.com> wrote:

> On 11/24/14, vedant agarwala <vedant.kota at gmail.com> wrote:
> > Hello Nitul,
> >
> > I am not thoroughly familiar with the code but here is my suggestion:
> >
> > The CollectionScanner should find files and represent them. So if it
> finds
> > a track it represents it as such. You should have a one Track object
> > representing a track and a new CueSheet (or similar) class object to
> > represent a cue sheet.
> >
>
> Won't that make two entries for the same file, ie one for the music
> file and one for the corresponding cue sheet, since the music file
> will be scanned as such and the cue sheet as well? The solution to
> this seems to be that we only store the cue sheet file and not the
> corresponding track. This seems like a better approach than what I was
> trying earlier. What do you think?
>

I thought that a cue sheet is a separate file. Anyway, you are right: only
one object should represent one file. So we should have just a CueSheet
file.


>
> > But when these files are shown in the Database collection/playlist etc.
> in
> > amarok they should be shown as completely different (according to the
> music
> > file and cue sheet) files (i.e. different TrackPtr 's should be created
> for
> > each).
> >
>
> Yes, that may be achieved using timecode tracks. I haven't been able
> to find how timecode tracks are represented in the database though.
>
> > The thing that I haven't figured out is that how will amarok "dissect"
> > the CollectionScanner::Track
> > if it is a cue-sheeted track. Can you place some link in the
> > CollectionScanner::Track
> > to the corresponding CollectionScanner::CueSheet without modifying it? Or
> > when the CollectionScanner::CueSheet is accessed by Amarok, it should
> > dissect the CollectionScanner::Track into many TrackPtr 's.
> >
>
> I'll look into it.
>
> > It really depends on how the information fed by the collection scanner to
> > amarok is parsed by amarok. Look into and figure it out.
> >
> > Ask me again if needed.
> >
> > And, always 'cc' amarok-devel. Maybe someone would help you. Even if they
> > don't, they will be glad to know how the amarok community is bonding and
> > growing.
> >
> > Regards,
> > Vedant.
> >
> > On Sun, Nov 23, 2014 at 10:41 PM, Nitul Datt <nitul1991 at gmail.com>
> wrote:
> >
> >> Hey!
> >>
> >> I've run into a bit of a roadblock and require your guidance. How
> >> should I represent the individual tracks in a cue sheeted file? At
> >> present, the collection scanner does not include start and end times
> >> for regular tracks. The existing cue sheet related code creates
> >> timecode tracks for the individual songs in a cue sheeted file. That
> >> being said, should I thus create a new CollectionScanner::TimcodeTrack
> >> class or make changes to the existing CollectionScanner::Track class?
> >>
> >> --
> >> Regards,
> >> Nitul Datt
> >>
> >
>
> --
> Regards,
> Nitul Datt
>

Try and keep things on schedule. You wasted quite some time initially
"going through code".
I still don't have any reviewable code.

Regards,
Vedant.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/amarok-devel/attachments/20141124/9b87c107/attachment.html>


More information about the Amarok-devel mailing list