Bug #2443 for MP3-Info: Patch to overlay proper tracknumbers when id3v2 is of the form m/n

Sun Apr 27 11:17:12 2003 Guest - Ticket created

Subject:

Patch to overlay proper tracknumbers when id3v2 is of the form m/n

Some encoders (itunes, for instance) write the id3v2 tracknumber field as 1/n, 2/n, etc. where n is the number of tracks on the cd. As far as I can tell, perl interprets this as a string representation of a fraction, and then rounds it to integer, so that all the track numbers are 0 or 1. This patch chops off the tail of the tracknumber string, leaving only the number (and not the "/n") before writing it over the id3v1 value.

Download Info.pm.diff.2.2
application/octet-stream 445b

Message body not shown because it is not plain text.

Sun Apr 27 11:27:04 2003 cnandor [...] cpan.org - Correspondence added

[guest - Sun Apr 27 11:17:12 2003]: Show quoted text

> Some encoders (itunes, for instance) write the id3v2 tracknumber field > as 1/n, 2/n, etc. where n is the number of tracks on the cd. > > As far as I can tell, perl interprets this as a string representation > of a fraction, and then rounds it to integer, so that all the track > numbers are 0 or 1. > > This patch chops off the tail of the tracknumber string, leaving only > the number (and not the "/n") before writing it over the id3v1 value.

Oh, you mean "perl interprets this as a string representation" specifically when writing ID3v1?

Sun Apr 27 11:53:55 2003 Guest - Correspondence added

[CNANDOR - Sun Apr 27 11:27:04 2003]: Show quoted text

> [guest - Sun Apr 27 11:17:12 2003]: >

> > Some encoders (itunes, for instance) write the id3v2 tracknumber

> field

> > as 1/n, 2/n, etc. where n is the number of tracks on the cd. > > > > As far as I can tell, perl interprets this as a string

> representation

> > of a fraction, and then rounds it to integer, so that all the track > > numbers are 0 or 1. > > > > This patch chops off the tail of the tracknumber string, leaving

> only

> > the number (and not the "/n") before writing it over the id3v1

> value. > > Oh, you mean "perl interprets this as a string representation" > specifically when writing ID3v1?

Sorry for not being clear. This is when the calling app calls get_mp3tag without setting version. When version=2, it returns the id3v2 as a string, and with version = 1 it returns the id3v1 as an integer. The problem comes when it tries to return the id3v2 string as an integer.

Sun Apr 27 13:34:21 2003 cnandor [...] cpan.org - Correspondence added

[guest - Sun Apr 27 11:53:55 2003]: Show quoted text

> This is when the calling app calls get_mp3tag without setting version. > When version=2, it returns the id3v2 as a string, and with version = 1 > it returns the id3v1 as an integer. The problem comes when it tries to > return the id3v2 string as an integer.

As best I can tell, it doesn't return it as an integer. Your caller may *treat* it as an integer, and that is worth a documentation patch, but I don't see a problem with returning the track number as "n/m".

Sun Apr 27 14:36:16 2003 Guest - Correspondence added

[CNANDOR - Sun Apr 27 13:34:21 2003]: Show quoted text

> [guest - Sun Apr 27 11:53:55 2003]: >

> > This is when the calling app calls get_mp3tag without setting > > version. When version=2, it returns the id3v2 as a string, and with > > version = 1 it returns the id3v1 as an integer. The problem comes > > when it tries to return the id3v2 string as an integer.

> > As best I can tell, it doesn't return it as an integer. Your caller > may *treat* it as an integer, and that is worth a documentation patch, > but I don't see a problem with returning the track number as "n/m".

A problem I can see is that calling get_mp3tag without specifying a version is most likely to happen with legacy programs, which are the least likely to be able to cope with getting a string as a track number. I don't understand perl's typing philosophy very well, so I could be very wrong.