Bug 1290 - can't get cuesheet from invalid flac files (was: flac produced from eac won't parse)
: can't get cuesheet from invalid flac files (was: flac produced from eac won't...
Status: RESOLVED WONTFIX
Product: Logitech Media Server
Classification: Unclassified
Component: Formats
: 6.0.0
: PC All
: P2 enhancement (vote)
: 8.0.0
Assigned To: Andy Grundman
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2005-04-02 14:24 UTC by michael
Modified: 2009-11-11 14:35 UTC (History)
3 users (show)

See Also:
Category: ---


Attachments
lax checking for parsing with less paranoia (6.53 KB, patch)
2005-07-01 02:40 UTC, michael
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description michael 2005-04-02 14:24:02 UTC
Slimserver currently can't correctly parse the cuesheet block from files with
cuesheets embedded that were created by eac.  
Cuesheets created by cdrdao/cueconvert work fine.
Comment 1 Blackketter Dean 2005-04-04 14:36:54 UTC
michael:  can you attach a sample track to this bug?
Comment 2 michael 2005-04-05 02:37:10 UTC
I think I've figured out what was happening here.
eac was spitting out wavs that didn't align to cd sector boundries.
Audio::FLAC::Header treats this as an error which is technically the right thing
to do. 
However, it appears that only the lead-out in the flac comments at the end of
the cuesheet is affected, and slimserver doesn't actually use that value for
anything.
So it's probably worthwhile to see about being able to suck in the cuesheet in
spite of this error, and process the rest of the data as normal.
I'm not really sure how to do that without stomping on the error checking in the
CPAN module, but I'll take another look at it after I've closed some of the more
pressing bugs.
Comment 3 michael 2005-04-19 13:37:56 UTC
updating summary to better reflect the true core problem.
Comment 4 Blackketter Dean 2005-04-20 15:28:33 UTC
michael:  what's the status of this?
Comment 5 michael 2005-04-20 18:49:11 UTC
I think the status of this would be best described as "enhancement request".

We could add some sort of "lax checking" option to the cpan module, but I'm not
sure if it's really the best idea.

If no one else has stumbled across this yet besides myself, it may not be worth
worrying about.  I'm open to suggestions though.

Comment 6 michael 2005-07-01 02:07:24 UTC
I have a fix for this, but it involves adding a "less paranoia" option to the
preferences (which will hopefully be useful to other areas of the app as well).
 I'm going to hold on to this one until after 6.1 is out, and folks have a
chance to debate adding another pref, and produce translated strings, etc.
Comment 7 michael 2005-07-01 02:40:29 UTC
Created attachment 588 [details]
lax checking for parsing with less paranoia

attaching the fix so it doesn't get lost. (could probably still be polished a
bit.)
Comment 8 Ben Sandee 2005-11-18 07:56:39 UTC
is patch still valid?  anything holding it back from 6.5?
Comment 9 Andy Grundman 2009-07-29 14:58:23 UTC
Moving 7.4 bugs to 8.0.
Comment 10 Andy Grundman 2009-11-11 14:35:07 UTC
This is no longer relevant as we no longer use Audio::FLAC::Header.  Please reopen if you still have problems under 7.4+.