(regression) Huge delay in analyze phase

[GoogleCodeExporter]

Using rubyripper 0.5.7-0.0 on Ubuntu 10.04 amd64:

rrip_gui stalls for 1-5 minutes to compare identical .wav files.
It says it is in the analyzing phase, but there must be a bug because the 
loooong delay occurs even if the two ripped .wav files are exactly identical 
(this can be verified by using a terminal window to find the temp .wav files 
and running cmp on them).

Last year, with an older version of rubyripper, I ripped my entire collection 
of 200+ CDs and never noticed this problem.  Now it happens all the time and it 
can take over an hour to rip a single CD which has no actual errors.  In other 
words, I think this is a regression.

Perhaps rubyripper is jumping to some expensive differencing algorithm without 
first checking that the files are actually different....?

To reproduce:
1. Start rrip_gui, set Match all chunks=2 times, Match erroneous chunks=3 times 
Maximum trials=5
2. Begin ripping a CD.
3. Wait until cdparanoia has finished (twice) and rrip_gui says it is entering 
the checking phase
4. In another terminal, find the two temporary .wav files and run "cmp" on 
them.  Usually they will be identical.  If so, and if rrip_gui is still 
thinking, then the bug has been demonstrated.

Original issue reported on code.google.com by jim.av...@gmail.com on 8 Aug 2010 at 11:31

• Merged into: #348

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Dupe. Please see issue #348.  Basically use ruby 1.9 if possible.

Original comment by mordbr...@gmail.com on 9 Aug 2010 at 4:15

[GoogleCodeExporter]

Original comment by boukewou...@gmail.com on 29 Aug 2010 at 9:49

• Changed state: Duplicate