2nd pass and -destpath parameter

Island · Post by **Island** » Sun Jun 23, 2024 1:39 pm

Hello!
There is an error while eac3to is trying to strip superfluous zero bytes in 2nd pass. That happens only when there is a -destpath parameter in eac3to.ini.
Here is a log file:

eac3to v3.51
command line: eac3to audio.thd+ac3 test.flac -removedialnorm
------------------------------------------------------------------------------
Running in normal mode
Removing dialnorm
TrueHD/AC3, 5.1 channels, 48kHz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48kHz, dialnorm: -27dB)
Extracting TrueHD stream...
Removing TrueHD dialog normalization...
Decoding with libav/ffmpeg...
Encoding FLAC with libFlac...
Creating file "E:\Projects\eac3to jobs\test.flac"...
The original audio track has a constant bit depth of 16 bits.
Superfluous zero bytes detected, will be stripped in 2nd pass.
The source file "test.pass1.flac" could not be opened. <ERROR>
The source file "test.pass1.flac" could not be opened. <ERROR>

Here is settings from eac3to.ini:

Code: Select all

-normal
-keepDialnorm
-progressnumbers
-showall
-minDuration=002
-destpath="E:\Projects\eac3to jobs\"

Post by **Curly** » Sun Jun 23, 2024 3:55 pm

Welcome to the forum and nice avatar!

Please provide a source file that I can use to duplicate the issue and test the fix I have in mind.

Island · Post by **Island** » Sun Jun 23, 2024 4:37 pm

Curly wrote: ↑
Sun Jun 23, 2024 3:55 pm
Welcome to the forum and nice avatar!

Thank you!

I sent you a link in a private message.

Post by **Curly** » Sun Jun 23, 2024 4:50 pm

Thank you. Investigating...

Post by **Curly** » Sun Jun 23, 2024 7:08 pm

Got it fixed but I have to do regression testing and Rocky wants to do a code review. Tomorrow.

Some .wav operations may also be similarly bugged. I'll check that.

Post by **Rocky** » Tue Jun 25, 2024 3:56 pm

Sorry, it's going to be a few more days as we are facing a crisis with Sweetie Pie. Thank you for your patience.

Post by **Curly** » Fri Jul 12, 2024 8:58 am

Guys, please test this new version. It adds a -nolog INI file option, and the bug of this thread should be fixed.

http://rationalqm.us/eac3to/eac3to_3.52_test.rar

Island · Post by **Island** » Fri Jul 12, 2024 9:22 am

Unfortunately I have a malfunction in my computer's power supply unit. Had to take it to repairs. Therefore I can't test the new version right now. Repairs may take about a week. I'm sorry

Post by **Curly** » Fri Jul 12, 2024 10:54 am

No problem, sir. Report back when you can.

Island · Post by **Island** » Wed Jul 17, 2024 12:13 pm

With 3.52 test version my problem was solved. Thank you!

eac3to v3.52
command line: eac3to audio.thd+ac3 test.flac -removedialnorm
------------------------------------------------------------------------------
Running in normal mode
Removing dialnorm
TrueHD/AC3, 5.1 channels, 48kHz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48kHz, dialnorm: -27dB)
Extracting TrueHD stream...
Removing TrueHD dialog normalization...
Decoding with libav/ffmpeg...
Encoding FLAC with libFlac...
Creating file "E:\Projects\eac3to jobs\test.flac"...
The original audio track has a constant bit depth of 16 bits.
Superfluous zero bytes detected, will be stripped in 2nd pass.
Starting 2nd pass...
Decoding FLAC...
Reducing depth from 24 to 16 bits...
Encoding FLAC with libFlac...
Creating file "E:\Projects\eac3to jobs\test.flac"...
The processed audio track has a constant bit depth of 16 bits.
eac3to processing took 25 seconds.
Done.

Post by **Curly** » Wed Jul 17, 2024 2:34 pm

That's great. Thank you for the update. If possible please continue using that version and let me know if you run into any regressions for other things. If all is well after a few days, I'll release it.