[X] My Assistant

[mmortal03]

It would be awesome to have it eventually handle FLAC files, but no rush. Excellent work!

[Vietwoojagig]

CUE Tools v1.4.1:

• Added: Hidden option to write PREGAP instead of INDEX 00 for the first track of a single file CUE sheet (only when the "Create CUE sheet only" option is used).

The settings are located in "%AppData%\CUE Tools\settings.txt", set UsePregapForFirstTrackInSingleFile to 1 to enable this new option.  See 2 posts above for screenshot and download link.

Nice Tool!

Two things I want to mention:

1. International characters in the filename still not work (éàç...)
2. What happens, if you choose "Gaps Left Out", and the gaps are not silent? If you remove them, could you add an option to add them in the previous track, if they are note empty and mark them somehow in the cue (e.g REM PAUSE 04:03:23)

thanks

[Moitah]

CUE Tools v1.4.2:

• Fixed: Extended ASCII characters weren't working properly.
• Rejects CUE sheets with POSTGAP command.

[Link removed, see here for updated link] - .NET Framework 2.0 required

@Vietwoojagig: Thanks for the bug report, I should have looked into this when DrGreen said something on the first page. It should be fine now. Regarding "Gaps Left Out", it will remove them even if they contain audio. This is the same behavior as EAC, and I would suggest using one of the other CUE styles if you want to keep them.

This post has been edited by Moitah: Nov 12 2006, 07:58

[Moitah]

CUE Tools v1.4.3:

• Fixed: Small range of extended ASCII characters were being changed for filename creation.

[Link removed, see here for updated link] - .NET Framework 2.0 required

Oops, so it wasn't completely fixed in 1.4.2 (certain characters such as € and ™ didn't work).

This post has been edited by Moitah: Nov 12 2006, 07:59

[djet]

Great!! Thank you!

[Moitah]

FLAC input files are now fully supported (output is converted to WAV, though). A new version will be released after I do some benchmarking and optimization if needed. For those curious, I accomplished this by writing a wrapper for libFLAC in Managed C++ for simple decoding (no encoding, reading/writing tags, etc).

[mmortal03]

I am assuming you haven't posted the FLAC decoding one? Either way, thanks a lot!

[Moitah]

Not yet, it needs more testing. But either tonight or tomorrow I'll release it.

[Moitah]

CUE Tools v1.5.0:

• Added: Support for reading audio from FLAC files.

[Link removed, see here for updated link] - .NET Framework 2.0 required

This post has been edited by Moitah: Nov 12 2006, 07:59

[mmortal03]

There is something about parenthesis that this new version doesn't like.

[Moitah]

Only for FLAC files? You are talking about the filename, right? Can you give me an example?

EDIT: Probably unrelated, but it doesn't properly close FLAC files. This will be fixed tomorrow.

This post has been edited by Moitah: Nov 4 2006, 05:43

[Vietwoojagig]

Regarding "Gaps Left Out", it will remove them even if they contain audio.  This is the same behavior as EAC, and I would suggest using one of the other CUE styles if you want to keep them.

OK. But maybe you could easily add two new Output syles:
* Gaps Left Out (if not silent: Append)
* Gaps Left Out (if not silent: Prepend)

This would ensure, that only silent gaps would be removed.

You know, I already have these "Gaps Appended"-Cue-Sheeds and the only thing I need is the removement of only the silent Gaps at the end of each file, and keep the non-silent gaps. I know of CD's where both types of gaps (silent & non-silent) occure.

Please. You've done a great job so far and this shouldn't be too complicated for someone with your programming-skills.

[Moitah]

CUE Tools v1.5.1:

• Fixed: FLAC files weren't being closed completely.
• Fixed: Possible problem opening FLAC files with strange characters in the filename.
• Optimized FLAC sample copy loop.

[Link removed, see here for updated link] - .NET Framework 2.0 required

For FLAC decompression, the loop that copies the samples to a managed Int32 array has been optimized (takes less than half the time it used to), which saved about 1.5 seconds of CPU time when I tested decompressing a ~41 minute CD image.

This post has been edited by Moitah: Nov 12 2006, 07:59

[mmortal03]

CUE Tools v1.5.1:[list]
[*]Fixed: Possible problem opening FLAC files with strange characters in the filename.

By the way, I am really sorry if you spent a lot of time looking for this "bug", because there actually wasn't a bug regarding parantheses, as I instead miscontrued a misnamed file in a cuesheet that ALSO contained parantheses as the bug that I reported to you. Wow, and the funny thing is that I picked one with a wrong name in it randomly, out of all my cuesheets on my computer. Thanks again for all your hard work!

[Moitah]

It's okay, the fact that libFLAC didn't accept Unicode filenames bothered me when I started working on FLAC support, and even if it's not noticable in CUE Tools (I'm not sure...), I'm glad to have Unicode filenames working in case I or anyone else reuse FLACDotNet in other software.

This post has been edited by Moitah: Mar 2 2006, 05:28

[djet]

Is it possible to enable WavPack output?

[Moitah]

I'd rather work on FLAC output first, but if I get that done I'll see about supporting WavPack.

[Vietwoojagig]

CUE Tools v1.4.2:

• Rejects CUE sheets with POSTGAP command.

Why do you reject CUEs with POSTGAP? What is the reason?

[Moitah]

I didn't realize there was a POSTGAP command until recently after I read the CDRWIN manual. The older versions will just pass it through directly which is probably wrong. Instead of figuring out how to handle it, I made it reject CUE sheets with POSTGAP because I've never seen one. Do you know of any application that creates CUE sheets like this?

[Fandango]

I get this error: "Index length cannot be 0."

This is the cue sheet, what's wrong?

CODE

PERFORMER "Can"
TITLE "Sacrilege CD1"
FILE "01 - Pnoom, Moon Up Mix.flac" WAVE
TRACK 01 AUDIO
TITLE "Pnoom / Moon Up Mix"
PERFORMER "Can"
INDEX 01 00:00:00
TRACK 02 AUDIO
TITLE "Spoon / Sonic Youth Mix"
PERFORMER "Can"
INDEX 00 00:54:40
FILE "02 - Spoon, Sonic Youth Mix.flac" WAVE
INDEX 01 00:00:00
TRACK 03 AUDIO
TITLE "Blue Bag (Inside Paper) / Toroid Mix"
PERFORMER "Can"
INDEX 00 06:05:53
FILE "03 - Blue Bag (Inside Paper), Toroid Mix.flac" WAVE
INDEX 01 00:00:00
TRACK 04 AUDIO
TITLE "Tango Whiskeyman / A Guy Called Gerald Mix"
PERFORMER "Can"
INDEX 00 06:52:23
FILE "04 - Tango Whiskeyman, A Guy Called Gerald Mix.flac" WAVE
INDEX 01 00:00:00
TRACK 05 AUDIO
TITLE "TV Spot / Bruce Gilbert Mix"
PERFORMER "Can"
INDEX 00 05:10:52
FILE "05 - TV Spot, Bruce Gilbert Mix.flac" WAVE
INDEX 01 00:00:00
TRACK 06 AUDIO
TITLE "Vitamin C / U.N.K.L.E. Mix"
PERFORMER "Can"
INDEX 00 03:50:03
FILE "06 - Vitamin C, U.N.K.L.E. Mix.flac" WAVE
INDEX 01 00:00:00
TRACK 07 AUDIO
TITLE "Halleluhwah / Halleluwa Orbus 2"
PERFORMER "Can"
INDEX 00 08:17:00
FILE "07 - Halleluhwah, Halleluwa Orbus 2.flac" WAVE
INDEX 01 00:00:00
TRACK 08 AUDIO
TITLE "Oh Yeah / Sunroof Mix"
PERFORMER "Can"
INDEX 00 09:18:50
FILE "08 - Oh Yeah, Sunroof Mix.flac" WAVE
INDEX 01 00:00:00

Btw, what's up with cue sheets that contain data tracks. Did you get my PM? Will you support them in some way? Either pass the file+track through or omit the data track completely?

[valnar]

I get this error: "Index length cannot be 0."

I am having the same problem. Is there a fix (or workaround)?

Robert

[Fandango]

seems like Moitah's taking a break... ^^

[Moitah]

I get this error: "Index length cannot be 0."

This is the cue sheet, what's wrong?

Let's assume the length of the first file was 00:54:40. Converting it to a single-file CUE would result in:

CODE

TRACK 01 AUDIO
  INDEX 01 00:00:00
TRACK 02 AUDIO
  INDEX 00 00:54:40
  INDEX 01 00:54:40

Which doesn't make any sense to me (that's why I throw an error in this case). Was this file written by EAC? If you rip the CD in single-file mode does EAC or whatever write any duplicate timestamps like that in the cuesheet?

Btw, what's up with cue sheets that contain data tracks. Did you get my PM? Will you support them in some way? Either pass the file+track through or omit the data track completely?

Yeah I got it, sorry for not responding until now . Is the data track saved to a file somewhere during the ripping process?

[Fandango]

If you rip the CD in single-file mode does EAC or whatever write any duplicate timestamps like that in the cuesheet?

I don't know...

I got that rip from a good friend... so I can only speculate about the way it was created, sorry. It could be that the cue sheet got messed up in a text editor. Maybe. At least that's my guess.

I think valnar might be able to test this... in case he didn't get his rip from a good friend, too.

Is the data track saved to a file somewhere during the ripping process?

No.

Well, this problem here is rooted in EAC's non-compliant way of creating cue sheets and its incapability of extracting data tracks. When creating all four kinds of cue sheets for a Audio+Data CD, the only cue sheet with this "TRACK 08 MODEx/2xxx" in it was the one made with the method labeled "(non-compliant)". There's also no corresponding "FILE", which also brakes the standard apart from the MODEx/2xxx and the obvious fact that mixing single WAV and a raw data track in one cue isn't really supported by a lot of apps (I know none).

Anyway this is what EAC gives us for such a Multi-Session CD in case the non-compliant cue sheet was created:

TRACK 08 MODEx/2xxx
TITLE "CD-Rom Track For PC and Mac "
PERFORMER "Creation Rebel"
INDEX 00 04:10:15

But a slightly more compliant data track entry would look like this:

FILE "Track 8.bin" BINARY
TRACK 08 MODE2/2352
TITLE "CD-Rom Track For PC and Mac "
PERFORMER "Creation Rebel"
ISRC 000000000000
INDEX 00 04:10:15

Since the data track won't be extracted by EAC, when using copy - EAC has no data extraction capabilities whatsoever, thus creating only a pseudo data track entry - it could at least write the cue sheet differently, commenting the data track entry out for instance.

Ok, so back to CUETools:

The error messages of CUETools are as follows: When using the unmodified cue sheet I get the "Track type must be audio." which makes sense because CUETools think TRACK 08 belongs to the last FILE which is audio.

When correcting the cue sheet my way, but entering a bogus FILE path, I get the error message that the file is missing. That's correct, too.

When providing a .bin file. I get the error message "Unsupported audio input file". Which doesn't make sense since FILE "" BINARY is binary and not audio. At least CUETools is trying!

So, what to do?

That's very, very tricky IMHO.

Deleting the last TRACK by hand works just fine, but of course it's more comfortable if CUETools would ignore or correct them in some way. Since EAC does create these broken cue sheets and you can't expect a newbie to edit his cue sheets... it's important that you do something: Either omitting, appending or automatically correcting these MODEx/2xxx tracks... simply leaving them out would be the easiest way and EAC's standard behaviour anyway when converting from non-compliant to any of the other three cue sheet modes.

In case you want to go beyond EAC compatibility, for example putting Audio+Data in one raw image file in case the Data Session was extracted with another ripping application other than EAC or was recreated with mkisofs from the copied files, then here are some thoughts of mine:

I'm not that familiar with the cue sheet standards, but still I thought a fixed FILE '' BINARY won't do any sane application any harm. So I've tested it with two "single WAV" cue sheets - one appended with the broken and unchanged data track and one with a fixed and compliant one - and loaded them into foobar2000 0.9: got me the complaint about data tracks not being supported. Then I've tested the same two cue sheets with Burrrn, both gave me erros. Even IsoBuster does not support single audio wav + data track. Well, this is not a surprise at all, since it's not much different from a non-compliant cue sheet like EAC is popular for, with the exception that not the INDEX/TRACK order is the problem but two image files in one cue sheet being of incompatible format, one WAV and the other one RAW (or BINARY).

Nevertheless my suggestion is to support at least two of the things I mentioned above, making it selectable via a check box: omitting the data track or commenting it out (for EAC compliance and even 100% reconstruction of a non-compliant cue sheet later with that MODEc/2xxx if you like) is a must IMHO. Then either simply appending + autofixing the TRACK ?? MODE2/2352 (like I did in my second code snippet - FILE path might be requested via a "open file dialog" or simply "" used) or the making of a true single RAW image and compliant cue sheet (like IsoBuster creates them for example). Personally I would go for implementing the latter since it is more widely supported than this double image cue sheet I made by hand. All applications I tested it with (foobar2000, Burrrn, IsoBuster and burnatonce) refused to accept it for the various reasons. So people who want to make a true 1:1 backup can use the single RAW image + cue, and those wanting to do stuff with the audio can shoose the EAC (omitting) approach.

Cheers!

This post has been edited by Fandango: Aug 2 2006, 21:56

[greynol]

CODE

TRACK 01 AUDIO
  INDEX 01 00:00:00
TRACK 02 AUDIO
  INDEX 00 00:54:40
  INDEX 01 00:54:40

It looks like you are working with wave files that were created with gaps left out (EDIT: or gaps appended to the next track) and are trying to use a noncompliant CUE to put them back together.

My guess is the person that you downloaded them from either ripped them incorrectly or split an image with the wrong option.

This post has been edited by greynol: Aug 2 2006, 21:59