I have created a 4.31 beta version of WavPack for Unix and Linux. The biggest change is that now wildcards are expanded by the shell (as God intended) and so many filename and filepath related problems are fixed. Unfortunately this required a small syntax change (for the *nix version only) that -o is now used when specifying an output file name or path.

Also, I have attempted to eliminate the compiler requirement that the char type default to signed (thanks Sebastian!) and have attempted to portably implement the -t (timestamp) option (thanks Graue!)

And, of course, WvGain is added (thanks David Robinson!)

http://www.wavpack.com/wavpack-4.31.tar.bz2

Please let me know about any troubles...

Everything works fine, as far as I can tell

Tried to compile on Ubuntu.

./configure went okay

During compile I got
/usr/bin/ld: cannot find -lcurses
collect2: ld returned 1 exit status
make: *** [wavpack] Error 1

BTW this version is much welcomed, since at the moment I have to use a for loop to supply wavpack with more than one file.

looks good.

packages are uploading to rarewares/debian atm

thanks bryant!


later

I suppose you're talking about Ken Thompson...



You have to install ncurses-dev.


I'll try to provide compiles for as many Unixen as possible at RareWares ASAP

Will be there a Windows-Version availible ??

Greets

so... here's a report (i'm assuming you wanted one ? )

wvgain is working great!, the library is working (xmms-wavpack still works ) and i haven't had any problems with the frontends.

some comments:

1. replaygain doesn't seem to be detected by the xmms-plugin - i'll test some more, and also with LAMIP ("wvunpack -ss" shows replaygain info, so it looks like it should work)
2. will replaygain analysis be added to the wavpack frontend? i.e. by default or by a switch?
3. a tool like ogginfo would be nice for wavpack files (otherwise it is hunting down all these options in wavpack/wvunpack/wvgain)
4. wvgain should be able to output WHAT gain is present without analyzing it (i.e. output the appropriate information from "wvunpack -ss" and maybe even deliberately steal from the musepack replaygain tool

just some ideas! wavpack is becoming more and more my lossless encoder of choice


later

WavPack has been working for a week, with foobar2000 transcoding a HDD of Monkey's Audio files. It went smoothly, and there are three more drives waiting. Indeed, WavPack is the lossless format of my choice, too.

Now I was able to compile 4.31b statically on a dual P3 testdrive server, so it will work on an old box with glibc 2.2.4. So far everything works, and foobar is happy with the generated replaygain tags.This would certainly be an appreciated feature, ie WavPack encoding and replaygain calculation in one step, with a single binary. This would also be nice in the windows version.

thanks!

minor nit: on your download page it says static compiles when at least the netbsd compiles are dynamic.

edit: disregard that, it's only half true. the builds are static in that they don't need libwavpack.so, but they still need the usual system libs.

this is not a problem in this case. the binaries run on netbsd 3.0 as well. however, if they had used some outdated system lib, one would need to install compat libs on a higher version of netbsd. the builds are thus only "half-static", so to speak.

Indeed, I call them "static" because they don't require libwavpack.so and other libraries placed at strategical locations.

Also, if you are willing to contribute NetBSD 3 compiles, I would be most pleased to host them at RareWares

None of the development systems I have access to offer NetBSD 3. And matter of factly, Sourceforge's compile farm systems are so outdated, it's depressing:

FreeBSD 4.11
OpenBSD 3.4
NetBSD 2.0.2
Fedora Core 2 - Kernel 2.6
Debian Linux 2.2 - Kernel 2.4

BTW, Sourceforge is precisely the reason why I didn't upload the binaries yet: neither of their two MacOS machines are working, and I didn't want to do the release without MacOS X binaries.


So, if there is some kind soul out there willing to offer me ssh/telnet access to a *nix machine (MacOS, Solaris, Irix, AIX, *BSD...) with development tools, please PM me! I swear I will only use it for development purposes. I won't run DoS or mass mailer tools from there

Regards;

Roberto.

netbsd 3.0 is currently nearing release. the 5th release candidate is out and being polished.

when it's finished, i'd be glad to provide some builds.

That would probably be because it's not implemented yet.

It's on my list of things to do.

ah - well that would make sense then

i should have known that... but until recently haven't been able to replaygain my wavpack files under linux!

so is it something that is planned? or on the backburner...

btw, nice to see ya again, kuniklo -


later

It's definitely planned. I'm hoping it won't be that tricky, actually.

I haven't been devoting much time to this kind of thing recently but since the xmms-wavpack plugin is something I use pretty regularly I have some incentive to make it work.

The Windows version was ready first:

http://www.hydrogenaudio.org/forums/index....ndpost&p=344536

Hi xmixahlx,

Thanks for the feedback (and the debian packages); glad to hear WavPack is working out for you!

There's a complication with adding ReplayGain to the frontend (I assume you mean wavpack.exe and not the Windows GUI "frontend"). Track mode works fine, but I can't figure out a good way to do Album mode. All the files must be processed together, but then how do you add specific tags that are different from file to file (i.e. "Title")? I figured that the stand-alone processor made the most sense.

I do like the idea of a wvgain option (-s for show?) to display the stored ReplayGain information; I'll try to throw that in.

I had thought about adding a feature to wvunpack.exe so that if it is renamed wvinfo.exe it forces -ss and if it is renamed wvcheck.exe it forces -vm. Do you think this would be worth it?

Thanks. Compiles fine now.

thanx for the reply, david -

ok...

oh, btw... wavpack is now in debian unstable (v4.3)! - i'll still be packaging betas, etc. for rarewares, but congrats - i think adoption will go much quicker.

re: wavpack frontend (not the win32 gui or wavpack.exe <-- ;P )
i think that adding only track gain to the frontend is enough (and doesn't need to be default).

re: wvgain
the "-s" option for show sounds like a good idea (and what i was thinking of, too). and the existing album-gain-only option compliments the above addition to the frontend nicely...

re: whatever that binary would be called, eheh
mmm the more i play with wvunpack the more i think another frontend would be unnecessary - if it did exist, perhaps it just collected the options that exclude actual encoding/decoding.



later

*xmixahlx patiently waits*


later

Applying track gain on the fly will actually result in album gain when using image+cuesheet. The WavPack encoder will only see one file, and that one is the whole album. This implies that there are actually choices for everyone; if album gain is desired, image+cuesheet mode can be used. If track gain is desired, one-track-per-file mode is the choice.

As a follow up to this thread, I have finished the update to 4.31 final for both the *nix and win32 versions (including source).

The only significant change was adding the discussed -s command to WvGain to display exiting ReplayGain information to stdout. I decided that adding any ReplayGain scanning to the encoder would be too much trouble for now (I have been trying to get this update behind me for months so I can work on fun stuff!)

Thanks again to everyone for suggestions and testing and packaging and building and hosting. WavPack would be nothing without you all.

Pardon my ignorance, but what Linux player is supporting wavpack ?

http://article.gmane.org/gmane.comp.video.mplayer.user/41135

Christian
matroska project admin

LAMIP, xmms, and there's a gstreamer plugin, used by Quod Libet and amaroK, as far as I know.

And WV works great with Quodlibet (RG, tagging and playing), even though the author doesn't want to advertise the support yet.

Im so badly waiting for mpd to support wavpack

This thread is kind of old, please avoid replying to old threads in the future.
If you want support badly, then you should do it yourself or find someone who will do it for you.