lame 3.98 beta 5

[Robert Hegemann-2]

Hello,

we are at 3.98 beta 5. If no more serious bugs pop-up,
then I would say this is a release candidate for a
full featured 3.98 stable (with sourceforge announcement,
tar ball for download etc.).

I would like the Debian/RPM packaging gurus have a look at
the current packaging status. Are they in good shape and working?


Ciao Robert

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Rogério Brito-2]

Hi, Robert.

On Aug 12 2007, Robert Hegemann wrote:
> we are at 3.98 beta 5. If no more serious bugs pop-up, then I would
> say this is a release candidate for a full featured 3.98 stable (with
> sourceforge announcement, tar ball for download etc.).

Really nice.

> I would like the Debian/RPM packaging gurus have a look at the current
> packaging status. Are they in good shape and working?

No, the Debian packaging is not in good shape (sorry, I was too busy
having a full week and I decided for the first time to participate on
the listening test, which required quite a lot of time from my side, but
it was enlightening).  It seems that compiling after running configure
is now working (well, at least under a pure 64 bit system here).

I will try to update the Debian packaging side of things this weekend,
if all is well with you guys.


Regards, Rogério Brito.

--
Rogério Brito : rbrito@{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Rogério Brito-2]

Ooops, it seems that I spoke too soon.

On Aug 17 2007, Rogério Brito wrote:
(...)
> It seems that compiling after running configure is now working (well,
> at least under a pure 64 bit system here).

That is not true if I tell lame to use libsndfile to take input with the
configure script. OTOH, if I *don't* use configure and just use
Makefile.unix adding the proper information for libsndfile, then it gets
compiled.

BTW, this is the problem that I'm getting after "./configure --with-fileio=sndfile":

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
(...)
 gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame -I.. -O3 -ffast-math -funroll-loops -Wall -pipe -MT decode_i386.lo -MD -MP -MF .deps/decode_i386.Tpo -c decode_i386.c -o decode_i386.o >/dev/null 2>&1
mv -f .deps/decode_i386.Tpo .deps/decode_i386.Plo
/bin/sh ../libtool --tag=CC   --mode=compile gcc -DHAVE_CONFIG_H  -I. -I.. -I  -I../include -I. -I../libmp3lame -I..    -O3  -ffast-math                     -funroll-loops -Wall -pipe  -MT interface.lo -MD -MP -MF .deps/interface.Tpo -c -o interface.lo interface.c
 gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame -I.. -O3 -ffast-math -funroll-loops -Wall -pipe -MT interface.lo -MD -MP -MF .deps/interface.Tpo -c interface.c  -fPIC -DPIC -o .libs/interface.o
interface.c:14:18: error: lame.h: No such file or directory
In file included from ../libmp3lame/machine.h:25,
                 from interface.c:15:
../libmp3lame/version.h:54: error: expected ')' before '*' token
In file included from interface.c:16:
../libmp3lame/VbrTag.h:73: error: expected ')' before '*' token
../libmp3lame/VbrTag.h:74: error: expected ')' before 'const'
../libmp3lame/VbrTag.h:75: error: expected ')' before 'const'
../libmp3lame/VbrTag.h:77: error: expected ')' before '*' token
make[2]: *** [interface.lo] Error 1
make[2]: Leaving directory `/tmp/lame/mpglib'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/tmp/lame'
make: *** [all] Error 2
rbrito@chagas:/tmp/lame$
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Regards, Rogério Brito.

--
Rogério Brito : rbrito@{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting Rogério Brito <[hidden email]> (from Fri, 17 Aug 2007  
12:20:18 -0300):

> Ooops, it seems that I spoke too soon.
>
> On Aug 17 2007, Rogério Brito wrote:
> (...)
>> It seems that compiling after running configure is now working (well,
>> at least under a pure 64 bit system here).
>
> That is not true if I tell lame to use libsndfile to take input with the
> configure script. OTOH, if I *don't* use configure and just use
> Makefile.unix adding the proper information for libsndfile, then it gets
> compiled.
>
> BTW, this is the problem that I'm getting after "./configure  
> --with-fileio=sndfile":

If you don't specify the fileio option for configure, it compiles just fine?

Bye,
Alexander.

--
To stand and be still,
At the Birkenhead drill,
Is a damned tough bullet to chew.
                -- Rudyard Kipling

http://www.Leidinger.net    Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org       netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Rogério Brito-2]

Hi, Alexander and other fellow developers.

Sorry for the late reply, but I had problems with real life (having
quite some problems with my health and also trying to get back to my
studies on Algebraic structures):

On Aug 20 2007, Alexander Leidinger wrote:
> If you don't specify the fileio option for configure, it compiles just
> fine?

Yes, I can compile just fine. There is some problem with the way that
the configure script tells the compiler to glue libsndfile, which is a
pity, because it is more flexible than the default fileio shipped with
lame.

Again, just for reference, here is the output that I get when I am using
--with-fileio=sndlame:

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
 gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame -I.. -O3 -ffast-math -funroll-loops -Wall -pipe -MT interface.lo -MD -MP -MF .deps/interface.Tpo -c interface.c  -fPIC -DPIC -o .libs/interface.o
interface.c:14:18: error: lame.h: No such file or directory
In file included from ../libmp3lame/machine.h:25,
                 from interface.c:15:
../libmp3lame/version.h:54: error: expected ')' before '*' token
In file included from interface.c:16:
../libmp3lame/VbrTag.h:73: error: expected ')' before '*' token
../libmp3lame/VbrTag.h:74: error: expected ')' before 'const'
../libmp3lame/VbrTag.h:75: error: expected ')' before 'const'
../libmp3lame/VbrTag.h:77: error: expected ')' before '*' token
make[2]: *** [interface.lo] Error 1
make[2]: Leaving directory `/home/rbrito/src/sf/lame/mpglib'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/rbrito/src/sf/lame'
make: *** [all] Error 2
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Notice that the problem is caused by the file lame.h not being found,
which is in <include/lame.h>. Obviously, VbrTag.h doesn't have any
syntax problems (as far as I can see in a quick glance).

Just as a simple test, changing the include #include "lame.h" to
#include "include/lame.h" makes it compile some more files and then end
with the same problem a few files later.

(I reverted these changes back now, as I want to have something as close
to the repository as possible while debuggin things)

I just thought that -I../include would be sufficient.

But I finally figured out the problem: it's the "empty" -I option to
gcc: if I manually change I don't have the Makefile, the things in the
mpglib directory all get compiled without errors (but other
directories---like libmp3lame---have problems exactly due to the "bare"
-I option to gcc).


Regards, Rogério Brito.

--
Rogério Brito : rbrito@{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Robert Hegemann-2]

Am 23.08.2007, 14:23 Uhr, schrieb Rogério Brito <[hidden email]>:

> Hi, Alexander and other fellow developers.
>
> Again, just for reference, here is the output that I get when I am using
> --with-fileio=sndlame:
>
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>  gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame -I..  
> -O3 -ffast-math -funroll-loops -Wall -pipe -MT interface.lo -MD -MP -MF  
> .deps/interface.Tpo -c interface.c  -fPIC -DPIC -o .libs/interface.o

Why are the "-I." and "-I.." include directives showing up twice on the
commandline??

The empty "-I" seems to eat the following "-I../include" where the
lame.h is located. There must be some automatism ending up in an empty
variable, or maybe a variable containing white spaces only.

> interface.c:14:18: error: lame.h: No such file or directory
> In file included from ../libmp3lame/machine.h:25,
>                  from interface.c:15:
> ../libmp3lame/version.h:54: error: expected ')' before '*' token
> In file included from interface.c:16:
> ../libmp3lame/VbrTag.h:73: error: expected ')' before '*' token
> ../libmp3lame/VbrTag.h:74: error: expected ')' before 'const'
> ../libmp3lame/VbrTag.h:75: error: expected ')' before 'const'
> ../libmp3lame/VbrTag.h:77: error: expected ')' before '*' token
> make[2]: *** [interface.lo] Error 1
> make[2]: Leaving directory `/home/rbrito/src/sf/lame/mpglib'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/home/rbrito/src/sf/lame'
> make: *** [all] Error 2
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>
> Notice that the problem is caused by the file lame.h not being found,
> which is in <include/lame.h>. Obviously, VbrTag.h doesn't have any
> syntax problems (as far as I can see in a quick glance).
>
> Just as a simple test, changing the include #include "lame.h" to
> #include "include/lame.h" makes it compile some more files and then end
> with the same problem a few files later.
>
> (I reverted these changes back now, as I want to have something as close
> to the repository as possible while debuggin things)
>
> I just thought that -I../include would be sufficient.
>
> But I finally figured out the problem: it's the "empty" -I option to
> gcc: if I manually change I don't have the Makefile, the things in the
> mpglib directory all get compiled without errors (but other
> directories---like libmp3lame---have problems exactly due to the "bare"
> -I option to gcc).
>
>
> Regards, Rogério Brito.
>

--


Ciao Robert

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting robert <[hidden email]> (from Thu, 23 Aug 2007  
15:16:12 +0200):

> Am 23.08.2007, 14:23 Uhr, schrieb Rogério Brito <[hidden email]>:
>
>> Hi, Alexander and other fellow developers.
>>
>> Again, just for reference, here is the output that I get when I am using
>> --with-fileio=sndlame:
>>
>> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>>  gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame -I..
>> -O3 -ffast-math -funroll-loops -Wall -pipe -MT interface.lo -MD -MP -MF
>> .deps/interface.Tpo -c interface.c  -fPIC -DPIC -o .libs/interface.o
>
> Why are the "-I." and "-I.." include directives showing up twice on the
> commandline??

Automatism of the autotools, I think.

> The empty "-I" seems to eat the following "-I../include" where the
> lame.h is located. There must be some automatism ending up in an empty
> variable, or maybe a variable containing white spaces only.

Yes, and I try to get time next week to have a look at it.

Bye,
Alexander.

--
The Russians have put a small ball up in the air.
That does not raise my apprehensions one iota.
                -- Dwight D. Eisenhower

http://www.Leidinger.net    Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org       netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting Rogério Brito <[hidden email]> (from Thu, 23 Aug 2007  
09:23:51 -0300):

> Hi, Alexander and other fellow developers.
>
> Sorry for the late reply, but I had problems with real life (having
> quite some problems with my health and also trying to get back to my
> studies on Algebraic structures):

I hope all goes well...

> On Aug 20 2007, Alexander Leidinger wrote:
>> If you don't specify the fileio option for configure, it compiles just
>> fine?
>
> Yes, I can compile just fine. There is some problem with the way that
> the configure script tells the compiler to glue libsndfile, which is a
> pity, because it is more flexible than the default fileio shipped with
> lame.
>
> Again, just for reference, here is the output that I get when I am using
> --with-fileio=sndlame:
>
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
>  gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame

> But I finally figured out the problem: it's the "empty" -I option to
> gcc: if I manually change I don't have the Makefile, the things in the
> mpglib directory all get compiled without errors (but other
> directories---like libmp3lame---have problems exactly due to the "bare"
> -I option to gcc).

I committed a change which is supposed to fix this. Please test.

Bye,
Alexander.

--
At these prices, I lose money -- but I make it up in volume.
                -- Peter G. Alaquon

http://www.Leidinger.net    Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org       netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Robert Hegemann-2]

Hello Rogério,

does the configure stuff work for you now, with Alexanders latest changes?

Ciao Robert


Am Mittwoch, 29. August 2007 15:44 schrieb Alexander Leidinger:

> Quoting Rogério Brito <[hidden email]> (from Thu, 23 Aug 2007
>
> 09:23:51 -0300):
> > Hi, Alexander and other fellow developers.
> >
> > Sorry for the late reply, but I had problems with real life (having
> > quite some problems with my health and also trying to get back to my
> > studies on Algebraic structures):
>
> I hope all goes well...
>
> > On Aug 20 2007, Alexander Leidinger wrote:
> >> If you don't specify the fileio option for configure, it compiles just
> >> fine?
> >
> > Yes, I can compile just fine. There is some problem with the way that
> > the configure script tells the compiler to glue libsndfile, which is a
> > pity, because it is more flexible than the default fileio shipped with
> > lame.
> >
> > Again, just for reference, here is the output that I get when I am using
> > --with-fileio=sndlame:
> >
> > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> >  gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame
> >
> > But I finally figured out the problem: it's the "empty" -I option to
> > gcc: if I manually change I don't have the Makefile, the things in the
> > mpglib directory all get compiled without errors (but other
> > directories---like libmp3lame---have problems exactly due to the "bare"
> > -I option to gcc).
>
> I committed a change which is supposed to fix this. Please test.
>
> Bye,
> Alexander.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Rogério Brito]

Oh, sorry. I tested but didn't report. :-(

On Oct 7, 2007, at 9:10 AM, Robert Hegemann wrote:
> does the configure stuff work for you now, with Alexanders latest  
> changes?

No, it does not. An empty "-I  " is still generated when libsndfile  
is chosen instead of that of the input routines of lame proper.

BTW, I did the same test on a MacOS X running on Intel (with  
libraries installed by Fink) and the result that I get is close to  
that of the computer under Debian, but not the same (see below).

> Am Mittwoch, 29. August 2007 15:44 schrieb Alexander Leidinger:
>> Quoting Rogério Brito <[hidden email]> (from Thu, 23 Aug 2007
>>> On Aug 20 2007, Alexander Leidinger wrote:
>>>> If you don't specify the fileio option for configure, it  
>>>> compiles just
>>>> fine?
>>>
>>> Yes, I can compile just fine.
(...)
>>> Again, just for reference, here is the output that I get when I  
>>> am using
>>> --with-fileio=sndlame:
>>>
>>> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -  
>>> - - -
>>>  gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame

On the machine runnning Darwin, I get that empty -I "glued" on the  
next -I and everything compiles fine. On the Debian box, this doesn't  
happen and I still get the problem of compilation. Again, changing  
the built Makefiles by hand (after they are generated by config) and  
eliminating manually the empty -I switches makes compilation succeed.

This GCC on Dardwin is the GCC "instrumented" by Apple (which,  
despite it having its number as 4.0.1, seems to have (some) patches  
backported from the GCC 4.1---this is a wild guess of mine).

Is lame getting released soon? I will try to update the debian  
packaging of it. Just please let me know. I don't know how to best  
package this in respect to this build...

>> I committed a change which is supposed to fix this. Please test.

Thank you very much for your efforts, Alexander. For the next  
version, I have some questions regarding the optimization levels of  
lame and compiler versions (some switches seem to have been dropped  
in recent GCCs). I can modify configure.in, but I don't know how to  
generate the configure script from that... :-(


Thanks, Rogério Brito.

--
Rogério Brito : rbrito@{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Robert Hegemann-2]

Hello Rogério!

Am Mittwoch, 10. Oktober 2007 03:39 schrieb Rogério Brito:
> Oh, sorry. I tested but didn't report. :-(
>
> On Oct 7, 2007, at 9:10 AM, Robert Hegemann wrote:
> > does the configure stuff work for you now, with Alexanders latest
> > changes?
>
> No, it does not. An empty "-I  " is still generated when libsndfile
> is chosen instead of that of the input routines of lame proper.

Seeing that the configure files doing some tests like:
        test -n "$SNDFILE_CFLAGS"
I'm wondering, is there some way to let the test fail
if the variable contains white spaces only?
Because, test -n " " will never fail.

> BTW, I did the same test on a MacOS X running on Intel (with
> libraries installed by Fink) and the result that I get is close to
> that of the computer under Debian, but not the same (see below).
>
> > Am Mittwoch, 29. August 2007 15:44 schrieb Alexander Leidinger:
> >> Quoting Rogério Brito <[hidden email]> (from Thu, 23 Aug 2007
> >>
> >>> On Aug 20 2007, Alexander Leidinger wrote:
> >>>> If you don't specify the fileio option for configure, it
> >>>> compiles just
> >>>> fine?
> >>>
> >>> Yes, I can compile just fine.
>
> (...)
>
> >>> Again, just for reference, here is the output that I get when I
> >>> am using
> >>> --with-fileio=sndlame:
> >>>
> >>> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> >>> - - -
> >>>  gcc -DHAVE_CONFIG_H -I. -I.. -I -I../include -I. -I../libmp3lame
>
> On the machine runnning Darwin, I get that empty -I "glued" on the
> next -I and everything compiles fine. On the Debian box, this doesn't
> happen and I still get the problem of compilation. Again, changing
> the built Makefiles by hand (after they are generated by config) and
> eliminating manually the empty -I switches makes compilation succeed.
>
> This GCC on Dardwin is the GCC "instrumented" by Apple (which,
> despite it having its number as 4.0.1, seems to have (some) patches
> backported from the GCC 4.1---this is a wild guess of mine).
>
> Is lame getting released soon? I will try to update the debian
> packaging of it. Just please let me know. I don't know how to best
> package this in respect to this build...

The release of 3.98 stable if overdue.

> >> I committed a change which is supposed to fix this. Please test.
>
> Thank you very much for your efforts, Alexander. For the next
> version, I have some questions regarding the optimization levels of
> lame and compiler versions (some switches seem to have been dropped
> in recent GCCs). I can modify configure.in, but I don't know how to
> generate the configure script from that... :-(
>
>
> Thanks, Rogério Brito.

Ciao Robert

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting Robert Hegemann <[hidden email]> (from Thu, 1 Nov 2007  
13:22:57 +0100):

> Hello Rogério!
>
> Am Mittwoch, 10. Oktober 2007 03:39 schrieb Rogério Brito:
>> Oh, sorry. I tested but didn't report. :-(
>>
>> On Oct 7, 2007, at 9:10 AM, Robert Hegemann wrote:
>> > does the configure stuff work for you now, with Alexanders latest
>> > changes?
>>
>> No, it does not. An empty "-I  " is still generated when libsndfile
>> is chosen instead of that of the input routines of lame proper.
>
> Seeing that the configure files doing some tests like:
> test -n "$SNDFILE_CFLAGS"
> I'm wondering, is there some way to let the test fail
> if the variable contains white spaces only?
> Because, test -n " " will never fail.

test -n "`echo   `" && echo hh

produces no "hh" output. So I think
   test -n "`echo $SNDFILE_CFLAGS`"
should do the trick. I try to get some time later do change configure  
accordingly.

>> Thank you very much for your efforts, Alexander. For the next
>> version, I have some questions regarding the optimization levels of
>> lame and compiler versions (some switches seem to have been dropped
>> in recent GCCs). I can modify configure.in, but I don't know how to
>> generate the configure script from that... :-(

In case I haven't answered this already: just run autoconf (a recent  
version of it) after changing configure.in.

Bye,
Alexander.

--
meeting, n:
        An assembly of people coming together to decide what person or
        department not represented in the room must solve a problem.

http://www.Leidinger.net    Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org       netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Robert Hegemann-2]

>> Seeing that the configure files doing some tests like:
>> test -n "$SNDFILE_CFLAGS"
>> I'm wondering, is there some way to let the test fail
>> if the variable contains white spaces only?
>> Because, test -n " " will never fail.
>
> test -n "`echo   `" && echo hh
>
> produces no "hh" output. So I think
>    test -n "`echo $SNDFILE_CFLAGS`"
> should do the trick. I try to get some time later do change configure  
> accordingly.

I don't know where SNDFILE_CFLAGS originates from, but
it would probably be better to just strip whitespaces only once
from it:
SNDFILE_CFLAGS=`echo $SNDFILE_CFLAGS`


Ciao Robert


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Robert Hegemann-2]

> I don't know where SNDFILE_CFLAGS originates from, but
> it would probably be better to just strip whitespaces only once
> from it:
> SNDFILE_CFLAGS=`echo $SNDFILE_CFLAGS`

Just found some bash example:
# Strip the whitespace from the beginning and end.
LRFDATE=`expr "$LRFDATE" : '[[:space:]]*\(.*\)[[:space:]]*$'`




Ciao Robert


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting "Robert Hegemann" <[hidden email]> (Fri, 2 Nov 2007 12:19:31 +0100):

> > I don't know where SNDFILE_CFLAGS originates from, but
> > it would probably be better to just strip whitespaces only once
> > from it:
> > SNDFILE_CFLAGS=`echo $SNDFILE_CFLAGS`
>
> Just found some bash example:
> # Strip the whitespace from the beginning and end.
> LRFDATE=`expr "$LRFDATE" : '[[:space:]]*\(.*\)[[:space:]]*$'`

Sorry, no cookie for you. We can't use bash syntax. Apart from that (I
haven't checked if my plain sh is able to do this), the echo is just
enough I think, no need for fancy things which need a lot of docs to
understand.

Bye,
Alexander.

--
A princess should not be afraid -- not with a brave knight to protect
her.
                -- McCoy, "Shore Leave", stardate 3025.3
http://www.Leidinger.net  Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org     netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting "Robert Hegemann" <[hidden email]> (Fri, 2 Nov 2007 12:13:53 +0100):

> >> Seeing that the configure files doing some tests like:
> >> test -n "$SNDFILE_CFLAGS"
> >> I'm wondering, is there some way to let the test fail
> >> if the variable contains white spaces only?
> >> Because, test -n " " will never fail.
> >
> > test -n "`echo   `" && echo hh
> >
> > produces no "hh" output. So I think
> >    test -n "`echo $SNDFILE_CFLAGS`"
> > should do the trick. I try to get some time later do change configure  
> > accordingly.
>
> I don't know where SNDFILE_CFLAGS originates from, but
> it would probably be better to just strip whitespaces only once
> from it:
> SNDFILE_CFLAGS=`echo $SNDFILE_CFLAGS`

I give it a try...

Bye,
Alexander.

--
To see how long it takes a command to run, type the word "time" before the
command name.
                -- Dru <[hidden email]>
http://www.Leidinger.net  Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org     netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Rogério Brito-2]

Hi Robert and Alexander.

On Aug 17 2007, Rogério Brito wrote:
> On Aug 12 2007, Robert Hegemann wrote:
> > I would like the Debian/RPM packaging gurus have a look at the
> > current packaging status. Are they in good shape and working?
>
> No, the Debian packaging is not in good shape

I tweaked the Debian packaging (more to come in the near future), but I
guess that, at least, it is in a releaseable state. Oh, and the dreaded
problem with libsndfile seems to be gone.


Regards, Rogério Brito.

--
Rogério Brito : rbrito@{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev

[Alexander Leidinger]

Quoting Rogério Brito <[hidden email]> (Wed, 9 Jan 2008 22:00:56 -0200):

> Hi Robert and Alexander.
>
> On Aug 17 2007, Rogério Brito wrote:
> > On Aug 12 2007, Robert Hegemann wrote:
> > > I would like the Debian/RPM packaging gurus have a look at the
> > > current packaging status. Are they in good shape and working?
> >
> > No, the Debian packaging is not in good shape
>
> I tweaked the Debian packaging (more to come in the near future), but I
> guess that, at least, it is in a releaseable state. Oh, and the dreaded
> problem with libsndfile seems to be gone.

Nice to hear that my changes seem to work. :)

Bye,
Alexander.

--
Ferengi Rule of Acquisition #60:
        Keep your lies consistent.
                -- ST: Legends of the Ferengi
http://www.Leidinger.net  Alexander @ Leidinger.net: PGP ID = B0063FE7
http://www.FreeBSD.org     netchild @ FreeBSD.org  : PGP ID = 72077137

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Lame-dev mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/lame-dev