peep-develop — Mailing list for the usage and development of Peep

[Peep-develop] Small project page update

[Michael G. <mg...@ee...>]

  I've removed the "forums" and "task manager" from the sourceforge
project page.  We weren't using them and they were kinda clutter. The
mailing lists are far superior to the forums, which haven't been used
in a while anyway.

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] Re: Aha!

[Michael G. <mg...@ee...>]

  Heh. Of course not! "Probably" here was a figure of speech.  I have
a habit of avoiding total commitment :)

                   -- Mike

On Wed, Sep 11 @ 02:02, xavier renaut wrote:
> On Tue, Sep 10, 2002 at 02:38:42PM -0400, Michael Gilfix wrote:
> |  Got this a little late so I missed the fun. We should probably
> |setup another mailing list for commit emails. 
> 
> of course we should set up a different mailing list ?
> 
> you weren't thinking that all cvs commit would go into peep-develop,
> were you ?
> 
> 
> That's going to annoy
> |some people pretty fast and make weeding through the list stuff
> |significantly harder.
> 
> 
> yeah, and you want to sort it easily, 
> and you may be not interested, even if you're on peep-develop.
> 
> bye

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] Re: Aha!

[xavier r. <xa...@ha...>]

On Tue, Sep 10, 2002 at 02:38:42PM -0400, Michael Gilfix wrote:
|  Got this a little late so I missed the fun. We should probably
|setup another mailing list for commit emails. 

of course we should set up a different mailing list ?

you weren't thinking that all cvs commit would go into peep-develop,
were you ?


That's going to annoy
|some people pretty fast and make weeding through the list stuff
|significantly harder.


yeah, and you want to sort it easily, 
and you may be not interested, even if you're on peep-develop.

bye




-- 
xavier

Re: [Peep-develop] Re: Aha!

[Michael G. <mg...@ee...>]

  Hey Collin, can you test out the clients running over tcp? I think you broke tcp
by accident at some point. I mentioned this before this last commit and it still
doesn't work. The clients don't seem to send to the server. I'm testing by
running keytest with:

keytest --server=localhost --port=2001 --noautodiscovery
--config=/home/mgilfix/peep/peep-0.5.x/conf/mypeep.conf --protocol=tcp

  I watched with tcpdump and didn't see anything.

                       -- Mike

On Tue, Sep 10 @ 23:24, Collin Starkweather wrote:
> tir, 2002-09-10 kl. 11:16 skrev xavier renaut:
> > cool, i'm waiting for your commit.
> > 
> > -- 
> > xavier
> 
> I just did a fresh commit of logparser.  You may also want to update
> Net::Peep::BC.
> 
> The change fixed the tailing problem and the "Something is not right
> ..." warning we were getting earlier.
> 
> My favorite perlism from the patch (from the tail() method):
> 
>   chomp for @food;
> 
> I tested logparser with:
> 
> logparser --nodaemon --noautodiscovery  --server=localhost --port=2001
> --debug=0 --config=./conf/peep.conf
> --logfile=/home/olsonco/junk/logparser-test.txt  --protocol=udp
> 
> (I had some debugging output in logparser that indicated which lines it
> was catching at debug level 0.)
> 
> I then did a 
> 
> perl -e 'open(F,">>logparser-test.txt"); for ($i=1;$i<1000;$i++) { print
> F "$i:  pam_unix: authentication failure\n" }; close F'
> 
> and all the lines were caught.
> 
> It sounded like a rerun of Hitchcock's "The Birds" ;-)
> 
> Regards,
> 

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] Re: Aha!

[Collin S. <col...@co...>]

tir, 2002-09-10 kl. 11:16 skrev xavier renaut:
> cool, i'm waiting for your commit.
> 
> -- 
> xavier

I just did a fresh commit of logparser.  You may also want to update
Net::Peep::BC.

The change fixed the tailing problem and the "Something is not right
..." warning we were getting earlier.

My favorite perlism from the patch (from the tail() method):

  chomp for @food;

I tested logparser with:

logparser --nodaemon --noautodiscovery  --server=localhost --port=2001
--debug=0 --config=./conf/peep.conf
--logfile=/home/olsonco/junk/logparser-test.txt  --protocol=udp

(I had some debugging output in logparser that indicated which lines it
was catching at debug level 0.)

I then did a 

perl -e 'open(F,">>logparser-test.txt"); for ($i=1;$i<1000;$i++) { print
F "$i:  pam_unix: authentication failure\n" }; close F'

and all the lines were caught.

It sounded like a rerun of Hitchcock's "The Birds" ;-)

Regards,

-Collin

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [Peep-develop] Re: Aha!

[Collin S. <col...@co...>]

tir, 2002-09-10 kl. 12:38 skrev Michael Gilfix:
>   Got this a little late so I missed the fun. We should probably
> setup another mailing list for commit emails. That's going to annoy
> some people pretty fast and make weeding through the list stuff
> significantly harder.

I set up pee...@li... for the commit e-mails (i.e.,
the CVS watches will be directed towards it).  Were you thinking another
list would be in order?  I figure the list will be high volume once in a
while but low volume most of the time.

Once I figure out how to set up a watch on SF, we'll see how the e-mails
look coming from CVS.

(Last time I set up watches on CVS, I remember a file in
$CVSROOT/CVSROOT called notify where you set up information for
watches.  I didn't think we'd have access to CVSROOT but I did a
checkout and voila!  There it is.  Those wily SF hackers!)

>   By the way, that "push @lines, $line while ..." bit is a wicked
> use of perl :) But maybe we should flesh it out so it's a little
> more explicit. After all, this was a problem area before.

Thanks!

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [Peep-develop] Re: Aha!

[Michael G. <mg...@ee...>]

  Got this a little late so I missed the fun. We should probably
setup another mailing list for commit emails. That's going to annoy
some people pretty fast and make weeding through the list stuff
significantly harder.

  By the way, that "push @lines, $line while ..." bit is a wicked
use of perl :) But maybe we should flesh it out so it's a little
more explicit. After all, this was a problem area before.

                      -- Mike

On Tue, Sep 10 @ 11:49, Collin Starkweather wrote:
> tir, 2002-09-10 kl. 11:16 skrev xavier renaut:
> > (do we want a mailing list for the commits which would send a mail for every commit
> > with the files modified and the logs ?)
> 
> I just set up pee...@li....  I'll set up some watches
> to direct CVS commits and whatnot to the list when I get CVS watches all
> figured out.

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] Re: Aha!

[Collin S. <col...@co...>]

tir, 2002-09-10 kl. 11:16 skrev xavier renaut:
> (do we want a mailing list for the commits which would send a mail for every commit
> with the files modified and the logs ?)

I just set up pee...@li....  I'll set up some watches
to direct CVS commits and whatnot to the list when I get CVS watches all
figured out.

-me

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[Peep-develop] Re: Aha!

[xavier r. <xa...@ha...>]

On Tue, Sep 10, 2002 at 10:47:08AM -0600, Collin Starkweather wrote:
|tir, 2002-09-10 kl. 09:26 skrev xavier renaut:
|> as you can see, it's missing a lot of them. 
|> 
|> any idea on how to fix that problem ?  
|
|I made another pass and I think I located the problem:
|
|In logparser:
|
|124  &tail($client,$conf,$filetail->read) unless $filetail->predict;
|
|The read only returns one line.  So I think I need to do something like:
|
|  if (not $filetail->predict) {
|
|    my ($line,@lines);
|    push @lines, $line while $line = $filetail->read;
|    &tail($client,$conf,@lines);
|
|  }
|
|then adjust the tail method to take an array of lines.
|
|I'll test it using your example this evening and make sure that's in
|fact the problem.


cool, i'm waiting for your commit.

(do we want a mailing list for the commits which would send a mail for every commit
with the files modified and the logs ?)

bye


-- 
xavier

[Peep-develop] Aha!

[Collin S. <col...@co...>]

tir, 2002-09-10 kl. 09:26 skrev xavier renaut:
> as you can see, it's missing a lot of them. 
> 
> any idea on how to fix that problem ?  

I made another pass and I think I located the problem:

In logparser:

124  &tail($client,$conf,$filetail->read) unless $filetail->predict;

The read only returns one line.  So I think I need to do something like:

  if (not $filetail->predict) {

    my ($line,@lines);
    push @lines, $line while $line = $filetail->read;
    &tail($client,$conf,@lines);

  }

then adjust the tail method to take an array of lines.

I'll test it using your example this evening and make sure that's in
fact the problem.

Regards,

-Collin

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [Peep-develop] missing lines in logparser

[Collin S. <col...@co...>]

tir, 2002-09-10 kl. 09:26 skrev xavier renaut:
> On Thu, Sep 05, 2002 at 10:41:35AM -0600, Collin Starkweather wrote:
> |tor, 2002-09-05 kl. 08:40 skrev xavier renaut:
> i've been playing with it a bit : 
> 
> 
> i=1;while (( ( i += 1 ) <= 100000 )); do echo $i; sleep 0.1;done > /tmp/test
> 
> so the numbers here are tenth of seconds.
> 
> [Tue Sep 10 09:28:40 2002] Checking [1359] ...
> [Tue Sep 10 09:28:40 2002] Checking group [sudo] against group and excluded group lists ...
> ...
> [Tue Sep 10 09:28:40 2002] Checking [1383] ...
> [Tue Sep 10 09:28:40 2002] Checking group [sudo] against group and excluded group lists ...
> ...
> 
> as you can see, it's missing a lot of them. 
> 
> any idea on how to fix that problem ?  
> 
> are you able to reproduce it ?
> (i was on the 2 machines i tried)

Nicely done.  I was able to reproduce the result, but I didn't think of
doing it the way you did which naturally makes things *much* easier to
figure out.

I'm going to take another look at it tonight and see if I can get a
handle on it.  A preliminary look didn't reveal anything obvious.

-me

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [Peep-develop] missing lines in logparser

[xavier r. <xa...@ha...>]

On Thu, Sep 05, 2002 at 10:41:35AM -0600, Collin Starkweather wrote:
|tor, 2002-09-05 kl. 08:40 skrev xavier renaut:
|> 
|> hello,
|> 
|> is it possible that logparser does not print everyline it parses ?
|> 
|> [Thu Sep  5 10:37:51 2002] Checking [@400000003d776c472a513aa4 tcpserver: status: 2/60] ...
|> [Thu Sep  5 10:37:51 2002] Checking group [sudo] against group and excluded group lists ...
|> [Thu Sep  5 10:37:51 2002] Checking group [incoming] against group and excluded group lists ...
|> [Thu Sep  5 10:37:51 2002]      Trying to match regex [sudo] for event [sudo]
|> [Thu Sep  5 10:37:51 2002]      Trying to match regex [ok] for event [incoming]
|> [Thu Sep  5 10:37:59 2002] Checking [@400000003d776c4f1fb56c8c tcpserver: status: 2/60] ...
|> [Thu Sep  5 10:37:59 2002] Checking group [sudo] against group and excluded group lists ...
|> [Thu Sep  5 10:37:59 2002] Checking group [incoming] against group and excluded group lists ...
|> [Thu Sep  5 10:37:59 2002]      Trying to match regex [sudo] for event [sudo]
|> [Thu Sep  5 10:37:59 2002]      Trying to match regex [ok] for event [incoming]
|> 
|> 
|> in my log file, there is few lines between thoses 2. (that is, around 4 lines)
|> is that fine ?
|> 
|> bye
|
|Curious.  It shouldn't be dropping any lines, though the debugging
|statements should reflect all the lines.  
|
|I use the File::Tail module to get the tails.  I'll have to do some
|stress testing to what's going on.
|


i've been playing with it a bit : 


i=1;while (( ( i += 1 ) <= 100000 )); do echo $i; sleep 0.1;done > /tmp/test

so the numbers here are tenth of seconds.

[Tue Sep 10 09:28:40 2002] Checking [1359] ...
[Tue Sep 10 09:28:40 2002] Checking group [sudo] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] Checking group [procmail] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] Checking group [inrx] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] Checking group [incoming] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] [incoming] will be ignored:  Host [ibis] is not in the host pool [inet-cou2].
[Tue Sep 10 09:28:40 2002] Checking group [pop] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] [pop] will be ignored:  Host [ibis] is not in the host pool [inet-cou2].
[Tue Sep 10 09:28:40 2002] Checking group [nrpe] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] [nrpe] will be ignored:  Host [ibis] is not in the host pool [inet-cou2].
[Tue Sep 10 09:28:40 2002]      Trying to match regex [sudo] for event [sudo]
[Tue Sep 10 09:28:40 2002]      Trying to match regex [D=procmail] for event [procmail]
[Tue Sep 10 09:28:40 2002]      Trying to match regex [LASTFOLDER=IN.rx] for event [inrx]
[Tue Sep 10 09:28:40 2002] Checking [1383] ...
[Tue Sep 10 09:28:40 2002] Checking group [sudo] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] Checking group [procmail] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] Checking group [inrx] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] Checking group [incoming] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] [incoming] will be ignored:  Host [ibis] is not in the host pool [inet-cou2].
[Tue Sep 10 09:28:40 2002] Checking group [pop] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] [pop] will be ignored:  Host [ibis] is not in the host pool [inet-cou2].
[Tue Sep 10 09:28:40 2002] Checking group [nrpe] against group and excluded group lists ...
[Tue Sep 10 09:28:40 2002] [nrpe] will be ignored:  Host [ibis] is not in the host pool [inet-cou2].
[Tue Sep 10 09:28:40 2002]      Trying to match regex [sudo] for event [sudo]
[Tue Sep 10 09:28:40 2002]      Trying to match regex [D=procmail] for event [procmail]
[Tue Sep 10 09:28:40 2002]      Trying to match regex [LASTFOLDER=IN.rx] for event [inrx]
[Tue Sep 10 09:28:40 2002] Checking [1391] ...



as you can see, it's missing a lot of them. 

any idea on how to fix that problem ?  

are you able to reproduce it ?
(i was on the 2 machines i tried)


thanks.

ps : i think file tail is fine : 

#  i=0;while (( ( i += 1 ) <= 100000 )); do echo $i; sleep 0.01;done > /tmp/test


/tmp/test.pl :

 use File::Tail;

$i=1;
         $file=File::Tail->new("/tmp/test");
         while (defined($line=$file->read)) {
             print $i++." $line";
         }



gives the expected result.

-- 
xavier


(needless to say, that's hard to do something with peep with that problem)

Re: [Peep-develop] recording / retransmitting peep

[Michael G. <mg...@ee...>]

  I just realized this email came out wrong :) The first two should
be combined into one or it sounds rude. I mean't "that won't help him
stream though, just listen". Just in case you thought I was having a
bad day or something.

On Mon, Sep 09 @ 14:39, Michael Gilfix wrote:
>   That won't help him stream though. Just listen. However,
> I think some streaming software should support raw sound
> format. It's kinda the lowest common denominator...
> 
> On Mon, Sep 09 @ 12:40, Collin Starkweather wrote:
> > You may have to check the docs to make sure I'm right (it's been a while
> > since I've used it) but with a raw sound file I think you can listen to
> > it with something like
> > 
> >   bplay -S -s 44100 -b 16 output-stream.snd
> > 
> > Check out
> > 
> >   http://www.amberdata.demon.co.uk/bplay/
> > 
> > for more info.
> 
> -- 
> Michael Gilfix
> mg...@ee...
> 
> For my gpg public key:
> http://www.eecs.tufts.edu/~mgilfix/contact.html
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Peep-develop mailing list
> Pee...@li...
> https://lists.sourceforge.net/lists/listinfo/peep-develop

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] recording / retransmitting peep

[Michael G. <mg...@ee...>]

  That won't help him stream though. Just listen. However,
I think some streaming software should support raw sound
format. It's kinda the lowest common denominator...

On Mon, Sep 09 @ 12:40, Collin Starkweather wrote:
> You may have to check the docs to make sure I'm right (it's been a while
> since I've used it) but with a raw sound file I think you can listen to
> it with something like
> 
>   bplay -S -s 44100 -b 16 output-stream.snd
> 
> Check out
> 
>   http://www.amberdata.demon.co.uk/bplay/
> 
> for more info.

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] recording / retransmitting peep

[Collin S. <col...@co...>]

man, 2002-09-09 kl. 12:13 skrev Michael Gilfix:
>   Hey Xavier. This is on my TODO list as one of the things I'd like
> to do. I think I can give you a quick hack though that should get you
> outputing to a file that you can use for the transmission.
> 
>   Here are the steps:
> 
> * Edit oss.c (0.5.x)
> * Put a #if 0/ #endif around the contents of soundSetFormat()
> * Recompile peepd
> * Run with: ./peepd --snd-device=output-stream.snd
> 
>   That will output raw output to the "output-stream.snd" file.  You
> can then an adapter application do a tail on the file and convert
> the stream as it comes in for broadcasting. The format is 44100 Khz,
> stereo, signed 16-bit raw (.snd) format.
> 
>   One reason this didn't exist in peep before though is that you
> can use the existing infrastructure to accomplish the same thing.
> Just run a peep server at each location where you want to listen
> and configure the clients to talk to each of those servers. That
> configuration works beautifully as auto-discovery will take care
> of it all and little configuration effort is necessary.
> 
>   This, however, is not an option if your clients are windows as peepd
> doesn't run on windows. The hack should get you up and running. If
> you're interested in integrating this functionality, I'd love a patch
> :) Lemme know how it works out.
> 
>                           -- Mike

You may have to check the docs to make sure I'm right (it's been a while
since I've used it) but with a raw sound file I think you can listen to
it with something like

  bplay -S -s 44100 -b 16 output-stream.snd

Check out

  http://www.amberdata.demon.co.uk/bplay/

for more info.

-me

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [Peep-develop] recording / retransmitting peep

[Michael G. <mg...@ee...>]

  Hey Xavier. This is on my TODO list as one of the things I'd like
to do. I think I can give you a quick hack though that should get you
outputing to a file that you can use for the transmission.

  Here are the steps:

* Edit oss.c (0.5.x)
* Put a #if 0/ #endif around the contents of soundSetFormat()
* Recompile peepd
* Run with: ./peepd --snd-device=output-stream.snd

  That will output raw output to the "output-stream.snd" file.  You
can then an adapter application do a tail on the file and convert
the stream as it comes in for broadcasting. The format is 44100 Khz,
stereo, signed 16-bit raw (.snd) format.

  One reason this didn't exist in peep before though is that you
can use the existing infrastructure to accomplish the same thing.
Just run a peep server at each location where you want to listen
and configure the clients to talk to each of those servers. That
configuration works beautifully as auto-discovery will take care
of it all and little configuration effort is necessary.

  This, however, is not an option if your clients are windows as peepd
doesn't run on windows. The hack should get you up and running. If
you're interested in integrating this functionality, I'd love a patch
:) Lemme know how it works out.

                          -- Mike

On Mon, Sep 09 @ 12:22, xavier renaut wrote:
> Hello, 
> 
> i'd like to record (wav) / retransmit peep (ogg broadcasting)
> 
> is there already something existing before i search 
> for my self ?
> 
> (i once looked at dspproxy, a proxy for the sound, which
> was i think able to record everything going to the
> sound card.)
> 
> thanks

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] recording / retransmitting peep

[Collin S. <col...@co...>]

man, 2002-09-09 kl. 10:22 skrev xavier renaut:
> Hello, 
> 
> i'd like to record (wav) / retransmit peep (ogg broadcasting)
> 
> is there already something existing before i search 
> for my self ?
> 
> (i once looked at dspproxy, a proxy for the sound, which
> was i think able to record everything going to the
> sound card.)
> 
> thanks
> 
> -- 
> xavier

There is something similar in concept but entirely different in
implementation (?!?):  You can have the server record its play for later
playback.

  ./peepd --help | grep "record\|playback"

I've used it successfully to record the sound of Code Red and Nimbda. 
Michael could prob'ly give you more detail.

I don't think there's anything that is part of Peep that will record
your audio output, though that seems like just the kind of thing that
someone out there undoubtedly would have tackled.  After all, you can't
copy-protect your audio device ;-)

-me

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[Peep-develop] recording / retransmitting peep

[xavier r. <xa...@ha...>]

Hello, 

i'd like to record (wav) / retransmit peep (ogg broadcasting)

is there already something existing before i search 
for my self ?

(i once looked at dspproxy, a proxy for the sound, which
was i think able to record everything going to the
sound card.)

thanks

-- 
xavier

[Peep-develop] Document Searcher

[Michael G. <mg...@ee...>]

  The document searching form is now available on the new homepage
under the documents section. Give it a try. The end result was
that I had to use a ZCatalog and then some dtml to generate the
query results. Thanks goes to Bastian for pointing me in that
direction. Next stop will be getting the user contrib part of the site
up and running...

                    -- Mike

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] A Peep audio-visualization teaser

[Michael G. <mg...@ee...>]

  Sounds good. It's still a good-looking GUI. We just have to figure
out how to get useful patterns in there. That part will be more of
a test/try thing though. If you need any GTK pointers, lemme know.
I'm familiar with the pyGTK bindings and the GTK+ C docs as well,
so I know how the whole thing works.

                         -- Mike

On Sun, Sep 08 @ 19:30, Collin Starkweather wrote:
> s=F8n, 2002-09-08 kl. 17:17 skrev Michael Gilfix:
> >   Hey Collin, this look pretty cool. Have you been able to discern
> > patterns when watching the GUI over time? How was it tying in to the
> > server XML hooks? Lemme know when you can check it into CVS (perhaps
> > cvsroot/peep/visualization is a good place) so we can give it a test
> > drive.
> >=20
> >                      -- Mike
>=20
> I haven't hooked it into eXtace yet.  The screenshot has some cheats in
> it, one of them being that I took a screenshot of eXtace and just put
> the PNG in a canvas for the time being as a placeholder.
>=20
> The other is that the event hooks aren't in place, though I'm getting
> close to being able to put them in.  I simply generated the screenshot
> from a set of canned (i.e., hardcoded) notice structs.
>=20
> Since step 1 of the process was "Figure out Gtk+", step 2 was "Figure
> out that you'll learn faster if you use Glade" and step 3 was "Figure
> out Glade," going has been slow though I think I'm just cresting the
> learning curve.
>=20
> I think it'll be a few weeks before I get the event hooks in place and
> working properly.  I'm still getting things like sorting of GtkCList
> widget columns, er, sorted out.  I'll commit it to the CVS tree after I
> get the hooks worked out (or just before) so that I'm really close to
> having something that will actually do something before turning it loose
> on the general public. =20
>=20
> I'm saving integration of eXtace till last because I'd like to show the
> eXtace author (who earlier seemed to be open to the idea) a more-or-less
> complete framework simply awaiting a nestable component.  I figure it'll
> lower his costs of lending me a hand where necessary.
>=20
> Regards,
>=20
> -me
>=20
> --=20
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Collin Starkweather              http://www.collinstarkweather.com
> Consulting Software Engineer      col...@co...
> Ph.D. Candidate     University of Colorado Department of Economics
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>=20

--=20
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] A Peep audio-visualization teaser

[Collin S. <col...@co...>]

s=F8n, 2002-09-08 kl. 17:17 skrev Michael Gilfix:
>   Hey Collin, this look pretty cool. Have you been able to discern
> patterns when watching the GUI over time? How was it tying in to the
> server XML hooks? Lemme know when you can check it into CVS (perhaps
> cvsroot/peep/visualization is a good place) so we can give it a test
> drive.
>=20
>                      -- Mike

I haven't hooked it into eXtace yet.  The screenshot has some cheats in
it, one of them being that I took a screenshot of eXtace and just put
the PNG in a canvas for the time being as a placeholder.

The other is that the event hooks aren't in place, though I'm getting
close to being able to put them in.  I simply generated the screenshot
from a set of canned (i.e., hardcoded) notice structs.

Since step 1 of the process was "Figure out Gtk+", step 2 was "Figure
out that you'll learn faster if you use Glade" and step 3 was "Figure
out Glade," going has been slow though I think I'm just cresting the
learning curve.

I think it'll be a few weeks before I get the event hooks in place and
working properly.  I'm still getting things like sorting of GtkCList
widget columns, er, sorted out.  I'll commit it to the CVS tree after I
get the hooks worked out (or just before) so that I'm really close to
having something that will actually do something before turning it loose
on the general public. =20

I'm saving integration of eXtace till last because I'd like to show the
eXtace author (who earlier seemed to be open to the idea) a more-or-less
complete framework simply awaiting a nestable component.  I figure it'll
lower his costs of lending me a hand where necessary.

Regards,

-me

--=20
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: [Peep-develop] A Peep audio-visualization teaser

[Michael G. <mg...@ee...>]

  Hey Collin, this look pretty cool. Have you been able to discern
patterns when watching the GUI over time? How was it tying in to the
server XML hooks? Lemme know when you can check it into CVS (perhaps
cvsroot/peep/visualization is a good place) so we can give it a test
drive.

                     -- Mike

On Sun, Sep 08 @ 15:22, Collin Starkweather wrote:
> s=F8n, 2002-09-08 kl. 04:12 skrev Michael Gilfix:
> > To get yourself up and running quickly, just have the server dump
> > event output to a log (flat file format is probably easiest, although
> > XML is always a good candidate for this sort of wet-work).
>=20
> I'll send along a more detailed reply to Michael's thread in a bit, but
> for now just a teaser.
>=20
> I posted a screenshot of the prototype of a visualization component for
> the server I've been playing with.  The idea is to combine a
> visualization framework like eXtace (http://extace.sourceforge.net)=20
> with easy browsing of event and state notices and whatnot in a UI.
>=20
> There's still *alot* of work to do and some uncertainty to be resolved
> before it's ready for prime time (e.g., the hooks are not implemented
> yet and I'm going to have to upgrade everything to Gtk+ 2.0 before too
> long) but I thought I'd throw a screenshot up there for your viewing
> pleasure:
>=20
>   http://www.auralizer.com
>=20
> -me
>=20
> P.S.  I know there must be a drop-dead simple way to do this, but I
> can't seem to find it:  How do you convert a tm struct (used to populate
>  notice->date) into a char array?

--=20
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] A Peep audio-visualization teaser

[Michael G. <mg...@ee...>]

  ... Or asctime() does the trick if you want to get quick n' dirty.
But strftime will give you greater control and will be more portable.

               -- Mike

On Sun, Sep 08 @ 17:13, Andy Davidoff wrote:
> ITYM strftime(3) HTH.
> 
> 
> #if Collin Starkweather /* Sep 08, 15:22 */
> > P.S.  I know there must be a drop-dead simple way to do this, but
> > I can't seem to find it:  How do you convert a tm struct (used to
> > populate notice->date) into a char array?
> #endif /* col...@co... */
> 
> -- 
> Andy Davidoff
> Sen. Unix SysAdmin
> Tufts University

-- 
Michael Gilfix
mg...@ee...

For my gpg public key:
http://www.eecs.tufts.edu/~mgilfix/contact.html

Re: [Peep-develop] A Peep audio-visualization teaser

[Andy D. <de...@po...>]

ITYM strftime(3) HTH.


#if Collin Starkweather /* Sep 08, 15:22 */
> P.S.  I know there must be a drop-dead simple way to do this, but
> I can't seem to find it:  How do you convert a tm struct (used to
> populate notice->date) into a char array?
#endif /* col...@co... */

-- 
Andy Davidoff
Sen. Unix SysAdmin
Tufts University

[Peep-develop] A Peep audio-visualization teaser

[Collin S. <col...@co...>]

s=F8n, 2002-09-08 kl. 04:12 skrev Michael Gilfix:
> To get yourself up and running quickly, just have the server dump
> event output to a log (flat file format is probably easiest, although
> XML is always a good candidate for this sort of wet-work).

I'll send along a more detailed reply to Michael's thread in a bit, but
for now just a teaser.

I posted a screenshot of the prototype of a visualization component for
the server I've been playing with.  The idea is to combine a
visualization framework like eXtace (http://extace.sourceforge.net)=20
with easy browsing of event and state notices and whatnot in a UI.

There's still *alot* of work to do and some uncertainty to be resolved
before it's ready for prime time (e.g., the hooks are not implemented
yet and I'm going to have to upgrade everything to Gtk+ 2.0 before too
long) but I thought I'd throw a screenshot up there for your viewing
pleasure:

  http://www.auralizer.com

-me

P.S.  I know there must be a drop-dead simple way to do this, but I
can't seem to find it:  How do you convert a tm struct (used to populate
 notice->date) into a char array?

--=20
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Collin Starkweather              http://www.collinstarkweather.com
Consulting Software Engineer      col...@co...
Ph.D. Candidate     University of Colorado Department of Economics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~