Re: Debian and GNOME, partnership with Helixcode?

[Jules Bean]

On Tue, Mar 21, 2000 at 08:31:01AM -0500, Miguel de Icaza wrote:
>  
> > The Debian packages, however, are *not* found in the GNOME ftp
> > site. Why aren't they included and mirrored?
> 
> Because nobody did contribute them.  Every binary on the site was
> contributed by someone.  For example, recently we got mail from a
> TurboLinux user, and he provided us with the packages for TurboLinux. 
> 

We have done, in the past.  I started a movement which ended with
gnome 1.0 debs on gnome.org, if I remember correctly. Miguel - if, at
release time (e.g. of 1.2 or 2.0) you want debian packages, the place
to bug is [EMAIL PROTECTED]

Jules

-- 
Jules Bean  |Any sufficiently advanced 
[EMAIL PROTECTED],jellybean.co.uk}  |  technology is indistinguishable
[EMAIL PROTECTED]  |   from a perl script

Re: Debian and GNOME, partnership with Helixcode?

[Martin Bialasinski]

* "Miguel" == Miguel de Icaza <[EMAIL PROTECTED]> wrote:

>> The Debian packages, however, are *not* found in the GNOME ftp
>> site. Why aren't they included and mirrored?

Miguel> Because nobody did contribute them.  Every binary on the site
Miguel> was contributed by someone. 

They could be mirrored from ftp.debian.org, but to me, it doesn't make 
much sense.

If you (preferably) make the dir on ftp.gnome.org aptable, the user
has to put a line for a regular ftp.debian.org mirror into his apt
sources.list nevertheless, so that when he installs a GNOME package,
the needed libs get pulled from the Debian mirror automatically.

But then, he could also fetch the GNOME packages from the Debian
mirror right away.

So a note that GNOME packages are available from the regular Debian
mirrors would be sufficient, no?

Ciao,
Martin

PS: Reply-to is set to debian-gtk-gnome

Re: Debian and GNOME, partnership with Helixcode?

[Miguel de Icaza]

> Hello Miguel, I would have prefered to know you taking some "tapas"
> in Madrid, oh welll..

A large group went on my last day in Madrid for tapas ;-)

> I disagree. The fragment is not "completely wrong",
> however, I read it now and find a *big* mistake.
> Where I said GNOME developers
> I should have said "HELIX developers"

Exactly, hence the paragraph is wrong it in entirety :-)

Now, even then, it was our plan all along (and I believe the FAQ on
our web site mentions it), that we wanted to provide Debian packages.

> I'm sorry to disagree with you, the GNOME project *does*
> distribute binaries (and packages too), looking
> in http://www.gnome.org/start/ will give you pointers
> to packages for Caldera, RedHat and SuSE distributed
> from the GNOME site (ftp://ftp.gnome.org/pub/GNOME/stable/latest).

Sadly, those packages are seldome updated, and there is not an ongoing
effort to keep them up to date.  I tried to assemble such a team, in
the gnome-packaging-list, but that group never produced binaries.

>From time to time we do get contributed packages, but it is not
something users can really rely on having.

> The Debian packages, however, are *not* found in the GNOME ftp
> site. Why aren't they included and mirrored?

Because nobody did contribute them.  Every binary on the site was
contributed by someone.  For example, recently we got mail from a
TurboLinux user, and he provided us with the packages for TurboLinux. 

> I brought this thing first in Debian lists since I thought some
> discussion should be needed first, not just by discussing the
> issue of GNOME/HELIX but also whether if our package system
> was not docummented enough for others to make packages for Debian.

I am sorry, because many people got the wrong impression about GNOME
and Helix due to the non-factual post.

> My point was that maybe Debian is not getting enough attention
> from GNOME and other related projects (like HELIX) and that maybe
> we had a public-relations problem. This thread might not have helped too
> much :)

It did not hurt, as I know this is not how the Debian developer feel.
I regret that you posted inacurrate information, when you had a chance
to talk to me directly instead of coming up with some conspiracy
theories. 

> you are quite popular .. I only attended the exposition to hear you speak :)

I had a lot of free time :-)

Re: noninteractive upgrades of packages

[Robert Thomson]

On Tue, Mar 21, 2000 at 05:02:31PM -0500, Nick Cabatoff wrote:
> I'm interested in being able to upgrade packages noninteractively,
> assuming that the postinst script itself is noninteractive.  Search as
> I might however, I can't find any easy way around dpkg's handling of
> conffiles.  I'm quite happy to blindly accept the one provided in the
> newer package, but cannot guarantee that the version on disk will be
> unmodified.
> 
> Any suggestions?

In /etc/apt/apt.conf
DPkg 
{
Options {"--force-confdef";}
}

This will automatically choose the default action.. if the conffile has been
modified, the default is 'N'. If it hasn't, the default is 'Y' (99% sure)

Regards,
rmt.

-- 
zninski, 30, had been drinking with friends when it was suggested they strip
naked and play some "men's games." Initially they hit each other over the head
with frozen icicles, but then one man seized a chainsaw and cut off the end of
his foot. Not to be outdone, Azninski grabbed the saw and crying "Watch this
then!" -- then swung at his own head and chopped it off.

noninteractive upgrades of packages

[Nick Cabatoff]

I'm interested in being able to upgrade packages noninteractively,
assuming that the postinst script itself is noninteractive.  Search as
I might however, I can't find any easy way around dpkg's handling of
conffiles.  I'm quite happy to blindly accept the one provided in the
newer package, but cannot guarantee that the version on disk will be
unmodified.

Any suggestions?

Bash, Keys, Potato

[Rodrigo Castro]

Hello,

I am running the lastest potato version and since sometime ago
my key E (yeah, E upcase) stopped working. But only in bash. If I run
csh or ksh, it works. What happens is the following, I type and I dont
get the key displayed. It works as a dead key, waiting for any other
key. When I do type other key, it beeps and I get no output on
screen. I reinstalled bash, libncurses5, libc6 and already trying
changing my keymap, but I wasn't sucessful. I am getting crazy. 

PS: Please, send a copy of the answer to me because I am not a
subscriber of debian lists.

Thank you for any help,
-- 
Rodrigo Castro <[EMAIL PROTECTED]> 
Computer Science undergraduate student - University of Sao Paulo

I do not fear computers.  I fear the lack of them.
-- Isaac Asimov

Re: Bug#60753: mutt: /etc/Muttrc should not use colors

[Filip Van Raemdonck]

On Mon, Mar 20, 2000 at 02:38:12PM -0600, Steve Greenland wrote:
> On 20-Mar-00, 01:46 (CST), Nicolás Lichtmaier <[EMAIL PROTECTED]> wrote: 
> >  Using the Space and the Backspace keys for up and down movement is absurd,
> > it's even stupid. Backspace is back-space. Those keybindings where thought
> > for keyboards without arrows, and those keyboards no longer exists...
> 
> While I agree with most of what you wrote, you're wrong on this one.
> There's a *lot* of history of using the space bar to "do the next thing"
> in Unix console programs (more/less, most news readers and existing mail
> readers). And there's no reason *not* to use them -- what else would you
> bind to the space key? You're right: the defaults should cater to the
> new user, but there's no reason to deliberatly aggravate the experienced
> user.
> 
In fact, this has at least worked for a long time (and probably still works -
but I don't do any helpdesk anymore so I don't have to use those programs) with
a few internet related windows programs as well, as there are: netscape (not
strictly windows, I know), for both navigator and messenger, interest exploder^H
^H^Hwhatever, outlook express. And who are we to correct Micro$oft? :-)

Regards,

Filip

Re: Novell: NDS eDirectory for Linux

[Erik]

On Tue, Mar 21, 2000 at 09:59:49AM +0100, Andreas Tille wrote:
> On Tue, 21 Mar 2000, Lauri Tischler wrote:
> 
> > Load NLDAP.NLM on one of your Netware servers.
> > I did just that and wrote some WEB-applications with php3 to get phone and
> > mailinggrop information from NDS.  
> Please excuse my ignorance about LDAP.  I really don't know how it works
> and how can I profit from it.
> 
> My intention is to enable normal Novell users access to linux boxes
> without triggering my /etc/passwd with the Novell user accounts.  I
> hope that there is any simple possibility for Netware users to get
> a login on some (one to three) Linux boxes with their Novell password
> and home directory on the NCPFS mounted Novell volume.  This would
> save my time to install NIS and adding and removing users on my
> Linux boxes.
> 
> May be I'm dreaming of something which is not possible, yet.
> 
> Kind regards
> 
>Andreas.

Well, i know that you can get usernames/passwords through ldap using a pam
module .. libpam-ldap probably.  So if you can get the users names and
passwords over this ldap interface then it should be completely possible

As for the file mounting, i'm not sure ... i know there is stuff in the
kernel for it, but i've never used it (and like samba it probably needs
some help from userland applications).

Erik Bernhardson
[EMAIL PROTECTED]
--
It is better to remain silent and be considered a fool, than to speak and
remove all doubt.
-- Mark Twain


pgp0Z1Lc3pXJH.pgp
Description: PGP signature

Re: ITP: GOB [Was: Re: GOB package?]

[Jules Bean]

On Tue, Mar 21, 2000 at 04:26:37PM +, Mark Brown wrote:
> On Tue, Mar 21, 2000 at 09:38:50AM -0500, Peter Teichman wrote:
> 
> > A quick search through debian-devel and wnpp yields no ITP for
> 
> Indeed.  I'd been wondering if it were packaged under an unusual name or
> something.
> 
> > gob.. Maybe someone would like to pick it up?
> 
> I'll do it myself if nobody else has a burning desire to, although quite
> probably on the basis that anyone else who wants to take it is welcome 
> to.
> 
> >From the web page (http://www.5z.com/jirka/gob.html):
> 
>   GOB is a preprocessor for making GTK+ objects with inline C code so
>   that generated files are not editted. Syntax is inspired
>   by java and yacc or lex. The implementation is intentionaly kept simple,
>   and no C actual code parsing is done. 
> 
> The license is GPL.

This would be mildy convenient for me, since the next major release of
balsa may require, or simply prefer, GOB, to build.

[*sigh* why not just blasted use C++]

Jules

-- 
Jules Bean  |Any sufficiently advanced 
[EMAIL PROTECTED],jellybean.co.uk}  |  technology is indistinguishable
[EMAIL PROTECTED]  |   from a perl script

Re: SuSe proxy suite

[David Coe]

Michael Meskes <[EMAIL PROTECTED]> writes:

> What happened to the ITP for it? I didn't see a package yet.

I started working on it, but didn't get very far (due to lack
of time, not any particular problem).  If you'd like to take
it over I'll pass what I've done to you; otherwise I will get
back to it when I can.

Re: cannot login in xdm anymore (upgrade potato -> potato)

[Carlo Segre]

On Fri, Mar 10, 2000 at 09:03:25PM +0100, Richard P. Groenewegen wrote:
> [2] Logging in is still impossible: my password is accepted but
> apparently I cannot connect to the X-server (here is my
> .xsession-errors:)
> 
>   Xlib: connection to ":0.0" refused by server
>   Xlib: Client is not authorized to connect to Server
>   xrdb: Can't open display ':0'
>   Xlib: connection to ":0.0" refused by server
>   ...
> After less than a second, the login screen is there again.
> 
> This is both a cry for help as a (not so precise) bug-report.  I'm
> happy to give whatever information you might need about (configuration
> files on my) computer.

I have seen a similar symptom on 1 out of three computers that I have
running
potato.  In my case, I am not running xdm but gdm and I am able to log
in and 
run any X applications from the GUI but if I try to run one from a
command line 
in a a terminal, I get the same message.

This happens only when I am using a Gnome session.  With a Debian
session it works 
fine.  Furthermore, as root there are no problems.  Since I am seeing
this on one
of three machines which are presumabley set up in the same way, I am
presuming that 
it is some incorrect configuration file somewhere.  If anyone has an
insight, I would appreciate it.  My next move is to purge all X packages
and reinstall...

Carlo Segre
Assoc. Prof. of Physics
Illinois Institute of Technology

ITP: GOB [Was: Re: GOB package?]

[Mark Brown]

On Tue, Mar 21, 2000 at 09:38:50AM -0500, Peter Teichman wrote:

> A quick search through debian-devel and wnpp yields no ITP for

Indeed.  I'd been wondering if it were packaged under an unusual name or
something.

> gob.. Maybe someone would like to pick it up?

I'll do it myself if nobody else has a burning desire to, although quite
probably on the basis that anyone else who wants to take it is welcome 
to.

>From the web page (http://www.5z.com/jirka/gob.html):

  GOB is a preprocessor for making GTK+ objects with inline C code so
  that generated files are not editted. Syntax is inspired
  by java and yacc or lex. The implementation is intentionaly kept simple,
  and no C actual code parsing is done. 

The license is GPL.

-- 
Mark Brown  mailto:[EMAIL PROTECTED]   (Trying to avoid grumpiness)
http://www.tardis.ed.ac.uk/~broonie/
EUFShttp://www.eusa.ed.ac.uk/societies/filmsoc/

Re: Bug#60399: crashes on installation

[Jacob Kuntz]

Daniel Martin ([EMAIL PROTECTED]) wrote:
> I seem to remember once upon a time that man would crash and burn if
> one somehow had a corrupt index.bt file.  Could it be that a partially
> completed install produces such a corrupt file (i.e. could it be that
> whatever mandb does in the background is rebuilding this, and being
> only half-way done means the resulting file seems corrupt), and that
> an existing but corrupt index file causes havoc?
> 

this would be unlikely. on all three of the machines i experienced this bug
on, man-db had never been installed before. completely virgin systems. i
tend to lean toward the idea that this is a kernel bug. i seem to remember
something on kernel traffic (kt.linuxcare.com) about a very rare bug that
rears it's ugly head with dpkg. can't find it now. then again, i guess
that's about as likely as solar flares...

anyone have any other guesses?

-- 
(jacob kuntz)[EMAIL PROTECTED] [EMAIL 
PROTECTED],underworld}.net
(megabite systems)   "think free speech, not free beer."

Re: Non-Debian Debian Packages (maelstrom and libsdl)

[Jules Bean]

On Tue, Mar 21, 2000 at 12:42:52PM +, Christoph Baumann wrote:
> Hi!
> 
> I just came across some strange debs on
> http://www.devolution.com/~slouken/SDL/download-1.0.html  
>   
> and
> http://www.devolution.com/~slouken/Maelstrom/binary.html  
>   
> 
> I downloaded and extraced them. They aren't policy compliant and contain
> evil dependencies. e.g. the maelstrom package has this:
> Depends: sdl1(>=3D1.0)
> sdl1 isn't a official Debian package it should be libsdl1.
> As I'm the maintainer of maelstrom I'm rather concerned. I will now
> contact this "maintainer".

Well, be polite.

There's nothing wrong with sam (or someone he knows) doing unofficial
debs of hist software. If I were you, I'd just make him aware of the
duplication of effort.

Jules



-- 
Jules Bean  |Any sufficiently advanced 
[EMAIL PROTECTED],jellybean.co.uk}  |  technology is indistinguishable
[EMAIL PROTECTED]  |   from a perl script

Re: Null Pointer exceptions in servlets since latest Apache/Jserv update

[Stefan Gybas]

Andreas Tille wrote:

> java.lang.NullPointerException:
> where line 70 is something like
> 
>   rs = stmt.executeQuery("SELECT * FROM Table");

This simply means that stmt is NULL at this point, i.e. the previous call
to Connection.createStatement() failed. We can't help you unless you post
the whole source file.

> I have included the line
> 
>repositories=/usr/share/java/servlets,/usr/share/java/freetds_jdbc.jar
> 
> in /etc/jserv/zones/root.properties.

You should not do this as the classes in the propsitory are loaded using a
special classloader which detects changes files. You should rather put
freetds_jdbc.jar into wrapper.classpath in /etc/jserv/jserv.properties and
make sure that the system classes (classes.zip for JDK 1.1) are also included
in this classpath.

BTW: This is the wrong mailinglist for this question, please use debian-user
next time (it's also off topic on debian-java).

-- 
Stefan Gybas

Re: blue on black is unreadable (was Re: Bug#60753: mutt: /etc/Muttrc should not use colors)

[Lauri Tischler]

Junichi Uekawa wrote:
> 
> garabik> > lynx has the same problem. hyper links are blue on black, which 
> makes it
> garabik> > very difficult to see where you are going. fixed with:
> garabik> >
> garabik> >  COLOR:1:cyan:black
> garabik> >  COLOR:5:brightcyan:black
> 
> The same can be said about the default "ls" colors.
> It shows directory names with blue on black.

These must be set up by some bug-eyed alien with colour-resolution going
well into ultraviolet. :)

--
Lauri Tischler, Network Admin
Tel:+358-9-47846331*   Mouse movement detected  *
Fax:+358-9-47846500* Reboot Windows to activate changes *
Mobile: +358-40-5569010
EMail:  [EMAIL PROTECTED]

Re: blue on black is unreadable (was Re: Bug#60753: mutt: /etc/Muttrc should not use colors)

[Junichi Uekawa]

In Mon, 20 Mar 2000 09:40:53 +0100, de profundis Radovan Garabik <[EMAIL 
PROTECTED]> cum veritas scribat

garabik> > lynx has the same problem. hyper links are blue on black, which 
makes it
garabik> > very difficult to see where you are going. fixed with:
garabik> > 
garabik> >  COLOR:1:cyan:black
garabik> >  COLOR:5:brightcyan:black

The same can be said about the default "ls" colors.
It shows directory names with blue on black.



--
dancer, a.k.a. Junichi Uekawa   http://www.netfort.gr.jp/~dancer
 Dept. of Knowledge Engineering and Computer Science, Doshisha University.
... Long Live Free Software, LIBERTAS OMNI VINCIT.

PDQ?

[John Goerzen]

Hi,

There is apparently a printing system out there that is designed to
replace lpr-based ones, called PDQ.  I notice this is not yet in
Debian.  Is anyone planning to package it?  Does anyone have any
experience with it?  If so, how do you like it?

Thanks,
John

Re: OT: Transmeta

[Steve Gore]

esoR ocsirF wrote:
> On Mon, Mar 20, 2000 at 09:15:02PM -0800, Scott Simmons wrote:
> > 
> > Note: Since this message is of a commercial nature, we cleared it
> > with Wichert Akkerman first and we're also donating $1000 to
> > Debian.
> > 
> Wowzers! Wouldn't it be swell if every commercial email that made it to
> this list was also accompanied by a cool grand. 
> 

Yeah, the donation is very cool.  But what impresses me more is the courtesy
and respect that is demonstrated.  Seems like an ethical company (ie, one
that I wouldn't mind working for :).

Regards,

Steve

Re: Null Pointer exceptions in servlets since latest Apache/Jserv update

[Andreas Tille]

On Tue, 21 Mar 2000, Andreas Tille wrote:

> Hello,
> 
> I've run into big trouble since I updatet do the latest Apache and
> Jserv packages:
>  ...
> For JDBC access to an MS SQL 7.0 server I use freetds-jdbc.  This
> works well in standalone Java applications but fails in servlets:
Furthermore I've found in the logs the following information:

[Tue Mar 21 11:25:22 2000] [error] JServ: ajp12: Servlet Error:
java.lang.NoSuchMethodError: com.internetcds.jdbc.tds.Statement: method
(Ljava/sql/Connection;Lcom/internetcds/jdbc/tds/Tds;)V not found:
com.internetcds.jdbc.tds.Statement: method
(Ljava/sql/Connection;Lcom/internetcds/jdbc/tds/Tds;)V not found
[Tue Mar 21 11:25:22 2000] [error] JServ: an error returned handling request
via protocol "ajpv12"


I'm really in trouble and have no idea why the communication between
Jserv and FreeTDS-JDBC did not work as before :-(((.

Any help is really appreciated!

Kind regards
 
Andreas.

Re: Null Pointer exceptions in servlets since latest Apache/Jserv update

[Andreas Tille]

On Tue, 21 Mar 2000, Lauri Tischler wrote:

> > For JDBC access to an MS SQL 7.0 server I use freetds-jdbc.  This
> > works well in standalone Java applications but fails in servlets:
> 
> Would some knowledgable guru do Debian package on freetds.
> It would make nice addition to database access, php3-freetds comes to mind :)
I did an ITP on this for FreeTDS version 0.51, because there were
config files on strange locations in 0.50.  I suggested patches to
the authors which told me, that they wanted to include it.

If there is a *more knowledgable guru* than me (my knowledge is
nearly zero :-( but I would do it because I need it) I would
greatly appreciate if someone else would take the package!  The
same for my FreeTDS-JDBC package.  By the way I've packaged the
latest FreeTDS-JDBC shnapshot, but there where problems with the
examples testsuite.

Kind regards

 Andreas.

Non-Debian Debian Packages (maelstrom and libsdl)

[Christoph Baumann]

Hi!

I just came across some strange debs on
http://www.devolution.com/~slouken/SDL/download-1.0.html
and
http://www.devolution.com/~slouken/Maelstrom/binary.html

I downloaded and extraced them. They aren't policy compliant and contain
evil dependencies. e.g. the maelstrom package has this:
Depends: sdl1(>=3D1.0)
sdl1 isn't a official Debian package it should be libsdl1.
As I'm the maintainer of maelstrom I'm rather concerned. I will now
contact this "maintainer".

Christoph

-- 
* Christoph Baumann  *
* [EMAIL PROTECTED]  *
* www.rzuser.uni-heidelberg.de/~cbauman1/welcome.html*
* "External Error : INTELLIGENCE not found !"*



pgpH8BGs1iGEM.pgp
Description: PGP signature

Re: Null Pointer exceptions in servlets since latest Apache/Jserv update

[Lauri Tischler]

Andreas Tille wrote:
> 
> For JDBC access to an MS SQL 7.0 server I use freetds-jdbc.  This
> works well in standalone Java applications but fails in servlets:

Would some knowledgable guru do Debian package on freetds.
It would make nice addition to database access, php3-freetds comes to mind :)

--
Lauri Tischler, Network Admin
Tel:+358-9-47846331*   Mouse movement detected  *
Fax:+358-9-47846500* Reboot Windows to activate changes *
Mobile: +358-40-5569010
EMail:  [EMAIL PROTECTED]

Re: Libc6 functions such as wprintf() (widechar/multibyte/UNICODE support)

[Jerry Lundström]

On Tue, 21 Mar 2000 10:26:34 +0900, Tomohiro KUBOTA <[EMAIL PROTECTED]> wrote:

>Hi,
>
>From: Jerry Lundström <[EMAIL PROTECTED]>
>Subject: Libc6 functions such as wprintf() (widechar/multibyte/UNICODE support)
>Date: Mon, 20 Mar 2000 17:29:29 +0100
>
>> Ive been looking around the manual pages of wprintf and
>> include files of the latest potato.
>> Widechar/Multibyte support exist (wchar.h) but almost all
>> the usefull function is not in the libc distribution of
>> potato. Functions like wprintf that a vital is no where
>> to be found and still they have manual pages for the
>> function.
>> Has the widechar/multibyte printing support been removed
>> for som resone or just missed?
>> 
>> Im hopeing for a quick response on this matter because its
>> very VITAL for my project!
>
>Yes, these functions are very vital also for multibyte-language-
>speaking (CJK) people, like me.  However, as far as I know, 
>these functions are not implemented yet.  (I hope glibc 2.2 will have.)
>On the other hand, conversion functions between multibyte characters
>and wide characters are available now.  Can you use wcstombs() and 
>printf() instead of wprintf() ?
>

Well, that was why I asked, because all the convertion functions seams to
exist and that got me clueless on why the wprintf etc functions aint.
Maybe this is a matter to take up with the libc ppl and not you folks, just
wanted to know if it may not been compiled into libc because of stability
problems or whatever.


On Mon, 20 Mar 2000 21:19:09 +, Edmund GRIMLEY EVANS <[EMAIL PROTECTED]> 
wrote:

>Until glibc-2.2 comes out you could use Bruno Haible's libutf8.
>

Okey, thanks, will check that lib.

--
Name:Work:  Work phone:Mobile:
Jerry Lundström  www.DaCapo.se  +46 (0)31 - 710 72 00  +46 (0)739 87 60 53
Occupation:Knowledge:
System Developer   C/C++, COM/DCOM/CORBA, ASP, SQL, PHP, MySQL, Linux/UNIX

Re: Novell: NDS eDirectory for Linux

[Andreas Tille]

On Tue, 21 Mar 2000, Lauri Tischler wrote:

> Load NLDAP.NLM on one of your Netware servers.
> I did just that and wrote some WEB-applications with php3 to get phone and
> mailinggrop information from NDS.  
Please excuse my ignorance about LDAP.  I really don't know how it works
and how can I profit from it.

My intention is to enable normal Novell users access to linux boxes
without triggering my /etc/passwd with the Novell user accounts.  I
hope that there is any simple possibility for Netware users to get
a login on some (one to three) Linux boxes with their Novell password
and home directory on the NCPFS mounted Novell volume.  This would
save my time to install NIS and adding and removing users on my
Linux boxes.

May be I'm dreaming of something which is not possible, yet.

Kind regards

   Andreas.

Re: [transcript] source package formats

[Martin Bialasinski]

* "Manoj" == Manoj Srivastava <[EMAIL PROTECTED]> wrote:

Manoj> One of the feature I like of a monolithic patch, as you call
Manoj> it, is that when upstream incorporates changes previously made
Manoj> by me or others, it is automatically handled on upgrade; the
Manoj> monolithic patch just gets smaller.

OTOH, I find it easier to keep track of the changes I made to a source 
with multiple patches. You keep them in logical units, can review them
and keep track of them easyly. For me, dbs made the mc package far
easier to maintain.

If upstream incorporates a change, I simply remove the of the diff
from the patches dir.

I also use cvs-buildpackage with other packages and it is very good in
its field. Version control of the diffs would be a very cool
thing. Hmm, maybe I could just manually CVS control the debian/ dir.

Ciao,
Martin

OT: Transmeta

[esoR ocsirF]

On Mon, Mar 20, 2000 at 09:15:02PM -0800, Scott Simmons wrote:
> 
> Note: Since this message is of a commercial nature, we cleared it
> with Wichert Akkerman first and we're also donating $1000 to
> Debian.
> 
Wowzers! Wouldn't it be swell if every commercial email that made it to
this list was also accompanied by a cool grand. 

Sorry for this intrusion by an IANAD. I just couldn't resist. :-)

-- 
Frisco Rose "By any other name, I would smell the same"
E.O.U. Student  [EMAIL PROTECTED] (541) 962-2987

Science Journal Ed. [EMAIL PROTECTED]
EOU Hoke Center 307 (541) 962-3787
La Grande, OR. 97850

Re: Debian and GNOME, partnership with Helixcode?

[Javier Fdz-Sanguino Pen~a]

Hello Miguel, I would have prefered to know you taking some "tapas"
in Madrid, oh welll..

On Mon, Mar 20, 2000 at 08:17:56AM -0500, Miguel de Icaza wrote:
> 
> Hello guys,
> 
>   I just got forwarded a few messages from a discussion that is going
> on at the Debian lists, let me reply:
> 
> First of all, this fragment --which started the whole debate-- is
> completely wrong:
> 
> > I assisted today to a conference by Miguel de Icaza here in Madrid,
> > it seems he is running a new business namde Helixcode
> > (http://www.helixcode.com) which will be working for GNOME. In the
> > conference he said that the GNOME developers do not want to make debian
> > packages because "they are too difficult". But he also did not understand
> > why distributions carry out old version of GNOME.
> 
> What I said was that Helix could not make GNOME packages for Debian
> for the "Helix GNOME Preview Release 1", not that the GNOME developers
> did not want to make them.

I disagree. The fragment is not "completely wrong",
however, I read it now and find a *big* mistake.
Where I said GNOME developers
I should have said "HELIX developers"

> 
> I am going to repeat because it seems Javier did not understand even
> when I spoke in spanish:
> 
>   Helix Code could not produce Debian packages for Helix GNOME Preview 1
>   because it was too hard to get them on time.
> 
>   The GNOME project is a completely different entity.  
> 
>   GNOME is the project to bring new technologies to Unix
> 
>   Helix Code is the company that is writing GNOME-based productivity
>   applications for GNOME, and happens to have a binary distribution of
>   GNOME for various GNU/Linux distributions.
> 
>   Helix Code is the one that could not produce the Debian packages on
>   time. 
> 
>   The GNOME project does not produce binaries.  Not for Debian, not
>   for Red Hat, not for anything else.  From time to time we get .spec
>   files and debian/ directories, but we do not maintain them, nor do
>   we keep track of them.

I'm sorry to disagree with you, the GNOME project *does*
distribute binaries (and packages too), looking
in http://www.gnome.org/start/ will give you pointers
to packages for Caldera, RedHat and SuSE distributed
from the GNOME site (ftp://ftp.gnome.org/pub/GNOME/stable/latest).
The Debian packages, however, are
*not* found in the GNOME ftp site. Why aren't they included and mirrored?


> 
> More:
> I have to point out a few things:
> 
>   1. Debian directories exist for all packages that the Debian
>  maintainers have requested to maintain in the GNOME CVS.
> 
>   2. I do not have anything to do with those packages.
> 
>   3. My comment on Debian packages that I made in the Spain conference
>  was that Helix did not release in the "Helix GNOME Preview 1"
>  GNOME packages for Debian because it was too hard for us to make
>  them.  We did not have the resources to attack the problem,
>  although that was one of the original plans.
> 
> If you want to help the Helix hackers to get Helix GNOME packages for
> Debian, then subscribe to the spidermonkey mailing list at Helix Code,
> and offer your help to our developers.
> 
> Our developers (the Helix developers) were working around the clock
> (Jacob even spent one day 56 hours awake) to get Helix GNOME out.  And
> yes, supporting Debian for preview 1 would have just delayed Preview 1
> a few weeks for no good reason.  In the future, there will be time to
> handle Debian.
> 
> I do not know why you guys did not contact me directly if you had any
> doubts. 

I brought this thing first in Debian lists since I thought some
discussion should be needed first, not just by discussing the
issue of GNOME/HELIX but also whether if our package system
was not docummented enough for others to make packages for Debian.
That's why the mail was not sent to HELIX. I also made
the mistake to CC: this to a public list, when it should be
kept private.

It was not my intention to disregard the efforts made by GNOME
and HELIX in order to make good productivity tools available.
However I felt that GNOME/HELIX disregard the effort made by
Debian in the comments people make and the actions they take.

Take for example the Linux Journal's article signed by you et al,
in issue 70 (I read it yesterday in the spanish translation issue 5),
also available at http://www.linuxjournal.com/lj-issues/issue70/3754.html.
Specifically the "popup" Resources
(http://www.linuxjournal.com/lj-issues/issue70/3754s1.html) many
distributions are mentioned but... where is Debian?

My point was that maybe Debian is not getting enough attention
from GNOME and other related projects (like HELIX) and that maybe
we had a public-relations problem. This thread might not have helped too
much :)

It is my opinion that GNOME and Debian should work more closely together,
for this reason was this discussion started, not to drive a flame war or to
pun anybody. It was just circumstances that your conference started t

Null Pointer exceptions in servlets since latest Apache/Jserv update

[Andreas Tille]

Hello,

I've run into big trouble since I updatet do the latest Apache and
Jserv packages:

Package: apache
Status: install ok installed
Priority: optional
Section: web
Installed-Size: 651
Maintainer: Johnie Ingram <[EMAIL PROTECTED]>
Version: 1.3.9-12

Package: jserv
Status: install ok installed
Priority: optional
Section: contrib/web
Installed-Size: 830
Maintainer: Stefan Gybas <[EMAIL PROTECTED]>
Version: 1.1-2

For JDBC access to an MS SQL 7.0 server I use freetds-jdbc.  This
works well in standalone Java applications but fails in servlets:

java.lang.NullPointerException:
at Hallo.service(Hallo.java:70)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:309)
at org.apache.jserv.JServConnection.processRequest(JServConnection.java:314)
at org.apache.jserv.JServConnection.run(JServConnection.java:188)
at java.lang.Thread.run(Thread.java)

where line 70 is something like

  rs = stmt.executeQuery("SELECT * FROM Table");
  
I have included the line

   repositories=/usr/share/java/servlets,/usr/share/java/freetds_jdbc.jar

in /etc/jserv/zones/root.properties.  Do I have to change anything else
to get the thing working?  I really have no clue what happened and what
to do.

Any help is really appreciated.

Kind regards

   Andreas.

Re: Novell: NDS eDirectory for Linux

[Lauri Tischler]

Andreas Tille wrote:
> 
> On Tue, 21 Mar 2000, Lauri Tischler wrote:
> 
> > Load NLDAP.NLM on one of your Netware servers.
> > I did just that and wrote some WEB-applications with php3 to get phone and
> > mailinggrop information from NDS.
> Please excuse my ignorance about LDAP.  I really don't know how it works
> and how can I profit from it.
> 
> My intention is to enable normal Novell users access to linux boxes
> without triggering my /etc/passwd with the Novell user accounts.  I
> hope that there is any simple possibility for Netware users to get
> a login on some (one to three) Linux boxes with their Novell password
> and home directory on the NCPFS mounted Novell volume.  This would
> save my time to install NIS and adding and removing users on my
> Linux boxes.

Netware users can connect to Linux three ways,
- use 'marsnw' netware daemon on Linux, emulates Netware 3.10
- use Calderas Netware on Linux, that costs money, normal Novell licences
- add Micro$oft networking to your Netware users and use Samba to connect
  to Linux.

If they allready are Netware users I fail too see reason to mount Novell
volumes to Linux, unless you want to give Linux-Users access to Netware.
 
--
Lauri Tischler, Network Admin
Tel:+358-9-47846331*   Mouse movement detected  *
Fax:+358-9-47846500* Reboot Windows to activate changes *
Mobile: +358-40-5569010
EMail:  [EMAIL PROTECTED]

Re: Bug#60753: mutt: /etc/Muttrc should not use colors

[Radovan Garabik]

On Tue, Mar 21, 2000 at 04:33:13PM +1100, Brian May wrote:
> For instance I push the up arrow numerous times to get my editor to go
> to the top of the file, and all of a sudden, the escape comes detached
> from the [D, and I suddenly end up altering my file, inserting invalid
> data, perhaps without even realizing it.
> 
> IMHO, the sooner this dead escape standard is dropped in favour of the
> better standard (IIRC: set the high bit on control characters), the

are you sure this would be better? you need to type 8-bit characters too...

-- 
 ---
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__garabik @ melkor.dnp.fmph.uniba.sk |
 ---
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!

Re: [transcript] source package formats

[Adam Heath]

On 21 Mar 2000, Brian May wrote:

> > "Adam" == Adam Heath <[EMAIL PROTECTED]> writes:
> 
> How would you create these diffs?

Please reread the mails.  I said it only extracts.  I've said nothing about
building anything yet.

> Also, have you considered how such a system could be integrated
> with CVS?

No, I haven't.  At this early point, I don't want to think about it.  And,
what I have done is just a test, to see if it is actually
possible.  Everything I have done will NOT appear in any final version.

> Who would this work without CVS? If the maintainer wanted to create
> the diff files manually (is this the only way?), could he/she include
> the diffs somewhere in the source tree, or would he/she have to
> manually create and update the *.diffs.tar.gz file?

There will be an automated tool to do it, it just haven't been
designed/created yet.

> Adam> Actually, it will make things 'less trustworthy,' to quote
> Adam> Ian.  What is to keep a script in debian/ from editting the
> Adam> files that exist in debian/diffs?
> 
> I don't understand the security concern - what is to stop the
> script from patching any number of files in the source tree, during
> the standard build process?

This is no different than the current dpkg-source -x.  Ian's concern(please
reread the transcript) was that to extract the source(and apply the
patches) it(dbs) had to run shell code that was inside each pkg.  This shell
code could be 'tainted' with a trojan horse.

The new way has the code responsible for extracting and patching the source
part of dpkg-source itself.  This means someone only has to do one audit, on
dpkg-source, and once satisfied, knw the series of steps that are taken to
extract a source archive.

BEGIN GEEK CODE BLOCK
Version: 3.12
GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS--
PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z?
-END GEEK CODE BLOCK-
BEGIN PGP INFO
Adam Heath <[EMAIL PROTECTED]>Finger Print | KeyID
67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP
AD46 C888 F587 F8A3 A6DA  3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG
-END PGP INFO-

Intent to package: eieio

[Takao KAWAMURA]

Hi,

I am working on the package eieio, an Emacs lisp program
which implements a controlled object-oriented programming
methodology following the CLOS standard.

Eieio is licensed under the GPL.

You can get more information about eieio at

http://www.ultranet.com/~zappo/eieio.shtml

-- 
Takao KAWAMURA

Re: ITP: solfege

[Jim Lynch]

Hi,

Gary Willis (www.garywillis.com) did a little ear training thing on
his web site; he is a master level bassist (really, master among
masters) and is also very generous with info about how to play. The
ear training drills are geared towards the instrument, and go
primarilary to a "hear note ==> know position of note instantly" kind
of approach. Would this kind of thing fit in with solfege?

Also, I wrote some c++ abstractions that might be useful: Note,
Interval and Scale. The latter is presently implemented as a class
containing a vector. A Chord can be done using a Scale;
a Voicing might be a bit different.

Last, my music teacher is presently writing an ear training book; he
has already wrote two beginning theory books, one on fundamentals
and the other on chords. Other than the ear training requirement, 
the study of these can be used as prereq to functional harmony.
I bring him up because a collaboration may be possible. If there
is interest, I will contact him to see if he is also interested.

-Jim

---
Jim Lynch   Finger for pgp key
as Laney College CIS admin:  [EMAIL PROTECTED]   http://www.laney.edu/~jim/
as Debian developer: [EMAIL PROTECTED]  http://www.debian.org/~jwl/

Re: [transcript] source package formats

[Brian May]

> "Adam" == Adam Heath <[EMAIL PROTECTED]> writes:

How would you create these diffs?

Also, have you considered how such a system could be integrated
with CVS?

I guess cvs could be used to automatically generate the required
diffs, given labels of key versions...

Who would this work without CVS? If the maintainer wanted to create
the diff files manually (is this the only way?), could he/she include
the diffs somewhere in the source tree, or would he/she have to
manually create and update the *.diffs.tar.gz file?

Adam> Actually, it will make things 'less trustworthy,' to quote
Adam> Ian.  What is to keep a script in debian/ from editting the
Adam> files that exist in debian/diffs?

I don't understand the security concern - what is to stop the
script from patching any number of files in the source tree, during
the standard build process?

Adam> And the diffs do live in a standard spot, in
Adam> ../*.diffs.tar.gz.

I meant in uncompressed form.

I have not gone into depth into any of the above issues, but will
leave it at that until I understand the proposal a bit better.
-- 
Brian May <[EMAIL PROTECTED]>

Re: Bug#60753: mutt: /etc/Muttrc should not use colors

[Brian May]

> "Nicolás" == Nicolás Lichtmaier <[EMAIL PROTECTED]> writes:

Nicolás>  Using the Space and the Backspace keys for up and down
Nicolás> movement is absurd, it's even stupid. Backspace is
Nicolás> back-space. Those keybindings where thought for keyboards
Nicolás> without arrows, and those keyboards no longer exists...

This is OK, when you are on a system with relatively high latency.

However, arrow keys are usually represented as several keystrokes, eg
escape, followed by [D, for the left arrow. This represents problems
on telnet/ssh sessions even on local networks at times.

For instance I push the up arrow numerous times to get my editor to go
to the top of the file, and all of a sudden, the escape comes detached
from the [D, and I suddenly end up altering my file, inserting invalid
data, perhaps without even realizing it.

IMHO, the sooner this dead escape standard is dropped in favour of the
better standard (IIRC: set the high bit on control characters), the
better. However, I cannot see a lot of progress from here...
-- 
Brian May <[EMAIL PROTECTED]>

Re: ITP: texguy

[Masayuki Hatta]

[Currently I'm not on this list.  Please Cc: me]

Hi there,

I heard Ryuichi Arafune has ITP'd texguy, which has been 
already ITP'd by me(I guess around in the early March).

After uploaded vflib3 my laptop got trouble and had been under repair 
for several weeks, but it's back now, so new tex-guy packages should be 
uploaded pretty soon.

If you want, you can still get my older packages(based on 1.1.4) 
from one of those Debian JP mirrors.

Anyway, please wait a little bit ;-)

Regards,

--
Masayuki Hatta
University of Tokyo
[EMAIL PROTECTED] / [EMAIL PROTECTED]
[EMAIL PROTECTED] / [EMAIL PROTECTED]

Transmeta

[Scott Simmons]

Come work with the Linux development team at Transmeta. We're
looking for Linux developers interested in working on Mobile
Linux (which is based on Debian) and related projects. Linux
software developed at Transmeta is open-sourced under the GNU
Public License.

We are interested in hiring Linux developers with software
development and integration skills, C programming experience, and
experience building packages with dpkg and RPM. Previous
experience with Embedded systems and mobile devices is ideal.
System administration skills are a plus.

Note: Since this message is of a commercial nature, we cleared it
with Wichert Akkerman first and we're also donating $1000 to
Debian.

For more information or consideration, please contact:
Scott Simmons,
Employment Manager
Transmeta Corporation
[EMAIL PROTECTED]
408.919.3003
begin:vcard 
n:Simmons;Scott
tel;fax:408.919.1069
tel;work:408.919.3003
x-mozilla-html:TRUE
adr:;;
version:2.1
email;internet:[EMAIL PROTECTED]
fn:Scott Simmons
end:vcard

Re: Novell: NDS eDirectory for Linux

[Lauri Tischler]

Stephen Zander wrote:
> 
> AFAICT, eDirectory either provides a complete LDAPv3 implementation or
> at leats provides an LDAv3 front-end to NDS.  Now, while I'd like LDAP
> access to the NDS direcorty currently in place where I work, it
> doesn't strike me as anything new or outlandish.
> 
> Have any of the LDAP gurus (BenC?) looked at NDS and LDAP and how they
> interconnect?

Load NLDAP.NLM on one of your Netware servers.
I did just that and wrote some WEB-applications with php3 to get phone and
mailinggrop information from NDS.  

--
Lauri Tischler, Network Admin
Tel:+358-9-47846331*   Mouse movement detected  *
Fax:+358-9-47846500* Reboot Windows to activate changes *
Mobile: +358-40-5569010
EMail:  [EMAIL PROTECTED]

Re: FreeBSD suggests not to use lynx.

[Joey Hess]

Andreas Tille wrote:
> >   if [ -f /tmp/$$.html ]; then
> >   rm -f /tmp/$$.html
> >   fi
> >   /usr/lib/cgi-bin/dsearch $1 | \
> > sed 's/\/doc/\/usr\/share\/doc/g' > /tmp/$$.html
> Could you be so kind for unclever people like me to explain this
> anywhere (may be on devel) how this would work.  I don't see the relation
> between /tmp/$$.html and would like to understand the problem to avoid
> such cases i my own programs.

It's quite simple. An attacker merely guesses what pid the script will run
with, and makes a /tmp/.html file that is a link to, say, /etc/passwd.

The attacker then enters a tight loop, looking at the file (it might help
you to thisk of it forking off about 50 attackers, all doing this). If it is
deleted, it replaces it with the same link.

1 When the script runs, it deletes the file. 
2 With luck, the attacker runs at the right time, and replaces it.
3 The script echos some text to the file. Which is a symlink to /etc/passwd, and
  results in /etc/passwd being trashed.

See any introductory security text for details.

-- 
see shy jo

Re: [transcript] source package formats

[Manoj Srivastava]

>>"Adam" == Adam Heath <[EMAIL PROTECTED]> writes:

 Adam> In the transcript, it is mentioned that I wrote dbs.  It is a
 Adam> multi-tarball, multi-patch system, that is currently in use by
 Adam> several pkgs.  However, it has its drawbacks, the 2 most
 Adam> glaring that it hides the source in subtrees, all packed up,
 Adam> and that it doesn't extract into - directly.

Another drawback I find is that this is very difficult to
 use with a source code control system. espescially ODE and eve
 CVS. The native source format is no longer one which one can directly
 edit.

One of the feature I like of a monolithic patch, as you call
 it, is that when upstream incorporates changes previously made by me
 or others, it is automatically handled on upgrade; the monolithic
 patch just gets smaller. 

I think that keeping patches separate would involve work
 everytime the upstream sources incorporates parts of the patches. 

Despite clains that dbs does much the same that CVS does, I
 think the people making the claim really do not know version control
 systems (I am not guessing, I am quoting things from IRC). Don't take
 my word for it, try both systems and see.

manoj
-- 
 What excuses stand in your way?  How can you eliminate them? Roger
 von Oech
Manoj Srivastava   <[EMAIL PROTECTED]>  
1024R/C7261095 print CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C

Re: ITP: Netrek Vanilla server and COW (Client of Windows)

[Luca Filipozzi]

On Tue, Mar 21, 2000 at 10:13:49AM +0800, Neil Hunt wrote:
> I intend to package the Netrek vanilla server, and the COW client.

Cool. Are you also going to include instructions on how to play it from
behind a masquerading firewall? That would be the kind thing to do.

-- 
Luca Filipozzi


pgpHzJ7sGAaeI.pgp
Description: PGP signature

Re: Bug#60399: crashes on installation

[Daniel Martin]

Brian May <[EMAIL PROTECTED]> writes:

> False conclusion #1: now, it seems to crash regardless of if mandb is
> running or not. ARGGHH!! However, I have left in that in the message,
> in case it gives anybody else some ideas.

I seem to remember once upon a time that man would crash and burn if
one somehow had a corrupt index.bt file.  Could it be that a partially
completed install produces such a corrupt file (i.e. could it be that
whatever mandb does in the background is rebuilding this, and being
only half-way done means the resulting file seems corrupt), and that
an existing but corrupt index file causes havoc?

I would suggest backing up all your index.bt files, rm'ing them and
then attempt installing again (this should be doable with
   tar czf index.bt.tar.gz `find /var/cache/man -name index.bt`
   rm `find /var/cache/man -name index.bt`
)

If you discover that you can make the install succeed when you have
rm'ed your index.bt files, and yet fail again when you restore
them...

ITP: Netrek Vanilla server and COW (Client of Windows)

[Neil Hunt]

I intend to package the Netrek vanilla server, and the COW client.
These are available from http://vanilla.us.netrek.org/ and 
http://cow.netrek.org/ respectivly.

Netrek is a 16-player graphical real-time battle simulation with a Star Trek 
theme. The game is divided into two teams of 8 (or less),
who dogfight each other and attempt to conquer each others planets. There are 
several different types of ships, from fast, fragile scouts
up to big, slow battleships; this allows a great deal of variance in play 
styles.

It is played over the Internet, against real human opponents. If you do not 
have a computer on the internet, or connected via SLIP or
PPP, you will not be able to play.

These packages are being sponsered by Peter Crystal ([EMAIL PROTECTED]), and 
will be uploaded later today.

-- 
Neil Hunt   
HuntCorp Enterprises
[EMAIL PROTECTED]   
0414 306 238



pgpQfepsfJN3b.pgp
Description: PGP signature

Re: Libc6 functions such as wprintf() (widechar/multibyte/UNICODE support)

[Tomohiro KUBOTA]

Hi,

From: Jerry Lundström <[EMAIL PROTECTED]>
Subject: Libc6 functions such as wprintf() (widechar/multibyte/UNICODE support)
Date: Mon, 20 Mar 2000 17:29:29 +0100

> Ive been looking around the manual pages of wprintf and
> include files of the latest potato.
> Widechar/Multibyte support exist (wchar.h) but almost all
> the usefull function is not in the libc distribution of
> potato. Functions like wprintf that a vital is no where
> to be found and still they have manual pages for the
> function.
> Has the widechar/multibyte printing support been removed
> for som resone or just missed?
> 
> Im hopeing for a quick response on this matter because its
> very VITAL for my project!

Yes, these functions are very vital also for multibyte-language-
speaking (CJK) people, like me.  However, as far as I know, 
these functions are not implemented yet.  (I hope glibc 2.2 will have.)
On the other hand, conversion functions between multibyte characters
and wide characters are available now.  Can you use wcstombs() and 
printf() instead of wprintf() ?

---
Tomohiro KUBOTA <[EMAIL PROTECTED]>
http://surfchem0.riken.go.jp/~kubota/

Re: [transcript] source package formats

[Adam Heath]

On 21 Mar 2000, Brian May wrote:
> I am not sure I like (or understand for that matter) the input tar.gz
> files:
> 
>  93c152103ca081bd993bb216952a61be 293809 xawtv_3.07.orig.tar.gz
>  49d09b3edefb8279dc0ae1df3a42d15b 15908 xawtv_3.07-2.debian.tar.gz
>  fbb0a2fdc26934037ad11553f8f19814 2624 xawtv_3.07-2.diffs.tar.gz
> 
> The first one is obvious, it contains the original source code, same
> with the last, it contains the diffs. I will assume that
> xawtv_3.07-2.debian.tar.gz can contain binary files, that can't be
> patched (true/false?).

Not quite true.  The code I have does not treat *.orig.tar.gz specially.  It
does, however, treat the first *.tar.gz as the orig file.

> I wonder, if, rather then treating files ending in *.diffs.tar.gz as a
> special case, perhaps it might be better for these files to untar in a
> special location, eg xawtv-3.07/debian/diffs.  That way, all Debian
> specific stuff could all go into xawtv_3.07-2.debian.tar.gz, too.

Think multiple patch archives, and multiple file archives.  You can have 20
file archives, and 20 patch archives, and my code will cope.  You can even
share archives between source pkgs, which can be very helpful.

> I think this approach would allow greatest flexibility, should we ever
> decide we need to change in the future. Also, I suspect it might make
> maintaining these packages easier, if diff files can live in a
> standard spot within the source structure.

Actually, it will make things 'less trustworthy,' to quote Ian.  What is to
keep a script in debian/ from editting the files that exist in
debian/diffs?  And the diffs do live in a standard spot, in ../*.diffs.tar.gz.

BEGIN GEEK CODE BLOCK
Version: 3.12
GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS--
PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z?
-END GEEK CODE BLOCK-
BEGIN PGP INFO
Adam Heath <[EMAIL PROTECTED]>Finger Print | KeyID
67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP
AD46 C888 F587 F8A3 A6DA  3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG
-END PGP INFO-

Re: Novell: NDS eDirectory for Linux

[Stephen Zander]

Andreas recently commented elsewhere about Novell's latest foray into
directory management and Linux.  See:

http://www.novell.com/lead_stories/2000/mar13/linux/index.html>

for details.  My question, after reading various documents attached to
that URL, is:

Does this really matter?

AFAICT, eDirectory either provides a complete LDAPv3 implementation or
at leats provides an LDAv3 front-end to NDS.  Now, while I'd like LDAP
access to the NDS direcorty currently in place where I work, it
doesn't strike me as anything new or outlandish.

Have any of the LDAP gurus (BenC?) looked at NDS and LDAP and how they
interconnect?

-- 
Please respect the privacy of this mailing list.

To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]




-- 
Stephen

"Farcical aquatic ceremonies are no basis for a system of government!"

Re: [transcript] source package formats

[Brian May]

> "Adam" == Adam Heath <[EMAIL PROTECTED]> writes:

Adam> The script looks for all files matching the pattern
Adam> *.diffs.tar.gz(yes, it allows multiple patch archives.  This
Adam> could be used to share between sources), then everything
Adam> else that matches *.tar.gz is placed into a tarballs
Adam> variable.  The first listed tarball is assumed to be
Adam> orig.tar.gz, and special care is taken to ensure that it
Adam> unpacks into a properly named directory.

Adam> All other tarballs and patches are extracted/applied by
Adam> cd'ing into the target directory first.

Adam> Each .diffs.tar.gz contains a file named 'dsc' which holds
Adam> the Patches field from the first version.

Adam> Comments?  Suggestions?


I am not sure I like (or understand for that matter) the input tar.gz
files:

 93c152103ca081bd993bb216952a61be 293809 xawtv_3.07.orig.tar.gz
 49d09b3edefb8279dc0ae1df3a42d15b 15908 xawtv_3.07-2.debian.tar.gz
 fbb0a2fdc26934037ad11553f8f19814 2624 xawtv_3.07-2.diffs.tar.gz

The first one is obvious, it contains the original source code, same
with the last, it contains the diffs. I will assume that
xawtv_3.07-2.debian.tar.gz can contain binary files, that can't be
patched (true/false?).

I wonder, if, rather then treating files ending in *.diffs.tar.gz as a
special case, perhaps it might be better for these files to untar in a
special location, eg xawtv-3.07/debian/diffs.  That way, all Debian
specific stuff could all go into xawtv_3.07-2.debian.tar.gz, too.

I think this approach would allow greatest flexibility, should we ever
decide we need to change in the future. Also, I suspect it might make
maintaining these packages easier, if diff files can live in a
standard spot within the source structure.

However, I may have lost the plot somewhere here ;-)
-- 
Brian May <[EMAIL PROTECTED]>

Re: Bug#60753: mutt: /etc/Muttrc should not use colors

[Josip Rodin]

On Mon, Mar 20, 2000 at 02:38:12PM -0600, Steve Greenland wrote:
> There's a *lot* of history of using the space bar to "do the next thing"
> in Unix console programs (more/less, most news readers and existing mail
> readers). And there's no reason *not* to use them -- what else would you
> bind to the space key? You're right: the defaults should cater to the
> new user, but there's no reason to deliberatly aggravate the experienced
> user.

BTW dselect keybinding for exiting the help screen shouldn't be .
Hopefully Wichert will include my patch for that (and the rest of the help
stuff) in woody (he said he would).

-- 
enJoy -*/\*- don't even try to pronounce my first name

[no, there was no purpose to this post other than to say that exiting help
system with  sucks ;>]

Re: Apt-Problem

[Brian May]

> "Andreas" == Andreas Tille <[EMAIL PROTECTED]> writes:

Andreas> On 20 Mar 2000, Brian May wrote:
>> I have to agree with Jason here, I was confused. In this case
>> the error is generating by apt-get, in my case the error was
>> generated by dpkg.
>> 
>> I will take Jason's word for it that a deb file with bytes
>> missing can still be valid...
Andreas> OK, I take the word as you, but what should I do if I
Andreas> detect such kind of problem.  Who should I inform and
Andreas> which information should I ship.

Andreas> By the way.  Shouldn't dpkg at least warn that md5 sums
Andreas> are wrong?

The md5sums are included in the *.changes files, not the *.deb file.
dpkg never sees the *.changes file...
-- 
Brian May <[EMAIL PROTECTED]>

Re: Release-critical Bugreport for March 18, 2000

[Adam Heath]

On Mon, 20 Mar 2000, Josip Rodin wrote:

> On Mon, Mar 20, 2000 at 02:41:45AM -0800, Kevin Dalley wrote:
> [sth]
> 
> Is it just me or did something make this message go out like 12 times?
> 
> Each time it had more and more of these:
> 
> MBOX-Line: From [EMAIL PROTECTED]  Mon Mar 20 13:55:55 2000
> X-Envelope-Sender: [EMAIL PROTECTED]
> MBOX-Line: From [EMAIL PROTECTED]  Mon Mar 20 13:55:50 2000
> X-Envelope-Sender: [EMAIL PROTECTED]
> MBOX-Line: From [EMAIL PROTECTED]  Mon Mar 20 13:55:45 2000
> X-Envelope-Sender: [EMAIL PROTECTED]
> MBOX-Line: From [EMAIL PROTECTED]  Mon Mar 20 13:55:41 2000
> X-Envelope-Sender: [EMAIL PROTECTED]
> MBOX-Line: From [EMAIL PROTECTED]  Mon Mar 20 13:55:36 2000
> X-Envelope-Sender: [EMAIL PROTECTED]
> 
> WTF? :)

This was my fault.  I was playing around with my procmail filters, and it was
broken.  Sorry everyone. :(

BEGIN GEEK CODE BLOCK
Version: 3.12
GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS--
PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z?
-END GEEK CODE BLOCK-
BEGIN PGP INFO
Adam Heath <[EMAIL PROTECTED]>Finger Print | KeyID
67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP
AD46 C888 F587 F8A3 A6DA  3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG
-END PGP INFO-

Re: [transcript] source package formats

[Joey Hess]

Adam Heath wrote:
> I say again.  This is a .dsc unpacker, in the same way dpkg-source is.

I really think you're going about this backwards, or at least sideways..

If you have an idea for a new source package format, write up a document
explaining it in detail. Let people consider the implications. Then, when
people have a bit of a consensus that this could be a good thing, write an
implementation.

-- 
see shy jo