Re: /bin/sh has no man page.

[Bob Nielsen]

On Wed, 22 Apr 1998, Dale Scheetz wrote:

> /bin/sh is provided by bash, but doesn't come with its own man page.
> 
> How does one determine the differences between sh and bash?
> 
> Is there some documentation that I have missed?

It's a link:

$ ls -al /bin/sh
lrwxrwxrwx   1 root root4 Apr 19 09:12 /bin/sh -> bash

Bob


Bob Nielsen Internet: [EMAIL PROTECTED]
Tucson, AZ  AMPRnet:  [EMAIL PROTECTED]
http://www.primenet.com/~nielsen


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

Re: /bin/sh has no man page.

[Stephen Zander]

> "Martin" == Martin Schulze <[EMAIL PROTECTED]> writes:
>> Of course I could have just put both feet in my mouth too

Martin> Thing is that bash behaves different if called as sh or
Martin> bash.

You mean I was right about my feet? Cool! :)

-- 
Stephen
---
all coders are created equal; that they are endowed with certain
unalienable rights, of these are beer, net connectivity, and the
pursuit of bugfixes...  - Gregory R Block


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

Re: signals and atomicity

[Herbert Xu]

In article <[EMAIL PROTECTED]> you wrote:
> On Sat, 18 Apr 1998, Herbert Xu wrote:
>> This should work:
>> 
>> static int wait_or_timeout_retval = -1;
>> 
>> static void alarm_handler(int sig) {
>>  errno = ETIMEDOUT; 
>> }
>> 
>> int wait_or_timeout (int *status) {
>>  struct sigaction act;
>> 
>>  wait_or_timeout_retval = -1; 
>> 
>>  sigaction(SIGALRM, 0, &act);
>>  act.sa_handler = alarm_handler;
>>  act.sa_flags &= ~SA_RESTART;
>>  sigaction(SIGALRM, &act, 0);
>>  alarm(1);
>>  wait_or_timeout_retval = wait(status);
>>  alarm(0); 
>>  return wait_or_timeout_retval; 
>> }

> i do not think, this works.

> alarm() calls setitimer(). setitimer() modifies "errno".
> so, setting errno inside the signal handler does not work, i think.

That's easy to fix.  Just store ETIMEDOUT in some other variable that is
reset at the start of wait_or_timeout and store the result in errno if wait
fails with EINTR.

-- 
Debian GNU/Linux 1.3 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt


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

Re: /etc/passwd : which software does support this ?

[Herbert Xu]

In article <[EMAIL PROTECTED]> you wrote:
> On Tue, 21 Apr 1998, matthew.r.pavlovich.1 wrote:

>> xdm-shadow is already available.

> Yes, and I am using it. But the question is, does it support the "pri=",
> "umask=" and "ulimit=" fields in /etc/passwd? And do cron and at also
> know about these fields? If cron and at don't know about them and don't
> support them they're practically useless, like I said before.

Use the allow/deny files for cron and at.

-- 
Debian GNU/Linux 1.3 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt


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

Re: /etc/passwd : which software does support this ?

[Remco Blaakmeer]

On Thu, 23 Apr 1998, Herbert Xu wrote:

> In article <[EMAIL PROTECTED]> you wrote:
> > On Tue, 21 Apr 1998, matthew.r.pavlovich.1 wrote:
> 
> >> xdm-shadow is already available.
> 
> > Yes, and I am using it. But the question is, does it support the "pri=",
> > "umask=" and "ulimit=" fields in /etc/passwd? And do cron and at also
> > know about these fields? If cron and at don't know about them and don't
> > support them they're practically useless, like I said before.
> 
> Use the allow/deny files for cron and at.

But that won't be a real solution, of course. Suppose I want to impose
different priorities, different ulimits and different umasks to different
users while all of them are still able to run cron/at jobs. Now I don't
really want to do this since I am the only one who really uses my computer
(apart from some friends that log in occasionally), but IMHO it should be
possible if it can be done.

Should I file wishlist bugs against the packages that IMHO should, but
don't support these fields?

BTW, about the issue of programs breaking on 50 out of the 100 different
systems they normally run on: can't this be solved by some smart #ifdef's
in the code? I'd think that if you write a patch that adds a good feature
while not breaking other things, most upstream authors will accept it.

Remco


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

Re: elvis package

[Rev. Joseph Carter]

On Wed, Apr 22, 1998 at 06:27:03PM -0400, Raul Miller wrote:
> > I'm pretty sure that a program must be either entirely GPLed,
> > or contain no GPLed parts.  
> 
> More precisely, the non-gpled parts must not have terms which prevent
> compliance with the gpled parts.

Uhh, the GPL does not state that the software must simply be free, it
states that it must be free in a form that is compatible with the GPL.
The GPL has what some consider to be a "virus".  It is free, but
militantly so.  This is good sometimes, but bad others.  It would have
been bad for Netscape, which is why they opted to use their own.

What people do not realize is that an author can release code to be used
under more than once license, ie under GPL or MPL as you choose, as long
as you follow one of them.

I would be much happier with the GPL if it was a bit more compatible with
other free licenses, but it does serve its purpose for some things.  For
example, M$ can't run off with a derivative of Linux and sell it (both as
idea and product) as their own.


pgp1BlRlrCH0l.pgp
Description: PGP signature

Re: Intent to package moxa radius

[Rev. Joseph Carter]

> >>Anyway, could you explain to me how this advertising clause is so 
> harmful?
> > 
> > See http://www.gnu.org/philosophy/bsd.html.
> 
> Ok, this helps.  I am still at a loss why we mention BSD as one of the "free" 
> licenses in DFSG, and have no mention of this problem there.  I'll try to 
> contact Moxa about this problem, but I doubt a successful outcome, since I 
> think they really want to get some publicity out of making their software 
> free 
> one way or another.
> 
> Am I correct that this clause doesn't make software really non-free (DFSG 
> definition) ? Or am I missing something obvious in DFSG?

If you ask RMS, MANY licenses are not "free enough", including BSD,
Artistic, and others.  DFSG is not free enough for him, yet you can do
more with one of the other licenses.  Interesting how that works out.

RMS is pushing an ideal more than anything.


pgpiobXkCzdBn.pgp
Description: PGP signature

Re: less: extra entries for lesspipe

[Manoj Srivastava]

Hi,
>>"Carl" == Carl Mummert <[EMAIL PROTECTED]> writes:

Carl> Joey Hess wrote:

>> > I would also find it disorienting to less a binary executable and
>> get a long list of identifiers.  I _expect_ a screen full of
>> garbage, and that "/lib/ld-linux.so.2" in a particular position :-)

Carl> What would be the use of looking at a screen full of control
Carl> characters? I can see prepending the output of strings with
Carl> other useful stuff, though;

Just because you do not see a need does not mean that there is
 no use for that feature. If you want to run strings on a binary, you
 know how to do so. From the principle of least surprise, I say leave
 the default alone (I hate how it does tar zvvft already; I used to be
 able to look at text files in a tar.gz archive without running
 through hoops. But no, the default could not be left well enough
 alone). 

manoj
 who is tired of progras who think they are more intelligent than the humans
-- 
 Percentage of Redbook readers who say they would rather have their
 genitals permanently numbed than go deaf: 70 Harper's Index
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


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

Re: Intent to package moxa radius

[James A . Treacy]

> If you ask RMS, MANY licenses are not "free enough", including BSD,
> Artistic, and others.  DFSG is not free enough for him, yet you can do
> more with one of the other licenses.  Interesting how that works out.
> 
> RMS is pushing an ideal more than anything.
>
Please don't get into an argument about what license is more free.
It all depends on what you are trying to achieve. When I release code,
I find that the GPL preserves the rights that are important to me. Someone
who wants to use it in commercial software will complain that the GPL isn't
free enough because he can't use it. Many would say that the GPL is
more free than, for example, the BSD license because it guarantees that
modified versions remain free.

If you are going to argue, please explain where you are coming from so
you can get past the word 'free' and actually discuss something instead
of having everyone talk past each other.

Different people prefer different licenses. Why don't we all agree to
that and go drink some beers and work on 'free' software.

Jay Treacy


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

Re: Upgrading Debian 1.3.1r6 to frozen

[Manoj Srivastava]

Hi,

The emacsen problem is due to the fact that as distributed,
 elib is buggy. If you remove elib (which also nukes cvs-pvl ;-(),
 emacsen-common +emacs20 shall work.

emacs18 is also present in Incoming now, and has the same
 problem with elib (I locally hacked the Makefile to make it work).

Thanks for the report.

manoj
-- 
 "(The Chief Programmer) personally defines the functional and
 performance specifications, designs the program, codes it, tests it,
 and writes its documentation... He needs great talent, ten years
 experience and considerable systems and applications knowledge,
 whether in applied mathematics, business data handling, or whatever."
 Fred P. Brooks, _The Mythical Man Month_
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


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

Re: dpkg-perl should predepend on perl?

[Martin Mitchell]

Enrique Zanardi <[EMAIL PROTECTED]> writes:

> Wouldn't it be enough to make dpkg-perl predepend on perl-base instead of
> the whole perl? Or does it need some functionality not provided in
> perl-base?

If dpkg-perl depended on perl-base instead of perl, the problem would
disappear. However I don't know enough about dpkg-perl to be able to say
if it would work. The maintainer for dpkg-perl is listed as Klee Dienes,
but I haven't seen many mails from Klee for a while. Is anyone else
familiar with the dpkg-perl package?

Martin.


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

libfdisk problem in dinstall

[Martin Mitchell]

Another thing that caused trouble when I tried installing from the base
disks (1998-04-11) yesterday was a libfdisk error that prevented dinstall
from detecting that a swap partition had been created. An error message
was printed before returning to the dinstall menu, and it came from this
section of libfdisk.

int sread(int fd, unsigned long sector, void *buffer)
{
if (!sseek( fd, sector ))
return 0;
if (read( fd, buffer, 512 ) != 512) {
fprintf( stderr, "error reading sector %ld", sector );
perror("");
return 0;
}
return 1;
}

I was receiving the message "error reading sector 0" all the time, but
cfdisk handled the partitioning just fine, so I expect this is a problem
in libfdisk or dinstall somewhere.

I did reboot the machine after the 'partition' step, as suggested by
cfdisk, but even then this error occurred, and when I tried the 'Init
a swap partition step' dinstall reported a strange error message 'No DOS
FAT12 partitions were found that are not already mounted.' This made
no sense at all, as no partitions were yet mounted. (NB the partition
types were set correctly.)

In the end, I had to dinstall without swap setup, and set it up later
when the base system booted.

Has anyone else experienced these kind of problems?

Martin.


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

xteddy pixmap_path (Was: autoconf problem)

[Andreas Tille]

On 22 Apr 1998, Brederlow wrote:

> > request, how about if no -F is given, use argv[0].  So a hard link to
> > xpenguin gets the debian chap, etc...
> 
> A softlink would do. You can try it with gzip, gunzip, unzip and
> zcat.
Off course, that's what I wanted to write :).  It's done now.  But the
problem where to place the pixmaps is not solved.  I can't cope with the
solution to place it in $HOME because I want to only a view files there
and put all other stuff into subdirectories.

I would be happy if there would be a directory, say

$HOME/.configor$HOME/.etc

where ALL .files (.bashrc, .bash_profile, ...) could be stored and
$HOME would be as clean as possible.  So every new file which has to
be in $HOME by dedault would make it more ugly...

Regards

Andreas. 


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

Re: less: extra entries for lesspipe

[Darren/Torin/Who Ever...]

-BEGIN PGP SIGNED MESSAGE-

Richard Braakman, in an immanent manifestation of deity, wrote:
>One reason I like less is that it's so fast.  I sometimes use it on
>thousands of files at once.  (For example, when searching a source
>package for uses of a particular identifier).

I've been thinking about that.  (When not mono-focused on this grant.)
In slink, I'm thinking of re-writing lessopen in C so that it will be
faster.  Shell scripts are slow, so is starting up Perl a lot.  Loading
binaries is fast.

>This seems like a bad idea.  "strings" is not the obvious information
>to provide about an executable.  (Consider size, objdump, od, hexdump, 
>et cetera).

Okay.  Sounds good to me.  I wanted at least one objection though so
that I wasn't just disregarding the user's request out of hand.

Darren

P.S.  less is ready and waiting on master.  The new Perl package on the
other hand isn't but should be done this weekend if I can get this grant 
out of the way.
- -- 
<[EMAIL PROTECTED]>  <[EMAIL PROTECTED]> <[EMAIL 
PROTECTED]>
Darren Stalder/2608 Second Ave, @282/Seattle, WA 98121-1212/USA/+1-800-921-4996
@ Sysadmin, webweaver, postmaster for hire.  C/Perl/CGI programmer and tutor. @
@Make a little hot-tub in your soul.  @

-BEGIN PGP SIGNATURE-
Version: 2.6.3a
Charset: noconv
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface

iQCVAwUBNT7yk44wrq++1Ls5AQHhLAQAnJ64sJZHowv/oQPQMWjFrJaGpivRFlAs
ZYswO3REW0asoIgjWjYF6J7zYWddNJURXkYfKWpkbRXXgxfTpr36ocm083YyflKd
hJB3R+ghegapnkrO81sPon9KUQpWdcNnC3Y6+vi5qyGUEyOyffr+Ts4xJZHdkYVw
DgQbyaE6v5E=
=qMOJ
-END PGP SIGNATURE-


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

Re: less: extra entries for lesspipe

[Darren/Torin/Who Ever...]

-BEGIN PGP SIGNED MESSAGE-

Carl Mummert, in an immanent manifestation of deity, wrote:
>useful information when you use it on different types of files-- gzip for
>gz files, tar -t for tar files, groff for manpages, etc...  If I _wanted_
>to look at the raw data of a gzipped file, I could do it.  But how often

Note that while I have control of lessopen, it will never, ever
interpret any kind of text files such as manpages, html, etc.

Darren
- -- 
<[EMAIL PROTECTED]>  <[EMAIL PROTECTED]> <[EMAIL 
PROTECTED]>
Darren Stalder/2608 Second Ave, @282/Seattle, WA 98121-1212/USA/+1-800-921-4996
@ Sysadmin, webweaver, postmaster for hire.  C/Perl/CGI programmer and tutor. @
@Make a little hot-tub in your soul.  @

-BEGIN PGP SIGNATURE-
Version: 2.6.3a
Charset: noconv
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface

iQCVAwUBNT70U44wrq++1Ls5AQFuPwP/bvZW58h2v3lvn4oo2XX4FB/WqgNT8TCP
lJDYRW3/RuBoaBzbjLo48thtfkUFhR9AFkSCYmOlaLX17Qascak3StlXEkOoA1ct
4XpZ23Wm0l+EGfo0Y/3Cmth9bRHyH/n/eHg0cRTF5tU8dPR+8ypFk0Ir8+QfSnkb
pr1LPSdY86s=
=X96m
-END PGP SIGNATURE-


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

Re: less: extra entries for lesspipe

[Darren/Torin/Who Ever...]

-BEGIN PGP SIGNED MESSAGE-

Manoj Srivastava, in an immanent manifestation of deity, wrote:
> know how to do so. From the principle of least surprise, I say leave
> the default alone (I hate how it does tar zvvft already; I used to be
> able to look at text files in a tar.gz archive without running
> through hoops. But no, the default could not be left well enough

Uh, Manoj, if you don't want less to interpret things for, just don't
use lessopen.  If you still want to use it, you can just edit the
lessopen script.

That's a reason not to make it a binary.  Even reading a config file
would probably be too slow.  Hmm.  Maybe when I've got some time slices
in my brain, I can work on this...

Darren
- -- 
<[EMAIL PROTECTED]>  <[EMAIL PROTECTED]> <[EMAIL 
PROTECTED]>
Darren Stalder/2608 Second Ave, @282/Seattle, WA 98121-1212/USA/+1-800-921-4996
@ Sysadmin, webweaver, postmaster for hire.  C/Perl/CGI programmer and tutor. @
@Make a little hot-tub in your soul.  @

-BEGIN PGP SIGNATURE-
Version: 2.6.3a
Charset: noconv
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface

iQCVAwUBNT71JI4wrq++1Ls5AQF1WgP+MYu/cH3z07gfURvmtmumdSEA88trdozX
k2NjlydUvxlePS768g/M4Zhk3d94h/Rc9G53d4OiEOstTtrUtiiUVOyxSGHqyDe6
you7zjs/L0HHHavB1Qja0YAei6dOzTbCgswPyJyaIJS+pwd37/Bs/bOcLI5B6tOr
kiBW/DoRGb0=
=SmVS
-END PGP SIGNATURE-


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

License advice

[jdassen]

As far as I can tell this license is DFSG-free; please let me know if you
disagree.


Copyright (c) 1997-1998 by Armin Biere.

Author: Armin Biere.

Permission is granted to anyone to use this software for any
purpose on any computer system, and to redistribute it freely,
subject to the following restrictions:

1. The author is not responsible for the consequences of use of
   this software, no matter how awful, even if they arise
   from defects in it.  

2. The origin of this software must not be misrepresented, either  
   by explicit claim or by omission.

3. Altered versions must be plainly marked as such, and must not   
   be misrepresented as being the original software.


4. This notice may not be removed or altered.

Armin Biere
Thu Mar  5 16:48:52 EST 1998
-- 
J.H.M. Dassen | RUMOUR  Believe all you hear. Your world may  
[EMAIL PROTECTED]  | not be a better one than the one the blocks   
  | live in but it'll be a sight more vivid.  
  | - The Hipcrime Vocab by Chad C. Mulligan  


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

Re: libfdisk problem in dinstall

[Enrique Zanardi]

On Thu, Apr 23, 1998 at 04:07:13PM +1000, Martin Mitchell wrote:
> Another thing that caused trouble when I tried installing from the base
> disks (1998-04-11) yesterday was a libfdisk error that prevented dinstall
> from detecting that a swap partition had been created. An error message
> was printed before returning to the dinstall menu, and it came from this
> section of libfdisk.
> 
> int sread(int fd, unsigned long sector, void *buffer)
> {
> if (!sseek( fd, sector ))
> return 0;
> if (read( fd, buffer, 512 ) != 512) {
> fprintf( stderr, "error reading sector %ld", sector );
> perror("");
> return 0;
> }
> return 1;
> }
> 
> I was receiving the message "error reading sector 0" all the time, but
> cfdisk handled the partitioning just fine, so I expect this is a problem
> in libfdisk or dinstall somewhere.
> 
> I did reboot the machine after the 'partition' step, as suggested by
> cfdisk, but even then this error occurred, and when I tried the 'Init
> a swap partition step' dinstall reported a strange error message 'No DOS
> FAT12 partitions were found that are not already mounted.' This made
> no sense at all, as no partitions were yet mounted. (NB the partition
> types were set correctly.)
> 
> In the end, I had to dinstall without swap setup, and set it up later
> when the base system booted.
> 
> Has anyone else experienced these kind of problems?

Was your swap partition a logical partition? libfdisk had a bug reading
extended partitions in big disks. A fixed libfdisk will be included in next
boot-floppies release.

Thanks,
--
Enrique Zanardi[EMAIL PROTECTED]


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

Re: libfdisk problem in dinstall

[Roman Hodek]

> I was receiving the message "error reading sector 0" all the time,
> but cfdisk handled the partitioning just fine, so I expect this is a
> problem in libfdisk or dinstall somewhere.

That's really strange, since the message is about a real read error.
Is something special about the disk or the location of that swap
partition? Can you send me the partition layout (fdisk -l)?

Roman


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

xteddy

[Andreas Tille]

Alex Romosan wrote in PM (hope you can copy with posting this single
line to the list :):

> hi andreas, please find included the penguin pixmaps. thanks
Thank you very much.  I've included it in my xteddy Debian
package which is available under

ftp://bridge.physik.uni-halle.de/pub/debian-maintain/xteddy

or 

http://www.physik.uni-halle.de/~e2od5/debian/xteddy 

This is *no official* Debian package, because my application as
maintainer isn't completed.  But it should be the first upload
to the Debian mirror when I will be the official maintainer.

May be that up to this time I will found a better mechanism for
searching the pixmaps.

To get xpenguin simply 

  ln -s xteddy xpenguin
  xpenguin

OR

  xteddy -Fxpenguin

May be I will include the link in the Debian package.
Do you think that this makes sense?

Any welcomes to my first package are appreciated.

Regards

Andreas.


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

Re: less: extra entries for lesspipe

[Manoj Srivastava]

Hi,
>>"Darren" == Darren Stalder <[EMAIL PROTECTED]> writes:


Darren> Uh, Manoj, if you don't want less to interpret things for,
Darren> just don't use lessopen.  If you still want to use it, you can
Darren> just edit the lessopen script.

I know that. The accepted way of using less used to be to use
 a upstream filter. Right now, it is done like so:
__
__> printenv | grep -i less
PAGER=less -ciM
LESSOPEN=|lessopen %s
LESS=-ciMP?f%f :std in .?n?m(file %i of %m) ..?ltline %lt :byte %bB?s/%s 
..?e(END) ?x- Next\: %x.:?pB%pB\%..%t
LESSCHARSET=latin1
__> eval $(lesspipe)
__> printenv | grep -i less
PAGER=less -ciM
LESSCLOSE=/usr/bin/lesspipe '%s' '%s'
LESSOPEN=| /usr/bin/lesspipe '%s'
LESS=-ciMP?f%f :std in .?n?m(file %i of %m) ..?ltline %lt :byte %bB?s/%s 
..?e(END) ?x- Next\: %x.:?pB%pB\%..%t
LESSCHARSET=latin1
__

See how the distribution now uses lesspipe?

I am bemoaning the fact that lesspipe is not directly usable
 by me anymore, since all the stuff got added to it.

manoj

-- 
 Aren't you glad you're not getting all the government you pay for
 now?
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


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

Re: signals and atomicity

[erikyyy]

On Thu, 23 Apr 1998, Herbert Xu wrote:

> In article <[EMAIL PROTECTED]> you wrote:
> > On Sat, 18 Apr 1998, Herbert Xu wrote:
> >> This should work:
> >> 
> >> static int wait_or_timeout_retval = -1;
> >> 
> >> static void alarm_handler(int sig) {
> >>errno = ETIMEDOUT; 
> >> }
> >> 
> >> int wait_or_timeout (int *status) {
> >>struct sigaction act;
> >> 
> >>wait_or_timeout_retval = -1; 
> >> 
> >>sigaction(SIGALRM, 0, &act);
> >>act.sa_handler = alarm_handler;
> >>act.sa_flags &= ~SA_RESTART;
> >>sigaction(SIGALRM, &act, 0);
> >>alarm(1);
> >>wait_or_timeout_retval = wait(status);
> >>alarm(0); 
> >>return wait_or_timeout_retval; 
> >> }
> 
> > i do not think, this works.
> 
> > alarm() calls setitimer(). setitimer() modifies "errno".
> > so, setting errno inside the signal handler does not work, i think.
> 
> That's easy to fix.  Just store ETIMEDOUT in some other variable that is
> reset at the start of wait_or_timeout and store the result in errno if wait
> fails with EINTR.

this way you can fix the problem. but my initial problem was different:


2 processes. if (1) does a specific call (probably raising a signal), B's
next interruptible (whatever this means) system call must be interrupted.
the fact that (1) raised the signal mustn't get lost.

if you implement "interruptible" system calls this way:
1. UNBLOCK SIGNAL
2. SYSTEM CALL
3. BLOCK SIGNAL
it may happen that the signal handler is called just after unblocking the
signal but before the call. this way no EINTR happens, the signal is lost
and (2) is stuck in the system call.

because of this, i have to do a siglongjmp() in the signal handler.
now it isn't anymore possible that signals get lost. BUT ! i do not know
the return value of the system call, if it was interrupted.
remember, after the call , the signal is blocked again. now, if the call
succeeded, but still before the BLOCK SIGNAL command, if now again a
signal is received, siglongjmp jumps away, and the fact, that the system
call succeeded is just lost.

i am looking for an atomic way for a process to interrupt another
processes next system call, __that is marked as interruptible (whatever
this means)__. the interrupt __mustn't get lost__. afterwards i must
always know, wether the system call succeeded, or not.

i find no way of doing it. this is why i said "the whole system is
shi." i am sorry about this sentence, but i was in a very bad mood,
because for weeks i try to solve this thing and i do not succeed.


byebye
Erik

--
EMAIL: [EMAIL PROTECTED]  
   [EMAIL PROTECTED] o `QQ'_
IRC:   erikyyy/   __8
WWW:   http://wwwcip.rus.uni-stuttgart.de/~inf24628/  '  `
   http://tick.informatik.uni-stuttgart.de/~thieleek/


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

Re: Intent to package moxa radius

[James Troup]

Igor Grobman <[EMAIL PROTECTED]> writes:

> Am I correct that this clause doesn't make software really non-free
> (DFSG definition)?

Yes.  To quote from the DFSG itself:

|   10. Example Licenses
|The "GPL", "BSD", and "Artistic" licenses are examples of licenses
|that we consider "free".

Let's please end this FUD now; the advertising clause is extremely
obnoxious but it doesn't make software non-free.

-- 
James


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

Re: signals and atomicity

[Roman Hodek]

> if you implement "interruptible" system calls this way: 1. UNBLOCK
> SIGNAL 2. SYSTEM CALL 3. BLOCK SIGNAL it may happen that the signal
> handler is called just after unblocking the signal but before the
> call. this way no EINTR happens, the signal is lost and (2) is stuck
> in the system call.
> 
> because of this, i have to do a siglongjmp() in the signal handler.
> now it isn't anymore possible that signals get lost. BUT ! i do not
> know the return value of the system call, if it was interrupted.
> remember, after the call , the signal is blocked again. now, if the
> call succeeded, but still before the BLOCK SIGNAL command, if now
> again a signal is received, siglongjmp jumps away, and the fact,
> that the system call succeeded is just lost.

You can solve it this way (at least I hope... :-) :

static int timeout;

static void alarm_handler(int sig) {
timeout = 1; 
}

int wait_or_timeout (int *status) {
struct sigaction act;
int wait_retval;

timeout = 0;

sigaction(SIGALRM, 0, &act);
act.sa_handler = alarm_handler;
act.sa_flags &= ~SA_RESTART;
sigaction(SIGALRM, &act, 0);
alarm(1);
wait_retval = wait(status);
alarm(0); 
/* ... */
}

If 'timeout' is set at the end, SIGALRM was delivered before the
alarm(0). On the other hand, wait_retval is -EINTR if and only if the
system call itself has been interrupted. So you can distinguish
between the cases "system call interrupted" and "signal arrived, but
somewhere around the syscall".

Roman


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

Re: weird utmp/perl problem

[Guy Maor]

Roderick Schertler <[EMAIL PROTECTED]> writes:

> On 21 Apr 1998 08:55:51 -0700, [EMAIL PROTECTED] (Karl M. Hegbloom) said:
> > 
> >> It's just a silly bug.  It calls that code from some scripts which
> >> have fd0 dup'd elsewhere, so isatty(0) is false and getlogin() fails.
> > 
> > Will someone please fix it?  It's really annoying.  Is it in the bug
> > sys?
> 
> The bug is in libc6.  getlogin() is supposed to work even if isatty(0)
> is false.  It is bug #17528.

How can it?  If isatty() is false, then ttyname(3) is null, so
getlogin() can't look up the entry in utmp.

The bug is fixed in Klee's version.  He says that he will upload that
soon.


Guy


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

Re: /bin/sh has no man page.

[Guy Maor]

Martin Schulze <[EMAIL PROTECTED]> writes:

> Thing is that bash behaves different if called as sh or bash.

Yes, find the section in the manpage that starts, "If bash is invoked
with the name sh, it tries to mimic the".


Guy


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

Re: Blender 3D

[Herbert Xu]

In article <[EMAIL PROTECTED]> you wrote:
> I maintain the mesa package. As no packages (before this) depended on
> a libc5 version of mesa, I stopped including them. Adding them back
> seems like a step backwards. It would be much better if you could
> get blender recompiled with libc6 based libs.

What? I thought hamm is supposed to be libc5 compatible.  Would you remove
ncurses3.0 because no Debian packages depended on it? I think it should be
reintroduced and maybe removed for slink.

-- 
Debian GNU/Linux 1.3 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt


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

Re: Gnome debs?

[Herbert Xu]

In article <[EMAIL PROTECTED]> you wrote:
> I checked, Debian and Red Hat were not compatible.  (e.g. libpng and
> libjpeg have different sonames.)

How did this happen? Shouldn't we try to rectify this ASAP so that there is
binary compatibility?

-- 
Debian GNU/Linux 1.3 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt


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

Re: /bin/sh has no man page.

[Herbert Xu]

In article <[EMAIL PROTECTED]> you wrote:
> /bin/sh is provided by bash, but doesn't come with its own man page.

> How does one determine the differences between sh and bash?

> Is there some documentation that I have missed?

/usr/doc/bash/POSIX.NOTES.gz.

-- 
Debian GNU/Linux 1.3 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt


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

build: no utmp entry available

[Andreas Tille]

Hello,

when using /usr/bin/build from the devscripts package I get a lot
of Warnings
...
no utmp entry available, using value of LOGNAME ("e2od5") at
/usr/lib/dpkg/controllib.pl line 16.
...

How to suppress this more precisely how to make a valid "utmp entry"
because my LOGNAME shouldn't be used in any *.deb file.  It is a
confusing symbolic name in our internal network and I would like to
use my name in real live (tille) instead.

Regards

 Andreas.

PS: Should I post such kind of questions to debian-mentors?
I'm afraid an overflow of my mailbox if I subscribe more and
more mailing lists.



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

Re: elvis package

[Marcelo E. Magallon]

On Mon, 20 Apr 1998, Alexey Marinichev wrote:

> Anyway, I do not think we should have X support compiled in
> vim, rather it should be a different package or something like
> that.  It's pretty bad debian policy forbids it. 

Do like I do with wmaker: compile several binaries[1]. One with
X, one without. Use alternatives to manage the whole thing. 


Marcelo


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

Re: xteddy pixmap_path (Was: autoconf problem)

[Brederlow]

Andreas Tille <[EMAIL PROTECTED]> writes:

> On 22 Apr 1998, Brederlow wrote:
> 
> > > request, how about if no -F is given, use argv[0].  So a hard link to
> > > xpenguin gets the debian chap, etc...
> > 
> > A softlink would do. You can try it with gzip, gunzip, unzip and
> > zcat.
> Off course, that's what I wanted to write :).  It's done now.  But the

Nice.

> problem where to place the pixmaps is not solved.  I can't cope with the
> solution to place it in $HOME because I want to only a view files there
> and put all other stuff into subdirectories.
> 
> I would be happy if there would be a directory, say
> 
> $HOME/.configor$HOME/.etc
> 
> where ALL .files (.bashrc, .bash_profile, ...) could be stored and
> $HOME would be as clean as possible.  So every new file which has to
> be in $HOME by dedault would make it more ugly...

If you have more than one file, as this is the case, its a good idea
to have a subdir like netscape or xemacs has.
All files should thus be in $HOME/.xteddy/.

May the Source be with you.
Mrvn


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

Re: xteddy pixmap_path (Was: autoconf problem)

[Andreas Tille]

On 23 Apr 1998, Brederlow wrote:

> If you have more than one file, as this is the case, its a good idea
> to have a subdir like netscape or xemacs has.
> All files should thus be in $HOME/.xteddy/.
What about an ENVIRONMENT variable
XTEDDY=$HOME/.xteddy/
for you and
XTEDDY=/
for all who don't want to put this into $HOME/.xteddy?
So the search hirarchy could be:

 1)   XTEDDY
 2)   $(prefix)/include/X11/pixmap
 3)   .

By the way: Has anybody an idea how to mention the search path
$(prefix)/include/X11/pixmap in the manpage?  I hardcoded
/usr/X11R6/include/X11/pixmap because this is the standard
Debian location but if someone compiles for a non Debian system
may be the manpage is wrong?? Could you have a look at my
xteddy manpage which is part of the package under

  ftp://bridge.physik.uni-halle.de/pub/debian-maintain/xteddy

(this is the only lokation, because it isn't an official package at
this time.)

Regards

  Andreas.


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

Processed: f

[Ian Jackson]

Processing commands for [EMAIL PROTECTED]:

> reassign 21540 boot-floppies
Bug#21540: Doesn't recognize my disks
Bug reassigned from package `bootfloppies' to `boot-floppies'.

> reassign 21464 general
Bug#21464: bo -> hamm upgrade problems
Bug reassigned from package `dselect, general upgrading' to `general'.

> reassign 21385 emacs20
Bug#21385: emacs20_20.2-6.deb install failed
Bug reassigned from package `emacs20_20' to `emacs20'.

> reassign 21322 lg-base
Bug#21322: Problem STILL with lg-issue packages
Bug reassigned from package `lg-*' to `lg-base'.

> reassign 21246 imlib1
Bug#21246: non-maintainer upload; patches for arch Alpha
Bug reassigned from package `imlib' to `imlib1'.

> reassign 21245 libpng2
Bug#21245: non-maintainer upload; patches for arch Alpha
Bug reassigned from package `libpng' to `libpng2'.

> reassign 21179 ssh
Bug#21179: ssh generating new host key
Bug reassigned from package `ssy' to `ssh'.

> thanks
Stopping processing here.

Please contact me if you need assistance.

Ian Jackson
(administrator, Debian bugs database)


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

Processed: f

[Ian Jackson]

Processing commands for [EMAIL PROTECTED]:

> reassign 21253 general
Bug#21253: gperf can't be installed with libstdc++
Bug reassigned from package `libstdc++2.8-dev' to `general'.

> severity 21253 important
Bug#21253: gperf can't be installed with libstdc++
Severity set to `important'.

> merge 21253
Unknown command or malformed arguments to command.

> thanks
Stopping processing here.

Please contact me if you need assistance.

Ian Jackson
(administrator, Debian bugs database)


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

Re: Upgrading Debian 1.3.1r6 to frozen

[Bob Hilliard]

Hi,
 Did you really mean 18 or is that a typo for 19?  

 Is it going to hamm or slink?  It seems to me that emacs19 should
be in 2.0, but the release policy would seem to require it to go in
slink.

Bob

Manoj Srivastava <[EMAIL PROTECTED]>  wrote:
>   emacs18 is also present in Incoming now, and has the same
>  problem with elib (I locally hacked the Makefile to make it work).
-- 
   _
  |_)  _  |_   Robert D. Hilliard<[EMAIL PROTECTED]>
  |_) (_) |_)  Palm City, FL  USAPGP Key ID: A8E40EB9


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

Dictionary directory

[Bob Hilliard]

 When we discussed directory locations several weeks ago, I said I
would wait for some policy statements about the FHS.  After looking at
how much is already in /usr/share, I have changed my mind and now
agree with you.  I am almost finished with my dictionary packages, and
have built them to go into /usr/share/dict.

Bob
-- 
   _
  |_)  _  |_   Robert D. Hilliard<[EMAIL PROTECTED]>
  |_) (_) |_)  Palm City, FL  USAPGP Key ID: A8E40EB9


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

Re: License advice

[Marcus Brinkmann]

On Thu, Apr 23, 1998 at 09:53:54AM +0200, [EMAIL PROTECTED] wrote:
> As far as I can tell this license is DFSG-free; please let me know if you
> disagree.

This is weird, especially because of point 3. He means the right thing, but
he doesn't mention modification, redistribution of modificated versions and
selling. Please point him kindly to Debian, the dfsg and to a more explicit
license we require to put it in main. This would mean, that you point him to
Artistic License, GPL, Xfree's and so on.

Thank you,
Marcus

> Copyright (c) 1997-1998 by Armin Biere.
> 
> Author: Armin Biere.
> 
> Permission is granted to anyone to use this software for any
> purpose on any computer system, and to redistribute it freely,
> subject to the following restrictions:
> 
> 1. The author is not responsible for the consequences of use of
>this software, no matter how awful, even if they arise
>from defects in it.  
> 
> 2. The origin of this software must not be misrepresented, either  
>by explicit claim or by omission.
> 
> 3. Altered versions must be plainly marked as such, and must not   
>be misrepresented as being the original software.
> 
> 
> 4. This notice may not be removed or altered.
> 
> Armin Biere
> Thu Mar  5 16:48:52 EST 1998
> -- 
> J.H.M. Dassen | RUMOUR  Believe all you hear. Your world may  
> [EMAIL PROTECTED]  | not be a better one than the one the blocks   
>   | live in but it'll be a sight more vivid.  
>   | - The Hipcrime Vocab by Chad C. Mulligan  
> 
> 
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 
> 

-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


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

Re: weird utmp/perl problem

[Roderick Schertler]

On 23 Apr 1998 04:56:12 -0700, Guy Maor <[EMAIL PROTECTED]> said:
> Roderick Schertler <[EMAIL PROTECTED]> writes:
>> 
>> getlogin() is supposed to work even if isatty(0) is false.  It is bug
>> #17528.
> 
> How can it?  If isatty() is false, then ttyname(3) is null, so
> getlogin() can't look up the entry in utmp.

It's supposed to open /dev/tty instead of using stdin.  This is the way
it works on all the systems I could get people to check for me, which
are Linux with libc5, AIX 4.2, Solaris 2.4, Solaris 2.5, DG/UX 4.11 and
FreeBSD 2.2.2-R.

Here's an example (from dgux):

# who | grep 'pts/[45] '
roderick   pts/4Apr 23 02:01
ccost  pts/5Apr 23 09:11
# tty
/dev/pts/4
# cat getlogin
#!/usr/bin/perl -lw
$l = getlogin;
print defined($l) ? $l : 'undef';
# ./getlogin
roderick
# ./getlogin < /dev/pts/5
roderick
# ./getlogin < /dev/null
roderick
# setsid ./getlogin
undef
# 

-- 
Roderick Schertler
[EMAIL PROTECTED]


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

Re: libfdisk problem in dinstall

[Martin Mitchell]

Enrique Zanardi <[EMAIL PROTECTED]> writes:

> Was your swap partition a logical partition? libfdisk had a bug reading
> extended partitions in big disks. A fixed libfdisk will be included in next
> boot-floppies release.

Yes, it was a logical partition. It seems this problem is well known now.

Martin.


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

Re: libfdisk problem in dinstall

[Martin Mitchell]

Roman Hodek <[EMAIL PROTECTED]> writes:

> > I was receiving the message "error reading sector 0" all the time,
> > but cfdisk handled the partitioning just fine, so I expect this is a
> > problem in libfdisk or dinstall somewhere.
> 
> That's really strange, since the message is about a real read error.
> Is something special about the disk or the location of that swap
> partition? Can you send me the partition layout (fdisk -l)?

No, it isn't my machine, and I'm no longer near it. However I can say
that the swap partition was definitely over the first 2Gb on the drive.

Martin.


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

Re: /etc/passwd : which software does support this ?

[David Frey]

On Wed, 22 Apr 1998 02:41:27 +0200 Remco Blaakmeer wrote:
> On 21 Apr 1998, Guy Maor wrote:
[at, cron etc and priorities]
> I am not really a programer, but I wonder how hard it would be to put
> support for these fields in all programs like login, xdm, cron, at and
> anything I am forgetting. Could this be done?

Sure. It shouldn't be too difficult, since shadow-login is already able
to set the values.
The only point is not to forget anything.

David
-- 
David Frey (B98D36A9) = 51F35923114FC864 7D05FF173C61EFDE
Those who do not understand Unix are condemned to reinvent it, poorly.
  -- Henry Spencer




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

Re: Lists archives outside debian.org

[David Frey]

pgpXXqDighaAf.pgp
Description: PGP message

Paging Dan Quinlan

[Dale Scheetz]

I have been getting bounced mail from the testing mailing list sent to Dan
Quinlan. Can anyone verify that he has indeed left the project, or give me
a new address for him?

Thanks,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz   Phone:   1 (850) 656-9769
  Flexible Software  11000 McCrackin Road
  e-mail:  [EMAIL PROTECTED] Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-



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

Problems with ae and xterm

[Dale Scheetz]

Since ae has been converted to slang I have been having many problems with
the keybindings for ae.

When it was compiled against ncurses, the terminfo config files worked
fine with entries like:

.cursor_up $(kcuu1)

Which the termcap equivalent should be:

.cursor_up $(ku)

Except it doesn't work.

With this entry in the ae.rc file, pressing the up arrow produces ^[[A on
the screen. Replacing this with:

.cursor_up^[[A

causes the up arrow key to work (in both the console and an xterm)

The same problem occurs witht the function keys as well. Putting the
keycode into the ae.rc file allows them to work on the console, but they
don't work in an xterm.

This is where is gets even more bizzare. Checking the keybindings for the
function keys, the xterm defines F1 as ^[OP, but when I put that keycode
into the F1 entry, all I get when pressing F1 is the characters ^[OP get
echoed to the screen.

There are other problems in an xterm, like the fact that the screen is
black and the letters are white, with no cursor appearing on the screen. 

Outside of the function key problems, ae has never behaved this badly in
an xterm. Many of these difficulties appear to be involved with the slang
library. 

In addition the ^M problem also seems to be a slang problem. While there
is a bad declaration of nel in the terminfo files, changing this doesn't
seem to fix the problem (the current nel is defined to be ^M^J, which is
the DOS version of a ). It appears that slang is making this
happen.

I have tried to contact the slang maintainer, Chris Fearnley, but have
recieved no reply. Anyone know where he is?

The hardest part of this package has been to try to deal with the problems
of an xterm, without compiling ae as an X compatible program (something
which it was not originally designed to do)

>From my point of view, this editor is the fallback editor for the
installation, and as such, was not expected to be used in an xterm.
However, there are several shops that wish to manage these machines
remotely from an xterm using ae, so the problem remains...what should be
done about this problem?

My biggest problem right now is that I can't see any way for a non-X-aware
program to control its own keybindings when run in an xterm. Can anyone
provide a functional approach to this problem?

Waiting is,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz   Phone:   1 (850) 656-9769
  Flexible Software  11000 McCrackin Road
  e-mail:  [EMAIL PROTECTED] Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-


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

Re: /tmp exploits

[joost witteveen]

In an attempt to save the world from disaster, Manoj Srivastava wrote:
> Hi,
> >>"Ian" == Ian Jackson <[EMAIL PROTECTED]> writes:
> 
> Ian> I think noone, even humans, should type `echo blah > /tmp/junk'.
> Ian> Make a directory named after your userid, or use your home
> Ian> directory, or something.
> 
>   Right. But my machine, my rules -- and this proposal would
>  change the behaviour of Debian to be very different from _any_ other
>  UNIX I have been on. 

Actually, I agree with Manoj. But to make a long thread much short,
I've written a lib that contains a wrapper function for open().
(Yes, I know, I should also wrap fopen(), I'll do that tomorrow).

I'm sure that the following uuencoded .tar.gz (that contains the .deb,
the .changes, etc) is much shorter than the discussion. (Or, it
would have been if dpkg would have used bzip2 for it's compression).

It doesn't contain much at the moment -- just the lib, and a
/usr/doc/libstress/README file that explains how to use it.
But that should be just about enough, I'd say.

---from the README-
Hi, this package is intended to do several (at the moment only one)
stress-tests for you, by changing the way libc works.

If you do 
  export LD_PRELOAD=libstress.so.0
then you will run with the open() function modified, and it will now
test for open()'s of files in the /tmp dir (not subdirs of /tmp) that
forget to set the O_CREATE mode. This is usually a sign of a security bug.


If you find other things libstress could do, and you have a patch
for it, please mail me ([EMAIL PROTECTED]). Note: NEVER wait more
than 6 days for me to answer your post. If it takes more than 6 days,
assume I'm swamped with work, and just upload a non-maintainer
release of this package (and, please inform me about having done so).


Example of how to find security bugs:

   $ LD_PRELOAD=libstress.so.0.0.0 man man
   open called on /tmp without O_CREAT mode, file = /tmp/zman26517aaa, 
flags=1101, mode=666
   man: can't remove /tmp/zman26517aaa: Bad file descriptor

There we've got a man security bug. Let's file it:

   $ bug man
   Getting package's info...done.
   Please describe your problems using man in one line.
   Man still doens't specify O_CREAT when opening tmp files.

Now, let's find more...

Oh, note: I sometimes get:
  id: error in loading shared libraries
  /usr/lib/libstress.so.0.0.0: undefined symbol: dlsym
I also don't know what it means, but haven't looked at it too closely.
It did seem to go away when I added -ldl on the gcc command line, but
later it returned again. Hope you don't see it.

(note that due to a security bug in glibc2 (libc6), this also works for
setuid binaries, as long as libstress is installed in /usr/lib)

Thanks!


---end from the README-


begin 644 libstress.tar.gz
M'XL(`$!4/C4``^V[951$`\O$A`T#_=D_;P(`6LX>,$NH,-#A7WP%
MH/N[+VR=G82!?SH-0,K6"0SS^?@
MZ)\)=M!7$*B%+=B)PQEF+0Z0A%G:V+I#+=T]8+]K`COY`+3T+`/PW;9?4UM'\>[/_7:W\'#RV8("T#1CF!G47
M!CHY6SH[>0(`MAK2NI)>.E)JVGPZLDIJNF!Y02E;62E-20TY;1>OUYRZKYWD
M/-Q4]7RL;!4-)'G!&MJR5A`K!3U>2SDW=1\7(8">D(N;G:2BO9RODS7$`6(O
MH\5CI^L(T=54UI)R='>#V4XM61Y.?7UG%]K
M&?BX^FBZ:;]VM?)ULO)6T)-2>6T@"%#U5;;VA%GK^EI+B@'$)'6U=?_6*[)J
M,O]=G_Q?>^M?_*/Y[^?&/[:._W?\Y^([EMAIL PROTECTED])?[_WOZ,_[]?!'_%_W\&M-&8
M2.M:XGS(*`?&`3%4\KK'8E66SOV#VUII6_G&T-G!L4IKJV(_7_$&@63O;!,1
M-?[+(=K3,&6GI4DMGLF5#T=YK=:S[_'%QSI?HBK?/UE'[EMAIL PROTECTED]
MNS^A9I%I(BT%?W`A6O4=`/CBHBR^DH79XUO4K)
M2EW6/4`9*[]HX?-1TV=RRNUAZKI
M4IUYA%R]-CNX4I?]9?GG603M33VHJK9P]6P$KXG$]6M()[[4CI_#)(`*CNG2
MB<(*\^>$:(Q'I9.XH+=5MHZJ2:195X`+B=C=6SHWO/0'#L%K,HOENZVB;*4L
[EMAIL PROTECTED]:09![VWAHGL^`4?:-,/##Z++L+RWSK#CFJH//H8`C^/&L5MDSSD
MBA2/<7[:C]FG2=>]VPB<,1_Y70\R,7M;LVK=#AGNA7')1>TZ=%OZ]C?
M$0=!LW1P)-VB4]2AYK1H7W0*=FS+PBV*3%]&LRC(QL/[EMAIL PROTECTED]'#2+OO_:H#;.+AG860!>8[!C
[EMAIL PROTECTED])*HK;.>H.`]B>TH)G;1KP:LV1"EG*B"M)X&Y#L@)V2>0,NQ:-*!O249H
MQ#'SD[YG2I/NWBQ^KYYC8"(N(-!OZA<;*;?322[:*C&@RBGST.4NKC,*K%PJ
MZ#2_"BX,D_]\?]!]L)3U.>FS]`:F9;CY=R*-KSX72I3/;P<-QPG@@M-O0]9?R-%.^;M+8]4>5IFJHT,\_\+$9A
M:5.!:$1EA]^I6=D[N].SU4\_?$GFP(_=)B;FV@:]IO'[EMAIL PROTECTED]@Y)
M4#=3B%4:/PRQ`4C.R_1<,,NC]V6;E8G=E?D(#T72YV0UE`QK"A7+F3O*3Z2C
M8!BY#5M"'[JLUZY">Y_=!ZILL)+?GY'?I^`\EN$\.J'/,FK>VZ$C+-#A0B$=
MB5VKPVN+XHD;&#SZXVM6Z!U)7<[0DT7R>XYU#"P$VGH$)[EMAIL PROTECTED]'G'2$XJ(
M/:$/O3J4OB?OSGDARCF#&WH6&D_`05!)Q$%DRLB6RF$?*CMUS0:.!3&5$],0
M-K)FQ.1.RT).!QN5X98REF`
M(!Q?JK\,^=U;V.3WUN3^)Z^BTGBNZW$>#W$>T=?O,!$/0>Y4#\-_[/#!>1?_T*RVQX(J=)K,^DFZG7F^D^YJUEZ.KS>RWRQRHB4\)?
M3VAU>U'W6M_`<4%B*'A8.`)0I8U6R`T)0_;KBLZSTR#4U<0NKR&:5_YIFNHG
MR^=+3900&C0A#)<8!1R.?!640-&SD0",P)J1CN1QJ@>"?7_.;RK?#TV_WX3V
M1B2U'(W^\,F!P8-I!]8BN([EMAIL PROTECTED]@FW+8(E"AR8TA.I9Q5W_101:5BKZK?]
MPJLYWX3(*4_N?GY3_%3!=O[%)F_;7:2=ECU4K&.-.!E4#T#FAZ#5:$`!D#>!(4H/"W3XAGVK3`TZX8G+VGL1><33BM]-'Z"[EMAIL PROTECTED]
M4IRB$%HT&X:5',QM1"Q[`7AZT-0>EIU//[EMAIL PROTECTED])?+`1)Z(CR,OV97O
M`(C6.K.=%B9G^)%PQI6S>^K71W\#,]BM+JF+0;_,$)5>Y-?**.'-\D*\SR:J
MVC2[L2Q!NJ7*[/P!Q;B%?>[EMAIL PROT

Re: signals and atomicity

[erikyyy]

On Thu, 23 Apr 1998, Roman Hodek wrote:

> 
> > if you implement "interruptible" system calls this way: 1. UNBLOCK
> > SIGNAL 2. SYSTEM CALL 3. BLOCK SIGNAL it may happen that the signal
> > handler is called just after unblocking the signal but before the
> > call. this way no EINTR happens, the signal is lost and (2) is stuck
> > in the system call.
> > 
> > because of this, i have to do a siglongjmp() in the signal handler.
> > now it isn't anymore possible that signals get lost. BUT ! i do not
> > know the return value of the system call, if it was interrupted.
> > remember, after the call , the signal is blocked again. now, if the
> > call succeeded, but still before the BLOCK SIGNAL command, if now
> > again a signal is received, siglongjmp jumps away, and the fact,
> > that the system call succeeded is just lost.
> 
> You can solve it this way (at least I hope... :-) :
> 
> static int timeout;
> 
> static void alarm_handler(int sig) {
>   timeout = 1; 
> }
> 
> int wait_or_timeout (int *status) {
>   struct sigaction act;
>   int wait_retval;
> 
>   timeout = 0;
> 
>   sigaction(SIGALRM, 0, &act);
>   act.sa_handler = alarm_handler;
>   act.sa_flags &= ~SA_RESTART;
>   sigaction(SIGALRM, &act, 0);
>   alarm(1);
>   wait_retval = wait(status);
>   alarm(0); 
>   /* ... */
> }
> 
> If 'timeout' is set at the end, SIGALRM was delivered before the
> alarm(0). On the other hand, wait_retval is -EINTR if and only if the
> system call itself has been interrupted. So you can distinguish
> between the cases "system call interrupted" and "signal arrived, but
> somewhere around the syscall".
> 

no. this is no solution, because:

once you activate the signal with sigaction, the sender might send a
signal, and this one will be lost.

so the thing that i wanted (i wanted that NO signal gets lost) does not
work.

the problem cannot be solved without using siglongjmp().

i am sure, the problem cannot be solved with this library+operating
system :-(
(or does anyone know some assembler hacks ?)

byebye
Erik

--
EMAIL: [EMAIL PROTECTED]  
   [EMAIL PROTECTED] o `QQ'_
IRC:   erikyyy/   __8
WWW:   http://wwwcip.rus.uni-stuttgart.de/~inf24628/  '  `
   http://tick.informatik.uni-stuttgart.de/~thieleek/


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

Re: License advice

[Oliver Elphick]

Marcus Brinkmann wrote:
  >On Thu, Apr 23, 1998 at 09:53:54AM +0200, [EMAIL PROTECTED] wrote:
  >> As far as I can tell this license is DFSG-free; please let me know if you
  >> disagree.
  >
  >This is weird, especially because of point 3. He means the right thing, but
  >he doesn't mention modification, redistribution of modificated versions and
  >selling.

He says anyone can use it for any purpose!  That includes everything
you mention.  All he's asking is that people say if they have altered
it from his original version.  I can't see any problem.
...
  >> 
  >> Permission is granted to anyone to use this software for any
  >> purpose on any computer system, and to redistribute it freely,
  >> subject to the following restrictions:
  >> ...
  >> 3. Altered versions must be plainly marked as such, and must not  
  > 
  >>be misrepresented as being the original software.
-- 
Oliver Elphick[EMAIL PROTECTED]
Isle of Wight  http://www.lfix.co.uk/oliver

PGP key from public servers; key ID 32B8FAA1

 
Come to me, all you who labour and are heavily laden, and I will
give you rest. Take my yoke upon you, and learn from me; for I am
meek and lowly in heart, and you shall find rest for your souls.
For my yoke is easy and my burden is light.(Matthew 11: 28-30)



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

Re: Anyone want to make a Debian XDM login screen?

[Barak Pearlmutter]

Inspired by your cute reboot/halt button hack, I added it plus a
little background color to my own xdm.  Everyone here is pretty happy
with the result.

--BAP.

Added to the bottom of /etc/X11/xdm/Xsetup_0

  # reboot button
  /usr/bin/wish < /var/run/xdmbutton_0.pid
  
  # background
  /usr/bin/X11/xlock -inroot -delay 5 -mode bouboule &
  echo $! > /var/run/xdmlock_0.pid

Added to the bottom of /etc/X11/xdm/Xstartup_0

  # reboot button
  if test -r /var/run/xdmbutton_0.pid; then
kill `cat /var/run/xdmbutton_0.pid`
rm /var/run/xdmbutton_0.pid
  fi
  
  # background
  if test -r /var/run/xdmlock_0.pid; then
kill `cat /var/run/xdmlock_0.pid`
rm  /var/run/xdmlock_0.pid
  fi


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

Re: License advice

[Marcus Brinkmann]

On Thu, Apr 23, 1998 at 06:20:21PM +0200, Oliver Elphick wrote:
> Marcus Brinkmann wrote:
>   >On Thu, Apr 23, 1998 at 09:53:54AM +0200, [EMAIL PROTECTED] wrote:
>   >> As far as I can tell this license is DFSG-free; please let me know if you
>   >> disagree.
>   >
>   >This is weird, especially because of point 3. He means the right thing, but
>   >he doesn't mention modification, redistribution of modificated versions and
>   >selling.
> 
> He says anyone can use it for any purpose!  That includes everything
> you mention.  All he's asking is that people say if they have altered
> it from his original version.  I can't see any problem.

I know what he wants to say, and I understand that. However, the word "use"
is very well undefined, and I don't know how the court would see it. The
general consens seems to be that "use" does not imply everything I said.
The "use" of software is probably only running and, well, "using" it.

I'm not a lawyer, so this is no legal advice. But I would really recommend
you to get a better copyright from the author, especially as he seems to
want to share the code.

Marcus

-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


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

Re: build: no utmp entry available

[B. Bell]

Hi,
That's caused by a bug in the scripts (or possibly libc6), and there is
probably nothing wrong with your utmp.  look up the thread "weird
utmp/perl problem" in the archives.  When it is fixed, it still will not
do what you want, I think, because even if it does find the utmp entry, it
will use LOGNAME if they are not equal.

brad

On Thu, 23 Apr 1998, Andreas Tille wrote:

> Hello,
> 
> when using /usr/bin/build from the devscripts package I get a lot
> of Warnings
> ...
> no utmp entry available, using value of LOGNAME ("e2od5") at
> /usr/lib/dpkg/controllib.pl line 16.
> ...
> 
> How to suppress this more precisely how to make a valid "utmp entry"
> because my LOGNAME shouldn't be used in any *.deb file.  It is a
> confusing symbolic name in our internal network and I would like to
> use my name in real live (tille) instead.
> 
> Regards
> 
>  Andreas.
> 
> PS: Should I post such kind of questions to debian-mentors?
> I'm afraid an overflow of my mailbox if I subscribe more and
> more mailing lists.


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

Re: Dictionary directory

[Andreas Tille]

On Thu, 23 Apr 1998, Bob Hilliard wrote:

>  When we discussed directory locations several weeks ago, I said I
> would wait for some policy statements about the FHS.  After looking at
> how much is already in /usr/share, I have changed my mind and now
> agree with you.  I am almost finished with my dictionary packages, and
> have built them to go into /usr/share/dict.
OK, I decided to place the wordnet dictionary in /usr/share/dict/wordnet.

At the moment I try to Debianize wordnet and have some trouble splitting
the package.  The former maintainer has made tree parts BEFORE starting
deb-make, that ends in three different *.orig.tar.gz files.  This is
not what I intended to do.  I want to prepare the rules file in this
way that all tree packages come from only one source (as it normaly
should be).

The problem is the following:
Wordnet comes in different dirctories:
 dict/
 man/
 paper/
 src/
The contents of dict/ should go into wordnet-base, the content
of paper/ should go into wordnet-doc and the content of
src/ and man/ should go into wordnet (more precisely not the
content but what you get if you go into the directory and type
make install in this directory).

According to the manpage of debstd I listed the files of dict/
in wordnet-base.files and the files of paper/ in wordnet-doc.files.
But both, wordnet-base*.deb and wordnet-doc*.deb are empty
after ther build process.  I spend now the whole afternoon to
make my rules file working but without any success.
I've read some rules files of other packages too, but havn't got
any clue how to do that.

I'm sure that this seems to be a simple task but I failed every time.
PLEASE HELP.

Regards

Andreas.




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

Re: Anyone want to make a Debian XDM login screen?

[Paul Reavis]

> Subject: Re: Anyone want to make a Debian XDM login screen?
> Date: Thu, 23 Apr 1998 11:21:52 -0600 (MDT)
> From: Barak Pearlmutter <[EMAIL PROTECTED]>
> To: "David A. van Leeuwen" <[EMAIL PROTECTED]>
> CC: debian-devel@lists.debian.org
>
> Inspired by your cute reboot/halt button hack, I added it plus a
> little background color to my own xdm.  Everyone here is pretty happy
> with the result.
>
> --BAP.
>

That's cool. I hacked my xdm to move the login window to the top left-hand
corner, set the background black, and run xlock with a pretty anim in
the background. You have to kill it when folks log in. Here's da code
(apologies for volume):


--

Paul Reavis  [EMAIL PROTECTED]
Design Lead
Partner Software, Inc.http://www.partnersoft.com


! $XConsortium: Xresources,v 1.7 93/09/28 14:30:29 gildea Exp $
xlogin*login.translations: #override\
CtrlR: abort-display()\n\
F1: set-session-argument(failsafe) finish-field()\n\
CtrlReturn: set-session-argument(failsafe) finish-field()\n\
Return: set-session-argument() finish-field()
xlogin*borderWidth: 3
xlogin.Login.width: 400
xlogin.Login.height: 200
xlogin.Login.x: 0
xlogin.Login.y: 0
xlogin*greeting: I am CLIENTHOST
xlogin*namePrompt: what is your name:\ 
xlogin*passwdPrompt: what is your quest:\ 
xlogin*fail: Hey, no way, man...
#ifdef COLOR
xlogin*greetColor: orange
xlogin*failColor: red
*Foreground: orange
*Background: black
#else
xlogin*Foreground: black
xlogin*Background: white
#endif

XConsole.text.geometry: 480x130
XConsole.verbose:   true
XConsole*iconic:true
XConsole*font:  fixed

Chooser*geometry:   750x550+200+100
Chooser*allowShellResize:   false
Chooser*viewport.forceBars: true
Chooser*label.font: *-new century schoolbook-bold-i-normal-*-240-*
Chooser*label.label:XDMCP Host Menu  from CLIENTHOST
Chooser*list.font:  -*-*-medium-r-normal-*-*-230-*-*-c-*-iso8859-1
Chooser*Command.font:   *-new century schoolbook-bold-r-normal-*-180-*


#!/bin/sh
# $XConsortium: Xsetup_0,v 1.3 93/09/28 14:30:31 gildea Exp $

#if grep -q ^run-xconsole /etc/X11/config
#then
#  xconsole -geometry 480x130-0-0 -daemon -notify -verbose -fn fixed \
#-exitOnFail -file /dev/xconsole
#fi

if [ -x /usr/bin/X11/xsetroot ] ; then
  /usr/bin/X11/xsetroot -solid black
fi

if [ -x /usr/bin/X11/xlock ] ; then
  /usr/bin/X11/xlock -nolock -inroot -mode bouboule &
  echo $! > /var/run/LoginPretty.pid
fi
exit 0
#! /bin/sh
#
# This script is run as root after a user starts a session on :0.

# Call the global Xstartup script, if it exists
if [ -x /etc/X11/xdm/Xstartup ] ; then
  /etc/X11/xdm/Xstartup
fi

# :0 specific startup commands go here
start-stop-daemon --stop --quiet --pidfile /var/run/LoginPretty.pid

exit 0

FTP Tracking software

[Michel West]

Hi,
I need to find FTP Tracking software for our linux server and am having
trouble finding anything to fit the bill, can you help?

The actual request I received was to be able to:

1) Directory access for organizing files remotely (from win95 pc's)
2) FTP Tracking software.
3) password protected FTP access accounts.

Thanks,
Michel West
Kenwood ISD



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

Re: Intent to package moxa radius

[Rev. Joseph Carter]

On Thu, Apr 23, 1998 at 12:06:33AM -0400, James A.Treacy wrote:
> > If you ask RMS, MANY licenses are not "free enough", including BSD,
> > Artistic, and others.  DFSG is not free enough for him, yet you can do
> > more with one of the other licenses.  Interesting how that works out.
> > 
> > RMS is pushing an ideal more than anything.
> >
> Please don't get into an argument about what license is more free.
> It all depends on what you are trying to achieve. When I release code,
> I find that the GPL preserves the rights that are important to me. Someone
> who wants to use it in commercial software will complain that the GPL isn't
> free enough because he can't use it. Many would say that the GPL is
> more free than, for example, the BSD license because it guarantees that
> modified versions remain free.

It does that, but sometimes that is not always a good thing.  Take for
example the libreadline library.  It is GPL, not LGPL.  In order to link
this library which is somewhat standard (IMO at least) your software must
be GPL.  An example of this is ncftp which was using it--that's a nono,
even though it is a simple shared library.  In this instance, the GPL
actually hurt ncftp.

The program ncftp is freeware with source (think beer) so really anyone
can use it, even if they can't hack it to pieces.  But it is not
compatible with the GPL under which libreadline is released and therefore
cannot have libreadline dynamically linked.

This is a limitation on the GPL I think, and the reason I think so is that
while ncftp is not OpenSource/DFSG-free, Artistic for example would be.
However if ncftp were Artistic, it would still be incompatible with the
GPL license on libreadline.  As would BSD and a number of other licenses
which are OpenSource/DFSG-free.

In fact, the GPL is incompatible with every other free license I know of,
possibly in some places even its cousin, the LGPL.  So really, one must
ask:  who is incompatible with whom here?  I do personally consider this
a limitation in the GPL--whether caused by oversight or by idealism I
can't say for certain (reading the writings of RMS, I would say the
latter)


> If you are going to argue, please explain where you are coming from so
> you can get past the word 'free' and actually discuss something instead
> of having everyone talk past each other.

I'm certain RMS is not going to like my suggestion, but perhaps if the GPL
were to instead of saying that other parts must be GPL, etc that other
parts must fit under the DFSG?  I think most do not find this too
limiting, but I would like to hear others' opinions..

> Different people prefer different licenses. Why don't we all agree to
> that and go drink some beers and work on 'free' software.

sounds like a plan


pgpVauOy1BdaV.pgp
Description: PGP signature

Re: Upgrading Debian 1.3.1r6 to frozen

[Manoj Srivastava]

Hi,

I meant emacs19, and I think it is felt to be a bug fix
 release, (not working with the other debian emacsen setup is a pretty
 major bug) so it shall be in hamm (I think the release manager has
 already blessed this).

manoj

-- 
 "Your development gets rotten if you take too long to market it."
 Hitoshi Aoike, JVC Ltd., Tokyo
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


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

Re: Intent to package moxa radius

[Raul Miller]

Rev. Joseph Carter <[EMAIL PROTECTED]> wrote:
> be GPL.  An example of this is ncftp which was using it--that's a nono,
> even though it is a simple shared library.  In this instance, the GPL
> actually hurt ncftp.
...
> This is a limitation on the GPL I think, ...

It's a limitation of ncftp.

There's nothing preventing:

(a) The authors of ncftp releasing it under terms compatible with
GPL.  [This course of action has been taken by authors of other
pieces of software.]

(b) The authors of ncftp writing a workalike for readline.  [This
course of action has also been taken by a few authors of other
pieces of software.]

Now, if there is something about this situation that hurts the
functionality of readline, that would be a limitation of the GPL.
Whether this would be good or bad in some larger sense would be subject
for the gnu advocacy mailing list.

[Aside: I notice that a few people tend to be a lot more upset about the
terms of the GPL -- despite it having proved its worth many times --
than about the terms of licenses which wind up being fairly useless for
our purposes. I don't really understand why this is.]

-- 
Raul


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

Wordnet Package: (Was Re: Dictionary directory)

[Bob Hilliard]

Hi,

 DISCLAIMER: I am a new developer also, and my advice may not be
very good.  Also I am building my packages by hand, and have never
looked at any of the helper scripts, so I can't address how to make
debstd put the files where you want them.

 If I understand you correctly, you want to keep the original
source as one .orig.tar.gz, but make three binary packages, separated
as the earlier wordnet packages did.  The obvious solution would be to
make one package, but you may have good reasons for wanting to make
them separate.  

 When you say  
> Wordnet comes in different directories:
>  dict/
>  man/
>  paper/
>  src/

do you mean that the source unpacks into those three directories?
Does the upstream Makefile have separate targets to make in all four
directories, and does it have separate install targets for each
directory?

 To make multiple binary packages, you have to branch the debian
tree into debian/tmp, debian/tmp-2, debian/tmp-3, etc.  (You could
make them debian/tmp, debian/tmp-base and debian/tmp-doc if you like
mnemonic names).  I am aware of two techniques for getting the build
files into the proper debian/tmp* tree.  One is to let the Makefile
build and install into debian/tmp, them move each file from the
debian/tmp tree to the appropriate debian/tmp-? tree.  This is most
useful when only a few files need be moved.

 The other is to manipulate variables in a "$(MAKE) install"
statement in the rules file.  This can be done by hacking the variable
assignments in the Makefile, or by defining variables in the rules
file, then $(MAKE) install varname=$(VARNAME) where varname is the
variable used in the Makefile install target and VARNAME is the
variable defined in the rules file.  In this case, depending on how
the Makefile is set up, you may have to hack the Makefile to use var1,
var2, etc., and define VAR1, VAR2, etc. in the rules file.  Depending
on how the upstream package is set up, you may have to do this in the
Makefiles in each of the subdirectories, with different variables
defined. 

 Any modifications you make should be in Makefile.in, not the
Makefile itself, so they will be preserved when the configure target
is run.  Any such Makefile hacks should be documented in your
changelog file.

 I have one package that produces two .debs from one source, and I
am now working on one that will produce four .debs.  Later this
evening I will send you copies of the rules file and Makefile for
these to illustrate what I am talking about.

 I hope this helps.

Bob
-- 
   _
  |_)  _  |_   Robert D. Hilliard<[EMAIL PROTECTED]>
  |_) (_) |_)  Palm City, FL  USAPGP Key ID: A8E40EB9


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

Re: build: no utmp entry available

[Bob Hilliard]

Hi,

 You wrote:

> PS: Should I post such kind of questions to debian-mentors?
> I'm afraid an overflow of my mailbox if I subscribe more and
> more mailing lists.

 debian-mentors is relatively low volume, and will likely produce
better responses to new maintainer type questions than debian-devel.

Bob
-- 
   _
  |_)  _  |_   Robert D. Hilliard<[EMAIL PROTECTED]>
  |_) (_) |_)  Palm City, FL  USAPGP Key ID: A8E40EB9


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

Re: Intent to package moxa radius

[Scott Ellis]

On Thu, 23 Apr 1998, Raul Miller wrote:

> Rev. Joseph Carter <[EMAIL PROTECTED]> wrote:
> > be GPL.  An example of this is ncftp which was using it--that's a nono,
> > even though it is a simple shared library.  In this instance, the GPL
> > actually hurt ncftp.
> ...
> > This is a limitation on the GPL I think, ...
> 
> It's a limitation of ncftp.
> 
> There's nothing preventing:
> 
> (a) The authors of ncftp releasing it under terms compatible with
> GPL.  [This course of action has been taken by authors of other
> pieces of software.]

In fact, this has happened.  Ncftp 2.4.3 was released under the GPL, and
the author has told me that the 3.x releases will also have the same
licence.

-- 
Scott K. Ellis <[EMAIL PROTECTED]> http://www.gate.net/~storm/


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

Re: Upgrading Debian 1.3.1r6 to frozen

[Bob Hilliard]

Hi,

 Thanks and bravo!  I'm getting tired of fighting with dselect,
which insists on replacing my emacs 19.34-13 with emacs20.

Bob

> Hi,
> 
>   I meant emacs19, and I think it is felt to be a bug fix
>  release, (not working with the other debian emacsen setup is a pretty
>  major bug) so it shall be in hamm (I think the release manager has
>  already blessed this).
-- 
   _
  |_)  _  |_   Robert D. Hilliard<[EMAIL PROTECTED]>
  |_) (_) |_)  Palm City, FL  USAPGP Key ID: A8E40EB9


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

Re: /etc/passwd : which software does support this ?

[Marcus Brinkmann]

On Thu, Apr 23, 1998 at 03:59:18AM +0200, Remco Blaakmeer wrote:
> On Thu, 23 Apr 1998, Herbert Xu wrote:
> 
> BTW, about the issue of programs breaking on 50 out of the 100 different
> systems they normally run on: can't this be solved by some smart #ifdef's
> in the code? I'd think that if you write a patch that adds a good feature
> while not breaking other things, most upstream authors will accept it.

Sure, if it does not conflict with any of the other 250 #ifdef's you have in
to support all 100 operating systems (from zx spectrum [does anyone remember
those? I have one, is it valuable?] to Alpha station) ;)

Marcus

-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


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

Re: less: extra entries for lesspipe

[Marcus Brinkmann]

On Thu, Apr 23, 1998 at 12:49:43AM -0700, Darren/Torin/Who Ever... wrote:
> -BEGIN PGP SIGNED MESSAGE-
> 
> >This seems like a bad idea.  "strings" is not the obvious information
> >to provide about an executable.  (Consider size, objdump, od, hexdump, 
> >et cetera).
> 
> Okay.  Sounds good to me.  I wanted at least one objection though so
> that I wasn't just disregarding the user's request out of hand.

Please collect flashy examples and provide them with the package, possibly
with one line comments. Thsi would make life for newbies easier, for
interested people it would be something to get starting with and all the
prof's can improve it without caring about defaults ;)

Thank you,
Marcus

-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


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

Re: Intent to package moxa radius

[Manoj Srivastava]

Hi,
>>"Rev" == Rev Joseph Carter <[EMAIL PROTECTED]> writes:

Rev> It does that, but sometimes that is not always a good thing.
Rev> Take for example the libreadline library.  It is GPL, not LGPL.
Rev> In order to link this library which is somewhat standard (IMO at
Rev> least) your software must be GPL.  An example of this is ncftp
Rev> which was using it--that's a nono, even though it is a simple
Rev> shared library.  In this instance, the GPL actually hurt ncftp.

On the contrary. This is an excellent point you  made. ncftp
 is now under GPL!! Yay! libreadline not being under LGPL worked!
 Hurrah! 

So, it is indeed a good thing. Maybe more software would
 become DFSG free if we were a little more pushy about it, like RMS
 has been being.

I think it is about time we hardened our stance in favour of
 the GPL. 

manoj

-- 
 It is impossible to make anything foolproof, because fools are so
 ingenious.
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


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

Re: Gnome debs?

[Joel Klecker]

At 22:29 +1000 1998-04-23, Herbert Xu wrote:
>In article <[EMAIL PROTECTED]> you wrote:
>> I checked, Debian and Red Hat were not compatible.  (e.g. libpng and
>> libjpeg have different sonames.)
>
>How did this happen?

2 is the upstream soname for libpng 1.0, so we are doing the right thing there.

--
Joel "Espy" Klecker
Debian GNU/Linux Developer...



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