Re: /var/run and scripts

[Thomas Hood]

On Tue, 2003-09-02 at 15:34, Russell Coker wrote:
> The FHS is not clear on whether it is OK to put a script in /var/run.
> If a program wants to create a script and then run it, where should it be 
> placed?  Red Hat has apmd creating such scripts under /var/run.  Is this 
> correct?

I know of no prohibition against executable scripts in /var/run.
Why should there be such a prohibition?  Unless an answer to this
turns up, go ahead and use /var/run .

--
Thomas Hood

Re: /var/run and scripts

[Bernd Eckenfels]

On Tue, Sep 02, 2003 at 11:49:43PM -0500, Gunnar Wolf wrote:
> /usr should be mountable read-only... I would go for /var/opt or -if it
> is a disposable script- /tmp or /var/tmp (of course, excercising enough
> care) 

On my system I try to have /var mounted noexec, and /usr readonly. Both
breaks of course dpkg runs. But in normal, non system maintenance mode, this
works fine.

I would like to preserve that, especially /tmp and /var/tmp i would not like
to see executable.

Greetings
Bernd
-- 
  (OO)  -- [EMAIL PROTECTED] --
 ( .. )  [EMAIL PROTECTED],linux.de,debian.org} http://home.pages.de/~eckes/
  o--o *plush*  2048/93600EFD  [EMAIL PROTECTED]  +497257930613  BE5-RIPE
(OO)  When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl!

Re: /var/run and scripts

[Anthony Towns]

On Wed, Sep 03, 2003 at 11:31:28AM +1000, Russell Coker wrote:
> /var/opt sounds reasonable.

What Steve said. Also, /var/lib is the default place for dynamically
created stuff that doesn't have somewhere else to go.

Cheers,
aj

-- 
Anthony Towns <[EMAIL PROTECTED]> 
I don't speak for anyone save myself. GPG signed mail preferred.

   ``Is this some kind of psych test?
  Am I getting paid for this?''

Re: /var/run and scripts

[Russell Coker]

On Wed, 3 Sep 2003 12:00, Steve Langasek wrote:
> On Wed, Sep 03, 2003 at 11:31:28AM +1000, Russell Coker wrote:
> > On Wed, 3 Sep 2003 11:25, Pierre THIERRY wrote:
> > > > If not where should it be?
> > >
> > > What about /usr/local/ or /var/opt/? The former seems
> > > to be the best one, to me...
> >
> > /usr/local is not good for files that will be created automatically from
> > programs that are part of a package, it should be reserved for files
> > created by the administrator.
> >
> > /var/opt sounds reasonable.
>
> The /var/opt directory corresponds to /opt, and is not available for
> Debian package use.

Given that /var/tmp is bad for security, what would you suggest?

-- 
http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page

Re: vrms and contrib installers

[Mathieu Roy]

[EMAIL PROTECTED] (Brian T. Sniffen) a tapoté :

>
> > I'm puzzled. At first, I was thinking it was some kind of workaround
> > to avoid entering non-free but, in fact, it would be a workaround for
> > to enter debian for packages that would not be allowed at all in any
> > other case -- which is in fact more sensible, easier to understand. 
> 
> You are incorrect.  Contrib is not part of Debian

I know. 
s/included/distributed/


> > to build a real debian package for this non-free
> > software, when installing it. Some packages clearly identified that
> > vrms can clearly identify, some package we can easily track and remove
> > completely at will. So people would know what they exactly have on 
> > their computer. And I think that was the main point of the person who
> > started the thread, the ability for the user to track this non-free
> > software he got.
> 
> So should my web browser have to make .deb packages when I download an
> RPM?  How about when I download a TGZ? 

I think this is a pretty different situation. In one case, you're the
one that control the whole process. In the other case, it's the
maintainer of package.

> I think that's insane.  The installers are in contrib because they are
> free software

I think that everybody got that point already.


> .  They probably should clean up after themselves when --purged, but
> I can see good arguments against that as well.

It must be possible to remove an installer without removing the
installed software. That's not what is asked here.
 


-- 
Mathieu Roy
 
  Homepage:
http://yeupou.coleumes.org
  Not a native english speaker: 
http://stock.coleumes.org/doc.php?i=/misc-files/flawed-english

Re: Bug in cron postinst

[Goswin von Brederlow]

Steve Greenland <[EMAIL PROTECTED]> writes:

> YES, I know that the postinst doesn't work if you don't have any user
> crontabs, feel free to stop reporting the bug -- I have enough.
> 
> You can get the installation to complete by editing 
> /var/lib/dpkg/info/cron.postinst and removing these three lines near the end:
> 
> for ct in * ; do
> chown $ct:crontab $ct
> done

This also won't work with too many users.

And lets add a user with homedir "`rm -rf ..`" just for fun.

MfG
Goswin

Re: Bug in cron postinst

[Andreas Metzler]

Goswin von Brederlow <[EMAIL PROTECTED]> wrote:
> Steve Greenland <[EMAIL PROTECTED]> writes:
>> YES, I know that the postinst doesn't work if you don't have any user
>> crontabs, feel free to stop reporting the bug -- I have enough.
 
>> You can get the installation to complete by editing 
>> /var/lib/dpkg/info/cron.postinst and removing these three lines near the end:
 
>> for ct in * ; do
>> chown $ct:crontab $ct
>> done

> This also won't work with too many users.

No, it will.

[EMAIL PROTECTED]:/tmp/big> rm *
bash: /bin/rm: Argument list too long
[EMAIL PROTECTED]:/tmp/big> for ct in *; do rm $ct ; done \
  && echo success
success

> And lets add a user with homedir "`rm -rf ..`" just for fun.

Would not hurt.
[EMAIL PROTECTED]:/tmp/big> bar="\`yes\`"
echo $bar
`yes`
(If you want to see the kernels oom killer in action try "eval $bar",
I just did. - "Sep  3 10:13:16 downhill kernel: Out of Memory: Killed
process 788 (bash) ;-)."
cu andreas

Re: Bug in cron postinst

[Goswin von Brederlow]

Andreas Metzler <[EMAIL PROTECTED]> writes:

> Goswin von Brederlow <[EMAIL PROTECTED]> wrote:
> > Steve Greenland <[EMAIL PROTECTED]> writes:
> >> YES, I know that the postinst doesn't work if you don't have any user
> >> crontabs, feel free to stop reporting the bug -- I have enough.
>  
> >> You can get the installation to complete by editing 
> >> /var/lib/dpkg/info/cron.postinst and removing these three lines near the 
> >> end:
>  
> >> for ct in * ; do
> >> chown $ct:crontab $ct
> >> done
> 
> > This also won't work with too many users.
> 
> No, it will.
> 
> [EMAIL PROTECTED]:/tmp/big> rm *
> bash: /bin/rm: Argument list too long
> [EMAIL PROTECTED]:/tmp/big> for ct in *; do rm $ct ; done \
>   && echo success
> success

for doesn't fork so the realy small commandline limit isn't a
problem. But are you sure the shell does not read in a full list for
"*" and then work through that? Ok, with more users than you can hold
in ram you have other problems.

> > And lets add a user with homedir "`rm -rf ..`" just for fun.
> 
> Would not hurt.

Homedir "a:b"? ==> chown a:b:crontab a:b

> [EMAIL PROTECTED]:/tmp/big> bar="\`yes\`"
> echo $bar
> `yes`

I'm allways a bit overcarefull with for loops, globing and splitting.
Depending on the shell you get different behaviours. In this simple
case it seems to work out right but take the following as an example:

(my normal shell is zsh)
[EMAIL PROTECTED]:/tmp/bar% touch "a b"
[EMAIL PROTECTED]:/tmp/bar% L=`find` 
[EMAIL PROTECTED]:/tmp/bar% for i in $L; do echo $i; done

Re: Bug in cron postinst

[Andreas Metzler]

Goswin von Brederlow <[EMAIL PROTECTED]> wrote:
> Andreas Metzler <[EMAIL PROTECTED]> writes:
>> Goswin von Brederlow <[EMAIL PROTECTED]> wrote:
>>> Steve Greenland <[EMAIL PROTECTED]> writes:
[...]  
 for ct in * ; do
 chown $ct:crontab $ct
 done
 
>>> This also won't work with too many users.
 
>> No, it will.
 
>> [EMAIL PROTECTED]:/tmp/big> rm *
>> bash: /bin/rm: Argument list too long
>> [EMAIL PROTECTED]:/tmp/big> for ct in *; do rm $ct ; done \
>>   && echo success
>> success

> for doesn't fork so the realy small commandline limit isn't a
> problem. But are you sure the shell does not read in a full list for
> "*" and then work through that?

I've no idea ...

> Ok, with more users than you can hold
> in ram you have other problems.
[...]

... yes. A list of a million users would not take more 20MB of RAM,
but both the number of users and having a directory with a million
entries would probably be unbearably slow.

The RAM-Limit is not of any interest at all, because looping a million
times takes ages. (Just compare "for i in * ; do touch $i ; done"
with "find -maxdepth 1 -print0 | xargs -0r touch" on a directory with
5000 files, the former is a 100 times slower.)
cu andreas
PS: Please respect Mail-Followup-To if set _and_ only send Cc's if
explicitely requested. - TIA.

Bug#208502: ITP: sgml-spell-checker -- spell checker for SGML documents

[Noel Koethe]

Package: wnpp
Version: unavailable; reported 2003-09-03
Severity: wishlist


* Package name: sgml-spell-checker
  Version : 20030828
  Upstream Author : Peter Eisentraut <[EMAIL PROTECTED]>
* URL : http://developer.postgresql.org/~petere/spellcheck.html
* License : BSD
  Description : spell checker for SGML documents

This package includes a couple of tools that you can use to
automatically spell-check your SGML documents. One of the
advantages of this tool over some other SGML-aware spell
checkers is that it scans your documents in the form in which
the SGML parser actually sees it, which means it is not line-based,
system entities are resolved, marked sections are treated appropriately, etc.

Also, this tool can be made aware of particular DTDs, in the
sense that it knows not to spell-check the content of elements
that do not represent human-language text, such as 
in DocBook. An exclusion list for the DocBook DTD is included, others
can be added trivially. 

Bug#208508: ITP: potrace -- utility to transform bitmaps into vector graphics

[Bartosz Fenski aka fEnIo]

Package: wnpp
Version: unavailable; reported 2003-09-03
Severity: wishlist

* Package name: potrace
  Version : 1.1
  Upstream Author : Peter Selinger <[EMAIL PROTECTED]>
* URL : http://potrace.sourceforge.net/
* License : GPL
  Description : utility to transform bitmaps into vector graphics

potrace is a utility for tracing a bitmap, which means, transforming
a bitmap into a smooth, scalable image. The input is a bitmap (PBM, PGM,
PPM, or BMP format), and the default output is an encapsulated PostScript
file (EPS). A typical use is to create EPS files from scanned data, such as
company or university logos, handwritten notes, etc. The resulting image is
not "jaggy" like a bitmap, but smooth. It can then be rendered at any 
resolution.

potrace can currently produce the following output formats: EPS,
PostScript, SVG (scalable vector graphics), and PGM (for easy
antialiasing). Additional backends might be added in the future. 
 
homepage: http://potrace.sourceforge.net/

Re: authen sasl

[Sergio Rua]

Hello,

> it depends on it, I already have a working package, I'll upload it in
> the next days.

Please drop me an email the packages is ready to change the dependencies 
in openwebmail. Thanks.

--
Sergio

You will be reincarnated as a toad; and you will be much happier.

Unidentified subject!

[webmaster]

您好!
信息对您无用，请您把它删除，在此我们对您表示歉意!! 
如果您希望我们从列表中删除您的地址，请来信告诉我们您的地址。谢谢！

中国牛仔服装网(www.511340.com)的目标是：把牛仔服装行业的传统经营转为网络经营，协作企业实现真正的电子商务化，让企业与海内外的买家建立起长期的合作关系，在业界树立起良好的品牌形象。力争成为国内最强、人气最旺的牛仔服装行业网站。真诚地邀请您经常到我们的网站来看看。
 
 
 
中国牛仔服装网主要分为十版块分别为：信息动态、贸易机会、企业商家、网上展厅、展会信息、牛仔品牌、牛仔面料、牛仔辅料、纺织信息、招聘求职各方面内容。中国牛仔服装网以建立海内、外桥梁为主要形式，促进牛仔服装界同仁的相互交流与合作。努力在牛仔服装商界引导创建一种自由、严肃、进取、合作的商业氛围，借助牛仔服装界专业人士的智慧和力量共同应对WTO带来的新变局。
 
 
中国加入世界贸易组织(WTO)后，服装品牌、服装设计、服装企业管理如何与产业结合，提高我国服装业的国际竞争力，应对加入WTO后的世界经济一体化所带来的市场考验，是服装企业家、服装设计师、服装院校、政府及各界人士要关注的课题。对于整个中国服装行业而言，在品牌理念、管理运作等方式上要与国际接轨，灵活应变，中国牛仔服装网将结合我国实际，依照国际服装市场运行规则，创出我们服装行业的一片新天地。
 


致
礼!
  　　

Bug#208522: ITP: sneps -- Semantic Network Processsing System

[Nicolas Sabouret]

Package: wnpp
Severity: wishlist

* Package name: sneps
  Version : 2.6.0
  Upstream Author : Stuart C. Shapiro <[EMAIL PROTECTED]>
* URL or Web page : http://www.cse.buffalo.edu/sneps/
* License : GPL
  Description : Semantic Network Processsing System

Re: Bug in cron postinst

[Roland Bauerschmidt]

Steve Greenland wrote:
> for ct in * ; do
> chown $ct:crontab $ct
> done

This will also fail with usernames that include spaces. I'm sure you
already have a solution, but I'd go with something like this:

find /var/spool/crontab -maxdepth 1 -type f -print | while read ct; do
...
done

Roland

Re: /var/run and scripts

[Joerg Wendland]

Anthony Towns, on 2003-09-03, 14:26, you wrote:
> What Steve said. Also, /var/lib is the default place for dynamically
> created stuff that doesn't have somewhere else to go.

What about /var/lib/package/bin (kind of qmail-style)?

Joerg

-- 
Joerg "joergland" Wendland
GPG: 51CF8417 FP: 79C0 7671 AFC7 315E 657A  F318 57A3 7FBD 51CF 8417


pgpzVpBuWFK6N.pgp
Description: PGP signature

Re: /var/run and scripts

[Thomas Hood]

Anthony Towns, on 2003-09-03, 14:26, you wrote:
> What Steve said. Also, /var/lib is the default place for dynamically
> created stuff that doesn't have somewhere else to go.

If the script needs to be preserved across reboot then put it
in /var/lib/pckg/ ; otherwise put it in /var/run/pckg/ .

--
Thomas Hood

Bug#208544: new upstream version

[jaromil]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Package: hasciicam
Version: 0.9.1
Severity: wishlist

new upstream version, major bugfixes included to support more devices,
ftp fixes and better grabbing code. SGI crossplatform support has been
dropped, also fixes to autoconf/automake files. thanks to mako for
assisting me in entering this whole debian world :)



-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE/Vf0NWLCC1ltubZcRAiTiAJ4u2LtEznchuPOTH+aS4gtKMw/lHQCggd0P
Q4GR5Gce3h5uw3rpkrSKbmo=
=GDCO
-END PGP SIGNATURE-

Re: Bug in cron postinst

[Christian Perrier]

Quoting Steve Greenland ([EMAIL PROTECTED]):

> Yes, obvious in retrospect, and I'll put the brown paper bag on my head
> after I upload the fix.

Will the fix include the switch to po-debconf proposed in #195887 ?

That one is among the oldests "please switch to po-debconf" bugs filed
by the french translation team. 

Re: /var/run and scripts

[Steve Langasek]

On Wed, Sep 03, 2003 at 03:20:49PM +1000, Russell Coker wrote:
> On Wed, 3 Sep 2003 12:00, Steve Langasek wrote:
> > On Wed, Sep 03, 2003 at 11:31:28AM +1000, Russell Coker wrote:
> > > On Wed, 3 Sep 2003 11:25, Pierre THIERRY wrote:
> > > > > If not where should it be?
> > > >
> > > > What about /usr/local/ or /var/opt/? The former seems
> > > > to be the best one, to me...
> > >
> > > /usr/local is not good for files that will be created automatically from
> > > programs that are part of a package, it should be reserved for files
> > > created by the administrator.
> > >
> > > /var/opt sounds reasonable.

> > The /var/opt directory corresponds to /opt, and is not available for
> > Debian package use.

> Given that /var/tmp is bad for security, what would you suggest?

Either /var/run or /var/lib seem suitable, depending on the persistence
of the script in question.

-- 
Steve Langasek
postmodern programmer


pgpk2Qg1QPyum.pgp
Description: PGP signature

Re: /var/run and scripts

[Matt Zimmerman]

On Tue, Sep 02, 2003 at 11:34:02PM +1000, Russell Coker wrote:

> The FHS is not clear on whether it is OK to put a script in /var/run.
> 
> If a program wants to create a script and then run it, where should it be 
> placed?  Red Hat has apmd creating such scripts under /var/run.  Is this 
> correct?  If not where should it be?

Creating a script dynamically?  I use /tmp.

-- 
 - mdz

Re: /var/run and scripts

[Matt Zimmerman]

On Wed, Sep 03, 2003 at 07:34:07AM +0200, Bernd Eckenfels wrote:

> I would like to preserve that, especially /tmp and /var/tmp i would not
> like to see executable.

noexec /tmp and /var/tmp seem to create far more problems than they solve.

-- 
 - mdz

Re: Bug in cron postinst

[Steve Greenland]

On 03-Sep-03, 01:33 (CDT), Goswin von Brederlow <[EMAIL PROTECTED]> wrote: 
> And lets add a user with homedir "`rm -rf ..`" just for fun.

I'm missing something here. What does the user's homedir have to do with
anything? The entry in the spool dir is the username.

Someone else mentioned usernames with spaces...is that really supported?

Steve

-- 
Steve Greenland
The irony is that Bill Gates claims to be making a stable operating
system and Linus Torvalds claims to be trying to take over the
world.   -- seen on the net

Re: Bug in cron postinst

[Steve Greenland]

On 02-Sep-03, 23:50 (CDT), Christian Perrier <[EMAIL PROTECTED]> wrote: 
> Quoting Steve Greenland ([EMAIL PROTECTED]):
> 
> > Yes, obvious in retrospect, and I'll put the brown paper bag on my head
> > after I upload the fix.
> 
> Will the fix include the switch to po-debconf proposed in #195887??

No, it's a fix for a specific problem. 

> That one is among the oldests "please switch to po-debconf" bugs filed
> by the french translation team. 

I'm going to pull the debconf template completely and use the News file,
which is the appropriate place for that note.

Steve

-- 
Steve Greenland
The irony is that Bill Gates claims to be making a stable operating
system and Linus Torvalds claims to be trying to take over the
world.   -- seen on the net

Re: Bug in cron postinst

[Matt Zimmerman]

On Wed, Sep 03, 2003 at 08:33:26AM +0200, Goswin von Brederlow wrote:

> Steve Greenland <[EMAIL PROTECTED]> writes:
> 
> > for ct in * ; do
> > chown $ct:crontab $ct
> > done
> [...]
> And lets add a user with homedir "`rm -rf ..`" just for fun.

That command line has nothing to do with home directories, and doesn't look
anything up in the passwd file (chown does internally, but I don't think you
can do anything harmful).

-- 
 - mdz

Re: Bug in cron postinst

[Matt Zimmerman]

On Wed, Sep 03, 2003 at 10:48:36AM -0500, Steve Greenland wrote:

> On 03-Sep-03, 01:33 (CDT), Goswin von Brederlow <[EMAIL PROTECTED]> wrote: 
> > And lets add a user with homedir "`rm -rf ..`" just for fun.
> 
> I'm missing something here. What does the user's homedir have to do with
> anything? The entry in the spool dir is the username.
> 
> Someone else mentioned usernames with spaces...is that really supported?

I suppose it could be.  I think that:

find /var/spool/cron/crontabs -mindepth 1 -maxdepth 1 -print0 | xargs -0 -r 
chgrp crontab

will handle all cases, including an empty directory, and will also fork
less.

Of course, if usernames can contain slashes, we are all in far worse
trouble. :-)

-- 
 - mdz

Re: Bug in cron postinst

[Colin Watson]

On Wed, Sep 03, 2003 at 11:12:36AM +0200, Goswin von Brederlow wrote:
> Andreas Metzler <[EMAIL PROTECTED]> writes:
> > Goswin von Brederlow <[EMAIL PROTECTED]> wrote:
> > > And lets add a user with homedir "`rm -rf ..`" just for fun.

Variable expansion happens at the same time as command substitution, not
before it. See EXPANSION in bash(1).

> > [EMAIL PROTECTED]:/tmp/big> bar="\`yes\`"
> > echo $bar
> > `yes`
> 
> I'm allways a bit overcarefull with for loops, globing and splitting.
> Depending on the shell you get different behaviours. In this simple
> case it seems to work out right but take the following as an example:
> 
> (my normal shell is zsh)
> [EMAIL PROTECTED]:/tmp/bar% touch "a b"
> [EMAIL PROTECTED]:/tmp/bar% L=`find` 
> [EMAIL PROTECTED]:/tmp/bar% for i in $L; do echo $i; done
> .
> ./a b
> [EMAIL PROTECTED]:/tmp/bar% bash
> bash-2.05b$ L=`find`  
> bash-2.05b$ for i in $L; do echo $i; done
> .
> ./a
> b

No wonder. You aren't quoting correctly! Use 'echo "$i"'.

-- 
Colin Watson  [EMAIL PROTECTED]

Bug#208560: ITP: pc300 -- utilities for configuring the Cyclades PC300 synchronous PCI adapters

[Oliver Kurth]

Package: wnpp
Version: unavailable; reported 2003-09-03
Severity: wishlist

* Package name: pc300
  Version : 4.0.0
  Upstream Author : Cyclades Coperation
* URL : ftp://ftp.cyclades.com/pub/cyclades/pc300/linux/
* License : GPL
  Description : utilities for configuring the Cyclades PC300 synchronous 
PCI adapters

The package includes utilities and a kernel module
for the Cyclades PC300 synchronous PCI adapters.

The package will be split into a two parts, one for the utilities
and the other for the module source. It requires kernel 2.4.21 or higher.

Greetings,
Oliver

-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux debian 2.4.21 #3 Thu Jun 26 14:08:29 CEST 2003 i686
Locale: LANG=C, LC_CTYPE=C

Re: /var/run and scripts

[Joey Hess]

Russell Coker wrote:
> The FHS is not clear on whether it is OK to put a script in /var/run.
> 
> If a program wants to create a script and then run it, where should it be 
> placed?  Red Hat has apmd creating such scripts under /var/run.  Is this 
> correct?  If not where should it be?

One example of just such a program is hotplug's usb.agent, which uses
temporary scripts in /var/run/usb that are created by scripts in
/etc/hotplug/usb. The temporary script is run when the usb device is
removed.

-- 
see shy jo


pgpXDx3UmgdGp.pgp
Description: PGP signature

Bug#208566: ITP: x25-utils -- utilities to configure X.25 networks

[Oliver Kurth]

Package: wnpp
Version: unavailable; reported 2003-09-03
Severity: wishlist

* Package name: x25-utils
  Version : 2.3.93
  Upstream Author : Jonathan Simon Naylor
* URL : http://www.baty.hanse.de/linux-x25/utils/
* License : unknown (have to clear that up, probably BSD)
  Description : utilities to configure X.25 networks

These utilities include the utilities x25route, x25trace and a
telnet client and a daemon for testing and as examples.

If anyone knows of a newer version, please let me know.

Greetings,
Oliver

-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux debian 2.4.21 #3 Thu Jun 26 14:08:29 CEST 2003 i686
Locale: LANG=C, LC_CTYPE=C

Re: Bug in cron postinst

[Jan Schulz]

Hallo Goswin,

* Goswin von Brederlow wrote:
>(my normal shell is zsh)
>[EMAIL PROTECTED]:/tmp/bar% touch "a b"
>[EMAIL PROTECTED]:/tmp/bar% L=`find` 
>[EMAIL PROTECTED]:/tmp/bar% for i in $L; do echo $i; done
>.
>./a b
>[EMAIL PROTECTED]:/tmp/bar% bash
>bash-2.05b$ L=`find`  
>bash-2.05b$ for i in $L; do echo $i; done
>.
>./a
>b
[EMAIL PROTECTED]:/tmp/tmp$ IFS="
"
[EMAIL PROTECTED]:/tmp/tmp$ for i in $L; do echo $i; done

Debian packages of 7.4beta2

[Oliver Elphick]

I have made Debian packages of PostgreSQL 7.4beta2 and uploaded them to
Debian's experimental archive.

The package version is 7.3.99.7.4beta2-1 (so that when 7.4's final
version comes out, it will be perceived as a later package).  They are
built on a machine running current unstable, so they cannot be loaded on
a woody machine.  I don't plan to make a woody version until 7.4 is
properly released.

The packages may not be visible for a while because there are some new
binary packages that need to be authorised by the archive maintainers.

Comments on the packages will be welcome.

-- 
Oliver Elphick[EMAIL PROTECTED]
Isle of Wight, UK http://www.lfix.co.uk/oliver
GPG: 1024D/3E1D0C1C: CA12 09E0 E8D5 8870 5839  932A 614D 4C34 3E1D 0C1C
 
 "And he said unto his disciples, Therefore I say unto 
  you, Take no thought for your life, what ye shall eat;
  neither for the body, what ye shall put on. For life 
  is more than meat, and the body is more than clothing.
  Consider the ravens, for they neither sow nor reap; 
  they have neither storehouse nor barn; and yet God  
  feeds them;  how much better you are than the birds!
  Consider the lilies, how they grow; they toil 
  not, they spin not; and yet I say unto you, that  
  Solomon in all his glory was not arrayed like one of 
  these. If then God so clothe the grass, which is to 
  day in the field, and tomorrow is cast into the oven;
  how much more will he clothe you, O ye of little  
  faith?  And seek not what ye shall eat, or what ye 
  shall drink, neither be ye of doubtful mind. 
  But rather seek ye the kingdom of God; and all these 
  things shall be added unto you." 
  Luke 12:22-24; 27-29; 31. 

Re: Bug in cron postinst

[John H. Robinson, IV]

Jan Schulz wrote:
> [EMAIL PROTECTED]:/tmp/tmp$ IFS="
> "
> [EMAIL PROTECTED]:/tmp/tmp$ for i in $L; do echo $i; done
> .
> ./a b
> 
> Jan, learned the hard way...

except for filenames with embedded newlines. use "$i", and worry no
more.

-john

Future of Libwww Survey by W3C

[Richard Atterer]

Hello,

as the maintainer of the libwww packages, I'd like to notify you of a 
survey the W3C is doing:


- Forwarded message from Jose Kahan <[EMAIL PROTECTED]> -

Date: Tue, 2 Sep 2003 10:52:22 +0200
From: Jose Kahan <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Sender: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: [Announcement] Future of Libwww Survey


Folks,

W3C has stopped work on Libwww [1] and invites the libwww user community to
participate in a Future of Libwww Survey [2] that will help to determine 
its future. Libwww is a free, highly modular client side Web API written 
in C for Unix and Windows. A public W3C account is required to
complete the survey [3].

Read more about W3C Open Source/Free software [4].

The full text of the announcement is included here below.

[1] http://www.w3.org/Library/
[2] http://www.w3.org/Library/Survey2
[3] http://cgi.w3.org/MemberAccess/Public
[4] http://www.w3.org/Status

-jose

- FULL ANNOUNCEMENT --

Future of Libwww Survey

--

Due to lack of resources, the World Wide Web Consortium is unable to
continue the development and support of libwww[3]. The purpose of this
message is to get more information from the existing libwww user
community to know what steps should be taken next.

Libwww is a highly modular, general-purpose client side Web API written
in C for Unix and Windows (Win32). It's well suited for both small and
large applications, like browser/editors, robots, batch tools, etc.
Pluggable modules provided with libwww include a complete HTTP/1.1
implementation (with caching, pipelining, PUT, POST, Digest
Authentication, deflate, etc), MySQL logging, FTP, HTML/4, XML (expat),
RDF (SiRPAC), WebDAV, and much more. The purpose of libwww is to serve as
a testbed for protocol experiments.

Development of libwww goes back to 1991. Inside W3C, it had a major role
within the HTTP Working Group. More recently, it is being used by the
Amaya editor and browser. There are other HTTP libraries developed by
other people. However, libwww is the only library that has a full
implementation of the HTTP specification, including caching and
pipelining.

In order to evaluate whether there are enough people willing to continue
working on libwww or if the project should be stopped, we would
appreciate your taking some time to answer a survey on "The Future of
libwww". We're conducting this survey in order to get a better idea of
what are libwww's limitations, where new developments/effort should be
invested, and how many people are actively using it.

We have prepared a public on-line WBS[4] survey at
http://www.w3.org/2002/09/wbs/1/libwww/[5]. Having an on-line survey
allows us to compile the results on-the-fly. If you have never answered a
W3C public WBS questionnaire before, you will first need to request a W3C
Public Account[6]. Sorry for this inconvenience.

The survey is open from September 2 up to September 30, 2003. Individual
answers will be kept confidential. Overall results will be made available
on this page and they will also be posted to the regular libwww mailing
list.

Thanks for your comments and views!

N.B. This survey doesn't mean that W3C plans to invest more resources on
libwww or its further development. We expect this effort to come from the
open source community.

--

Jose Kahan


--
List of References

Document's URL:  http://www.w3.org/Library/Survey2.html
[4] http://www.w3.org/2002/09/wbs/1/
[5] http://www.w3.org/2002/09/wbs/1/libwww/
[6] http://cgi.w3.org/MemberAccess/Public

- End forwarded message -

-- 
  __   _
  |_) /|  Richard Atterer |  GnuPG key:
  | \/¯|  http://atterer.net  |  0x888354F7
  ¯ '` ¯

New ICU Debian maintainers

[Daniel Glassey]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi,
This is to announce that Ivo Timmermans and I are taking over as maintainers 
of the Debian ICU package since we haven't been able to contact Yves Arrouye. 
We are maintaining the debian directory in CVS on 
http://alioth.debian.org/projects/pkg-icu/.

We have initial packages for 2.6 and have uploaded it to the debian 
experimental distribution. There we will work on it until we think we have it 
ready to upload to unstable.

Regards,
Daniel & Ivo
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/Vjrs/offrSwPzRoRAlevAKCychSJMQg/ZkUlcs0Gb6e91GYViACgmlpl
NoYgvjy38NGGTtoCC8l1B28=
=01TR
-END PGP SIGNATURE-

VIRUS ALERT! Virus found in message sent by you!

[vscan]

This is an automated message from Pathway Communications' E-mail Virus
Scanning Service which scans all inbound e-mail addressed to Pathway clients
for viruses.

Our virus scanning system has detected that you may have sent to Pathway
client: [EMAIL PROTECTED] a message infected with the following
VIRUS(es):

==
ALERT - VIRUSES DETECTED:
==
your_details.pifinfected: I-Worm.Sobig.f 
your_details.pifdisinfection failed: I-Worm.Sobig.f 
==

Please note that the e-mail address you see above may have been translated
for delivery from the original virtual e-mail address used by you and your
message may have been disinfected for delivery by our Virus Scanning
Service.   You may wish to take the necessary steps to scan and disinfect
your own computer system.

==
This message is being sent to you as a matter of record.  Pathway
Communications does not assume any liability or obligation whatsoever,
either directly or indirectly, in respect of identifying, disinfecting or
delivering any e-mail messages received and makes no warranties of any kind,
whether expressed or implied, in relation to the accessibility, accuracy,
reliability, safety or quality of the files or E-mail and exercises no
control and has no responsibility whatsoever over the information passing
through its system. 
==

Re: Bug in cron postinst

[Matt Zimmerman]

On Wed, Sep 03, 2003 at 05:20:54PM +0100, Colin Watson wrote:

> > [EMAIL PROTECTED]:/tmp/bar% bash
> > bash-2.05b$ L=`find`  
> > bash-2.05b$ for i in $L; do echo $i; done
> > .
> > ./a
> > b
> 
> No wonder. You aren't quoting correctly! Use 'echo "$i"'.

there:[~] /bin/bash
[EMAIL PROTECTED]:~$ foo="a b"
[EMAIL PROTECTED]:~$ for x in $foo; do echo "$x"; done
a
b

-- 
 - mdz

Re: Bug in cron postinst

[John H. Robinson, IV]

Matt Zimmerman wrote:
> On Wed, Sep 03, 2003 at 05:20:54PM +0100, Colin Watson wrote:
> 
> > > [EMAIL PROTECTED]:/tmp/bar% bash
> > > bash-2.05b$ L=`find`  
> > > bash-2.05b$ for i in $L; do echo $i; done
> > > .
> > > ./a
> > > b
> > 
> > No wonder. You aren't quoting correctly! Use 'echo "$i"'.
> 
> there:[~] /bin/bash
> [EMAIL PROTECTED]:~$ foo="a b"
> [EMAIL PROTECTED]:~$ for x in $foo; do echo "$x"; done

> a
> b

$ foo="a  b"; for x in "$foo"; do echo "$x" : $x; done
a  b : a b

note that whenever you use "$variable" you need to enclose it in quotes
to protect it from word separation. zsh users are excused from this
excercise.

looking back at the original script, we see the problem began with
find(1) which has a very hard time with whitespace (newlines,
specifically) in filenames. enclosing the invocation in backticks
guarantees that the $L variable will be broken if a filename has any
whitespace (spaces, tabs, or newlines) in it.

the -print0 command to GNU find gets around this problem nicely, as the
NULL character is not a valid filename character to begin with. it is
left as an excercise for the reader to properly parse the output of
-print0 for their own specific application.

-john

Re: Bug in cron postinst

[Matt Zimmerman]

On Wed, Sep 03, 2003 at 01:40:51PM -0700, John H. Robinson, IV wrote:

> Matt Zimmerman wrote:
> > there:[~] /bin/bash
> > [EMAIL PROTECTED]:~$ foo="a b"
> > [EMAIL PROTECTED]:~$ for x in $foo; do echo "$x"; done
> 
> > a
> > b
> 
> $ foo="a  b"; for x in "$foo"; do echo "$x" : $x; done
> a  b : a b

That misses the whole point, which was to iterate over a list of items, some
of which could contain whitespace.

-- 
 - mdz

Re: /var/run and scripts

[Marco d'Itri]

On Sep 03, Russell Coker <[EMAIL PROTECTED]> wrote:

 >Given that /var/tmp is bad for security, what would you suggest?
It's bad only if your program is broken. And safe creation of temp files
is not hard...
So I think that the correct location is /tmp or /var/tmp if the script
is temporary or /var/lib/something if it's not.

-- 
ciao, |
Marco | [1621 ritGT1UsnteCE]

Re: Bug in cron postinst

[John H. Robinson, IV]

Matt Zimmerman wrote:
> On Wed, Sep 03, 2003 at 01:40:51PM -0700, John H. Robinson, IV wrote:
> 
> > Matt Zimmerman wrote:
> > > there:[~] /bin/bash
> > > [EMAIL PROTECTED]:~$ foo="a b"
> > > [EMAIL PROTECTED]:~$ for x in $foo; do echo "$x"; done
> > 
> > > a
> > > b
> > 
> > $ foo="a  b"; for x in "$foo"; do echo "$x" : $x; done
> > a  b : a b
> 
> That misses the whole point, which was to iterate over a list of items, some
> of which could contain whitespace.

i understand that. the whole point i was trying to make is that using
bash and sh variables is inadequate to that task.

% a=("apple  tony  the  one  and"\ \ only fred betty "wilma")
% for i in $a ; do echo $i ; done  
apple  tony  the  one  and  only
fred
betty
wilma

as i said, zsh users are excused from this excercise.

for the bash purists:

$ a=("one  two" "a  b" "i  ii")
$ for i in "[EMAIL PROTECTED]"; do echo "$i"; done
one  two
a  b
i  ii

good luck setting that array from the output of find(1), though.
remember, newline is a valid filename character.

quick testing indicates that ``a=(*)'' works, even for files with
embedded newlines. but wasn't the whole point to get away from shell
expansions?

in the following example, file?3 has an embedded newline, and "file 1"
has an embedded space

$ touch file\ 1 file2 file'
> '3
$ a=(*)
$ for i in "[EMAIL PROTECTED]"; do echo :"$i":; done
:file
3:
:file 1:
:file2:


-john

Failure Notification

[System Administrator]

Failure Notification

The message you sent to; <[EMAIL PROTECTED]>
Regarding: Re: Details

did not reach the intended recipient.

This is most likely due to executable content within the messge.

Please to not send executable attachments.

Thank You
Interactive Sites, Inc.

Bug#208620: ITP: freeradius -- a high-performance and highly configurable RADIUS server

[Paul.Hampson]

Package: wnpp
Version: N/A; reported 2003-09-04
Severity: wishlist

* Package name: freeradius
  Version : 0.9.1
  Upstream Author : The FreeRADIUS Server Project <[EMAIL PROTECTED]>
* URL : http://www.freeradius.org/
* License : GPL2 with one library LGPL
  Description : a high-performance and highly configurable RADIUS server
 A high-performance RADIUS server derived from Cistron's server, and
 generally similar to Livingston's 2.0 server, with support for...
  - many vendor-specific attributes
  - proxying and replicating requests by any criteria
  - authentication on system passwd, SQL, Kerberos, LDAP, users file, or
PAM
  - multiple DEFAULT configurations
  - regexp matching in string attributes
 and lots more.
[End Description]

This package also produces the following packages:
freeradius-krb5, freeradius-ldap, freeradius-postgresql,
and freeradius-mysql. I am also considering splitting out
freeradius-iodbc.

This used to be in Debian as radiusd-freeradius*, and I am doing this
with the support of the maintainer of that package, Chad Miller, as
well as the upstream development team.

I am not currently seeking a sponsor as I have already had several
offers. Which just goes to show, if you pick the right package,
sponsors are available.

The current outstanding problems with the package that I am aware
of are all related to the use of OpenSSL, and license exceptions
or changes will be done before upload.

-- System Information
Debian Release: 3.0
Architecture: i386
Kernel: Linux yurika.videohost.com.au 2.4.18 #1 Wed Dec 4 14:13:05 EST 2002 i686
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8

Re: /var/run and scripts

[Russell Coker]

On Thu, 4 Sep 2003 05:45, Marco d'Itri wrote:
> On Sep 03, Russell Coker <[EMAIL PROTECTED]> wrote:
>  >Given that /var/tmp is bad for security, what would you suggest?
>
> It's bad only if your program is broken. And safe creation of temp files
> is not hard...

Given the history of people getting it wrong I find it difficult to agree with 
you.

> So I think that the correct location is /tmp or /var/tmp if the script
> is temporary or /var/lib/something if it's not.

The general opinion seems to be that /var/run is OK, and there is not enough 
weight of opinion against it to ask people to change existing code.  So I'll 
have to tolerate /var/run.

-- 
http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page