headers
Bob Proulx | 1 Jun 2004 01:26
Favicon
Gravatar

Re: How to parse sections in INI-file? (existing tool?)

Rudi Effe wrote:
> I plan to interpret / convert an ini-file like this:
> 
> [SECTION 1]
> property 1 = value 1
> property 2 = value 2
> property 3 = value 3
> 
> [SECTION 2]
> property 1 = value 4
> ...
> 
> [...]
> Any hints welcome

You said any hints.  Not suggesting this is the best way.  But for
single items it is not bad.

Using /bin/sh you can extract individual properties this way.

  PROPERTY1=$(sed -n '/^\[SECTION 1\]/,/^$/{/property 1 = */s///p;}' cnffile)

Bob
Permalink | Reply | 1 comment |
headers
Wouter Verhelst | 1 Jun 2004 01:29
Picon
Gravatar

Re: ITP: vigra -- Generic Computer Vision Algorithms

On Mon, May 31, 2004 at 04:41:45PM -0400, Evan Prodromou wrote:
> >>>>> "OE" == Oliver Elphick <olly <at> lfix.co.uk> writes:
> 
>     >> Package name : vigra
> 
>     OE> Perhaps you could find a different package name?
> 
>     OE> Your mail almost got dumped in my spam file by reflex, because
>     OE> of its close similarity to a certain product which I won't
>     OE> spell out for fear of triggering spam traps.
> 
> When we can't use the word "Viagra" in normal conversation, the
> spammers have already won.

Let them win, I don't care. This is not a race.

Do you actually *need* to use that word?

--

-- 
         EARTH
     smog  |   bricks
 AIR  --  mud  -- FIRE
soda water |   tequila
         WATER
 -- with thanks to fortune
Permalink | Reply | 3 comments |
headers
Andrew Pollock | 1 Jun 2004 01:28
Picon
Favicon

Bug#251862: ITP: bicyclerepair -- A refactoring tool for python

On Mon, May 31, 2004 at 10:14:04PM +1000, Robert Collins wrote:
> Package: wnpp
> Severity: wishlist
> 
> * Package name    : bicyclerepair
>   Version         : 0.9
>   Upstream Author : Shae Erisson <shae <at> lapland.fi>, Phil Dawes
>   <pdawes <at> users.sourceforge.net>
> * URL             : http://bicyclerepair.sourceforge.net/
> * License         : Other
>   Description     : A refactoring tool for python
> 
> 
>  long description...
>  A framework and refactoring tool for Python. IDE Plugins are included for
>  Pymacs, IDLE and Vim. Using Bicycle Repair Man you can rename classes, methods
>  and variables, and all users of them are found and adjusted appropriately.

Can the long description include wtf refactoring is and why I might want
some?

Andrew
Permalink | Reply | 2 comments |
headers
Andrew Pollock | 1 Jun 2004 01:21
Picon
Favicon

Re: Lost Trust

On Mon, May 31, 2004 at 08:39:53PM +0200, Josip Rodin wrote:
> On Mon, May 31, 2004 at 03:45:44PM +1000, Andrew Pollock wrote:
> > I think you'd be justified in your stance if woody had tg3 support, and
> > you did a dist-upgrade to sarge once it had been released (after reading
> > the release notes) and your tg3 support disappeared without warning.
> 
> As if saying "we removed this and that, lalala" in the release notes will be
> much better.

I'm not saying it's a good way to be, but at least there's been a decent
attempt at warning the user through the appropriate channels.

regards

Andrew
Permalink | Reply | 1 comment |
headers
Adam Byrtek | 1 Jun 2004 01:06
Picon
Favicon

Bug#251980: ITP: hc-cron -- A cron daemon for home computers

Package: wnpp
Severity: wishlist

* Package name    : hc-cron
  Version         : 0.15
  Upstream Author : Felix Braun <fbraun <at> atdot.org>
* URL             : http://freshmeat.net/projects/hc-cron/
* License         : GPL
  Description     : A cron daemon for home computers

 It runs specified jobs at periodic intervals and will remember the
 time when it was shut down and catch up jobs that have occurred
 during down time when it is started again. Hc-cron is based on the
 widely used vixie-cron and uses the same crontab format so that it
 can be used as a drop-in replacement for that program.

Regards,
Adam

PS. Please Cc any comments to me as I'm not subscribed to debian-devel.

-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (900, 'testing'), (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.6
Locale: LANG=pl_PL, LC_CTYPE=pl_PL
(Continue reading)

Permalink | Reply | 0 comments |
headers
Eric Wong | 1 Jun 2004 01:43

Re: logitech mouse [was: Lost Trust]

Erik Steffl <steffl <at> bigfoot.com> wrote:
>   if the driver is available somewhere it is probably fairly easy to 
> make it work with debian kernel. sometime it's not that easy to get the 
> HW working, e.g. when I got the SATA HD I had to spend time on lkml, 
> apply patches (well, use ac kernel with manually applied libata5 
> patches) etc. lot of fun. and now e.g. extra button on logitech mouse 
> doesn't work (worked fine with 2.4.x but 2.6.x is too smart (more like 
> smart-ass)).

Which Logitech mouse?  I'm assuming it's PS/2 since USB mice work more
or less the same (evdev is recommended for USB).  I co-wrote the 2.6
Logitech PS/2++ driver and also maintain a Debian package[1] for the
logitech_applet which controls USB functions.

1 - http://des.petta-tech.com/~eric/debian (a sponsor would be helpful)

--

-- 
Eric Wong                normalperson <at> yhbt.net
Petta Technology, Inc      eric <at> petta-tech.com
Permalink | Reply | 9 comments |
headers
Daniel Ruoso | 1 Jun 2004 02:16
Favicon
Gravatar

Debian Maintainer Groups Research

Hi,

I'd like to make a small research to know how many (and which) groups of
package maintainership exists in debian, and which group of packages
they work on and if they use alioth as project home, just like the
following example (that is actually a real group):

Name	 	   Use Alioth?	Which type of packages
Debian Perl Group	y	Perl libraries (ITAs and RFPs)

Please answer me in private, as I don't want to pollute debian-devel,
and I'll publish the result of the research later.

Thanks in advance,

daniel
Permalink | Reply | 2 comments |
headers
Erik Steffl | 1 Jun 2004 02:05
Picon

Re: logitech mouse [was: Lost Trust]

Eric Wong wrote:
> Erik Steffl <steffl <at> bigfoot.com> wrote:
> 
>>  if the driver is available somewhere it is probably fairly easy to 
>>make it work with debian kernel. sometime it's not that easy to get the 
>>HW working, e.g. when I got the SATA HD I had to spend time on lkml, 
>>apply patches (well, use ac kernel with manually applied libata5 
>>patches) etc. lot of fun. and now e.g. extra button on logitech mouse 
>>doesn't work (worked fine with 2.4.x but 2.6.x is too smart (more like 
>>smart-ass)).
> 
> 
> Which Logitech mouse?  I'm assuming it's PS/2 since USB mice work more
> or less the same (evdev is recommended for USB).  I co-wrote the 2.6
> Logitech PS/2++ driver and also maintain a Debian package[1] for the
> logitech_applet which controls USB functions.

   what's the status of the logitech driver? on my system the side 
button doesn't work (xev thinks it's button 2). Any comments (should it 
work? known problem?)

   it's ps/2 logitech cordless mouseman wheel, this is on debian 
unstable, kernel  2.6.5, X settings: Option "Protocol" "ImPS/2", Option 
"Device" "/dev/psaux", here's what kernel says when I modprobe psmouse:

May 31 17:03:26 jojda kernel: input: ImPS/2 Logitech Wheel Mouse on 
isa0060/serio1

   it used to work with kernel 2.4.x
(Continue reading)

Permalink | Reply | 5 comments |
headers
Nathanael Nerode | 1 Jun 2004 02:07
Picon

Re: Lost Trust

else <at> austin.rr.com wrote: 
>I'm a recent convert to debian at home and I was considering switching
>to it for development at work.

A working patch existed to remove the non-free parts of the driver while 
keeping the driver.

The maintainer at the time preferred to remove the driver entirely, because he 
was not happy with the patch.

Now, an even better patch exists and has (finally) made it in.

>I've been tracking sarge weeklies with
>jigdo and lost my network device when the tg3 network driver was
>removed without warning.
>
>I have to say I'm very disappointed.  I had this implicit trust that
>releases would get progressively better.  On my part, I would test
>them and if anything broke, I would help debug it.
First of all, sarge isn't a release at the moment.  It is certainly not 
guaranteed that 'testing' will get progressively better.

Secondly, making Debian 'better' does involve removing functionality sometimes 
(though not in this case), rather than keeping around buggy, unsupportable, 
unmaintainable stuff.

>The lists don't offer any consolation - I now see it was deleted because
>a patch to fix broken hardware was not offered in source form.  Right.
You did read the Social Contract, didn't you?
(Continue reading)

Permalink | Reply | 0 comments |
headers
Nathanael Nerode | 1 Jun 2004 02:11
Picon

Re: Lost Trust

Wouter Verhelst wrote:
>Please, try not to give such black-and-white responses. There's a world
>of difference between "merging non-free right into main" and "ensuring a
>driver, which requires non-free firmware to be loaded, works, instead of
>just throwing the driver and the firmware entirely out of Debian, and
>telling users that they're unlucky, and that they should've bought free
>hardware, ha ha".

Please come up with a meaningful difference between putting non-free 
peripheral software needed only for a few rare pieces of hardware (the actual 
case with the tg3 driver) into 'main', and putting any other piece of 
non-free  software in 'main'.

"Lost Trust" is a good title; I don't trust most Debian Developers to follow 
the Social Contract any more.

I agree, of course, that Herbert should have patched the driver to remove the 
usually-unnecessary firmware rather than just removing the driver.
Permalink | Reply | 5 comments |

Gmane