Frans Englich | 1 Feb 01:18 2004
Picon

root kcms not working. Showstopper?


Sorry if this is just my setup and I'm spamming the list.. but _if_ this is 
serious, well then it's serious..

Regardless of what root kcm I try to open in admin mode, inside kcontrol I get 
a message similar to this:
Command 'kcmshell kdm --embed 35659633 --lang en_US' not found.
(System & Administration/Login Manager)
and it fails.

It's only my system which is wacko right?

			Frans

Frans Englich | 1 Feb 04:40 2004
Picon

Mixing with KCM KMix


Hi everyone,

In "KControl/Sound & Multimedia/Mixer" hides 3 settings, which I suggest we do 
the following with:

* Save/Load volumes
	Remove. They are obsolete(usability change pre KDE 3.2).

* Load volumes on Login
	Move to kmix's preference 

* "Maximum number of probed devices per mixer"
	Fascinating if users can figure out what this does. Remove it and set the 
default value to 3. kmixctrl will then loop 3 times(detecting at maximum 3 
sound cards). The option was there in the belief it affected performance. On 
my setup, there is no measurable difference between detecting for 15 or 1 
mixers. Although, it is best if someone elaborates on this..

The attached patch does these changes.

This leaves an empty kcm, except for its init routine. It could be replaced by 
doing a simple:

if kreadconfig --file kmixrc --group Misc --key startkdeRestore --default true 
--type bool; then kmixctrl --restore

(if this is the right way, where should it be executed? It must be quite early 
in startup)

(Continue reading)

Stephan Kulow | 1 Feb 10:06 2004
Picon

Re: root kcms not working. Showstopper?

Am Sonntag 01 Februar 2004 01:18 schrieb Frans Englich:
> It's only my system which is wacko right?
It works fine for me.

Greetings, Stephan

--

-- 
Pace Peace Paix Paz Frieden Pax Pokój Friður Fred Béke 和平
Hasiti Lapé Hetep Malu Mир Wolakota Santiphap Irini שלום Peoch
Shanti Vrede Baris Rój Mír Taika Rongo Sulh Py'guapy 평화

bj | 1 Feb 12:54 2004

KGpg decryption unusable in KDE 3.2

The focus stealing policy in KWin introduced a new bug in KGpg that makes 
decryption almost unusable.
In konqueror, left click on an encrypted file to decrypt it... and the 
password dialog appears behind the konqueror window. That looks really 
unprofessional and makes it a pain to use KGpg for decryption...

The bug was only reported yesterday and I added it to the side effect list of 
bug #69519 ... is there any chance to get a fix for KDE 3.2 (I don't have 
much hope, but would be really sad to see KGpg broken in KDE 3.2  :-/ ) ?

regards.
Frans Englich | 1 Feb 14:44 2004
Picon

Sanity check in startkde


If a curious user runs startkde in his/her kde session - the sessions is 
crashed.
This can be fixed by doing a sanity check. There's two candidates:

# check for already running KDE
dcop kdesktop KDesktopIface isVRoot >/dev/null 2>/dev/null;
if test "$?" = 0; then
  echo "ERROR: KDE seems already running on this display."
  xmessage -geometry 500x100 "ERROR: KDE seems already running on this 
display."
  exit 1
fi
(from SuSE's startkde, I'm told)

And:
if test "$KDE_FULL_SESSION" = "true"; then
  echo "ERROR: KDE seems already running on this display."
  xmessage -geometry 500x100 "ERROR: KDE seems already running on this 
display."
  exit 1
fi

I don't know why the former/latter should be chosen infront of the other, the 
latter is perhaps faster while the former is more safe - if KDE_FULL_SESSION 
for some reason is not unset it won't bail out..?

Code by Stephan, basically.

Cheers,
(Continue reading)

Jonathan Riddell | 1 Feb 16:40 2004

fixing typos in head


Throughout the string freeze en_GB converters and proofreaders have been
keeping a list of typos as we come across them.

http://www.kde.me.uk/index.php?page=KDE_Typos

Should it be OK to start commiting these back into head?

Jonathan Riddell

Benjamin Meyer | 1 Feb 17:50 2004
Picon

release scripts


I have noticed me and many others (on and off the list) doing tasks that could 
be done by scripts.  It might be a good idea in kdesdk to create a directory 
where scripts can reside that check over cvs for problems.  Some scripts that 
I can think of off the top of my head are:

-Check to make sure that icons that say they are size YxY really are that 
size.
-Check that for every icon foo there is all the required sizes (16,32,48...)
-Check every desktop file for duplicate entries and entries that are 
depreciated
-Check every desktop file to make sure that the icon that is specified exists
-Create a list of classes in kdelibs that aren't documented
-Create screenshots of widgets for apidocs (using the ui viewer etc) and 
create list of widgets that don't have screenshots
-Check the apidocs for spelling errors (hehe oh the evil!)
-Check every file that contains code also contains a license and copyright
-XML validator for the KConfigXT files

I am sure that everyone could come up with one or two scripts that would 
validate or create a list of bugs to fix.  What do you all think?  If 
anything it would certainly create an excellent list of simple problems for a 
janitors and would help to reduce the amount of time spent on these simple, 
but repetitive problems.   What other scripts could you think of?

-Benjamin Meyer

--

-- 
Public Key: http://www.csh.rit.edu/~benjamin/public_key.asc
(Continue reading)

Frans Englich | 1 Feb 18:28 2004
Picon

Re: release scripts

On Sunday 01 February 2004 17:50, Benjamin Meyer wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> I have noticed me and many others (on and off the list) doing tasks that
> could be done by scripts.  

Fully agreed. I think it will save us a lot of trouble. We need a 
regression/automating framework, and a place where such scripts are run 
periodically.
There must be some unified way of adding/"registering" regtests("regression 
tests"). In case the script can't fix the problem itself(it's an obvious 
s/regexp/regexp/ ie.) the regtest will have to send it's output somewhere. A 
script which runs over all kde modules(checking licenses for example) can't 
just send it to one person(or what do you say, Coolo? :). Perhaps each 
directory relevant for the regtest should contain a file where there's 
contact information for that dir(and if it is recursive, etc.). Perhaps it 
could be extracted from the apidocs. Whom to contact would could also depend 
on the tpe of output(critical vs coding style ie.). When contact persons 
can't be reached it must fallback to another contact(mailinglist, 
maintainer).

I have absolutely no experience in this field, does there already exist such 
software? Are there any plans for this already?

> It might be a good idea in kdesdk to create a 
> directory where scripts can reside that check over cvs for problems.  Some
> scripts that I can think of off the top of my head are:
>
> - -Check to make sure that icons that say they are size YxY really are that
(Continue reading)

Aaron J. Seigo | 1 Feb 19:03 2004
Picon

Re: fixing typos in head


On February 01, 2004 08:40, Jonathan Riddell wrote:
> Throughout the string freeze en_GB converters and proofreaders have been
> keeping a list of typos as we come across them.
>
> http://www.kde.me.uk/index.php?page=KDE_Typos
>
> Should it be OK to start commiting these back into head?

yes, head is unfrozen for string changes now that the 3.2 branch is around...

--

-- 
Aaron J. Seigo
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43
while (!horse()); cart();
Rob Kaper | 1 Feb 21:35 2004

Re: [Kopete-devel] 3.2 showstopper

On Fri, Jan 30, 2004 at 02:17:54PM -0400, Jason Keirstead wrote:
> I can tell you I use MSN every day and I have never experienced this...nor is 
> there a bug anything like it in BKO.

Olivier just committed a fix in HEAD/branch. I'll leave it up to the RC to
decide whether data loss is worth repackaging.

Rob
--

-- 
Rob Kaper     | 2/17 Blink-182, 2/29 Anti-Flag
cap <at> capsi.com | 3/5  Anti-Flag/Heideroosjes, 3/19 The Offspring
www.capsi.com | 3/24 Good Riddance/None More Black, 5/14 Disney

Gmane