On Friday 26 February 2010, Frans Pop wrote:
> > > Keeping the white as option could be useful for e.g. the "dark"
> > > theme, but that means we'd need a way to change the cursor.
> >
> > At (d-i, not udeb-providing-source-package) build time, I assume? In
> > both cases, it's just about a link to
On Friday 19 March 2010, Julien Cristau wrote:
> Looks like none of the Xlib locale data is used by gtk apps, so I got a
> libx11-6-udeb from the current
> Installed-Size: 3700
> to
> Installed-Size: 1344
> by removing that. Should help with the image size, I hope (although
> this is mostly text,
On Fri, Feb 19, 2010 at 11:16:43 +0100, Julien Cristau wrote:
> I'm pretty sure libx11 would benefit from this, as it has some code
> that's not used by anything, but has to stay for ABI reasons. I believe
> its locale data could be stripped down as well, somehow.
>
Looks like none of the Xlib l
On Mon, Mar 15, 2010 at 05:34:42PM +0100, Frans Pop wrote:
>
> This does happen. At least, with only console-setup-pc-ekmap loaded I see
> "Sun dead keys" listed for Dutch (hmmm, but no Sun keymaps for USA) and I
> also see Macintosh a lot (which is probably necessary).
This is not a bug. It is
On Saturday 27 February 2010, Anton Zinoviev wrote:
> On Fri, Feb 26, 2010 at 10:22:54AM +0100, Frans Pop wrote:
> > Besides that it offers choices for keymaps that are not even available
This does happen. At least, with only console-setup-pc-ekmap loaded I see
"Sun dead keys" listed for Dutch (hm
Frans Pop, le Mon 15 Mar 2010 11:41:17 +0100, a écrit :
> On Saturday 27 February 2010, Anton Zinoviev wrote:
> > On Sat, Feb 27, 2010 at 09:16:40PM +0100, Julien Cristau wrote:
> > > It'd be quite weird to have a keyboard layout configured in d-i, and
> > > ignored when switching VTs to get to a c
On Saturday 27 February 2010, Anton Zinoviev wrote:
> On Sat, Feb 27, 2010 at 09:16:40PM +0100, Julien Cristau wrote:
> > It'd be quite weird to have a keyboard layout configured in d-i, and
> > ignored when switching VTs to get to a console, IMO...
>
> Unfortunately this is inevitable. 'loadkeys'
On Thursday 11 March 2010, Cyril Brulebois wrote:
> To try and make sure I wasn't missing any occurrences, I've patched
> lintian and sent the preliminary patch as a bugreport against lintian:
> #573399. Feel free to voice in, improve the patch, etc. :)
The background to this is that until Lenny '
Frans Pop (21/02/2010):
> The advantage of the first is that it will prevent that line from
> being included in the udeb's control file and thus in dpkg status
> file (which saves some memory).
>
> The fact that it's a udeb is only relevant for the build process and
> not inside the udeb itself.
On Sat, Feb 27, 2010 at 09:32:35PM +0100, Julien Cristau wrote:
>
> oh, indeed, that sucks. Although we could run loadkeys when the user
> presses enter to get a shell on vt2/3 I suppose?
If c-s-udeb is configured 'setupcon' with no parameters will take care
to run loadkeys and even to load a f
On Sat, Feb 27, 2010 at 22:25:46 +0200, Anton Zinoviev wrote:
> On Sat, Feb 27, 2010 at 09:16:40PM +0100, Julien Cristau wrote:
> > It'd be quite weird to have a keyboard layout configured in d-i, and
> > ignored when switching VTs to get to a console, IMO...
>
> Unfortunately this is inevitable.
On Sat, Feb 27, 2010 at 09:16:40PM +0100, Julien Cristau wrote:
> It'd be quite weird to have a keyboard layout configured in d-i, and
> ignored when switching VTs to get to a console, IMO...
Unfortunately this is inevitable. 'loadkeys' and 'setfont' do not work
when X is active. A solution wou
On Sat, Feb 27, 2010 at 21:12:45 +0200, Anton Zinoviev wrote:
> On Fri, Feb 26, 2010 at 10:49:42AM +0100, Julien Cristau wrote:
> >
> > if xkb-data is available then these keymaps can be generated at
> > runtime by ckbcomp/setupcon (although I'm not sure ckbcomp-mini can do
> > it; Anton?).
>
>
On Fri, Feb 26, 2010 at 10:49:42AM +0100, Julien Cristau wrote:
>
> if xkb-data is available then these keymaps can be generated at
> runtime by ckbcomp/setupcon (although I'm not sure ckbcomp-mini can do
> it; Anton?).
No, ckbcomp-mini can not do this and ckbcomp would require Perl and a
new u
On Friday 26 February 2010, Cyril Brulebois wrote:
> OK. I think I'm going to pass for this round, and use no -theme-udeb
> for now. It looks like there are some default cursors available for at
> least the ordinary arrow and for the ‘I’-shaped text cursor.
Sounds sensible. That can easily be impr
Frans Pop (26/02/2010):
> I'm sorry, but I can't. I've simply looked at wht cursors I actually
> see when running an install.
>
> I guess it should be possible to derive it from the code of the
> frontend or cairo or gtk or whatever is responsible for changing the
> cursor. But the window resizin
Frans Pop writes:
> On Wednesday 24 February 2010, Ferenc Wagner wrote:
>
>> While I'm in Syslinux mood: version 3.85 has a full-featured gPXELinux
>> component, boasting HTTP support (among others). Image loading over
>> HTTP is *much* faster than over TFTP. On the other hand, it requires an
>
On Wednesday 24 February 2010, Cyril Brulebois wrote:
> Could you please establish a normative list of the cursors you're
> using or going to use (which would then include the “wait” one), so
> that we only keep those?
I'm sorry, but I can't. I've simply looked at wht cursors I actually see
when
On Wednesday 24 February 2010, Ferenc Wagner wrote:
> While I'm in Syslinux mood: version 3.85 has a full-featured gPXELinux
> component, boasting HTTP support (among others). Image loading over
> HTTP is *much* faster than over TFTP. On the other hand, it requires an
> additional HTTP server, an
On Wednesday 24 February 2010, Cyril Brulebois wrote:
> Frans Pop (19/02/2010):
> > 1) every MB pushes other packages off the 1st full CD and DVD.
>
> While I can understand the ultimate goal, I'm not sure saving a few kB
> here and there is going to make such a difference in the end. (As you
> no
On Fri, Feb 26, 2010 at 10:22:54 +0100, Frans Pop wrote:
> I was somewhat surprised to see x11-xkb-utils-udeb and especially the very
> large xkb-data-udeb. Isn't there an overlap with the console-setup udebs?
> If there is, it would be good to get rid of that overlap.
>
There is. The choices o
On Monday 08 February 2010, Cyril Brulebois wrote:
> - Maybe think of switching over from kbd-chooser to console-setup. It
>should work (almost) out of the box for X11 (that was at least OK
>with my very first image, before cleaning up my patches), while
>kbd-chooser would mean patchin
Frans Pop writes:
> Why image size is important
> [...]
> 2) makes loading the image for netboot installs slower
While I'm in Syslinux mood: version 3.85 has a full-featured gPXELinux
component, boasting HTTP support (among others). Image loading over
HTTP is *much* faster than over TFTP. On t
Frans Pop (21/02/2010):
> I forgot something here.
>
> Please use
>XC-Package-Type: udeb
> instead of
>Package-Type: udeb
> for all new udebs.
ACK. Will do.
Mraw,
KiBi.
signature.asc
Description: Digital signature
Frans Pop (19/02/2010):
> I'm saving the switch to console-setup for a last separate reply.
(It's late^Wearly but) Did I miss it?
> > libx11
>
> Is /usr/share/X11/locale needed? If it is, could some of its
> contents be excluded because they are not relevant for G-I?
As replied by Julien, so
Frans Pop (19/02/2010):
> Why image size is important
> ---
> 1) every MB pushes other packages off the 1st full CD and DVD This
> may not seem like a major issue, but it's something that causes the
> debian-cd team headaches. The space on especially the first CD is
> at a
Hello Frans,
On Sun, Feb 21, 2010 at 9:35 AM, Frans Pop wrote:
> On Friday 19 February 2010, Frans Pop wrote:
> > In this mail some more specific issues I noticed while
> > testing/reviewing.
>
> I forgot something here.
>
> Please use
> XC-Package-Type: udeb
> instead of
> Package-Type: ude
On Friday 19 February 2010, Frans Pop wrote:
> In this mail some more specific issues I noticed while
> testing/reviewing.
I forgot something here.
Please use
XC-Package-Type: udeb
instead of
Package-Type: udeb
for all new udebs.
The advantage of the first is that it will prevent that line
On Fri, Feb 19, 2010 at 11:52:42 +0100, Samuel Thibault wrote:
> Julien Cristau, le Fri 19 Feb 2010 11:38:22 +0100, a écrit :
> > On Fri, Feb 19, 2010 at 11:27:50 +0100, Samuel Thibault wrote:
> > > Julien Cristau, le Fri 19 Feb 2010 11:16:43 +0100, a écrit :
> > > > All X needs is an SHA1 impleme
Julien Cristau, le Fri 19 Feb 2010 11:38:22 +0100, a écrit :
> On Fri, Feb 19, 2010 at 11:27:50 +0100, Samuel Thibault wrote:
> > Julien Cristau, le Fri 19 Feb 2010 11:16:43 +0100, a écrit :
> > > All X needs is an SHA1 implementation, which is likely a very small part
> > > of libgcrypt.
> >
> >
On Fri, Feb 19, 2010 at 11:27:50 +0100, Samuel Thibault wrote:
> Julien Cristau, le Fri 19 Feb 2010 11:16:43 +0100, a écrit :
> > All X needs is an SHA1 implementation, which is likely a very small part
> > of libgcrypt.
>
> Mmm, why does it need it? (I guess the whole authentication mechanisms
On Fri, Feb 19, 2010 at 11:16:43AM +0100, Julien Cristau wrote:
> All X needs is an SHA1 implementation, which is likely a very small part
> of libgcrypt.
Hmm, it is used by the Render extension, how important is this
extension? gtk seems to use it, but I can run all the basic gnome
programs while
On Fri, Feb 19, 2010 at 07:47:29 +0100, Frans Pop wrote:
> In this mail some more specific issues I noticed while testing/reviewing.
>
> I'm saving the switch to console-setup for a last separate reply.
>
> On Monday 08 February 2010, Cyril Brulebois wrote:
> > libx11
>
> Is /usr/share/X11/lo
Julien Cristau, le Fri 19 Feb 2010 11:16:43 +0100, a écrit :
> All X needs is an SHA1 implementation, which is likely a very small part
> of libgcrypt.
Mmm, why does it need it? (I guess the whole authentication mechanisms
could just be disabled).
Samuel
--
To UNSUBSCRIBE, email to debian-boo
Hi Frans,
thanks for your feedback on this.
On Fri, Feb 19, 2010 at 07:46:37 +0100, Frans Pop wrote:
> How image size could be reduced
> ---
> In general: many small savings can add up to a significant reduction.
>
> * try to avoid some dependencies altogether
> Two
On Fri, Feb 19, 2010 at 07:44:34 +0100, Frans Pop wrote:
> On Monday 08 February 2010, Cyril Brulebois wrote:
> > - Even if someone steps in to fix some bugs (I heard somebody has
> > started to publish some patches recently), it might lead to a
>
> Is that true? Are there patches and do they
In this mail some more specific issues I noticed while testing/reviewing.
I'm saving the switch to console-setup for a last separate reply.
On Monday 08 February 2010, Cyril Brulebois wrote:
> libx11
Is /usr/share/X11/locale needed? If it is, could some of its contents be
excluded because the
On Monday 08 February 2010, Cyril Brulebois wrote:
> If you were to build a netboot-gtk image using DirectFB, and to
> compare with one using X11, one gets:
> DirectFB: 9191 extents written (17 MB)
> X11 : 11239 extents written (21 MB)
Let's first look at image size in a bit more detail.
On Monday 08 February 2010, Cyril Brulebois wrote:
> this is a summary of the needed steps to switch from DirectFB to X11.
> After some time hacking in the wild, I then looked into providing
> (possibly) clean patches for each package, which I've just finished.
Sorry for the delay in replying. Thi
Hi Frans its been a long time !
First I can understand the problems with using Gtk/DirectFB.
It seems that the DirectFB community has not rallied to support it.
Which is sad. One of the biggest reasons I don't do a lot of work any more
is the single maintainer problem. If the DirectFB community is
On Thursday 18 February 2010, Samuel Thibault wrote:
> Could you discuss directly on the bugzilla, maybe? Also, in my
> understanding it would be better to use the us layout by default as it
> seems to be the majority in NL?
Yes, the US layout definitely needs to be the default. That's why for
c
On Thursday 18 February 2010, Anton Zinoviev wrote:
> This is how I would fix the nl keyboard:
[...]
> I chose a variant without deadkeys in levels 1 and 2 (I suppose this is
> what the people want). AltGr combinations can be used to produce
> deadkeys.
Yes, that looks correct although I'd need t
Anton Zinoviev, le Thu 18 Feb 2010 18:13:47 +0200, a écrit :
> On Thu, Feb 18, 2010 at 02:26:59PM +0100, Samuel Thibault wrote:
> > Anton Zinoviev, le Tue 16 Feb 2010 23:58:06 +0200, a écrit :
> > >
> > > The only remaining problem that has to be fixed is in xkb-data - for
> > > many layouts an U
On Thu, Feb 18, 2010 at 02:26:59PM +0100, Samuel Thibault wrote:
> Anton Zinoviev, le Tue 16 Feb 2010 23:58:06 +0200, a écrit :
> >
> > The only remaining problem that has to be fixed is in xkb-data - for
> > many layouts an US variant must be added so the users get it as option
> > in the first
Anton Zinoviev, le Tue 16 Feb 2010 23:58:06 +0200, a écrit :
> On Thu, Feb 11, 2010 at 07:03:45AM +0100, Christian PERRIER wrote:
> >
> > There are general considerations about the user interface that were
> > collected by Frans and reported to the archive. The concerns were also
> > about size is
Frans Pop (13/02/2010):
> I have now built it myself and done some comparisons and mostly
> collected my thoughts. I'll be out most of today, but will write
> things up tomorrow.
Ping?
It’d be nice to know what needs work. :)
Mraw,
KiBi.
signature.asc
Description: Digital signature
On Thu, Feb 11, 2010 at 07:03:45AM +0100, Christian PERRIER wrote:
>
> There are general considerations about the user interface that were
> collected by Frans and reported to the archive. The concerns were also
> about size issues (because of l10n as there are a huge amount of
> strings that are
On Sunday 14 February 2010, Frans Pop wrote:
> Thanks a lot for looking into this. Enabling library reduction will
> reduce the image size by ~0.5GB.
Eh, 0.5MB rather.
--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@l
On Sun, Feb 14, 2010 at 11:11:39PM +0100, Frans Pop wrote:
> I think just excluding them in the file list is acceptable, especially as a
> temporary solution.
This should work. At least as long the modules don't need further parts
of the libc.
As later solution I think about forcing all symbols
On Sunday 14 February 2010, Cyril Brulebois wrote:
> I've tried another approach, see attached patch.
I think just excluding them in the file list is acceptable, especially as a
temporary solution.
But IMHO the patch is more complicated than it needs to be.
I don't think it needs to be optional
Julien Cristau (12/02/2010):
> It looks like the bug here is simply that
> console-setup/debian/console-setup-udeb.base-installer was not
> updated when /etc/default/keyboard was split out from
> /etc/default/console-setup so the keyboard layout doesn't get copied
> to the installed system anymore
Cyril Brulebois (14/02/2010):
> +# Exclude X modules from the mklibs call:
> +# MKLIBS_FIND_FILTER = ! -wholename '*/usr/lib/xorg/modules/*'
Looks like I forgot to uncomment that line once I was done checking
that defaulting to “-true” when no such variable is defined didn't
lead to any (addition
Cyril Brulebois (08/02/2010):
> To build my test image, I had to patch slightly config/amd64.cfg
> (partial revert of r61645) and pkg-lists/gtk-common, to enable the
> netboot-gtk again, and to add the needed X packages. Since there are
> some modules, mklibs can't really figure out where some sym
On Wednesday 10 February 2010, Frans Pop wrote:
> On Monday 08 February 2010, Frans Pop wrote:
> > Thanks a lot for this Cyril. I'll have a close look tomorrow and reply
> > in detail then.
>
> Sorry for the delay. I have been looking at this, but haven't had time
> yet to write things up.
Another
On Thu, Feb 11, 2010 at 00:43:02 +0100, Frans Pop wrote:
> On Thursday 11 February 2010, Cyril Brulebois wrote:
> > - Trying to integrate console-setup properly this time, so that once
> >one has set it up once in d-i, preferences can be fed directly into
> >the installed system, so that
Christian PERRIER, le Thu 11 Feb 2010 07:03:45 +0100, a écrit :
> I mostly remind a long mail that was summarizing all this.
http://lists.debian.org/debian-boot/2009/06/msg00759.html
Samuel
--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of "unsubscribe". Troubl
Quoting Cyril Brulebois (k...@debian.org):
> I guess it's never too late to propose help? Maybe there's a list of
> open issues that an outsider like me could work on?
There are general considerations about the user interface that were
collected by Frans and reported to the archive. The concerns
Frans Pop (11/02/2010):
> Thanks. I was going to ask for that. Only i386 should do fine for
> the time being.
Alright, I'll stick with this for now then.
> > I think my next moves are going to be:
> > - Tweaking lowmem case for X (should be easy once I've figured
> >out how this and that co
(Dropping everyone but -boot@)
Frans Pop (08/02/2010):
> [1] Main UI issue was that "help" dialog are not sized correctly. To
> see that, install in expert mode and click the help button on one of
> the locale selection screens.
The attach patch against cdebconf-gtk seems to fix this issue.
Mra
On Thursday 11 February 2010, Cyril Brulebois wrote:
> You can find the pool at:
> http://people.debian.org/~kibi/udebs-v1/
Thanks. I was going to ask for that. Only i386 should do fine for the time
being.
> I think my next moves are going to be:
> - Tweaking lowmem case for X (should be easy
Frans Pop (08/02/2010):
> I have given your image (it used console-setup) a quick try earlier
> today and was very impressed. I noticed a few minor issues [1], but
> in general it worked perfectly.
I'm glad it doesn't only work at home. :)
I've managed to resurrect an old box, and I rebuilt all
On Monday 08 February 2010, Frans Pop wrote:
> On Monday 08 February 2010, Cyril Brulebois wrote:
> > this is a summary of the needed steps to switch from DirectFB to X11.
> > After some time hacking in the wild, I then looked into providing
> > (possibly) clean patches for each package, which I've
On Monday 08 February 2010, Cyril Brulebois wrote:
> this is a summary of the needed steps to switch from DirectFB to X11.
> After some time hacking in the wild, I then looked into providing
> (possibly) clean patches for each package, which I've just finished.
Thanks a lot for this Cyril. I'll ha
Hi,
this is a summary of the needed steps to switch from DirectFB to X11.
After some time hacking in the wild, I then looked into providing
(possibly) clean patches for each package, which I've just finished.
I'd like to once again thank Josselin Mouette for the initial idea,
Julien Cristau for h
64 matches
Mail list logo
