[PATCH] Add required files in stage2.img

[Alexander Todorov]

Adds missing libraries and Python modules.
With the new Dogtail port based on pyatspi that will need to change a 
bit but I can't test it right now.

[Joel Andres Granados]

Alexander Todorov wrote:
> Adds missing libraries and Python modules.
> With the new Dogtail port based on pyatspi that will need to change a 
> bit but I can't test it right now.
>
> Anaconda-devel-list mailing list
> [e-mail]
> https://www.redhat.com/mailman/listinfo/anaconda-devel-list

A question was raised in the #anaconda channel regarding the size of these
additions in the image.  Can you estimate?

Regards

[Alexander Todorov]

Joel Andres Granados wrote:
> Alexander Todorov wrote:
> > Adds missing libraries and Python modules.
> > With the new Dogtail port based on pyatspi that will need to change a 
> > bit but I can't test it right now.
> >
> > Anaconda-devel-list mailing list
> > [e-mail]
> > https://www.redhat.com/mailman/listinfo/anaconda-devel-list
>
> A question was raised in the #anaconda channel regarding the size of these
> additions in the image.  Can you estimate?

Calculations based on RHEL 5 were around 2MB.
I don't expect to be much different.

We can get rid of GConf2 stuff and the XML settings if we patch Dogtail 
to not check gconf key for accessibility. This is a simple patch that 
will cause one of the Dogtail functions to return always true.
That was discussed long time ago (can't find the link) but there was no 
decision.

Greetings,
Alexander.

[Chris Lumens]

> Calculations based on RHEL 5 were around 2MB.
> I don't expect to be much different.
>
> We can get rid of GConf2 stuff and the XML settings if we patch Dogtail to 
> not check gconf key for accessibility. This is a simple patch that will 
> cause one of the Dogtail functions to return always true.
> That was discussed long time ago (can't find the link) but there was no 
> decision.

Yeah it'd probably be nice to get rid of the gconf stuff if possible.
Perhaps we can go ahead and include these patches, and you can work on
getting the simple patch included upstream.

What does everyone else think?

- Chris

[David Cantrell]

On Tue, 22 Jan 2008 13:30:05 -0500
Chris Lumens  wrote:
> > Calculations based on RHEL 5 were around 2MB.
> > I don't expect to be much different.
> >
> > We can get rid of GConf2 stuff and the XML settings if we patch Dogtail to 
> > not check gconf key for accessibility. This is a simple patch that will 
> > cause one of the Dogtail functions to return always true.
> > That was discussed long time ago (can't find the link) but there was no 
> > decision.
>
> Yeah it'd probably be nice to get rid of the gconf stuff if possible.
> Perhaps we can go ahead and include these patches, and you can work on
> getting the simple patch included upstream.
>
> What does everyone else think?

I'm confused.  Is this patching dogtail to make a special version to include in stage2 or just patching dogtail to not require gconf stuff?  I am in favor of adding dogtail support, but I don't think we should make a special version included for anaconda.

[Alexander Todorov]

David Cantrell wrote:
> On Tue, 22 Jan 2008 13:30:05 -0500
> Chris Lumens  wrote:
> > > Calculations based on RHEL 5 were around 2MB.
> > > I don't expect to be much different.
> > >
> > > We can get rid of GConf2 stuff and the XML settings if we patch Dogtail to 
> > > not check gconf key for accessibility. This is a simple patch that will 
> > > cause one of the Dogtail functions to return always true.
> > > That was discussed long time ago (can't find the link) but there was no 
> > > decision.
> >
> > Yeah it'd probably be nice to get rid of the gconf stuff if possible.
> > Perhaps we can go ahead and include these patches, and you can work on
> > getting the simple patch included upstream.

I'm also confused here. Which patch gets included where?
> > What does everyone else think?
>
> I'm confused.  Is this patching dogtail to make a special version to include in stage2 or just patching dogtail to not require gconf stuff?  I am in favor of adding dogtail support, but I don't think we should make a special version included for anaconda.

Yes, that's patching Dogtail to not check the gconf settings.

[Chris Lumens]

> > > > We can get rid of GConf2 stuff and the XML settings if we patch Dogtail 
> > > > to not check gconf key for accessibility. This is a simple patch that 
> > > > will cause one of the Dogtail functions to return always true.
> > > > That was discussed long time ago (can't find the link) but there was no 
> > > > decision.
> > >
> > > Yeah it'd probably be nice to get rid of the gconf stuff if possible.
> > > Perhaps we can go ahead and include these patches, and you can work on
> > > getting the simple patch included upstream.
>
> I'm also confused here. Which patch gets included where?

Sorry for being confusing.  What I meant to say was:

- I'll merge your dogtail patches as-is.
- Let's try to get this gconf key accessibility patch merged into gconf.
- After that's done, we can deal with removing the stuff from stage2
  that we no longer require.

Sound good?

- Chris

[Alexander Todorov]

Chris Lumens wrote:
> Sorry for being confusing.  What I meant to say was:
> I'll merge your dogtail patches as-is.
> Let's try to get this gconf key accessibility patch merged into gconf.

Sorry but I still don't understand you. The gconf "patch" is just 
creating the necessary gconf key for the root user in the anaconda 
environment. I don't see how that relates to gconf itself. Gconf is the 
tool to read/write that key but it's not related to whether 
accessibility is enabled or disabled. For Dogtail we want it enabled.

Please clarify what you mean.

Thanks,
Alexander.