last update 299 might be broken

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

last update 299 might be broken

Nicolas Cellier
Hi,
it seems that last update is loading Graphics first and thus removing CornerRounder before loading Morphic. Alas, the old Morphic might still be using CornerRounder.

In some old images still using some rounded things it happened that the update process failed with unresponsive image (recursive error or something...).

I disabled all the rounded* Preferences, but it does not seem to be enough (I think the progress dialog itself must be using rounded corners or some thing like that...).

Can we fix that?
(for example just delay inclusion of latest Graphics - that is exclude it from the latest update)


Reply | Threaded
Open this post in threaded view
|

Re: last update 299 might be broken

Levente Uzonyi-2
On Sat, 28 Feb 2015, Nicolas Cellier wrote:

> Hi,
> it seems that last update is loading Graphics first and thus removing CornerRounder before loading Morphic. Alas, the old Morphic might still be using CornerRounder.
>
> In some old images still using some rounded things it happened that the update process failed with unresponsive image (recursive error or something...).
>
> I disabled all the rounded* Preferences, but it does not seem to be enough (I think the progress dialog itself must be using rounded corners or some thing like that...).
>
> Can we fix that?
> (for example just delay inclusion of latest Graphics - that is exclude it from the latest update)

I've changed the last update map to load Graphics-kfr.305 instead of
Graphics-mt.307.

Levente

>
>

Reply | Threaded
Open this post in threaded view
|

Re: last update 299 might be broken

marcel.taeumel (old)
We could just re-introduce the CornerRounder until the dependencies are fixed. I thought that committing Morphic before Graphics ensures some kind of update order.

Why does this happen?

Best,
Marcel
Reply | Threaded
Open this post in threaded view
|

Re: last update 299 might be broken

marcel.taeumel (old)
In reply to this post by Nicolas Cellier
Yes, the SystemProgressMorph is a Singleton. Changing the preference for roundedMenuCorners does not affect it until calling "SystemProgressMorph reset"

Best,
Marcel
Reply | Threaded
Open this post in threaded view
|

Re: last update 299 might be broken

Nicolas Cellier
Levente, thanks it now works like a charm.
Marcel, nothing more needs to be done, it's now updating smoothly.
Well, of course, I mean a lot yet remain to be done to Morphic, but you had a great start ;)

2015-03-01 13:44 GMT+01:00 Marcel Taeumel <[hidden email]>:
Yes, the SystemProgressMorph is a Singleton. Changing the preference for
roundedMenuCorners does not affect it until calling "SystemProgressMorph
reset"

Best,
Marcel



--
View this message in context: http://forum.world.st/last-update-299-might-be-broken-tp4808593p4808666.html
Sent from the Squeak - Dev mailing list archive at Nabble.com.




Reply | Threaded
Open this post in threaded view
|

Re: last update 299 might be broken

Levente Uzonyi-2
In reply to this post by marcel.taeumel (old)
On Sun, 1 Mar 2015, Marcel Taeumel wrote:

> We could just re-introduce the CornerRounder until the dependencies are
> fixed. I thought that committing Morphic before Graphics ensures some kind
> of update order.
>
> Why does this happen?

The load order is defined by the update map (aka Monticello Confiruation,
aka .mcm file). Commit order has no effect on the load order.

Levente

>
> Best,
> Marcel
>
>
>
> --
> View this message in context: http://forum.world.st/last-update-299-might-be-broken-tp4808593p4808665.html
> Sent from the Squeak - Dev mailing list archive at Nabble.com.
>
>