[Mm10s] 2020-06-08

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

[Mm10s] 2020-06-08

Esteban Lorenzano

### Last week:


   * [Idle-VM] w/Pablo. Worker problem with delays
   worker version was not interrupting the aio. We spent some time here and this was a non existing
   problem at the end (but we found and fix a bug on Spec2-Gtk).
   * [Idle-VM] problem when save&quit image with gtk backend
   fix a bug happening when application was shutting down and some events want still happen, who
   causes a callback that is alredy discarded to be executed.
   * [Idle-VM] new version of memory access
   test, bench and merge of new memory access primitive
   (lowcode didn't applied, finally)
   * [spec2] preparation of a window box to make some tests there.
   * [spec2] gtk backend now has an image-side based tree store, which allows much better
   interoperability. Next task is to migrate the adaptors: list, table and treetable (not trivial,
   but needed)
   * [TFFI] added a new type: FFIBoolean32 to handle the fact that many libraries (in my case,
   Gtk), do not use bool=8bit implementation. Now, users need to map their boolean (in my case,
   gboolean) to this type by hand. I didn't find a better way to fix this issue :(
   * [spec2] gtk backend now answers propertly to keyboard events in windows.
   * [spec2] fix a (new) bug on GtkTreeView, because we do not need to unref stores manually. This
   bug was introduced by my recent fix on GtkTreeView, which initialize method was not sending
   super.
   * [spec2] avoid the traverse of roots when releasing a GtkTreeDataStore (Gtk does that)
   * [spec2] list/table/tree migrated to use the new GtkTreeDataStore (lots of simplifications,
   specially on GtkTreeTableAdapter)
   * [NewTools-Playground] bug: doit from toolbar does not uses interactionMode
   * [Spec2-doc] Add style section (reference) to SpApplication
   * [NewTools-Playground] bug: "implementors" is not working. I couldn’t reproduce the problem.
   Maybe is fixed now?

### This week (starting 2020-06-08):


- Mostly cleanups and re-organisations on Spec2 (preparing v0.7 for Pharo9)
- Preparing v0.1 of NewTools (also, to prepare an integration to Pharo9)
- Building VM on Fedora linux
- Testing Idle-VM on Windows



Reply | Threaded
Open this post in threaded view
|

Re: [Mm10s] 2020-06-08

tesonep@gmail.com
### Last week:

- Monday Holiday
- Multiple images test runners
- [VM] Writing tests for JIT generation
- Idle VM with Esteban: Gtk
- Missing parts in the Image to have Sista + FullBlockClosure.
- Reviewing PR & Issues

### This week (starting 2020-06-08):

- Releasing VM with LibGit 1.0.0
- Multiple images test runners
- [VM] Writing tests for JIT generation
- Idle VM with Esteban: Gtk

On Mon, Jun 8, 2020 at 10:03 AM Esteban Lorenzano <[hidden email]> wrote:

>
>
> ### Last week:
>
>
>    * [Idle-VM] w/Pablo. Worker problem with delays
>    worker version was not interrupting the aio. We spent some time here and this was a non existing
>    problem at the end (but we found and fix a bug on Spec2-Gtk).
>    * [Idle-VM] problem when save&quit image with gtk backend
>    fix a bug happening when application was shutting down and some events want still happen, who
>    causes a callback that is alredy discarded to be executed.
>    * [Idle-VM] new version of memory access
>    test, bench and merge of new memory access primitive
>    (lowcode didn't applied, finally)
>    * [spec2] preparation of a window box to make some tests there.
>    * [spec2] gtk backend now has an image-side based tree store, which allows much better
>    interoperability. Next task is to migrate the adaptors: list, table and treetable (not trivial,
>    but needed)
>    * [TFFI] added a new type: FFIBoolean32 to handle the fact that many libraries (in my case,
>    Gtk), do not use bool=8bit implementation. Now, users need to map their boolean (in my case,
>    gboolean) to this type by hand. I didn't find a better way to fix this issue :(
>    * [spec2] gtk backend now answers propertly to keyboard events in windows.
>    * [spec2] fix a (new) bug on GtkTreeView, because we do not need to unref stores manually. This
>    bug was introduced by my recent fix on GtkTreeView, which initialize method was not sending
>    super.
>    * [spec2] avoid the traverse of roots when releasing a GtkTreeDataStore (Gtk does that)
>    * [spec2] list/table/tree migrated to use the new GtkTreeDataStore (lots of simplifications,
>    specially on GtkTreeTableAdapter)
>    * [NewTools-Playground] bug: doit from toolbar does not uses interactionMode
>    * [Spec2-doc] Add style section (reference) to SpApplication
>    * [NewTools-Playground] bug: "implementors" is not working. I couldn’t reproduce the problem.
>    Maybe is fixed now?
>
> ### This week (starting 2020-06-08):
>
>
> - Mostly cleanups and re-organisations on Spec2 (preparing v0.7 for Pharo9)
> - Preparing v0.1 of NewTools (also, to prepare an integration to Pharo9)
> - Building VM on Fedora linux
> - Testing Idle-VM on Windows
>
>
>


--
Pablo Tesone.
[hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Mm10s] 2020-06-08

Sean P. DeNigris
Administrator
[hidden email] wrote
> - Missing parts in the Image to have Sista + FullBlockClosure.

Yay! Thanks :)



-----
Cheers,
Sean
--
Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

Cheers,
Sean