Moving closer to ETS release

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|

Moving closer to ETS release

Peter Wang
Hi everyone,

I would like to do the release this Wednesday.  Please look over the
TODO list and comment if there is anything else that needs to be
updated: https://svn.enthought.com/enthought/wiki/ETS3_4_0_Roadmap

Thanks to everyone who is pitching in!  There are just a few more
items remaining.

Tonight I will merge the relative_imports Traits branch into the
trunk.  If there are any objections, please let me know!


-Peter
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

Darren Dale
On Mon, Feb 22, 2010 at 1:18 PM, Peter Wang <[hidden email]> wrote:
> Hi everyone,
>
> I would like to do the release this Wednesday.  Please look over the
> TODO list and comment if there is anything else that needs to be
> updated: https://svn.enthought.com/enthought/wiki/ETS3_4_0_Roadmap
>
> Thanks to everyone who is pitching in!  There are just a few more
> items remaining.


The dependencies in SciMath/EnthoughtBase still seem off. SciMath's
install_requires includes EnthoughtBase[traits]. EnthoughtBase[traits]
requires TraitsBackendWX. SciMath also has a traits project extra,
which depends on TraitsBackendWX. Other projects use the ui project
extra to pull in TraitsGUI and Traits['ui']. So perhaps:

for EnthoughtBase:
Index: setup_data.py
===================================================================
--- setup_data.py (revision 25283)
+++ setup_data.py (working copy)
@@ -20,15 +20,10 @@
 ENVISAGECORE = etsdep('EnvisageCore', '3.1.2')  # -- all from
logger.(plugin|agent|widget)
 ENVISAGEPLUGINS = etsdep('EnvisagePlugins', '3.1.2')  # -- all from
logger.plugin
 TRAITS = etsdep('Traits', '3.2.1')
-TRAITSBACKENDWX = etsdep('TraitsBackendWX', '3.2.1')  # -- only from
e.util.traits.editor.parameter_choice_editor.py
 TRAITSGUI = etsdep('TraitsGUI', '3.1.1')  # -- from
logger.(agent|plugin|widget)
 TRAITS_UI = etsdep('Traits[ui]', '3.2.1')

-# The following "soft dependencies" are wrapped in try..except blocks
-#APPTOOLS -- util/wx/drag_and_drop
-#SCIMATH -- util/wx/spreadsheet/unit_renderer.py

-
 # A dictionary of the setup data information.
 INFO = {
     'extras_require' : {
@@ -46,10 +41,6 @@
             TRAITSGUI,
             TRAITS_UI,
             ],
-        'traits': [
-            TRAITS,
-            TRAITSBACKENDWX,
-            ],
         'ui': [    # -- this includes util.ui.* and util.wx.* (see extras.map)
             TRAITSGUI,
             TRAITS_UI,





and for SciMath:
Index: setup_data.py
===================================================================
--- setup_data.py (revision 25283)
+++ setup_data.py (working copy)
@@ -16,12 +16,10 @@
 # Declare our ETS project dependencies.
 APPTOOLS = etsdep('AppTools', '3.3.1')
 BLOCKCANVAS = etsdep('BlockCanvas', '3.1.1')
-ENTHOUGHTBASE_TRAITS = etsdep('EnthoughtBase[traits]', '3.0.4')
+ENTHOUGHTBASE = etsdep('EnthoughtBase', '3.0.4')
 ENVISAGECORE = etsdep('EnvisageCore', '3.1.2')
 ENVISAGEPLUGINS = etsdep('EnvisagePlugins', '3.1.2')
-#ETSDEVTOOLS -- all gotcha imports enclosed in try...except blocks so
not needed
 TRAITS = etsdep('Traits', '3.2.1')
-TRAITSBACKENDWX = etsdep('TraitsBackendWX', '3.2.1')
 TRAITSGUI = etsdep('TraitsGUI', '3.1.1')
 TRAITS_UI = etsdep('Traits[ui]', '3.2.1')

@@ -32,15 +30,16 @@
         'converter':[
             BLOCKCANVAS,
             ],
-        "envisage": [
+        'envisage': [
             APPTOOLS,
             ENVISAGECORE,
             ENVISAGEPLUGINS,
             TRAITSGUI,
             TRAITS_UI,
             ],
-        'traits': [
-            TRAITSBACKENDWX,
+        'ui': [
+            TRAITSGUI,
+            TRAITS_UI,
             ],

         # All non-ets dependencies should be in this extra to ensure users can
@@ -52,7 +51,7 @@
             ],
         },
     'install_requires': [
-        ENTHOUGHTBASE_TRAITS,
+        ENTHOUGHTBASE,
         TRAITS,
         ],
     'name': 'SciMath',



Cheers,
Darren
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

Gael Varoquaux
In reply to this post by Peter Wang
On Mon, Feb 22, 2010 at 12:18:58PM -0600, Peter Wang wrote:
> I would like to do the release this Wednesday.  Please look over the
> TODO list and comment if there is anything else that needs to be
> updated: https://svn.enthought.com/enthought/wiki/ETS3_4_0_Roadmap

wrt Mayavi, the broadcasting of mlab arguments won't be done. Too bad!
It's not the end of the world, it'll be done later.

While testing, I found out that the new color editor doesn't work with
wx2.6. I am not sure if we are still trying to support wx2.6, but I
believe that we should drop that claim.

Gaël
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

Peter Wang
On Mon, Feb 22, 2010 at 4:57 PM, Gael Varoquaux
<[hidden email]> wrote:
> While testing, I found out that the new color editor doesn't work with
> wx2.6. I am not sure if we are still trying to support wx2.6, but I
> believe that we should drop that claim.

Thanks, that's a good catch.  But since wx 2.8 has been out for three
years now, I think we're safe to break backwards compatibility with
2.6.

-Peter
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

Gael Varoquaux
In reply to this post by Gael Varoquaux
On Mon, Feb 22, 2010 at 11:57:22PM +0100, Gael Varoquaux wrote:
> While testing, I found out that the new color editor doesn't work with
> wx2.6. I am not sure if we are still trying to support wx2.6, but I
> believe that we should drop that claim.

Actually, we should try and make that transition a bit smoother. Right
now, if wx2.6 is installed, it gets loaded by default, even if 2.8 is
also installed. So with very simple code such as:
http://code.enthought.com/projects/mayavi/docs/development/html/mayavi/auto/example_mlab_interactive_dialog.html
I get the bugs of 2.6 for no good reason[*].

Is there anychance that we could check in the following code somewhere in
the TraitsBackendWx early import path:

try:
    # Try forcing the use of wx 2.8
    from enthought.etsconfig.api import ETSConfig
    if ETSConfig.toolkit in ('wx', ''):
        import wxversion
        if wxversion.checkInstalled('2.8'):
            try:
                wxversion.select('2.8')
            except wxversion.VersionError:
                """ Too late, wxPython has already been imported. """
except ImportError:
    """ wxversion not installed """

I am not sure exactly where this should be check in: it needs to be check
in before the first import of wx when someone is about to use TraitsUI or
pyface with wx.

Comments?

Gaël

[*] Removing wx2.6 on my system (latest Ubuntu) removes SPE, amongst
other things, so there are good reasons for which people may have wx2.6
installed.
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

Gael Varoquaux
Another remark, that's coming from some pre-release testing: under
latest Ubuntu, with Qt, the tabs of the Mayavi2 application look pretty
bad.

Not the end of the world, thought.

Gaël
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

bryce hendrix-2
I concur. A recent change to improve the appearance on OS X is the
cause, we should fix this before the release.

Bryce

Gael Varoquaux wrote:

> Another remark, that's coming from some pre-release testing: under
> latest Ubuntu, with Qt, the tabs of the Mayavi2 application look pretty
> bad.
>
> Not the end of the world, thought.
>
> Gaël
> _______________________________________________
> Enthought-Dev mailing list
> [hidden email]
> https://mail.enthought.com/mailman/listinfo/enthought-dev
>  

_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

bryce hendrix-2
I opened a ticket with notes about the 2 changesets that affect the tabs.

bryce

bryce hendrix wrote:

> I concur. A recent change to improve the appearance on OS X is the
> cause, we should fix this before the release.
>
> Bryce
>
> Gael Varoquaux wrote:
>  
>> Another remark, that's coming from some pre-release testing: under
>> latest Ubuntu, with Qt, the tabs of the Mayavi2 application look pretty
>> bad.
>>
>> Not the end of the world, thought.
>>
>> Gaël
>> _______________________________________________
>> Enthought-Dev mailing list
>> [hidden email]
>> https://mail.enthought.com/mailman/listinfo/enthought-dev
>>  
>>    
>
> _______________________________________________
> Enthought-Dev mailing list
> [hidden email]
> https://mail.enthought.com/mailman/listinfo/enthought-dev
>  

_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

fred-2
In reply to this post by Peter Wang
Peter Wang a écrit :

> Hi everyone,
>
> I would like to do the release this Wednesday.  Please look over the
> TODO list and comment if there is anything else that needs to be
> updated: https://svn.enthought.com/enthought/wiki/ETS3_4_0_Roadmap
>
> Thanks to everyone who is pitching in!  There are just a few more
> items remaining.
>
> Tonight I will merge the relative_imports Traits branch into the
> trunk.  If there are any objections, please let me know!

Hi Peter,

What do you mean by that:

range plot/range2D off-by-one discrepancy (reported by Fred, Warren)

TIA.


Cheers,

--
Fred
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

Peter Wang
On Tue, Feb 23, 2010 at 1:21 AM, fred <[hidden email]> wrote:
> Hi Peter,
>
> What do you mean by that:
> range plot/range2D off-by-one discrepancy (reported by Fred, Warren)

This is the old, old, old bug you reported where the interpretation of
xbounds and ybounds of an imageplot differ in two parts of the code,
so that you have to pass in (N+1), (M+1) for an image of size (NxM).

-Peter
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev
Reply | Threaded
Open this post in threaded view
|

Re: Moving closer to ETS release

fred-2
Peter Wang a écrit :
> On Tue, Feb 23, 2010 at 1:21 AM, fred <[hidden email]> wrote:
>> Hi Peter,
>>
>> What do you mean by that:
>> range plot/range2D off-by-one discrepancy (reported by Fred, Warren)
>
> This is the old, old, old bug you reported where the interpretation of
> xbounds and ybounds of an imageplot differ in two parts of the code,
> so that you have to pass in (N+1), (M+1) for an image of size (NxM).
Well... glad to hear that, Peter! :-)
I'm daily in trouble with this issue :-/


Cheers,

--
Fred
_______________________________________________
Enthought-Dev mailing list
[hidden email]
https://mail.enthought.com/mailman/listinfo/enthought-dev