Login  Register

Re: [ANN] More on Keymappings

Posted by Tudor Girba on Mar 13, 2011; 9:40pm
URL: https://forum.world.st/ANN-More-on-Keymappings-tp3325569p3352618.html

Excellent initiative Camillo!

Regarding multiple packages: Having multiple packages limits the conflicts, and saving them individually (for now) it's a small price to pay.

Cheers,
Doru


On 13 Mar 2011, at 22:28, Camillo Bruni wrote:

> furthermore, lets use a single repos/package (or whatever this is called in MC).
> I do not like to commit 3 times while refactoring.
> later on we can still split it up so people can actually decide on what to load.
>
> camillo
>
> On 2011-03-13, at 22:20, Camillo Bruni wrote:
>
>> I can push my changes. but I don't think we should rely too much on your old code.
>>
>> Im trying to keep the structure of the classes, that was already very nice IMO.
>> I manly adress the following issues:
>>
>> - use of arrays as result (dedicatet results object)
>> - string to match the shortcuts with the incoming keyboard event (dropped all of that and started to work on tests to use the shortcuts directly)
>> - weird event matching directly on morph (simplified and using a recursive function call now)
>> - horrible unreadable variable names (wherever I started I tried to put long names to make the code readable)
>>
>> I suggest we can work together on the new code base, since the interface will stay fairly compatible.
>>
>> camillo
>>
>> On 2011-03-13, at 22:08, Guillermo Polito wrote:
>>
>>> Camillo, I was fixing some tests and going to refactor some ugly parts of
>>> the package.
>>>
>>> Is there a way to join forces so we don't step into the other work?
>>>
>>> Guille
>>>
>>> On Sun, Mar 13, 2011 at 6:04 PM, Camillo Bruni <[hidden email]>wrote:
>>>
>>>> I started on the last Lille sprint a complete rewrite of the Keymapping
>>>> package. As of now it is not yet functional but the growing test-coverage
>>>> should help to solve this issue.
>>>>
>>>> m(^_-)m
>>>> camillo
>>>>
>>>> On 2011-03-03, at 15:16, Camillo Bruni wrote:
>>>>
>>>>> Right,
>>>>>
>>>>> the stable has a preconditio which limits it to pharo 1.2.
>>>>> Furthermore the initialization code seems to be incompatible as it uses
>>>> to:do: on Character which is AFAIK not implemented in the core image Pharo
>>>> 1.3. Hence apply the following changes:
>>>>>
>>>>> KMKeyEvent class >> initializeControlSequences
>>>>>
>>>>> ...
>>>>> $a asciiValue to: $z asciiValue do: [:each |
>>>>>    d add: each asCharacter -> (each - $a asciiValue + 1)].
>>>>> ...
>>>>>
>>>>> then it should work.
>>>>>
>>>>> m(^_-)m
>>>>> camillo
>>>>>
>>>>>
>>>>> On 2011-03-03, at 09:25, Tudor Girba wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> I am very interested to get Keymapping integrated into Glamour. Could
>>>> someone help me to load it?
>>>>>>
>>>>>> I tried:
>>>>>> - in Pharo 1.2:
>>>>>> Gofer it squeaksource: 'Keymapping'; package:
>>>> 'ConfigurationOfKeymapping'; load.
>>>>>> (ConfigurationOfKeymapping project version: #stable) load
>>>>>>
>>>>>> - in Pharo 1.3:
>>>>>> Gofer it squeaksource: 'Keymapping'; package:
>>>> 'ConfigurationOfKeymapping'; load.
>>>>>> (ConfigurationOfKeymapping project version: '1.5') load
>>>>>>
>>>>>>
>>>>>> Cheers,
>>>>>> Doru
>>>>>>
>>>>>>
>>>>>> On 27 Feb 2011, at 09:59, Tudor Girba wrote:
>>>>>>
>>>>>>> I also I cannot load Keymapping 1.5 in Pharo 1.2. I get DNU for
>>>> Character>>+. This is due to an initialization in KMKeyEvent (see the
>>>> attached debug log).
>>>>>>>
>>>>>>> I did the followings:
>>>>>>> Gofer it
>>>>>>> squeaksource: 'Keymapping';
>>>>>>> package: 'ConfigurationOfKeymapping';
>>>>>>> load.
>>>>>>>
>>>>>>> (ConfigurationOfKeymapping project version: #stable) load
>>>>>>>
>>>>>>> Am I missing something, or is this version not supposed to work in
>>>> Pharo 1.2?
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Doru
>>>>>>>
>>>>>>> <PharoDebug.log>
>>>>>>>
>>>>>>>
>>>>>>> On 26 Feb 2011, at 21:08, Francisco Ortiz Peñaloza wrote:
>>>>>>>
>>>>>>>> You're telling me that if i do a clean installation of 1.5 it would
>>>> work?
>>>>>>>>
>>>>>>>> Thanks in advance,
>>>>>>>> Francisco
>>>>>>>>
>>>>>>>> On Sat, Feb 26, 2011 at 3:50 PM, Guillermo Polito
>>>>>>>> <[hidden email]> wrote:
>>>>>>>>> Mmm, If you had 1.4 and updated to 1.5, you will have some problems
>>>> because
>>>>>>>>> I did some refactorings on that... :/.
>>>>>>>>>
>>>>>>>>> On Sat, Feb 26, 2011 at 10:12 AM, Francisco Ortiz Peñaloza
>>>>>>>>> <[hidden email]> wrote:
>>>>>>>>>>
>>>>>>>>>> Guille i was using 1.4 and worked excellent, just tried 1.5 and on
>>>>>>>>>> every stroke i made i've got a DNU on #realtarget
>>>>>>>>>>
>>>>>>>>>> Installed on last PharoCore 1.2, should i try it on 1.3?
>>>>>>>>>>
>>>>>>>>>> Great work,
>>>>>>>>>> Fran
>>>>>>>>>>
>>>>>>>>>> On Sat, Feb 26, 2011 at 5:29 AM, laurent laffont
>>>>>>>>>> <[hidden email]> wrote:
>>>>>>>>>>>
>>>>>>>>>>> On Sat, Feb 26, 2011 at 5:42 AM, Guillermo Polito
>>>>>>>>>>> <[hidden email]> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> What do we have now?
>>>>>>>>>>>>
>>>>>>>>>>>> - Can provide settings for a set of morphs
>>>>>>>>>>>>
>>>>>>>>>>>> - Can provide settings for a TextEditors (Smalltalk editor and
>>>> related)
>>>>>>>>>>>>
>>>>>>>>>>>> - Settings integration
>>>>>>>>>>>> I added some methods to the Settings Tree Builder in order to
>>>> avoid
>>>>>>>>>>>> references from the users code.
>>>>>>>>>>>>
>>>>>>>>>>>> - I ran Slint over it and cleaned it a lot more :). ( And learnt
>>>> that
>>>>>>>>>>>> Slint is there :P )
>>>>>>>>>>>>
>>>>>>>>>>>> More info in here:  http://guilleel3.blogspot.com/
>>>>>>>>>>>
>>>>>>>>>>> A new blog, cool !
>>>>>>>>>>> Can I have Emacs-like keybinding in code editor, to switch browser,
>>>> ...
>>>>>>>>>>> ?
>>>>>>>>>>> Laurent.
>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> Guille
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> www.tudorgirba.com
>>>>>>>
>>>>>>> "Every thing has its own flow."
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> www.tudorgirba.com
>>>>>>
>>>>>> "Every thing should have the right to be different."
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>
>>
>
>

--
www.tudorgirba.com

"Value is always contextual."