[squeak-dev] Bug in trunk? 'select all' in Workspace, line counting issue?

Nicolas Cellier nicolas.cellier.aka.nice at gmail.com
Tue Apr 26 12:43:03 UTC 2022


Le mar. 26 avr. 2022 à 14:40, Nicolas Cellier <
nicolas.cellier.aka.nice at gmail.com> a écrit :

>
>
> Le mar. 26 avr. 2022 à 11:37, Marcel Taeumel <marcel.taeumel at hpi.de> a
> écrit :
>
>> Hi Nicolas --
>>
>> > Also, does overwriting selection followed by ALT+j or CTRL+j work for
>> you?
>> > ...
>> > Cmd+j also works there, contrary to windows...
>>
>> Please verify that the following (default) is set:
>>
>> EventSensor mapControlCharactersToPrintableCharacters: true.
>>
>> Yes, my image has nil, which should fallback to true by default.
> Forcing true as above does not change anything.
>
> On Windows, CMD shortcuts might better be pressed via the ALT key.
>> However, in the default configuration, both CTRL and ALT modify will raise
>> the #commandKeyPressed flag. So both should work.
>>
>> See the documentation in KeyboardEvent >> #checkCommandKey.
>>
>> Yes I tried both ALT+j and CRTL+j
>
>
>> Additionally, CTRL+J can have the issue of not arriving as #keyCharacter
>> $j but als Character lf. You can use the KeyboardExerciser to check.
>>
>> I get CTRL+j as keystroke
>
>
>> With the following (default) preferences on Windows, CTRL+j can be used
>> instead of ALT+j for find/replace:
>>
>> [ x ] EventSensor mapAltKeysToOptionKeys
>> [ x ] EventSensor mapControlKeysToCommandKeys
>>
>> See EventSensor class >> #startUp:.
>>
>> On all platforms, #mapControlCharactersToPrintableCharacters is enabled.
>> We expect the current programming model to use #keyCharacter based on
>> physically printed labels. In the future, we will continue to improve this
>> with the newer KeyboardEvent >> #key protocol/message, which makes the
>> abstraction of "virtual keys" accessible.
>>
>> NOTE THAT, whether it is better to expect that the user knows how to
>> enter a particular #keyCharacter or just the #key as printed on the
>> physical keyboard, is subject of discussion. So it is about "I know how to
>> enter this character" vs. "I can only enter the characters that have labels
>> printed on my keyboard". Keyboard layout matching physical layout ...etc
>> ... :-)
>>
>> The problem I observe on this Windows VM/updated image pair is not that
> ALT+j or CTRL+j does not work.
> It works IF and ONLY IF I start the replacement with a separator.
> I have no specific change in the base packages in this image,
> But I have installed some extra packages - mainly RefactoringBrowser and
> Metacello.
> Maybe I should inquire for updates of the first one.
>
>

Ah, got it, after disabling Enable OCompletion in Preferences, ALT+j and
CTRL+j work again...
Nasty interaction.

Best,
>> Marcel
>>
>> Am 25.04.2022 23:05:07 schrieb Nicolas Cellier <
>> nicolas.cellier.aka.nice at gmail.com>:
>> Ah, Shift+Cmd+arrow is now working on macos.
>> Cmd+j also works there, contrary to windows...
>>
>> Le lun. 25 avr. 2022 à 18:51, Nicolas Cellier <
>> nicolas.cellier.aka.nice at gmail.com> a écrit :
>>
>>> Also, does overwriting selection followed by ALT+j or CTRL+j work for
>>> you?
>>>
>>> On windows, both work when overwriting with backspace or some word
>>> beginning with a separator (CR, space, tab, ...) but not otherwise.
>>>
>>> Nicolas
>>>
>>> Le lun. 25 avr. 2022 à 14:20, Nicolas Cellier <
>>> nicolas.cellier.aka.nice at gmail.com> a écrit :
>>>
>>>> It did not work on macos last time I tried, maybe this combination is
>>>> captured at window manager level... I will retry this evening and report.
>>>>
>>>> Le lun. 25 avr. 2022, 14:07, Marcel Taeumel <marcel.taeumel at hpi.de> a
>>>> écrit :
>>>>
>>>>> Hi Nicholas --
>>>>>
>>>>> > what about shift+cmd+left/right to extend selection past
>>>>> previous/next word boundary?
>>>>>
>>>>> It's already this way. Or what do you mean?
>>>>>
>>>>> Best,
>>>>> Marcel
>>>>>
>>>>> Am 25.04.2022 12:53:04 schrieb Nicolas Cellier <
>>>>> nicolas.cellier.aka.nice at gmail.com>:
>>>>> Hi Marcel,
>>>>> what about shift+cmd+left/right to extend selection past previous/next
>>>>> word boundary?
>>>>>
>>>>>
>>>>> Le lun. 25 avr. 2022 à 11:20, Marcel Taeumel <marcel.taeumel at hpi.de>
>>>>> a écrit :
>>>>>
>>>>>> Hi Tim --
>>>>>>
>>>>>> Hopefully fixed via Morphic-mt.1970.
>>>>>>
>>>>>> Best,
>>>>>> Marcel
>>>>>>
>>>>>> snip...
>>>>>
>>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20220426/89b210f5/attachment.html>


More information about the Squeak-dev mailing list