[squeak-dev] The Trunk: CollectionsTests-ct.368.mcz

Marcel Taeumel marcel.taeumel at hpi.de
Tue Jan 11 09:52:50 UTC 2022


Hi Levente, hi Christoph --

Hmm... from a key-input handling perspective, I would prefer #return and #enter over #cr or #lf etc. Hmm.... there can only be one printString/storeString. If we would change #constantNames again ... I would have to duplicate it to make the virtual key stuff look as expected again.

About this test. I think it makes no sense to compare strings but the resulting objects. The test should test the storeOn:/evaluate: dualism. Older storeStrings can easily be tested as well.

self assert: Character cr equals: (Compiler evaluate: 'Character cr').

self assert: Character cr equals: (Compiler evaluate: 'Character return').

self assert: Character return equals: (Compiler evaluate: 'Character cr').

self assert: Character return equals: (Compiler evaluate: 'Character return').


The current results of #storeString should not matter.

Best,
Marcel
Am 10.01.2022 23:11:10 schrieb Thiede, Christoph <christoph.thiede at student.hpi.uni-potsdam.de>:
Hi Levente,

good question, I had assumed Marcel had also thought this through when renaming the constant. We cannot have two printString representations for the same character, obviously. In his original patch, Marcel wrote:

> One more thing: Also changes the printString of ASCII-13 to #return because it seems more readable than #cr.
> 
> Note that I pushed readability more into the direction of recognizing physical keys, not standard ASCII labels. I hope that's fine... In that sense, the printString for #lf might be also changed to #linefeed, which is already there.
Von: Squeak-dev <squeak-dev-bounces at lists.squeakfoundation.org> im Auftrag von Levente Uzonyi <leves at caesar.elte.hu>
Gesendet: Montag, 10. Januar 2022 22:21:36
An: squeak-dev at lists.squeakfoundation.org
Cc: packages at lists.squeakfoundation.org
Betreff: Re: [squeak-dev] The Trunk: CollectionsTests-ct.368.mcz
 
Hi Christoph,

On Mon, 10 Jan 2022, commits at source.squeak.org wrote:

> Christoph Thiede uploaded a new version of CollectionsTests to project The Trunk:
> http://source.squeak.org/trunk/CollectionsTests-ct.368.mcz [http://source.squeak.org/trunk/CollectionsTests-ct.368.mcz]
>
> ==================== Summary ====================
>
> Name: CollectionsTests-ct.368
> Author: ct
> Time: 10 January 2022, 8:03:34.822597 pm
> UUID: a071bc20-7ee8-c64b-a012-4d0b77d7317d
> Ancestors: CollectionsTests-ct.367
>
> Fixes failing collection tests.

Yes, the test passes now, but it marks a potentially breaking change
because it affects #printString and #storeString.
Do we really want it to be printed as Charcter return instead of Character cr?
If yes, why return?
Why not carriageReturn?


Levente

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20220111/9655b0be/attachment.html>


More information about the Squeak-dev mailing list