The Inbox: Kernel-ct.1258.mcz

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

The Inbox: Kernel-ct.1258.mcz

commits-2
A new version of Kernel was added to project The Inbox:
http://source.squeak.org/inbox/Kernel-ct.1258.mcz

==================== Summary ====================

Name: Kernel-ct.1258
Author: ct
Time: 15 August 2019, 12:11:10.7644 am
UUID: fbc76d48-44ce-1c40-91f2-2c73171c7981
Ancestors: Kernel-mt.1257

[Proposal] Implement #isEmptyOrNil on Object

Allows extended polymorphy.

=============== Diff against Kernel-mt.1257 ===============

Item was added:
+ ----- Method: Object>>isEmptyOrNil (in category 'testing') -----
+ isEmptyOrNil
+
+ ^ false!


Reply | Threaded
Open this post in threaded view
|

Re: The Inbox: Kernel-ct.1258.mcz

Jakob Reschke-2
When would you use it?

Two scenarios in which this might get used: 1. I expect a variable to contain a collection or nil (signifying that there is nothing to be enumerated); 2. I expect "any" kind of object to arrive in that variable.

If my expectation in scenario 1 is not met due to a programming error, I would have previously noticed it early because of the ensuing MessageNotUnderstood: isEmptyOrNil. Not any more with this addition. Instead I will just get MessageNotUnderstood: do:, for example, when I try to use this non-Collection like a Collection. This might happen at a much later time if the object gets stored somewhere due to "not being empty".

In scenario 2, what do I learn about my object if this answers false? That it is not nil and not an empty collection; there is still nothing that I can safely send to this kind of object except messages understood by all objects. To send the latter, I wouldn't have needed to run this test first. If the object is in fact empty or nil, I probably wouldn't keep it or send any further messages to it either because there is nothing in it.

Hence I'm wondering when this would ever be a useful response by a non-Collection. And whether it is worth hiding or delaying some errors, as explained above.

Am Do., 15. Aug. 2019 um 00:11 Uhr schrieb <[hidden email]>:
A new version of Kernel was added to project The Inbox:
http://source.squeak.org/inbox/Kernel-ct.1258.mcz

==================== Summary ====================

Name: Kernel-ct.1258
Author: ct
Time: 15 August 2019, 12:11:10.7644 am
UUID: fbc76d48-44ce-1c40-91f2-2c73171c7981
Ancestors: Kernel-mt.1257

[Proposal] Implement #isEmptyOrNil on Object

Allows extended polymorphy.

=============== Diff against Kernel-mt.1257 ===============

Item was added:
+ ----- Method: Object>>isEmptyOrNil (in category 'testing') -----
+ isEmptyOrNil
+
+       ^ false!




Reply | Threaded
Open this post in threaded view
|

Re: The Inbox: Kernel-ct.1258.mcz

marcel.taeumel
In reply to this post by commits-2
-1 :-)

Yet, it reminds of the interface I added in Vivide:

anObject asList "=> { anObject }"
nil asList "=> #()"
someObjects asList "=> someObjects"

aDictionary asList "=> { aDictionary }"
aString asList "=> { aString }"
aText asList "=> { aText }"
aByteArray asList "=> "{ aByteArray }"
aHashedCollection asList "=> aSequenceableCollection"
aCharacterSet asList "=> aCharacterSet or aSequenceableCollection if not enumerable"

And similar for #isList. :-) Makes working with collections of domain objects much easier.

Best,
Marcel

Am 15.08.2019 00:11:23 schrieb [hidden email] <[hidden email]>:

A new version of Kernel was added to project The Inbox:
http://source.squeak.org/inbox/Kernel-ct.1258.mcz

==================== Summary ====================

Name: Kernel-ct.1258
Author: ct
Time: 15 August 2019, 12:11:10.7644 am
UUID: fbc76d48-44ce-1c40-91f2-2c73171c7981
Ancestors: Kernel-mt.1257

[Proposal] Implement #isEmptyOrNil on Object

Allows extended polymorphy.

=============== Diff against Kernel-mt.1257 ===============

Item was added:
+ ----- Method: Object>>isEmptyOrNil (in category 'testing') -----
+ isEmptyOrNil
+
+ ^ false!




Reply | Threaded
Open this post in threaded view
|

Re: The Inbox: Kernel-ct.1258.mcz

Christoph Thiede

Okay, thanks for your feedback :)


-Christoph


Von: Squeak-dev <[hidden email]> im Auftrag von Taeumel, Marcel
Gesendet: Donnerstag, 15. August 2019 10:39:18
An: gettimothy via Squeak-dev
Betreff: Re: [squeak-dev] The Inbox: Kernel-ct.1258.mcz
 
-1 :-)

Yet, it reminds of the interface I added in Vivide:

anObject asList "=> { anObject }"
nil asList "=> #()"
someObjects asList "=> someObjects"

aDictionary asList "=> { aDictionary }"
aString asList "=> { aString }"
aText asList "=> { aText }"
aByteArray asList "=> "{ aByteArray }"
aHashedCollection asList "=> aSequenceableCollection"
aCharacterSet asList "=> aCharacterSet or aSequenceableCollection if not enumerable"

And similar for #isList. :-) Makes working with collections of domain objects much easier.

Best,
Marcel

Am 15.08.2019 00:11:23 schrieb [hidden email] <[hidden email]>:

A new version of Kernel was added to project The Inbox:
http://source.squeak.org/inbox/Kernel-ct.1258.mcz

==================== Summary ====================

Name: Kernel-ct.1258
Author: ct
Time: 15 August 2019, 12:11:10.7644 am
UUID: fbc76d48-44ce-1c40-91f2-2c73171c7981
Ancestors: Kernel-mt.1257

[Proposal] Implement #isEmptyOrNil on Object

Allows extended polymorphy.

=============== Diff against Kernel-mt.1257 ===============

Item was added:
+ ----- Method: Object>>isEmptyOrNil (in category 'testing') -----
+ isEmptyOrNil
+
+ ^ false!