Re: Searching senders of a method

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

Re: Searching senders of a method

Bert Freudenberg
(moving to squeak-dev)

On 03.03.2012, at 17:39, Levente Uzonyi wrote:

> On Sat, 3 Mar 2012, Bert Freudenberg wrote:
>
>> On 03.03.2012, at 14:07, deejay220989 wrote:
>>
>>> I'm using 4.2..now downloading 4.3 and will try to download 4.4 alpha..will
>>> let you know later.
>>
>>
>> It looks like the behavior was changed in 4.3. It now excludes subclasses. Not sure why. Levente?
>
> I guess it's the lack of tests and me not being careful enough. I'll check it out soon. Should we keep the old behavior?

Both seem to be reasonable. If in doubt I'd match Pharo's behavior. No need to differ.

- Bert -



Reply | Threaded
Open this post in threaded view
|

Re: Searching senders of a method

Eliot Miranda-2


On Sat, Mar 3, 2012 at 11:26 AM, Bert Freudenberg <[hidden email]> wrote:
(moving to squeak-dev)

On 03.03.2012, at 17:39, Levente Uzonyi wrote:

> On Sat, 3 Mar 2012, Bert Freudenberg wrote:
>
>> On 03.03.2012, at 14:07, deejay220989 wrote:
>>
>>> I'm using 4.2..now downloading 4.3 and will try to download 4.4 alpha..will
>>> let you know later.
>>
>>
>> It looks like the behavior was changed in 4.3. It now excludes subclasses. Not sure why. Levente?
>
> I guess it's the lack of tests and me not being careful enough. I'll check it out soon. Should we keep the old behavior?

What behavior?  Which method(s) are we talking about?
 

Both seem to be reasonable. If in doubt I'd match Pharo's behavior. No need to differ.

- Bert -






--
best,
Eliot



Reply | Threaded
Open this post in threaded view
|

Re: Searching senders of a method

Bert Freudenberg
On 03.03.2012, at 20:35, Eliot Miranda wrote:

> On Sat, Mar 3, 2012 at 11:26 AM, Bert Freudenberg <[hidden email]> wrote:
> (moving to squeak-dev)
>
> On 03.03.2012, at 17:39, Levente Uzonyi wrote:
>
> > On Sat, 3 Mar 2012, Bert Freudenberg wrote:
> >
> >> It looks like the behavior was changed in 4.3. It now excludes subclasses. Not sure why. Levente?
> >
> > I guess it's the lack of tests and me not being careful enough. I'll check it out soon. Should we keep the old behavior?
>
> What behavior?  Which method(s) are we talking about?

Sorry, snipped too much from the conversation on the newbies list. The method in question is

        Behavior>>allCallsOn:

which used to traverse subclasses, but doesn't anymore. Also see forwarded message below.

- Bert -

Begin forwarded message:

> From: Bert Freudenberg <[hidden email]>
> Subject: Re: [Newbies] Re: Searching senders of a method
> Date: 3. März 2012 14:17:32 MEZ
> To: "A friendly place to get answers to even the most basic questions about Squeak." <[hidden email]>
>
> On 03.03.2012, at 14:07, deejay220989 wrote:
>
>> I'm using 4.2..now downloading 4.3 and will try to download 4.4 alpha..will
>> let you know later.
>
>
> It looks like the behavior was changed in 4.3. It now excludes subclasses. Not sure why. Levente?
>
> To get the "old" behavior you now would need to do
>
> SystemNavigation default allCallsOn: #subclassResponsibility localTo: ProtoObject
>
> The size difference comes from duplicates - if you convert the result of this to a Set and compare sizes, they do match.
>
> (SystemNavigation default allCallsOn: #subclassResponsibility localTo: ProtoObject) size
> ==> 391
>
> (SystemNavigation default allCallsOn: #subclassResponsibility localTo: ProtoObject) asSet size
> ==> 358
>
> This may be considered a bug, I guess.
>
> - Bert -
>