Smalltalk › Pharo › Pharo Smalltalk Developers

Remove rule "References an abstract class"

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

10 messages Options

Damien Cassou-2

Remove rule "References an abstract class"

This lint rule pops up all the times and my code is always ok. I
typically have to reference an abstract class when:

- the abstract class takes care of choosing the concrete class to
instantiate

- the abstract class takes care of some shared variable for its
subclasses (e.g., a cache or an shared external object)

I propose to remove this rule which has never been useful for me.

--
Damien Cassou
http://damiencassou.seasidehosting.st

"Success is the ability to go from one failure to another without
losing enthusiasm." --Winston Churchill

Uko2

Re: Remove rule "References an abstract class"

Hi.

First of all: thank you for the feedback Damien!
It makes sense, the rule also has higher amount of negative feedback. I will remove it from QA, let it stay in the image, it actually can be useful.

Uko

> On 17 Sep 2015, at 06:49, Damien Cassou <[hidden email]> wrote:
>
> This lint rule pops up all the times and my code is always ok. I
> typically have to reference an abstract class when:
>
> - the abstract class takes care of choosing the concrete class to
> instantiate
>
> - the abstract class takes care of some shared variable for its
> subclasses (e.g., a cache or an shared external object)
>
> I propose to remove this rule which has never been useful for me.
>
> --
> Damien Cassou
> http://damiencassou.seasidehosting.st
>
> "Success is the ability to go from one failure to another without
> losing enthusiasm." --Winston Churchill
>

Peter Uhnak

Re: Remove rule "References an abstract class"

You can probably remove it directly from Pharo instead

https://pharo.fogbugz.com/f/cases/15858/review-rule-RBAbstractClassRule

Peter

On Thu, Sep 17, 2015 at 9:09 AM, Yuriy Tymchuk <[hidden email]> wrote:

Hi.

First of all: thank you for the feedback Damien!
It makes sense, the rule also has higher amount of negative feedback. I will remove it from QA, let it stay in the image, it actually can be useful.

Uko

> On 17 Sep 2015, at 06:49, Damien Cassou <[hidden email]> wrote:
>
> This lint rule pops up all the times and my code is always ok. I
> typically have to reference an abstract class when:
>
> - the abstract class takes care of choosing the concrete class to
> instantiate
>
> - the abstract class takes care of some shared variable for its
> subclasses (e.g., a cache or an shared external object)
>
> I propose to remove this rule which has never been useful for me.
>
> --
> Damien Cassou
> http://damiencassou.seasidehosting.st
>
> "Success is the ability to go from one failure to another without
> losing enthusiasm." --Winston Churchill
>

Uko2

Re: Remove rule "References an abstract class"

Ok, it can be removed from Pharo… I still find it useful as well as missing yourself, but I can keep track of then individually :)

On 17 Sep 2015, at 09:39, Peter Uhnák <[hidden email]> wrote:

You can probably remove it directly from Pharo instead
https://pharo.fogbugz.com/f/cases/15858/review-rule-RBAbstractClassRule

Peter

On Thu, Sep 17, 2015 at 9:09 AM, Yuriy Tymchuk <[hidden email]> wrote:
Hi.

First of all: thank you for the feedback Damien!
It makes sense, the rule also has higher amount of negative feedback. I will remove it from QA, let it stay in the image, it actually can be useful.

Uko

> On 17 Sep 2015, at 06:49, Damien Cassou <[hidden email]> wrote:
>
> This lint rule pops up all the times and my code is always ok. I
> typically have to reference an abstract class when:
>
> - the abstract class takes care of choosing the concrete class to
> instantiate
>
> - the abstract class takes care of some shared variable for its
> subclasses (e.g., a cache or an shared external object)
>
> I propose to remove this rule which has never been useful for me.
>
> --
> Damien Cassou
> http://damiencassou.seasidehosting.st
>
> "Success is the ability to go from one failure to another without
> losing enthusiasm." --Winston Churchill
>

demarey

Re: Remove rule "References an abstract class"

yes the problem with rules comes when the rule add more noise than is helpful.

If you do not want to remove this rule because it is useful in some cases, maybe the rule could check only these specific situations to do not add noise in others.

Le 17 sept. 2015 à 10:37, Yuriy Tymchuk a écrit :

Ok, it can be removed from Pharo… I still find it useful as well as missing yourself, but I can keep track of then individually :)

On 17 Sep 2015, at 09:39, Peter Uhnák <[hidden email]> wrote:

You can probably remove it directly from Pharo instead
https://pharo.fogbugz.com/f/cases/15858/review-rule-RBAbstractClassRule

Peter

On Thu, Sep 17, 2015 at 9:09 AM, Yuriy Tymchuk <[hidden email]> wrote:
Hi.

First of all: thank you for the feedback Damien!
It makes sense, the rule also has higher amount of negative feedback. I will remove it from QA, let it stay in the image, it actually can be useful.

Uko

> On 17 Sep 2015, at 06:49, Damien Cassou <[hidden email]> wrote:
>
> This lint rule pops up all the times and my code is always ok. I
> typically have to reference an abstract class when:
>
> - the abstract class takes care of choosing the concrete class to
> instantiate
>
> - the abstract class takes care of some shared variable for its
> subclasses (e.g., a cache or an shared external object)
>
> I propose to remove this rule which has never been useful for me.
>
> --
> Damien Cassou
> http://damiencassou.seasidehosting.st
>
> "Success is the ability to go from one failure to another without
> losing enthusiasm." --Winston Churchill
>

smime.p7s (5K) Download Attachment

Uko2

Re: Remove rule "References an abstract class"

Well, what I am saying is that I agree that people should not be bothered with this rule by QualityAssistant. But why we should remove the rule from the image?

Uko

On 17 Sep 2015, at 12:11, Christophe Demarey <[hidden email]> wrote:

yes the problem with rules comes when the rule add more noise than is helpful.
If you do not want to remove this rule because it is useful in some cases, maybe the rule could check only these specific situations to do not add noise in others.

Le 17 sept. 2015 à 10:37, Yuriy Tymchuk a écrit :

Ok, it can be removed from Pharo… I still find it useful as well as missing yourself, but I can keep track of then individually :)

On 17 Sep 2015, at 09:39, Peter Uhnák <[hidden email]> wrote:

You can probably remove it directly from Pharo instead
https://pharo.fogbugz.com/f/cases/15858/review-rule-RBAbstractClassRule

Peter

On Thu, Sep 17, 2015 at 9:09 AM, Yuriy Tymchuk <[hidden email]> wrote:
Hi.

First of all: thank you for the feedback Damien!
It makes sense, the rule also has higher amount of negative feedback. I will remove it from QA, let it stay in the image, it actually can be useful.

Uko

> On 17 Sep 2015, at 06:49, Damien Cassou <[hidden email]> wrote:
>
> This lint rule pops up all the times and my code is always ok. I
> typically have to reference an abstract class when:
>
> - the abstract class takes care of choosing the concrete class to
> instantiate
>
> - the abstract class takes care of some shared variable for its
> subclasses (e.g., a cache or an shared external object)
>
> I propose to remove this rule which has never been useful for me.
>
> --
> Damien Cassou
> http://damiencassou.seasidehosting.st
>
> "Success is the ability to go from one failure to another without
> losing enthusiasm." --Winston Churchill
>

Marcus Denker-4

Re: Remove rule "References an abstract class"

> On 17 Sep 2015, at 12:50, Yuriy Tymchuk <[hidden email]> wrote:
>
> Well, what I am saying is that I agree that people should not be bothered with this rule by QualityAssistant. But why we should remove the rule from the image?
>
Because it is wrong? If we keep it, people will see it when they use the Critic Browser… what would be the reason to keep a rule that everyone things is not good?

Marcus

Uko2

Re: Remove rule "References an abstract class"

Because people can choose in critic browser what they want to see. At least this is my opinion. As a sneak peek to the survey results I can say that there are people who find the removed of “missing yourself” rule as a negative change. Of course they are only a few of them, but as you are not able to set up what rules you want to have in QualityAssistant, it makes sense to remove ones that the majority does not like. But in critics browser we can specify which rules you want to run. So I don’t see a need in a complete removal. Yes, maybe form UX perspective critic browser should have some button like: “Run only the important ones”…

Uko

> On 17 Sep 2015, at 13:43, Marcus Denker <[hidden email]> wrote:
>
>
>> On 17 Sep 2015, at 12:50, Yuriy Tymchuk <[hidden email]> wrote:
>>
>> Well, what I am saying is that I agree that people should not be bothered with this rule by QualityAssistant. But why we should remove the rule from the image?
>>
> Because it is wrong? If we keep it, people will see it when they use the Critic Browser… what would be the reason to keep a rule that everyone things is not good?
>
> Marcus
>
>
>

Marcus Denker-4

Re: Remove rule "References an abstract class"

> On 17 Sep 2015, at 14:08, Yuriy Tymchuk <[hidden email]> wrote:
>
> Because people can choose in critic browser what they want to see. At least this is my opinion. As a sneak peek to the survey results I can say that there are people who find the removed of “missing yourself” rule as a negative change

When I learned something than this: Whatever you do, someone will think it is wrong. The largest mistake that you can do is to only do things that nobody objects to… because the only thing that fulfils that is to do nothing.

Marcus

Uko2

Re: Remove rule "References an abstract class"

Well, for me it’s ok. Then I don’t have to create a new version in conf :).

> On 17 Sep 2015, at 14:17, Marcus Denker <[hidden email]> wrote:
>
>
>> On 17 Sep 2015, at 14:08, Yuriy Tymchuk <[hidden email]> wrote:
>>
>> Because people can choose in critic browser what they want to see. At least this is my opinion. As a sneak peek to the survey results I can say that there are people who find the removed of “missing yourself” rule as a negative change
>
> When I learned something than this: Whatever you do, someone will think it is wrong. The largest mistake that you can do is to only do things that nobody objects to… because the only thing that fulfils that is to do nothing.
>
> Marcus