recent buglet

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

recent buglet

Bob Arning-2

'From Squeak4.5 of 16 September 2013 [latest update: #12968] on 17 September 2013 at 5:31:18 pm'!

!StringHolder methodsFor: '*Tools' stamp: 'fbs 9/13/2013 17:29'!
browseMethodFull
    "Create and schedule a full Browser and then select the current class and message."
    ^ self selectedClassOrMetaClass ifNotNil: [ :cls |
        self selectedMessageName
            ifNotNil: [ :sel |
                ToolSet browse: cls selector: sel]
            ifNil: [ToolSet browserClass: cls]].! !

#browserClass: is not understood here.
#browseClass: will work

Cheers,
Bob


Reply | Threaded
Open this post in threaded view
|

Re: recent buglet

Frank Shearar-3
On 17 September 2013 22:33, Bob Arning <[hidden email]> wrote:

>
> 'From Squeak4.5 of 16 September 2013 [latest update: #12968] on 17 September
> 2013 at 5:31:18 pm'!
>
> !StringHolder methodsFor: '*Tools' stamp: 'fbs 9/13/2013 17:29'!
> browseMethodFull
>     "Create and schedule a full Browser and then select the current class
> and message."
>     ^ self selectedClassOrMetaClass ifNotNil: [ :cls |
>         self selectedMessageName
>             ifNotNil: [ :sel |
>                 ToolSet browse: cls selector: sel]
>             ifNil: [ToolSet browserClass: cls]].! !
>
> #browserClass: is not understood here.
> #browseClass: will work

Oh dear. I'll fix it up. We seriously need more tests for core
infrastructure like this.

Thanks for the catch!

frank

> Cheers,
> Bob
>
>
>

Reply | Threaded
Open this post in threaded view
|

Re: recent buglet

Nicolas Cellier
What is the difference between:
    ToolSet browse: Object selector: nil
and
    ToolSet browseClass: Object
Is it just more explicit?

There are other senders of browse:selector: with nil selector, should they be turned into browseClass: ?



2013/9/17 Frank Shearar <[hidden email]>
On 17 September 2013 22:33, Bob Arning <[hidden email]> wrote:
>
> 'From Squeak4.5 of 16 September 2013 [latest update: #12968] on 17 September
> 2013 at 5:31:18 pm'!
>
> !StringHolder methodsFor: '*Tools' stamp: 'fbs 9/13/2013 17:29'!
> browseMethodFull
>     "Create and schedule a full Browser and then select the current class
> and message."
>     ^ self selectedClassOrMetaClass ifNotNil: [ :cls |
>         self selectedMessageName
>             ifNotNil: [ :sel |
>                 ToolSet browse: cls selector: sel]
>             ifNil: [ToolSet browserClass: cls]].! !
>
> #browserClass: is not understood here.
> #browseClass: will work

Oh dear. I'll fix it up. We seriously need more tests for core
infrastructure like this.

Thanks for the catch!

frank

> Cheers,
> Bob
>
>
>




Reply | Threaded
Open this post in threaded view
|

Re: recent buglet

Frank Shearar-3
It's more explicit, and often I want to browse the class definition.
This way the tools will open a Browser on a particular method _if
possible_, and simply Browser the class definition otherwise (instead
of doing nothing at all) when I alt-b the list item.

frank

On 17 September 2013 23:51, Nicolas Cellier
<[hidden email]> wrote:

> What is the difference between:
>     ToolSet browse: Object selector: nil
> and
>     ToolSet browseClass: Object
> Is it just more explicit?
>
> There are other senders of browse:selector: with nil selector, should they
> be turned into browseClass: ?
>
>
>
> 2013/9/17 Frank Shearar <[hidden email]>
>>
>> On 17 September 2013 22:33, Bob Arning <[hidden email]> wrote:
>> >
>> > 'From Squeak4.5 of 16 September 2013 [latest update: #12968] on 17
>> > September
>> > 2013 at 5:31:18 pm'!
>> >
>> > !StringHolder methodsFor: '*Tools' stamp: 'fbs 9/13/2013 17:29'!
>> > browseMethodFull
>> >     "Create and schedule a full Browser and then select the current
>> > class
>> > and message."
>> >     ^ self selectedClassOrMetaClass ifNotNil: [ :cls |
>> >         self selectedMessageName
>> >             ifNotNil: [ :sel |
>> >                 ToolSet browse: cls selector: sel]
>> >             ifNil: [ToolSet browserClass: cls]].! !
>> >
>> > #browserClass: is not understood here.
>> > #browseClass: will work
>>
>> Oh dear. I'll fix it up. We seriously need more tests for core
>> infrastructure like this.
>>
>> Thanks for the catch!
>>
>> frank
>>
>> > Cheers,
>> > Bob
>> >
>> >
>> >
>>
>
>
>
>