OmniBrowser support now in trunk

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

OmniBrowser support now in trunk

Colin Putney-3
Hi all,

Just wanted give a bit more context for the changes I've just uploaded
to trunk. They fall into two categories:

One is adding support for icons in lists. OmniBrowser uses this to
indicate things like methods being overridden and classes inheriting
from certain important superclasses, like Collection or Exception.

The other is making the current selection available when building
context menus for text panes. Previously the current selection wasn't
available when the menu was being built, though you could get it when
actually executing a menu command. Having the selection lets us
determine which refactorings are applicable, and indicate that
correctly in the menu.

I probably should have submitted these changes to the Inbox for review
before pushing them to trunk, but I didn't think of it until I had
already uploaded the first few versions. I don't think there's much
risk of breaking anything, though. There's very little change to
existing code, and that's only triggered by the model indicating that
it wants to use the new functionality when it builds a window in
ToolBuilder.

The next release of OmniBrowser should be available pretty soon too,
just a few more bugs to fix.

Colin

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Hannes Hirzel
Good to have now  this update in the trunk. Thank you!

-- Hannes

On 4/3/12, Colin Putney <[hidden email]> wrote:

> Hi all,
>
> Just wanted give a bit more context for the changes I've just uploaded
> to trunk. They fall into two categories:
>
> One is adding support for icons in lists. OmniBrowser uses this to
> indicate things like methods being overridden and classes inheriting
> from certain important superclasses, like Collection or Exception.
>
> The other is making the current selection available when building
> context menus for text panes. Previously the current selection wasn't
> available when the menu was being built, though you could get it when
> actually executing a menu command. Having the selection lets us
> determine which refactorings are applicable, and indicate that
> correctly in the menu.
>
> I probably should have submitted these changes to the Inbox for review
> before pushing them to trunk, but I didn't think of it until I had
> already uploaded the first few versions. I don't think there's much
> risk of breaking anything, though. There's very little change to
> existing code, and that's only triggered by the model indicating that
> it wants to use the new functionality when it builds a window in
> ToolBuilder.
>
> The next release of OmniBrowser should be available pretty soon too,
> just a few more bugs to fix.
>
> Colin
>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

David T. Lewis
In reply to this post by Colin Putney-3
On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:

> Hi all,
>
> Just wanted give a bit more context for the changes I've just uploaded
> to trunk. They fall into two categories:
>
> One is adding support for icons in lists. OmniBrowser uses this to
> indicate things like methods being overridden and classes inheriting
> from certain important superclasses, like Collection or Exception.
>
> The other is making the current selection available when building
> context menus for text panes. Previously the current selection wasn't
> available when the menu was being built, though you could get it when
> actually executing a menu command. Having the selection lets us
> determine which refactorings are applicable, and indicate that
> correctly in the menu.

Excellent!

>
> I probably should have submitted these changes to the Inbox for review
> before pushing them to trunk, but I didn't think of it until I had
> already uploaded the first few versions. I don't think there's much
> risk of breaking anything, though. There's very little change to
> existing code, and that's only triggered by the model indicating that
> it wants to use the new functionality when it builds a window in
> ToolBuilder.

No worries, if any issues appear we can fix them in trunk. That way
they will be addressed as quickly as possible.

>
> The next release of OmniBrowser should be available pretty soon too,
> just a few more bugs to fix.
>
> Colin

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Hannes Hirzel
Hello

How do I load OmniBrowser these days into a fully updated Squeak 4.4 image?

--Hannes

On 4/3/12, David T. Lewis <[hidden email]> wrote:

> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>> Hi all,
>>
>> Just wanted give a bit more context for the changes I've just uploaded
>> to trunk. They fall into two categories:
>>
>> One is adding support for icons in lists. OmniBrowser uses this to
>> indicate things like methods being overridden and classes inheriting
>> from certain important superclasses, like Collection or Exception.
>>
>> The other is making the current selection available when building
>> context menus for text panes. Previously the current selection wasn't
>> available when the menu was being built, though you could get it when
>> actually executing a menu command. Having the selection lets us
>> determine which refactorings are applicable, and indicate that
>> correctly in the menu.
>
> Excellent!
>
>>
>> I probably should have submitted these changes to the Inbox for review
>> before pushing them to trunk, but I didn't think of it until I had
>> already uploaded the first few versions. I don't think there's much
>> risk of breaking anything, though. There's very little change to
>> existing code, and that's only triggered by the model indicating that
>> it wants to use the new functionality when it builds a window in
>> ToolBuilder.
>
> No worries, if any issues appear we can fix them in trunk. That way
> they will be addressed as quickly as possible.
>
>>
>> The next release of OmniBrowser should be available pretty soon too,
>> just a few more bugs to fix.
>>
>> Colin
>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Hannes Hirzel
To load OmniBrowser

The code snippet which is in the workspace accessible under 'Help',
'Extending the system'

    "Omnibrowser, including Refactoring engine"
    (Installer ss project: 'MetacelloRepository') install:
'ConfigurationOfOmniBrowser'.
    ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
#lastVersion) load: #( Dev ).

Still does the job.

--Hannes


--Hannes

On 9/11/12, H. Hirzel <[hidden email]> wrote:

> Hello
>
> How do I load OmniBrowser these days into a fully updated Squeak 4.4 image?
>
> --Hannes
>
> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>> Hi all,
>>>
>>> Just wanted give a bit more context for the changes I've just uploaded
>>> to trunk. They fall into two categories:
>>>
>>> One is adding support for icons in lists. OmniBrowser uses this to
>>> indicate things like methods being overridden and classes inheriting
>>> from certain important superclasses, like Collection or Exception.
>>>
>>> The other is making the current selection available when building
>>> context menus for text panes. Previously the current selection wasn't
>>> available when the menu was being built, though you could get it when
>>> actually executing a menu command. Having the selection lets us
>>> determine which refactorings are applicable, and indicate that
>>> correctly in the menu.
>>
>> Excellent!
>>
>>>
>>> I probably should have submitted these changes to the Inbox for review
>>> before pushing them to trunk, but I didn't think of it until I had
>>> already uploaded the first few versions. I don't think there's much
>>> risk of breaking anything, though. There's very little change to
>>> existing code, and that's only triggered by the model indicating that
>>> it wants to use the new functionality when it builds a window in
>>> ToolBuilder.
>>
>> No worries, if any issues appear we can fix them in trunk. That way
>> they will be addressed as quickly as possible.
>>
>>>
>>> The next release of OmniBrowser should be available pretty soon too,
>>> just a few more bugs to fix.
>>>
>>> Colin
>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Frank Shearar-3
On 11 September 2012 14:37, H. Hirzel <[hidden email]> wrote:

> To load OmniBrowser
>
> The code snippet which is in the workspace accessible under 'Help',
> 'Extending the system'
>
>     "Omnibrowser, including Refactoring engine"
>     (Installer ss project: 'MetacelloRepository') install:
> 'ConfigurationOfOmniBrowser'.
>     ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
> #lastVersion) load: #( Dev ).

Cool! But why the #perform? Couldn't you just say (Smalltalk at:
#ConfigurationOfOmniBrowser) project lastVersion load: #( Dev ) ?

frank

> Still does the job.
>
> --Hannes
>
>
> --Hannes
>
> On 9/11/12, H. Hirzel <[hidden email]> wrote:
>> Hello
>>
>> How do I load OmniBrowser these days into a fully updated Squeak 4.4 image?
>>
>> --Hannes
>>
>> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>>> Hi all,
>>>>
>>>> Just wanted give a bit more context for the changes I've just uploaded
>>>> to trunk. They fall into two categories:
>>>>
>>>> One is adding support for icons in lists. OmniBrowser uses this to
>>>> indicate things like methods being overridden and classes inheriting
>>>> from certain important superclasses, like Collection or Exception.
>>>>
>>>> The other is making the current selection available when building
>>>> context menus for text panes. Previously the current selection wasn't
>>>> available when the menu was being built, though you could get it when
>>>> actually executing a menu command. Having the selection lets us
>>>> determine which refactorings are applicable, and indicate that
>>>> correctly in the menu.
>>>
>>> Excellent!
>>>
>>>>
>>>> I probably should have submitted these changes to the Inbox for review
>>>> before pushing them to trunk, but I didn't think of it until I had
>>>> already uploaded the first few versions. I don't think there's much
>>>> risk of breaking anything, though. There's very little change to
>>>> existing code, and that's only triggered by the model indicating that
>>>> it wants to use the new functionality when it builds a window in
>>>> ToolBuilder.
>>>
>>> No worries, if any issues appear we can fix them in trunk. That way
>>> they will be addressed as quickly as possible.
>>>
>>>>
>>>> The next release of OmniBrowser should be available pretty soon too,
>>>> just a few more bugs to fix.
>>>>
>>>> Colin
>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Hannes Hirzel
If you remove the
    perform: #lastVersion


and execute

(Installer ss project: 'MetacelloRepository') install:
'ConfigurationOfOmniBrowser'.
((Smalltalk at: #ConfigurationOfOmniBrowser) project lastVersion)
load: #( Dev ).

in a fresh Squeak4.4. trunk image

you get a feedback 'unknown selector'  / 'lastVersion'



On the other side the current load script

"Omnibrowser, including Refactoring engine"
(Installer ss project: 'MetacelloRepository') install:
'ConfigurationOfOmniBrowser'.
((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
#lastVersion) load: #( Dev ).

just works fine.

--Hannes






On 9/11/12, Frank Shearar <[hidden email]> wrote:

> On 11 September 2012 14:37, H. Hirzel <[hidden email]> wrote:
>> To load OmniBrowser
>>
>> The code snippet which is in the workspace accessible under 'Help',
>> 'Extending the system'
>>
>>     "Omnibrowser, including Refactoring engine"
>>     (Installer ss project: 'MetacelloRepository') install:
>> 'ConfigurationOfOmniBrowser'.
>>     ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>> #lastVersion) load: #( Dev ).
>
> Cool! But why the #perform? Couldn't you just say (Smalltalk at:
> #ConfigurationOfOmniBrowser) project lastVersion load: #( Dev ) ?
>
> frank
>
>> Still does the job.
>>
>> --Hannes
>>
>>
>> --Hannes
>>
>> On 9/11/12, H. Hirzel <[hidden email]> wrote:
>>> Hello
>>>
>>> How do I load OmniBrowser these days into a fully updated Squeak 4.4
>>> image?
>>>
>>> --Hannes
>>>
>>> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>>>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>>>> Hi all,
>>>>>
>>>>> Just wanted give a bit more context for the changes I've just uploaded
>>>>> to trunk. They fall into two categories:
>>>>>
>>>>> One is adding support for icons in lists. OmniBrowser uses this to
>>>>> indicate things like methods being overridden and classes inheriting
>>>>> from certain important superclasses, like Collection or Exception.
>>>>>
>>>>> The other is making the current selection available when building
>>>>> context menus for text panes. Previously the current selection wasn't
>>>>> available when the menu was being built, though you could get it when
>>>>> actually executing a menu command. Having the selection lets us
>>>>> determine which refactorings are applicable, and indicate that
>>>>> correctly in the menu.
>>>>
>>>> Excellent!
>>>>
>>>>>
>>>>> I probably should have submitted these changes to the Inbox for review
>>>>> before pushing them to trunk, but I didn't think of it until I had
>>>>> already uploaded the first few versions. I don't think there's much
>>>>> risk of breaking anything, though. There's very little change to
>>>>> existing code, and that's only triggered by the model indicating that
>>>>> it wants to use the new functionality when it builds a window in
>>>>> ToolBuilder.
>>>>
>>>> No worries, if any issues appear we can fix them in trunk. That way
>>>> they will be addressed as quickly as possible.
>>>>
>>>>>
>>>>> The next release of OmniBrowser should be available pretty soon too,
>>>>> just a few more bugs to fix.
>>>>>
>>>>> Colin
>>>>
>>>>
>>>
>>
>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Frank Shearar-3
On 11 September 2012 15:28, H. Hirzel <[hidden email]> wrote:

> If you remove the
>     perform: #lastVersion
>
>
> and execute
>
> (Installer ss project: 'MetacelloRepository') install:
> 'ConfigurationOfOmniBrowser'.
> ((Smalltalk at: #ConfigurationOfOmniBrowser) project lastVersion)
> load: #( Dev ).
>
> in a fresh Squeak4.4. trunk image
>
> you get a feedback 'unknown selector'  / 'lastVersion'
>
>
>
> On the other side the current load script
>
> "Omnibrowser, including Refactoring engine"
> (Installer ss project: 'MetacelloRepository') install:
> 'ConfigurationOfOmniBrowser'.
> ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
> #lastVersion) load: #( Dev ).
>
> just works fine.

Ah, right of course. A vanilla image has no such sender.

frank

> --Hannes
>
>
>
>
>
>
> On 9/11/12, Frank Shearar <[hidden email]> wrote:
>> On 11 September 2012 14:37, H. Hirzel <[hidden email]> wrote:
>>> To load OmniBrowser
>>>
>>> The code snippet which is in the workspace accessible under 'Help',
>>> 'Extending the system'
>>>
>>>     "Omnibrowser, including Refactoring engine"
>>>     (Installer ss project: 'MetacelloRepository') install:
>>> 'ConfigurationOfOmniBrowser'.
>>>     ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>>> #lastVersion) load: #( Dev ).
>>
>> Cool! But why the #perform? Couldn't you just say (Smalltalk at:
>> #ConfigurationOfOmniBrowser) project lastVersion load: #( Dev ) ?
>>
>> frank
>>
>>> Still does the job.
>>>
>>> --Hannes
>>>
>>>
>>> --Hannes
>>>
>>> On 9/11/12, H. Hirzel <[hidden email]> wrote:
>>>> Hello
>>>>
>>>> How do I load OmniBrowser these days into a fully updated Squeak 4.4
>>>> image?
>>>>
>>>> --Hannes
>>>>
>>>> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>>>>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>>>>> Hi all,
>>>>>>
>>>>>> Just wanted give a bit more context for the changes I've just uploaded
>>>>>> to trunk. They fall into two categories:
>>>>>>
>>>>>> One is adding support for icons in lists. OmniBrowser uses this to
>>>>>> indicate things like methods being overridden and classes inheriting
>>>>>> from certain important superclasses, like Collection or Exception.
>>>>>>
>>>>>> The other is making the current selection available when building
>>>>>> context menus for text panes. Previously the current selection wasn't
>>>>>> available when the menu was being built, though you could get it when
>>>>>> actually executing a menu command. Having the selection lets us
>>>>>> determine which refactorings are applicable, and indicate that
>>>>>> correctly in the menu.
>>>>>
>>>>> Excellent!
>>>>>
>>>>>>
>>>>>> I probably should have submitted these changes to the Inbox for review
>>>>>> before pushing them to trunk, but I didn't think of it until I had
>>>>>> already uploaded the first few versions. I don't think there's much
>>>>>> risk of breaking anything, though. There's very little change to
>>>>>> existing code, and that's only triggered by the model indicating that
>>>>>> it wants to use the new functionality when it builds a window in
>>>>>> ToolBuilder.
>>>>>
>>>>> No worries, if any issues appear we can fix them in trunk. That way
>>>>> they will be addressed as quickly as possible.
>>>>>
>>>>>>
>>>>>> The next release of OmniBrowser should be available pretty soon too,
>>>>>> just a few more bugs to fix.
>>>>>>
>>>>>> Colin
>>>>>
>>>>>
>>>>
>>>
>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Hannes Hirzel
Retest:

Installing Omnibrowser still works fine

"Omnibrowser, including Refactoring engine"
(Installer ss project: 'MetacelloRepository') install:
'ConfigurationOfOmniBrowser'.
((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
#lastVersion) load: #( Dev ).


Code snippet from the 'How to extend the system' workspace.

--Hannes

On 9/11/12, Frank Shearar <[hidden email]> wrote:

> On 11 September 2012 15:28, H. Hirzel <[hidden email]> wrote:
>> If you remove the
>>     perform: #lastVersion
>>
>>
>> and execute
>>
>> (Installer ss project: 'MetacelloRepository') install:
>> 'ConfigurationOfOmniBrowser'.
>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project lastVersion)
>> load: #( Dev ).
>>
>> in a fresh Squeak4.4. trunk image
>>
>> you get a feedback 'unknown selector'  / 'lastVersion'
>>
>>
>>
>> On the other side the current load script
>>
>> "Omnibrowser, including Refactoring engine"
>> (Installer ss project: 'MetacelloRepository') install:
>> 'ConfigurationOfOmniBrowser'.
>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>> #lastVersion) load: #( Dev ).
>>
>> just works fine.
>
> Ah, right of course. A vanilla image has no such sender.
>
> frank
>
>> --Hannes
>>
>>
>>
>>
>>
>>
>> On 9/11/12, Frank Shearar <[hidden email]> wrote:
>>> On 11 September 2012 14:37, H. Hirzel <[hidden email]> wrote:
>>>> To load OmniBrowser
>>>>
>>>> The code snippet which is in the workspace accessible under 'Help',
>>>> 'Extending the system'
>>>>
>>>>     "Omnibrowser, including Refactoring engine"
>>>>     (Installer ss project: 'MetacelloRepository') install:
>>>> 'ConfigurationOfOmniBrowser'.
>>>>     ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>>>> #lastVersion) load: #( Dev ).
>>>
>>> Cool! But why the #perform? Couldn't you just say (Smalltalk at:
>>> #ConfigurationOfOmniBrowser) project lastVersion load: #( Dev ) ?
>>>
>>> frank
>>>
>>>> Still does the job.
>>>>
>>>> --Hannes
>>>>
>>>>
>>>> --Hannes
>>>>
>>>> On 9/11/12, H. Hirzel <[hidden email]> wrote:
>>>>> Hello
>>>>>
>>>>> How do I load OmniBrowser these days into a fully updated Squeak 4.4
>>>>> image?
>>>>>
>>>>> --Hannes
>>>>>
>>>>> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>>>>>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>>>>>> Hi all,
>>>>>>>
>>>>>>> Just wanted give a bit more context for the changes I've just
>>>>>>> uploaded
>>>>>>> to trunk. They fall into two categories:
>>>>>>>
>>>>>>> One is adding support for icons in lists. OmniBrowser uses this to
>>>>>>> indicate things like methods being overridden and classes inheriting
>>>>>>> from certain important superclasses, like Collection or Exception.
>>>>>>>
>>>>>>> The other is making the current selection available when building
>>>>>>> context menus for text panes. Previously the current selection
>>>>>>> wasn't
>>>>>>> available when the menu was being built, though you could get it
>>>>>>> when
>>>>>>> actually executing a menu command. Having the selection lets us
>>>>>>> determine which refactorings are applicable, and indicate that
>>>>>>> correctly in the menu.
>>>>>>
>>>>>> Excellent!
>>>>>>
>>>>>>>
>>>>>>> I probably should have submitted these changes to the Inbox for
>>>>>>> review
>>>>>>> before pushing them to trunk, but I didn't think of it until I had
>>>>>>> already uploaded the first few versions. I don't think there's much
>>>>>>> risk of breaking anything, though. There's very little change to
>>>>>>> existing code, and that's only triggered by the model indicating
>>>>>>> that
>>>>>>> it wants to use the new functionality when it builds a window in
>>>>>>> ToolBuilder.
>>>>>>
>>>>>> No worries, if any issues appear we can fix them in trunk. That way
>>>>>> they will be addressed as quickly as possible.
>>>>>>
>>>>>>>
>>>>>>> The next release of OmniBrowser should be available pretty soon too,
>>>>>>> just a few more bugs to fix.
>>>>>>>
>>>>>>> Colin
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>>
>>
>
>

Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Chris Cunnington
On 2012-12-12 8:07 PM, H. Hirzel wrote:

> Retest:
>
> Installing Omnibrowser still works fine
>
> "Omnibrowser, including Refactoring engine"
> (Installer ss project: 'MetacelloRepository') install:
> 'ConfigurationOfOmniBrowser'.
> ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
> #lastVersion) load: #( Dev ).
>
>
> Code snippet from the 'How to extend the system' workspace.
>
> --Hannes
>
> On 9/11/12, Frank Shearar <[hidden email]> wrote:
>> On 11 September 2012 15:28, H. Hirzel <[hidden email]> wrote:
>>> If you remove the
>>>      perform: #lastVersion
>>>
>>>
>>> and execute
>>>
>>> (Installer ss project: 'MetacelloRepository') install:
>>> 'ConfigurationOfOmniBrowser'.
>>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project lastVersion)
>>> load: #( Dev ).
>>>
>>> in a fresh Squeak4.4. trunk image
>>>
>>> you get a feedback 'unknown selector'  / 'lastVersion'
>>>
>>>
>>>
>>> On the other side the current load script
>>>
>>> "Omnibrowser, including Refactoring engine"
>>> (Installer ss project: 'MetacelloRepository') install:
>>> 'ConfigurationOfOmniBrowser'.
>>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>>> #lastVersion) load: #( Dev ).
>>>
>>> just works fine.
>> Ah, right of course. A vanilla image has no such sender.
>>
>> frank
>>
>>> --Hannes
>>>
>>>
>>>
>>>
>>>
>>>
>>> On 9/11/12, Frank Shearar <[hidden email]> wrote:
>>>> On 11 September 2012 14:37, H. Hirzel <[hidden email]> wrote:
>>>>> To load OmniBrowser
>>>>>
>>>>> The code snippet which is in the workspace accessible under 'Help',
>>>>> 'Extending the system'
>>>>>
>>>>>      "Omnibrowser, including Refactoring engine"
>>>>>      (Installer ss project: 'MetacelloRepository') install:
>>>>> 'ConfigurationOfOmniBrowser'.
>>>>>      ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>>>>> #lastVersion) load: #( Dev ).
>>>> Cool! But why the #perform? Couldn't you just say (Smalltalk at:
>>>> #ConfigurationOfOmniBrowser) project lastVersion load: #( Dev ) ?
>>>>
>>>> frank
>>>>
>>>>> Still does the job.
>>>>>
>>>>> --Hannes
>>>>>
>>>>>
>>>>> --Hannes
>>>>>
>>>>> On 9/11/12, H. Hirzel <[hidden email]> wrote:
>>>>>> Hello
>>>>>>
>>>>>> How do I load OmniBrowser these days into a fully updated Squeak 4.4
>>>>>> image?
>>>>>>
>>>>>> --Hannes
>>>>>>
>>>>>> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>>>>>>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>>>>>>> Hi all,
>>>>>>>>
>>>>>>>> Just wanted give a bit more context for the changes I've just
>>>>>>>> uploaded
>>>>>>>> to trunk. They fall into two categories:
>>>>>>>>
>>>>>>>> One is adding support for icons in lists. OmniBrowser uses this to
>>>>>>>> indicate things like methods being overridden and classes inheriting
>>>>>>>> from certain important superclasses, like Collection or Exception.
>>>>>>>>
>>>>>>>> The other is making the current selection available when building
>>>>>>>> context menus for text panes. Previously the current selection
>>>>>>>> wasn't
>>>>>>>> available when the menu was being built, though you could get it
>>>>>>>> when
>>>>>>>> actually executing a menu command. Having the selection lets us
>>>>>>>> determine which refactorings are applicable, and indicate that
>>>>>>>> correctly in the menu.
>>>>>>> Excellent!
>>>>>>>
>>>>>>>> I probably should have submitted these changes to the Inbox for
>>>>>>>> review
>>>>>>>> before pushing them to trunk, but I didn't think of it until I had
>>>>>>>> already uploaded the first few versions. I don't think there's much
>>>>>>>> risk of breaking anything, though. There's very little change to
>>>>>>>> existing code, and that's only triggered by the model indicating
>>>>>>>> that
>>>>>>>> it wants to use the new functionality when it builds a window in
>>>>>>>> ToolBuilder.
>>>>>>> No worries, if any issues appear we can fix them in trunk. That way
>>>>>>> they will be addressed as quickly as possible.
>>>>>>>
>>>>>>>> The next release of OmniBrowser should be available pretty soon too,
>>>>>>>> just a few more bugs to fix.
>>>>>>>>
>>>>>>>> Colin
>>>>>>>
>>>>
>>
Well that's better. OmniBrowser to hack on Altitude seems a better idea.
Thank you, Hannes.

Chris


Reply | Threaded
Open this post in threaded view
|

Re: OmniBrowser support now in trunk

Hannes Hirzel
The search box does gives a walkback but otherwise it seems to work fine so far.

--Hannes

On 12/13/12, Chris Cunnington <[hidden email]> wrote:

> On 2012-12-12 8:07 PM, H. Hirzel wrote:
>> Retest:
>>
>> Installing Omnibrowser still works fine
>>
>> "Omnibrowser, including Refactoring engine"
>> (Installer ss project: 'MetacelloRepository') install:
>> 'ConfigurationOfOmniBrowser'.
>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>> #lastVersion) load: #( Dev ).
>>
>>
>> Code snippet from the 'How to extend the system' workspace.
>>
>> --Hannes
>>
>> On 9/11/12, Frank Shearar <[hidden email]> wrote:
>>> On 11 September 2012 15:28, H. Hirzel <[hidden email]> wrote:
>>>> If you remove the
>>>>      perform: #lastVersion
>>>>
>>>>
>>>> and execute
>>>>
>>>> (Installer ss project: 'MetacelloRepository') install:
>>>> 'ConfigurationOfOmniBrowser'.
>>>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project lastVersion)
>>>> load: #( Dev ).
>>>>
>>>> in a fresh Squeak4.4. trunk image
>>>>
>>>> you get a feedback 'unknown selector'  / 'lastVersion'
>>>>
>>>>
>>>>
>>>> On the other side the current load script
>>>>
>>>> "Omnibrowser, including Refactoring engine"
>>>> (Installer ss project: 'MetacelloRepository') install:
>>>> 'ConfigurationOfOmniBrowser'.
>>>> ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>>>> #lastVersion) load: #( Dev ).
>>>>
>>>> just works fine.
>>> Ah, right of course. A vanilla image has no such sender.
>>>
>>> frank
>>>
>>>> --Hannes
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On 9/11/12, Frank Shearar <[hidden email]> wrote:
>>>>> On 11 September 2012 14:37, H. Hirzel <[hidden email]> wrote:
>>>>>> To load OmniBrowser
>>>>>>
>>>>>> The code snippet which is in the workspace accessible under 'Help',
>>>>>> 'Extending the system'
>>>>>>
>>>>>>      "Omnibrowser, including Refactoring engine"
>>>>>>      (Installer ss project: 'MetacelloRepository') install:
>>>>>> 'ConfigurationOfOmniBrowser'.
>>>>>>      ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform:
>>>>>> #lastVersion) load: #( Dev ).
>>>>> Cool! But why the #perform? Couldn't you just say (Smalltalk at:
>>>>> #ConfigurationOfOmniBrowser) project lastVersion load: #( Dev ) ?
>>>>>
>>>>> frank
>>>>>
>>>>>> Still does the job.
>>>>>>
>>>>>> --Hannes
>>>>>>
>>>>>>
>>>>>> --Hannes
>>>>>>
>>>>>> On 9/11/12, H. Hirzel <[hidden email]> wrote:
>>>>>>> Hello
>>>>>>>
>>>>>>> How do I load OmniBrowser these days into a fully updated Squeak 4.4
>>>>>>> image?
>>>>>>>
>>>>>>> --Hannes
>>>>>>>
>>>>>>> On 4/3/12, David T. Lewis <[hidden email]> wrote:
>>>>>>>> On Mon, Apr 02, 2012 at 05:19:34PM -0700, Colin Putney wrote:
>>>>>>>>> Hi all,
>>>>>>>>>
>>>>>>>>> Just wanted give a bit more context for the changes I've just
>>>>>>>>> uploaded
>>>>>>>>> to trunk. They fall into two categories:
>>>>>>>>>
>>>>>>>>> One is adding support for icons in lists. OmniBrowser uses this to
>>>>>>>>> indicate things like methods being overridden and classes
>>>>>>>>> inheriting
>>>>>>>>> from certain important superclasses, like Collection or Exception.
>>>>>>>>>
>>>>>>>>> The other is making the current selection available when building
>>>>>>>>> context menus for text panes. Previously the current selection
>>>>>>>>> wasn't
>>>>>>>>> available when the menu was being built, though you could get it
>>>>>>>>> when
>>>>>>>>> actually executing a menu command. Having the selection lets us
>>>>>>>>> determine which refactorings are applicable, and indicate that
>>>>>>>>> correctly in the menu.
>>>>>>>> Excellent!
>>>>>>>>
>>>>>>>>> I probably should have submitted these changes to the Inbox for
>>>>>>>>> review
>>>>>>>>> before pushing them to trunk, but I didn't think of it until I had
>>>>>>>>> already uploaded the first few versions. I don't think there's
>>>>>>>>> much
>>>>>>>>> risk of breaking anything, though. There's very little change to
>>>>>>>>> existing code, and that's only triggered by the model indicating
>>>>>>>>> that
>>>>>>>>> it wants to use the new functionality when it builds a window in
>>>>>>>>> ToolBuilder.
>>>>>>>> No worries, if any issues appear we can fix them in trunk. That way
>>>>>>>> they will be addressed as quickly as possible.
>>>>>>>>
>>>>>>>>> The next release of OmniBrowser should be available pretty soon
>>>>>>>>> too,
>>>>>>>>> just a few more bugs to fix.
>>>>>>>>>
>>>>>>>>> Colin
>>>>>>>>
>>>>>
>>>
> Well that's better. OmniBrowser to hack on Altitude seems a better idea.
> Thank you, Hannes.
>
> Chris
>
>
>