MetacelloBrowser on Squeak

> Soooo, that means splitting things up into package-sized chucks, probably something along these lines:
>
>  - MetacelloBrowser-Core
>  - MetacelloBrowser-Tests
>  - MetacelloBrowser-Pharo-Core
>  - MetacelloBrowser-Pharo-Tests
>
> then I'll be able to add:
>
>  - MetacelloBrowser-Squeak-Core
>  - MetacelloBrowser-Squeak-Tests
>
> as I work on the port ...
>
> Does this make sense to you or do you think we should take a different approach to naming the packages? We just have to stay within the package naming guidelines, i.e., being careful where we put '-'.

>> While waiting for Seaside 3.0 to load (sweet that I just have to press buttons now ...) I tried loading MetacelloBrowser into Squeak and of course ran into issues, which means that we've got to go to multiple packages for the proejct ...
>
> Yes. We can do this. The only inconvenient is that it will significantly increase the loading time of MetacelloBrowser. What I did over the last two months, is to always take a Pharo1.2, install the browser and load what I have to load. This is usually fast since the browser holds in one single package.
>
>> Soooo, that means splitting things up into package-sized chucks, probably something along these lines:
>>
>>   - MetacelloBrowser-Core
>>   - MetacelloBrowser-Tests
>>   - MetacelloBrowser-Pharo-Core
>>   - MetacelloBrowser-Pharo-Tests
>>
>> then I'll be able to add:
>>
>>   - MetacelloBrowser-Squeak-Core
>>   - MetacelloBrowser-Squeak-Tests
>>
>> as I work on the port ...
>>
>> Does this make sense to you or do you think we should take a different approach to naming the packages? We just have to stay within the package naming guidelines, i.e., being careful where we put '-'.
>
>
> Yes, it makes sense. Now, we have to do such refactoring without opening the Monticello browser :-) The first step, is to create the packages from the class categories. Nice challenge isn't it ?
>

> On 03/01/2011 04:06 AM, Alexandre Bergel wrote:
>>> While waiting for Seaside 3.0 to load (sweet that I just have to press buttons now ...) I tried loading MetacelloBrowser into Squeak and of course ran into issues, which means that we've got to go to multiple packages for the proejct ...
>>
>> Yes. We can do this. The only inconvenient is that it will significantly increase the loading time of MetacelloBrowser. What I did over the last two months, is to always take a Pharo1.2, install the browser and load what I have to load. This is usually fast since the browser holds in one single package.
>>
>>> Soooo, that means splitting things up into package-sized chucks, probably something along these lines:
>>>
>>>  - MetacelloBrowser-Core
>>>  - MetacelloBrowser-Tests
>>>  - MetacelloBrowser-Pharo-Core
>>>  - MetacelloBrowser-Pharo-Tests
>>>
>>> then I'll be able to add:
>>>
>>>  - MetacelloBrowser-Squeak-Core
>>>  - MetacelloBrowser-Squeak-Tests
>>>
>>> as I work on the port ...
>>>
>>> Does this make sense to you or do you think we should take a different approach to naming the packages? We just have to stay within the package naming guidelines, i.e., being careful where we put '-'.
>>
>>
>> Yes, it makes sense. Now, we have to do such refactoring without opening the Monticello browser :-) The first step, is to create the packages from the class categories. Nice challenge isn't it ?
>>
>
> I confess that I had to resort to the Monticello Browser at times when working on Seaside30 yesterday:)  But we have to keep pushing this thing ...
>
>> Small things. When right-clicking on a baseline or version, there is two menu item 'add version' and 'add development version'. These are implemented in the classes MBAddVersionCommand and MBAddDevelopmentVersionCommand. I feel there is one which is redundant.
>> http://code.google.com/p/metacello/issues/detail?id=116
>>
>> Another bug:
>> Removing a version does not update the #development symbolic version.
>> http://code.google.com/p/metacello/issues/detail?id=117
>
> Another Seaside30 day for me, but I will have time while things are loading/running tests to poke at things...
>
> Dale
>

>> Here, Here! I don't think I really understood what this was for until I looked at the diffs and I agree that it should be used EVERYWHERE ... move a version of it to the command structure, because if you don't, I will ... I _am_ sick and tired of setting my username and password in the #storeVersion: method the debugger :)
>
>
> Ok, before I do so, I would like to understand how comes that you do not bump into this very problem of missing username.
>
> For example, when I want to load Mondrian, in essence, the following expression is executed when I press 'Load configuration':
> Gofer new
> squeaksource: 'Mondrian';
> package: 'ConfigurationOfMondrian';
> load.
> At that stage, the http repository is created but _without_ my username and password.
>
> When I want to create a new version of Mondrian or simply save the configuration, the browser tries to write in the Mondrian repository. But it fails since it does not know about my username/password. The idea of my fix, is to check if there is an user or not. And it asks for one if you want to add one.
>
> Don't you bump into this problem? Are you committing only in globally writable repositories?
>
> How could we accommodate this? Would it be sufficient to ask if you ever want to be asked for a password again when you say 'no' the first time a password is asked?
>
> Cheers,
> Alexandre

>>> Don't you bump into this problem? Are you committing only in globally writable repositories?
>>>
>>> How could we accommodate this? Would it be sufficient to ask if you ever want to be asked for a password again when you say 'no' the first time a password is asked
>
>
>> I see this all of the tim, so I agree with you that it is a problem...
>
> So, what would be a solution that fit both of us (and hopefully the whole Smalltalk community :-) ?
> When do you want to be asked for a username and password?
>
>
>> I don't use chat much, but there is a chat client somewhere around here and I have a chat id of some wort ... I'll dig around a bit and when I find it I'll let you know ...
>
>
> Ok, if you do not use chat client, then do not worry. We stick to emails then.
>
> Alexandre

>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>
>
> Version 1.52. I cleaned a lot as well.
> I added a new command: 'save all and add development version'
>
> Cheers,
> Alexandre
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>

> I'm planning to rename the classes in the Info hierarchy with and MB prefix MBInfo, etc. So let me know when would be a good time to do the rename since it can be a pain to merge...
>
> Dale
>
>
> On Mar 1, 2011, at 12:13 PM, Alexandre Bergel wrote:
>
>>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>>
>>
>> Version 1.52. I cleaned a lot as well.
>> I added a new command: 'save all and add development version'
>>
>> Cheers,
>> Alexandre
>>
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>>
>>
>
>

> Sure, please do. I won't touch the browser until your ping.
>
> Cheers,
> Alexandre
>
>
> On 1 Mar 2011, at 23:47, Dale Henrichs wrote:
>
>> I'm planning to rename the classes in the Info hierarchy with and MB prefix MBInfo, etc. So let me know when would be a good time to do the rename since it can be a pain to merge...
>>
>> Dale
>>
>>
>> On Mar 1, 2011, at 12:13 PM, Alexandre Bergel wrote:
>>
>>>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>>>
>>>
>>> Version 1.52. I cleaned a lot as well.
>>> I added a new command: 'save all and add development version'
>>>
>>> Cheers,
>>> Alexandre
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>>
>>>
>>
>>
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>

> done...
> On Mar 2, 2011, at 3:24 AM, Alexandre Bergel wrote:
>
>> Sure, please do. I won't touch the browser until your ping.
>>
>> Cheers,
>> Alexandre
>>
>>
>> On 1 Mar 2011, at 23:47, Dale Henrichs wrote:
>>
>>> I'm planning to rename the classes in the Info hierarchy with and MB prefix MBInfo, etc. So let me know when would be a good time to do the rename since it can be a pain to merge...
>>>
>>> Dale
>>>
>>>
>>> On Mar 1, 2011, at 12:13 PM, Alexandre Bergel wrote:
>>>
>>>>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>>>>
>>>>
>>>> Version 1.52. I cleaned a lot as well.
>>>> I added a new command: 'save all and add development version'
>>>>
>>>> Cheers,
>>>> Alexandre
>>>>
>>>> --
>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>> Alexandre Bergel  http://www.bergel.eu
>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>
>> --
>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>> Alexandre Bergel  http://www.bergel.eu
>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>
>>
>>
>>
>>
>
>

> Excellent. I created 1.53 that makes all the tests green.
> The new 'compare to current' function is amazing. Dale, this has really some value.
> For everybody who reads this, here an example:
>
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> OmniBrowser '1.2.4' to '1.2.1'
> Additions:
> Modifications:
> OCompletion
> '1.2.2' to '1.2.1'
> OB-Regex
> 'OB-Regex-lr.25' to 'OB-Regex-lr.24'
> OCompletion Tests
> '1.2.2' to '1.2.1'
> OB-Shout
> 'OB-Shout-GuillermoPolito.13' to 'OB-Shout-lr.8'
> OB-Morphic
> 'OB-Morphic-GuillermoPolito.153' to 'OB-Morphic-lr.145'
> OB-SUnitIntegration
> 'OB-SUnitIntegration-lr.39' to 'OB-SUnitIntegration-lr.37'
> OmniBrowser
> 'OmniBrowser-lr.503' to 'OmniBrowser-lr.493'
> OB-Refactory
> 'OB-Refactory-lr.244' to 'OB-Refactory-lr.232'
> OB-Standard
> 'OB-Standard-GuillermoPolito.531' to 'OB-Standard-lr.511'
> OB-Tests-Core
> 'OB-Tests-Core-lr.88' to 'OB-Tests-Core-lr.85'
> OB-Tests-Standard
> 'OB-Tests-Standard-lr.126' to 'OB-Tests-Standard-lr.123'
> OCForOB
> 'OCForOB-ul.5' to 'OCForOB-rr.2'
> Removals:
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> Cheers,
> Alexandre
>
> On 2 Mar 2011, at 13:17, Dale Henrichs wrote:
>
>> done...
>> On Mar 2, 2011, at 3:24 AM, Alexandre Bergel wrote:
>>
>>> Sure, please do. I won't touch the browser until your ping.
>>>
>>> Cheers,
>>> Alexandre
>>>
>>>
>>> On 1 Mar 2011, at 23:47, Dale Henrichs wrote:
>>>
>>>> I'm planning to rename the classes in the Info hierarchy with and MB prefix MBInfo, etc. So let me know when would be a good time to do the rename since it can be a pain to merge...
>>>>
>>>> Dale
>>>>
>>>>
>>>> On Mar 1, 2011, at 12:13 PM, Alexandre Bergel wrote:
>>>>
>>>>>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>>>>>
>>>>>
>>>>> Version 1.52. I cleaned a lot as well.
>>>>> I added a new command: 'save all and add development version'
>>>>>
>>>>> Cheers,
>>>>> Alexandre
>>>>>
>>>>> --
>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>>
>>>
>>
>>
>

> <GRIN> ... to go along with the compare to #stable menu item which produces something like the following, you can get mcz changes against #stable </GRIN>
>
> The update dev menu item now checks for dirtied packages that would be loaded over and prompts for several options to deal with this including allowing you to merge the modified packages and then update ...
>
> I think we're dang close to what Stef was looking for the other day
>
> Dale
>
>
> On 03/02/2011 09:50 AM, Alexandre Bergel wrote:
>> Excellent. I created 1.53 that makes all the tests green.
>> The new 'compare to current' function is amazing. Dale, this has really some value.
>> For everybody who reads this, here an example:
>>
>> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>> OmniBrowser '1.2.4' to '1.2.1'
>> Additions:
>> Modifications:
>> OCompletion
>> '1.2.2' to '1.2.1'
>> OB-Regex
>> 'OB-Regex-lr.25' to 'OB-Regex-lr.24'
>> OCompletion Tests
>> '1.2.2' to '1.2.1'
>> OB-Shout
>> 'OB-Shout-GuillermoPolito.13' to 'OB-Shout-lr.8'
>> OB-Morphic
>> 'OB-Morphic-GuillermoPolito.153' to 'OB-Morphic-lr.145'
>> OB-SUnitIntegration
>> 'OB-SUnitIntegration-lr.39' to 'OB-SUnitIntegration-lr.37'
>> OmniBrowser
>> 'OmniBrowser-lr.503' to 'OmniBrowser-lr.493'
>> OB-Refactory
>> 'OB-Refactory-lr.244' to 'OB-Refactory-lr.232'
>> OB-Standard
>> 'OB-Standard-GuillermoPolito.531' to 'OB-Standard-lr.511'
>> OB-Tests-Core
>> 'OB-Tests-Core-lr.88' to 'OB-Tests-Core-lr.85'
>> OB-Tests-Standard
>> 'OB-Tests-Standard-lr.126' to 'OB-Tests-Standard-lr.123'
>> OCForOB
>> 'OCForOB-ul.5' to 'OCForOB-rr.2'
>> Removals:
>> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>>
>> Cheers,
>> Alexandre
>>
>> On 2 Mar 2011, at 13:17, Dale Henrichs wrote:
>>
>>> done...
>>> On Mar 2, 2011, at 3:24 AM, Alexandre Bergel wrote:
>>>
>>>> Sure, please do. I won't touch the browser until your ping.
>>>>
>>>> Cheers,
>>>> Alexandre
>>>>
>>>>
>>>> On 1 Mar 2011, at 23:47, Dale Henrichs wrote:
>>>>
>>>>> I'm planning to rename the classes in the Info hierarchy with and MB prefix MBInfo, etc. So let me know when would be a good time to do the rename since it can be a pain to merge...
>>>>>
>>>>> Dale
>>>>>
>>>>>
>>>>> On Mar 1, 2011, at 12:13 PM, Alexandre Bergel wrote:
>>>>>
>>>>>>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>>>>>>
>>>>>>
>>>>>> Version 1.52. I cleaned a lot as well.
>>>>>> I added a new command: 'save all and add development version'
>>>>>>
>>>>>> Cheers,
>>>>>> Alexandre
>>>>>>
>>>>>> --
>>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>> --
>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>> Alexandre Bergel  http://www.bergel.eu
>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>
>
>

> Excellent. I created 1.53 that makes all the tests green.
> The new 'compare to current' function is amazing. Dale, this has really some value.
> For everybody who reads this, here an example:
>
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> OmniBrowser '1.2.4' to '1.2.1'
> Additions:
> Modifications:
> OCompletion
> '1.2.2' to '1.2.1'
> OB-Regex
> 'OB-Regex-lr.25' to 'OB-Regex-lr.24'
> OCompletion Tests
> '1.2.2' to '1.2.1'
> OB-Shout
> 'OB-Shout-GuillermoPolito.13' to 'OB-Shout-lr.8'
> OB-Morphic
> 'OB-Morphic-GuillermoPolito.153' to 'OB-Morphic-lr.145'
> OB-SUnitIntegration
> 'OB-SUnitIntegration-lr.39' to 'OB-SUnitIntegration-lr.37'
> OmniBrowser
> 'OmniBrowser-lr.503' to 'OmniBrowser-lr.493'
> OB-Refactory
> 'OB-Refactory-lr.244' to 'OB-Refactory-lr.232'
> OB-Standard
> 'OB-Standard-GuillermoPolito.531' to 'OB-Standard-lr.511'
> OB-Tests-Core
> 'OB-Tests-Core-lr.88' to 'OB-Tests-Core-lr.85'
> OB-Tests-Standard
> 'OB-Tests-Standard-lr.126' to 'OB-Tests-Standard-lr.123'
> OCForOB
> 'OCForOB-ul.5' to 'OCForOB-rr.2'
> Removals:
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> Cheers,
> Alexandre
>
> On 2 Mar 2011, at 13:17, Dale Henrichs wrote:
>
>> done...
>> On Mar 2, 2011, at 3:24 AM, Alexandre Bergel wrote:
>>
>>> Sure, please do. I won't touch the browser until your ping.
>>>
>>> Cheers,
>>> Alexandre
>>>
>>>
>>> On 1 Mar 2011, at 23:47, Dale Henrichs wrote:
>>>
>>>> I'm planning to rename the classes in the Info hierarchy with and MB prefix MBInfo, etc. So let me know when would be a good time to do the rename since it can be a pain to merge...
>>>>
>>>> Dale
>>>>
>>>>
>>>> On Mar 1, 2011, at 12:13 PM, Alexandre Bergel wrote:
>>>>
>>>>>> If I wasn't clear in my original message, I was in favor of your edits and I think that I like your solution and I think that it should applied to all of the places were we do saves ...
>>>>>
>>>>>
>>>>> Version 1.52. I cleaned a lot as well.
>>>>> I added a new command: 'save all and add development version'
>>>>>
>>>>> Cheers,
>>>>> Alexandre
>>>>>
>>>>> --
>>>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>>>> Alexandre Bergel  http://www.bergel.eu
>>>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>> --
>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
>>> Alexandre Bergel  http://www.bergel.eu
>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>>>
>>>
>>>
>>>
>>>
>>
>>
>
> --
> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
> Alexandre Bergel  http://www.bergel.eu
> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
>
>
>
>
>