Issue 4952 in pharo: Browse senders from a flat message list leads to MessageNotUnderstood

Status: Accepted
Owner: [hidden email]

New issue 4952 by [hidden email]: Browse senders from a flat message  
list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

Open a Workspace, type 'DiffChangeMorph', select it, browse class  
references (CMD+SHIFT+N)
In the message list, select pop-up menu 'Senders of...' and...
MessageNotUnderstood Array>>selector

Note that the short cut (CMD+N) seems to work...

Guys, I was trying to debug another tool, but it's just too upsetting...
I feel like endlessly whining ;)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #1 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

FlatMessageListBrowser>>browseSenders
        self browseSendersOfMessagesFrom: selectedItem

AbstractTool>>browseSendersOfMessagesFrom: aMethod

Hey it expects aMethod, but gets an Array in the selectedItem, something  
like {class. selector}
A little voice tells me: avoid programming with Array, these are poor man's  
objects.
(unless you program in Perl, Lisp or Haskell of course)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Cc: [hidden email]

Comment #2 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

Yes the change with the dual list to display class and selectors broke a  
lot of assumptions.
We should not have included it. Yes Arrays suck.
Benjamin why don't we use rgMethodDefinition?


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Labels: Importance-High

Comment #3 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

We should fix this situation it kills us when the tools do not work.


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #4 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

Ok I will buy some food and I will try to get this tools manipulating  
rgMethodDefinition.
Ben if you wake up/start to do something let me know


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #5 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

I will do a refactor to have an abstraction to hide the implementation of  
items (here, the array)

I think I can do it cleaner, I will have a look this evening (cause this  
weekend is the LIlle Comics Festival ;) )


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #6 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

I think the way to go is to keep the list of methods as it is (with  
RGMethod or whatever) and to have a method which can build the list of  
arrays for the representation.


So I think I should improve the listClass to have a way to ask the model  
how to "cut" the real object to have the columns.
Let's try this quickly


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: FixReviewNeeded

Comment #7 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

So here is a better implementation.

So the model keeps the real object (which is cleaner and cleaner) and have  
to provide a method to "cut" each object in columns.

And then the morph take the real object from the model and cut it to build  
is own list representation

So now, the FlatListMessageBrowser is manipulating RGMethod and have multi  
column display :)

(I am a bit hurry, so I may have forgot some reference to the former array  
struct (as "first" and so on) but I have tested menu entries and buttons)

Attachments:
        ImproveMultiColumn.1.cs  13.7 KB


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #8 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

cool idea
I was reading multiColumnLazyListMorph and I will have a look at your fix.
I like the idea that we pass the selectors of how to cut the object quite  
nice.
a real abstraction: I contained complex data you can cut me like that.  
Cool :)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #9 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

What I see is that it cleans a lot of points! Good!



_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #10 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

Ok I see you pass the name of the method that should return information and  
this method should pass an array (because Smalltalk does not have multiple  
result methods.

Now the method methodModified looks strange to me. Why selectedItem is  
still a pair?
I will load the system and play with it tomorrow morning (now I'm trashing).

In any case the level of complexity decreased so this is clearly showing  
that the solution is good.


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #11 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

You're right, it's because I was barely asleep this morning when I did it :)

I attached a new version which fix that, and remove the old one for a  
better visibility :)


I also think it's the good way to do it :)

Attachments:
        ImproveMultiColumn.3.cs  14.0 KB


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #12 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

Thanks I will integrate it now.

Stef


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: Closed

Comment #13 on issue 4952 by [hidden email]: Browse senders from a  
flat message list leads to MessageNotUnderstood
http://code.google.com/p/pharo/issues/detail?id=4952

in 14218


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker