Hi!
I think I reached a first milestone in my quest to have consistent and traceable version in Moose. Version 4.3.3 is now defined.
I produced the versions:
- DSM 2.1
- RPackage 1.0
- Glamour 2.1
- SmallDude 2.1
I created version 4.3.2. and 4.3.3 of Moose.
- 4.3.2 uses the 'default' version of SmallDude. It is not necessary, but it helped me to track down some problem.
- 4.3.3 does not depend on any baseline. This is an important step since it means that we should be able to load 4.3.3 of Moose in 10 years from now, and so, whatsoever you will commit in the dependent packages.
So, at the next commit of any packages, I will produce version 4.3.4 and so on.
The configurations were full of problems (e.g., PetitJava that does not require PetitParser for example). This means that the validator of Metacello seriously needs to be improved. It took me some time to identify and fix all this.
I worked with Pharo1.2OneClick image. I haven't tried with Core.
Now, the question is, should we create a hudson project to try the stable version? To make my life simpler, I always assign the last version as the stable version. I hope this is ok with you.
Comments are welcome
Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.
What is the purpose of the first argument of this method?
MOViewRendered>>
edges: aCollection from: aFromBlock toAll: aToBlock
"aToBlock should return a collection, edges will be drawn from the element returned by aFromBlock to this collection"
^self edges: aCollection from: aFromBlock toAll: aToBlock using: self edgeShape
nicolas
All work well with class refactoring, the problem is that when opening a
moose image, RPackageOrganizer has not yet registered to system
announcements. This should be done in the initialization of the class
RPackageOrganizer.
If I do myself a 'ConfigurationOfRPackage loadDefault' in a fresh pharo
image, the organizer correctly register to announcements.
Could it be a problem with metacello that does not correctly initialize
classes ?
2011/2/15 Tudor Girba <tudor(a)tudorgirba.com>
> Don't forget to write a test :)
>
> Doru
>
>
> On 15 Feb 2011, at 11:36, Cyrille Delaunay wrote:
>
> > Indeed there is something wrong. I will try to fix that today
> >
> > 2011/2/15 Tudor Girba <tudor(a)tudorgirba.com>
> > I did not try recently. But if I remembered correctly, I just performed a
> rename refactoring and then checked the RPackage model.
> >
> > Cheers,
> > Doru
> >
> >
> > On 15 Feb 2011, at 11:14, Cyrille Delaunay wrote:
> >
> > > Do you still have this problem?
> > >
> > > 2011/1/31 Cyrille Delaunay <cy.delaunay(a)gmail.com>
> > > Thanks, I will look at that
> > >
> > > 2011/1/31 Tudor Girba <tudor(a)tudorgirba.com>
> > >
> > > Hi Cyrille,
> > >
> > > I noticed a case in which the RPackage did not seem to refresh after
> performing a class rename refactoring.
> > >
> > > I did not have a chance to write a test case for this, but I think it
> probably affects the importing as well.
> > >
> > > Cheers,
> > > Doru
> > >
> > >
> > > --
> > > www.tudorgirba.com
> > >
> > > "No matter how many recipes we know, we still value a chef."
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> >
> > --
> > www.tudorgirba.com
> >
> > "Reasonable is what we are accustomed with."
> >
> >
>
> --
> www.tudorgirba.com
>
> "Every thing has its own flow."
>
>
>
>
>
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 530 by vuqui...(a)vub.ac.be: Glamour: highlighting source code
does not work anymore
http://code.google.com/p/moose-technology/issues/detail?id=530
Text presentations do not highlight Smalltalk methods' source code
anymore. All the content appears in red.
This happens with the last changes of PharoCore 1.2.
I am using both #forSmalltalk #forSmalltalk:
* Type-Defect
* Component-Glamour
Status: New
Owner: ----
CC: georgega...(a)gmail.com
Labels: Type-Defect Priority-Medium Component-ExternalTools
New issue 538 by tudor.gi...(a)gmail.com: inFusion should recover the
references between classes
http://code.google.com/p/moose-technology/issues/detail?id=538
When a class name appears in the source code (not in the type declaration),
it should be modeled as a FAMIX.Reference.
Status: New
Owner: ----
CC: georgega...(a)gmail.com
Labels: Type-Defect Priority-Medium Component-ExternalTools
New issue 534 by tudor.gi...(a)gmail.com: inFusion should export the
initialization of attributes in Java static constructs
http://code.google.com/p/moose-technology/issues/detail?id=534
static
{
RATE_CODES = new HashSet<String>(3);
RATE_CODES.add(RATE_ASSITANT_SERVICE);
RATE_CODES.add(RATE_MEDICAL_SERVICE);
RATE_CODES.add(RATE_TECHNICAL_SERVICE);
}
Hi,
Showing a method' source code in text presentations does not highlight anymore. All appear in red.
Indeed, this happens with the last actualizations of PharoCore 1.2.
I just noticed that this was already fixed in Mondrian-easel. Could you also fix it in Glamour? thanks :)
Regards,
Veronica
Hi!
Just wondering why a success is represented as a blue bullet and not a green one?
Cheers,
Alexandre
--
_,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:
Alexandre Bergel http://www.bergel.eu
^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.