Hi,
On 12 Feb 2011, at 21:17, Simon Denier wrote:
On 12 févr. 2011, at 21:09, Tudor Girba wrote:
I looked a bit into the problem, and the problem is generated by the changes in RPackage which now does not match sub-categpries. The behavior from RPackage is the wanted one, and thus, it is the tests that should be changed.
Could you be a bit more specific about the problem.
If I remember well, the purpose of the category importer is to consider each category and subcategory as a package.
Well, the problem was that calling addFromPackageNamed: on a CategoryImporter falls back to working with the PackageImporter, because the method is not implemented in the CategoryImporter. Before, it worked fine because it looked in the package and all subcategories. Now, there aren't any classes in the root package if the package is empty and the categories are not looked up anymore, so you have to explicitly look for all the category names, or use addFromCategoriesMatching:.
I think it's fine like it is now. Actually, I think that categories should go away now that we have RPackage which matches categories one-to-one.
Cheers, Doru
I fixed the problem in the MooseSqueakClassPackageImporterTest>>testCategoryImporter.
There are still 7 impacted tests in DSM. Jannik could you fix them please?
Cheers, Doru
On 12 Feb 2011, at 15:12, Tudor Girba wrote:
Hi,
It looks like the category importer is broken. Does anyone have time to take a look?
Cheers, Doru
-- www.tudorgirba.com
"Value is always contextual."
-- www.tudorgirba.com
"Relationships are of two kinds: those we choose and those that happen. They both matter."
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- Simon Denier
Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev
-- www.tudorgirba.com
"Reasonable is what we are accustomed with."