Hi guys

we have some deadlines coming and esteban got some deadline last week, igor is 
diving into the new text model and doing a large pass on it and we should be 
back on hacking mode.

Stef


On Jun 13, 2013, at 2:31 PM, Goubier Thierry <[email protected]> wrote:

> However, if you write the extension category as *announcements-core, then the 
> bug you describe does not happen.
> 
> (Note: with my AltBrowser, you don't write the '*announcements-core' 
> category, you drag the new method on the extension part of the 
> Announcements-Core package, and the right protocol will be created for you)
> 
> I think that overall we have a conflict with overlapping names in packages (a 
> package with subcategories or: two packages, the first one with as name the 
> prefix of another packages name) and I'd like a decisive resolution on that.
> 
> In short, when you write *announcements-core-subcategory, to you want it to 
> be:
> 1)a subcategory of *announcements-core, or
> 2)for a package named announcements-core-subcategory,
> 3)or a subcategory of announcements ?
> 
> (I guess you want 1), and the system 2), but your change may result in 3)
> 
> I like it that this code rely on category:matches: whose code is repeated 
> four times in my 2.0 image:
> 
> PackageInfo>>category:matches:
> RGContainer>>category:matches:
> RPackage>>category:matches:
> RPackageOrganizer>>category:matches:
> 
> Thierry
> 
> Le 11/06/2013 21:11, Sean P. DeNigris a écrit :
>> This is a nasty one. It may lead to lost code, without warning or
>> notification. Applies to latest 2.0 and 3.0:
>> 1. In Nautilus, in any class outside Announcements-Core, add protocol
>> '*announcements-core-subcategory'
>> 2. Open MCWrokingCopyBrowser. You will see that Announcements-Core is dirty
>> 3. Click the "Changes" button
>> 
>> It will say that there are no changes, and the package will be unmarked as
>> dirty. This is because 'announcements-core-subcategory' was created as a
>> separate RPackage and stole the extension method.
>> 
>> For 3.0,
>> https://pharo.fogbugz.com/f/cases/10887/Compound-Extension-Methods-Unpackaged
>> 
>>> Fix in inbox:
>>> SLICE-Issue-10887-Compound-Extension-Methods-Unpackaged-SeanDeNigris.1
>>> 
>>> Case 10887: Compound Extension Methods Unpackaged
>>> 
>>> * Fix RPackageOrganizer>>checkPackageExistsOrRegister:, which is sent when
>>> a class organization changes, to see if extension methods have been added
>>> * Add test which fails before fix and passes after
>> 
>> For 2.0,
>> https://pharo.fogbugz.com/f/cases/10888/Backport-2-0-Compound-Extension-Methods-Unpackaged
>> 
>>> Fix in inbox:
>>> SLICE-Issue-10888-Backport-20-Compound-Extension-Methods-Unpackaged-SeanDeNigris.1
>> 
>> 
>> 
>> 
>> 
>> -----
>> Cheers,
>> Sean
>> --
>> View this message in context: 
>> http://forum.world.st/BUG-Extension-Methods-Missing-from-Packages-tp4692842.html
>> Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.
>> 
>> 
> 
> -- 
> Thierry Goubier
> CEA list
> Laboratoire des Fondations des Systèmes Temps Réel Embarqués
> 91191 Gif sur Yvette Cedex
> France
> Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95
> 


Reply via email to