'dirty' Etoys package in 6704?

Adrian Lienhard adi at netstyle.ch
Thu Dec 8 19:19:56 UTC 2005


Probably it means that some other package removed/changed methods in  
EToys package, but an up to date version of EToys package was not  
saved and loaded. This should not be. Loading the last EToys package  
would revert the changes!

This is a packaging issue. However, the other dirty packages are  
probably caused by some MC bug. At least its not obvious why they  
should have the modified flag set. Just going through the list and  
click on "changes" does not solve the real problem, and, when loading  
a series of updates from the stream, it will still break (i.e., the  
user is asked if he really wants to override the changes). It would  
be nice if somebody could fix this, instead.

Something else I run into when working on traits: There are  
deprecated methods in Object that, in my (older) version of the  
kernel, are in 38Deprecated package but now are not anymore. Examples  
are: Object>>beep, Object>>isKindOf:thenDo:

So, that would be some work that is needed to be done - apart from  
the MC bug, each in the range of 15 minutes... ;-)

Adrian

On Dec 8, 2005, at 19:36 , Cees De Groot wrote:

> The published 6704 image has dirty packages - probably we should run
> 'Changes' on dirty package to remove the dirty flag before uploading,
> because most of the 'dirty' markers in the MC browser go away when you
> let MC do a compare.
>
> However, EToys-stephaneducasse.2 stays dirty and has an appreciable
> list of changes. How comes?




More information about the V3dot9 mailing list