Keep deprecated methods for one or two versions? (was Re: Aboutdeprecated methods)

Doug Way dway at riskmetrics.com
Fri Sep 12 22:52:26 UTC 2003


ducasse wrote:

>> We could wait two versions if others feel strongly about it, though.
>
>
> Me too :) 


By "Me too" I assume you'd vote to wait two versions?

How do others feel about this?  I still slightly favor waiting just one 
version, but if most folks think we should wait two versions, we could 
do that.

> In fact I would favor to remove them all and create a package with 
> them so that people can still load it for documentation purpose.


That might be a good idea.  I'd be willing to do that myself if we 
decide to keep deprecated methods for just one version.  If we keep them 
for two versions, I personally wouldn't bother to save them in a 
package, but someone else could do that if they wanted.

>> Speaking of deprecated methods, someone mentioned awhile ago that 
>> they thought the deprecation method should be renamed from 
>> #deprecatedExplanation: to just #deprecated:.  I think this would be 
>> a good idea.
>
>
> if someone wants no problem with us. But this will break our pending 
> changes :) but someone can do it.
> Now two methods for deprecation MUST be there one for ordinary methods 
> and one with a block as parameter for primitives because we cannot 
> insert in the method body before the primitive tag a deprecation 
> statement.
> In the next batch of changes I have move around primitives and you 
> will see some examples of the second methods.


Okay, we could do this then.  I could wait until your changes are 
incorporated before changing #deprecatedExplanation: to #deprecated, or 
I could update your changesets to use #deprecated: as I incorporate 
them, to make it easier for you.

Also, we should probably change the other method with the block 
parameter to be consistent with the simple one, so that the explanation 
is always the first parameter and then extra parameters follow.  Maybe 
change it to #deprecated:block:, or something like that.

- Doug




More information about the Squeak-dev mailing list