-dip25 switch: time to make it always on?

Dicebot via Digitalmars-d digitalmars-d at puremagic.com
Sat Jul 2 02:55:35 PDT 2016


On 06/29/2016 02:53 PM, Walter Bright wrote:
> On 6/29/2016 3:02 AM, Dicebot wrote:
>> See this d.learn discussion
>> (http://forum.dlang.org/thread/vtvtukooicwspxzzoxtt@forum.dlang.org)
>> for example
>> of diagnostics problems I am talking about.
> 
> Please post bug reports to bugzilla. Brad set up bugzilla because using
> the n.g. as a buglist was a complete failure.

I have submitted a short summary of that ng.learn thread here :
https://issues.dlang.org/show_bug.cgi?id=16228

My overall opinion of DIP25 is that it must not become part of language
mainline for two reasons:

1) Implementation is not sufficiently mature (mostly diagnostics issues)
2) It targets a very niche use case and is too much of a language burden
on its own - there are no interesting designs one can do with DIP25
alone to "sell" the feature.

In my opinion it should sit in a preview state until `return scope` or
whatever replaces it gets finalized and tested and merged in main
language simultaneously as parts of same feature.



More information about the Digitalmars-d mailing list