Implied readability + layers

[Peter Levart]

On 11/05/2015 09:42 PM, Ali Ebrahimi wrote:
> Hi,
>
> On Thu, Nov 5, 2015 at 11:43 PM, Peter Levart <peter.levart at gmail.com 
> <mailto:peter.levart at gmail.com>> wrote:
>
>
>
>     On 11/05/2015 07:39 PM, Ali Ebrahimi wrote:
>>     Hi,
>>     So with current implementation of Implied readability w.r.t. layers and
>>     unexpected consequences of its remove or addition, I strongly propose to
>>     drop it from current design.
>
>     Unless the rule is very simple. The implied readability should be
>     no different than explicit readability.
>
>     ...in all possible configurations of layers and classloaders.
>
> Exactly. I agree.
> Now suppose we have foo, bar at 1 in layer1 and baz and bar at 2 in layer2.
>
> situation1:
>
> module foo {
>     requires public bar;
>     ...
> }
>
> module baz {
>     requires foo;
>     ...
> }
>
> The readability graph for configuration1/layer1 is:
>   bar at 1 reads java.base
>   foo reads java.base
>   foo reads bar at 1
>
> The readability graph for configuration2/layer2 is:
>  bar at 2 reads java.base
>  baz reads foo
>  baz reads bar at 1==================
>
> But situation2:
> module foo {
> requires bar;
>     ...
> }
>
> module baz {
> requires foo;
> requires bar;
>     ...
> }
>
> The readability graph for configuration1/layer1 is:
>   bar at 1 reads java.base
>   foo reads java.base
>   foo reads bar at 1
>
> The readability graph for configuration2/layer2 is:
>  bar at 2 reads java.base
>  baz reads foo
>  baz reads bar at 2==================
>
>
>
>     If implied readability is taken "symbolically", the same caveats
>     apply to one or the other form - they are just different forms of
>     expressing the same thing.
>
> As you can see this is currently is not case.

What I'm trying to say is that it should be the same. situation1 should 
construct the same readability graph as situation2 (which is the right one).

>
> But with one simple rule "Always current layer's version of module win 
> over parent layer's one" we would have same readability graphs for 
> both situations.

Right. I would say it even simpler: take the implicitly and explicitly 
required module names "symbolically" and evaluate them in the context of 
each module.

This way adding a "public" keyword is always safe and the existence of 
"public" keyword can not harm anybody. Removing it is of course not safe 
and can not be made safe.

Regards, Peter

>
>
>
> -- 
>
> Best Regards,
> Ali Ebrahimi