ScopedValue.runWhere not returning scope

Andrew Haley aph-open at littlepinkcloud.com
Fri Jun 21 12:51:07 UTC 2024


On 6/21/24 09:11, Marcin Grzejszczak wrote:
 >  > Who did you think you'd been talking to?
 >
 > I'm sorry, but you don't have to be unpleasant.

I thought it was a fair question.

 > I thought that this mailing list should be used for exactly the
 > reason related to giving feedback on preview features such as scoped
 > values. I understand that this thread is extremely long but the
 > problem is complex in my opinion. If you're unwilling to continue
 > this discussion of get feedback about the features you're working on
 > then just say so.

It is. What you're hearing here is disagreement. That is the exact
opposite of unwillingness to have a discussion. It is participation.

 > Afair some people in this thread stated that they are not working on
 > the JDK. Some, like you, if I'm not mistaken, have not addressed
 > most of my arguments.

You are mistaken. I believe I've addressed them.

 > I've only heard that for security reasons (scopes could be opened
 > and closed out of order)

Yes. That is very important.

 > and immutability (I don't understand how the api changes would
 > modify that.

If you can set a value, then call a method that sets a new binding for
that value, and that value is different after the method returns, then
the value has changed. You've lost immutability.

 > It simply allows me to have more control.

The purpose of (shallow) immutability and one-way transmission of data
from caller to callee is to *restrict* and limit control. Just as the
purpose of structured programming is to restrict the generality of
goto statements in order to make programs easier to reason about.

 > What currently is called in the lambda would be manually called by
 > me. There are no changes here unless I'm mistaken) you are reluctant
 > to consider my feedback.

Of course I've considered it. Not only have I considered it, I have
responded here, at length.

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671





More information about the loom-dev mailing list