[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-lambda-dev
Subject:    Re: Lambda syntax puzzler
From:       Remi Forax <forax () univ-mlv ! fr>
Date:       2013-10-17 20:28:01
Message-ID: 52604851.2070500 () univ-mlv ! fr
[Download RAW message or body]

On 10/17/2013 12:10 PM, Sam Pullara wrote:
> I’m ok with this. It is trivial to do the transformation from an inner class that \
> mutates its parameters to a lambda that doesn’t. 
> I still would like the variable name scoping to change though at this late date I \
> don’t expect that to happen. It is very common in the Scala code bases I have seen \
> to reuse the outside name as the inside name, now there will be some ugly \
> convention that adds a number or a _ or something.

yes, I agree, it's very painful.
I fail to convince the other EG members that it was an important issue 
during the latest face to face meeting.
I still hope that when people will start using new APIs that massively 
use lambdas, this rule will be removed.

> 
> Sam

Rémi

> 
> On Oct 16, 2013, at 7:05 AM, Remi Forax <forax@univ-mlv.fr> wrote:
> 
> > On 10/16/2013 03:43 PM, Millies, Sebastian wrote:
> > > Yes, I like that proposal as well. Only thing could suffer would be automatic \
> > > conversion of inner classes to lambdas, if those class implementations mutated \
> > > their parameters. That would not concern me. -- Sebastian
> > I think that the main issue I see, is that there is no Java keyword for
> > saying 'not final',
> > so if by example you have a method that calls a lambda with the bounds
> > of a partition of an array, you can not write a code like this:
> > (array, i, end) -> {
> > for(; i < end; i++) {
> > ... array[i] ...
> > }
> > }
> > 
> > With the proposal of Stuart, you need to introduce a new local variable
> > (array, start, end) -> {
> > for(int i = start; i < end; i++) {
> > ... array[i] ...
> > }
> > }
> > 
> > That's said, it's not a big issue
> > 
> > cheers,
> > Rémi
> > 
> > > -----Original Message-----
> > > From: lambda-dev-bounces@openjdk.java.net \
> > >                 [mailto:lambda-dev-bounces@openjdk.java.net] On Behalf Of Zhong \
> > >                 Yu
> > > Sent: Wednesday, October 16, 2013 2:31 PM
> > > To: Stuart Marks
> > > Cc: lambda-dev@openjdk.java.net
> > > Subject: Re: Lambda syntax puzzler
> > > 
> > > On Tue, Oct 15, 2013 at 1:34 PM, Stuart Marks <stuart.marks@oracle.com> wrote:
> > > > Either that, or make lambda formal parameters implicitly final. I've
> > > > always
> > > That is apparently a good idea with no objections; how come EG isn't adopting \
> > > it? what are the concerns? 
> > > > hated code that mutated its parameters anyway.
> > > > 
> > > > s'marks
> > > > 
> > > > On 10/12/13 6:50 AM, Samir Talwar wrote:
> > > > > > -D
> > > > > 
> > > > > Now that traditional C-style `for` loops are a thing of the past, I
> > > > > think we should expunge the increment and decrement operators from the \
> > > > > language. Python had the right idea.
> > > > > 
> > > > > — Samir.
> > > > > 
> > > > > 
> > > > > On Sat, Oct 12, 2013 at 1:27 PM, Remi Forax <forax@univ-mlv.fr> wrote:
> > > > > 
> > > > > > I'm sure that people of this list will be able to see the beauty of
> > > > > > the following code produced by one of my students.
> > > > > > IntPredicate p = i -> i --> 0;
> > > > > > 
> > > > > > Rémi
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > > > > 
> > > Software AG – Sitz/Registered office: Uhlandstraße 12, 64297 Darmstadt, Germany \
> > > – Registergericht/Commercial register: Darmstadt HRB 1562 - Vorstand/Management \
> > > Board: Karl-Heinz Streibich (Vorsitzender/Chairman), Dr. Wolfram Jost, Arnd \
> > > Zinnhardt; - Aufsichtsratsvorsitzender/Chairman of the Supervisory Board: Dr. \
> > > Andreas Bereczky - http://www.softwareag.com 
> > > 
> > 


[prev in list] [next in list] [prev in thread] [next in thread]