Send NumPy-Discussion mailing list submissions to
numpy-discussion@scipy.org
To subscribe or unsubscribe via the World Wide Web, visit
http://mail.scipy.org/mailman/listinfo/numpy-discussion
or, via email, send a message with subject or body 'help' to
numpy-discussion-request@scipy.org
You can reach the person managing the list at
numpy-discussion-owner@scipy.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of NumPy-Discussion digest..."
Today's Topics:
1. Re: [help needed] associativity and precedence of '@'
(Nathaniel Smith)
2. Re: GSoC project: draft of proposal (Julian Taylor)
3. Re: [help needed] associativity and precedence of '@'
(Christophe Bal)
4. Re: [help needed] associativity and precedence of '@'
(Alexander Belopolsky)
5. Re: [help needed] associativity and precedence of '@' (Bago)
6. Re: [help needed] associativity and precedence of '@'
(Christophe Bal)
7. Re: [help needed] associativity and precedence of '@'
(Christophe Bal)
8. Re: [help needed] associativity and precedence of '@'
(Nathaniel Smith)
----------------------------------------------------------------------
Message: 1
Date: Mon, 17 Mar 2014 22:02:33 +0000
From: Nathaniel Smith <njs@pobox.com>
Subject: Re: [Numpy-discussion] [help needed] associativity and
precedence of '@'
To: Discussion of Numerical Python <numpy-discussion@scipy.org>
Message-ID:
<CAPJVwB=zBazN+fiYWJeiWOL=4a9Bf2XGxJGoTT8GFTt-kDUDZw@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
On Mon, Mar 17, 2014 at 9:38 PM, Christophe Bal <projetmbc@gmail.com> wrote:
Here is the translation. ;-)
Hello,
and what about something like that ?
a @ b @ c -> (a @ b) @ c
a * b @ c -> (a * b) @ c
a @ b * c -> a @ (b * c)
Easy to remember: the *-product has priority regarding to the @-product, and
we just do @-product from left to right.
In the terminology we've been using in this thread, this is "weak-left".
An advantage of this is that most parsers do analyze from left to right.
So I really think that it is a better choice than the weak-right one.
We've mostly ignored this option because of assuming that if we want
left-associativity, we should go with "same-left" instead of
"weak-left". Same-left is:
a @ b @ c -> (a @ b) @ c
a * b @ c -> (a * b) @ c
a @ b * c -> (a @ b) * c
i.e., even more left-to-right than weak-left :-)
Do you think weak-left is better than same-left?