How do you think ruby parses this expression?
a = 1 + b = 2 + c = 4 + d = 8
Originally I thought that operator precedence would result in
a = (1 + b) = (2 + c) = (4 + d) = 8
and since an assignment requires a variable, that would result in a
SyntaxError.
But instead it seems that everything on the right side of the assignment
operator is evaluated first:
a = (1 + (b = (2 + (c = (4 + (d = 8))))))
That had me really puzzled at first but I guess the behavior makes a
certain
sense, and it’s more useful than a syntax error. Ruby’s tolerant parser
wins again!
Daniel
Daniel DeLorme wrote:
How do you think ruby parses this expression?
a = 1 + b = 2 + c = 4 + d = 8
Per…
http://phrogz.net/ProgrammingRuby/language.html#operatorexpressions
…the + operator has much higher precedence than the assignment
operator. That’s interesting that it doesn’t parse as you expected.
It’s what I would have expected, too.
Hi,
In message “Re: operator precedence of assignment”
on Tue, 2 Jan 2007 22:26:53 +0900, Daniel DeLorme [email protected]
writes:
|
|How do you think ruby parses this expression?
| a = 1 + b = 2 + c = 4 + d = 8
|
|Originally I thought that operator precedence would result in
| a = (1 + b) = (2 + c) = (4 + d) = 8
|and since an assignment requires a variable, that would result in a SyntaxError.
|But instead it seems that everything on the right side of the assignment
|operator is evaluated first:
| a = (1 + (b = (2 + (c = (4 + (d = 8))))))
|
|That had me really puzzled at first but I guess the behavior makes a certain
|sense, and it’s more useful than a syntax error. Ruby’s tolerant parser wins again!
Ah, for your information, that surprised me as well. Perhaps bison
generated parser is smarter and eagerer than I expected. I have no
idea what to do.
matz.
On 02.01.2007 17:56, Yukihiro M. wrote:
|and since an assignment requires a variable, that would result in a SyntaxError.
|But instead it seems that everything on the right side of the assignment
|operator is evaluated first:
| a = (1 + (b = (2 + (c = (4 + (d = 8))))))
|
|That had me really puzzled at first but I guess the behavior makes a certain
|sense, and it’s more useful than a syntax error. Ruby’s tolerant parser wins again!
Ah, for your information, that surprised me as well. Perhaps bison
generated parser is smarter and eagerer than I expected. I have no
idea what to do.
I think Daniel did not want you to do anything - in fact he seems rather
positively amazed. 
(The only thing that comes to mind is thank him for the praise. :-))
Kind regards
robert
On 02.01.2007 18:09, Yukihiro M. wrote:
|(The only thing that comes to mind is thank him for the praise. :-))
The praise should go to yacc, not me. 
But you chose it, didn’t you? So…
robert
Hi,
In message “Re: operator precedence of assignment”
on Wed, 3 Jan 2007 02:05:06 +0900, Robert K.
[email protected] writes:
|I think Daniel did not want you to do anything - in fact he seems rather
|positively amazed. 
That made me wonder what I should do.
|(The only thing that comes to mind is thank him for the praise. :-))
The praise should go to yacc, not me. 
matz.