Erlang › Erlang Questions

If one guard expression in the guard fails, are the remaining expressions still evaluated?

Classic

List

Threaded

5 messages Options

Alexey Romanov

If one guard expression in the guard fails, are the remaining expressions still evaluated?

It seems natural to me that the answer should be no, because this
evaluation can't change the result and it doesn't have any side
effects by definition of guards. But while the reference manual
(http://www.erlang.org/doc/reference_manual/expressions.html)
explicitly says the rest of guards are not evaluated if one of them
succeeds, it doesn't say whether guards fail fast.

Yours, Alexey Romanov

________________________________________________________________
erlang-questions (at) erlang.org mailing list.
See http://www.erlang.org/faq.html
To unsubscribe; mailto:[hidden email]

Steve Davis

Re: If one guard expression in the guard fails, are the remaining expressions still evaluated?

Take a look at section 7.14 (Short-Circuit Expressions) as well as 7.24 of that chapter. It should clarify.

/s

Alexey Romanov

Re: If one guard expression in the guard fails, are the remaining expressions still evaluated?

I've asked precisely because reading section 7.24 didn't clarify it
for me. The relevant sentences are "The guard is true if all guard
expressions evaluate to true." and "If an arithmetic expression, a
boolean expression, a short-circuit expression, or a call to a guard
BIF fails (because of invalid arguments), the entire guard fails." The
first sentence suggests to me that all expressions are evaluated; the
second that expressions after the one that failed aren't. But both
aren't explicit.

I know about short-circuit expressions, but 7.14 doesn't say anything
about evaluation of guards (unsurprisingly).

Yours, Alexey Romanov

On Sun, Feb 20, 2011 at 4:31 PM, Steve Davis
<[hidden email]> wrote:

>
> Take a look at section 7.14 (Short-Circuit Expressions) as well as 7.24 of
> that chapter. It should clarify.
>
> /s
>
> --
> View this message in context: http://erlang.2086793.n4.nabble.com/If-one-guard-expression-in-the-guard-fails-are-the-remaining-expressions-still-evaluated-tp3315062p3315110.html
> Sent from the Erlang Questions mailing list archive at Nabble.com.
>
> ________________________________________________________________
> erlang-questions (at) erlang.org mailing list.
> See http://www.erlang.org/faq.html
> To unsubscribe; mailto:[hidden email]
>
>

________________________________________________________________
erlang-questions (at) erlang.org mailing list.
See http://www.erlang.org/faq.html
To unsubscribe; mailto:[hidden email]

Lukas Larsson-4

Re: If one guard expression in the guard fails, are the remaining expressions still evaluated?

The second sentance refers to the fact that if something in a guard throws an exception (i.e. fails) then the entire guard will fail. It does not refer to a guard statement returning false. If a guard statement does return false it will not continue executing if , are used, but it will of course do so if ; are used in the guard. You could of course skip the usage of , and ; and use andalso and orelse if you want control yourself of what will happen.

Lukas
----- Original Message -----
From: "Alexey Romanov" <[hidden email]>
To: "Steve Davis" <[hidden email]>
Cc: [hidden email]
Sent: Sunday, February 20, 2011 3:46:27 PM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
Subject: Re: [erlang-questions] Re: If one guard expression in the guard fails, are the remaining expressions still evaluated?

I've asked precisely because reading section 7.24 didn't clarify it
for me. The relevant sentences are "The guard is true if all guard
expressions evaluate to true." and "If an arithmetic expression, a
boolean expression, a short-circuit expression, or a call to a guard
BIF fails (because of invalid arguments), the entire guard fails." The
first sentence suggests to me that all expressions are evaluated; the
second that expressions after the one that failed aren't. But both
aren't explicit.

I know about short-circuit expressions, but 7.14 doesn't say anything
about evaluation of guards (unsurprisingly).

Yours, Alexey Romanov

On Sun, Feb 20, 2011 at 4:31 PM, Steve Davis
<[hidden email]> wrote:

________________________________________________________________
erlang-questions (at) erlang.org mailing list.
See http://www.erlang.org/faq.html
To unsubscribe; mailto:[hidden email]

________________________________________________________________
erlang-questions (at) erlang.org mailing list.
See http://www.erlang.org/faq.html
To unsubscribe; mailto:[hidden email]

Alexey Romanov

Re: If one guard expression in the guard fails, are the remaining expressions still evaluated?

> If a guard statement does return false it will not continue executing if , are used

That's precisely what I was asking about. Thank you!

Yours, Alexey Romanov

> ----- Original Message -----
> From: "Alexey Romanov" <[hidden email]>
> To: "Steve Davis" <[hidden email]>
> Cc: [hidden email]
> Sent: Sunday, February 20, 2011 3:46:27 PM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna
> Subject: Re: [erlang-questions] Re: If one guard expression in the guard fails, are the remaining expressions still evaluated?
>
> I've asked precisely because reading section 7.24 didn't clarify it
> for me. The relevant sentences are "The guard is true if all guard
> expressions evaluate to true." and "If an arithmetic expression, a
> boolean expression, a short-circuit expression, or a call to a guard
> BIF fails (because of invalid arguments), the entire guard fails." The
> first sentence suggests to me that all expressions are evaluated; the
> second that expressions after the one that failed aren't. But both
> aren't explicit.
>
> I know about short-circuit expressions, but 7.14 doesn't say anything
> about evaluation of guards (unsurprisingly).
>
> Yours, Alexey Romanov
>
>
>
> On Sun, Feb 20, 2011 at 4:31 PM, Steve Davis
> <[hidden email]> wrote:
>>
>> Take a look at section 7.14 (Short-Circuit Expressions) as well as 7.24 of
>> that chapter. It should clarify.
>>
>> /s
>>
>> --
>> View this message in context: http://erlang.2086793.n4.nabble.com/If-one-guard-expression-in-the-guard-fails-are-the-remaining-expressions-still-evaluated-tp3315062p3315110.html
>> Sent from the Erlang Questions mailing list archive at Nabble.com.
>>
>> ________________________________________________________________
>> erlang-questions (at) erlang.org mailing list.
>> See http://www.erlang.org/faq.html
>> To unsubscribe; mailto:[hidden email]
>>
>>
>
> ________________________________________________________________
> erlang-questions (at) erlang.org mailing list.
> See http://www.erlang.org/faq.html
> To unsubscribe; mailto:[hidden email]
>
>

________________________________________________________________
erlang-questions (at) erlang.org mailing list.
See http://www.erlang.org/faq.html
To unsubscribe; mailto:[hidden email]