Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Ben Bacarisse <ben.usenet@bsb.me.uk>
Newsgroups: comp.lang.javascript
Subject: Re: x === x
Date: Wed, 31 May 2017 21:52:31 +0100
Organization: A noiseless patient Spider
Lines: 68
Message-ID: <87efv4zr3k.fsf@bsb.me.uk>
References: <JavaScript-20170530123255@ram.dialup.fu-berlin.de> <874lw231ui.fsf@bsb.me.uk> <context-20170530144535@ram.dialup.fu-berlin.de> <2872662.aeNJFYEL58@PointedEars.de> <87inkh28sf.fsf@bsb.me.uk> <19437984.EfDdHjke4D@PointedEars.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Injection-Info: mx02.eternal-september.org; posting-host="04e45becd6848d7651d9f559a5a1bcb5"; logging-data="25362"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19nU57ScX8RSBVKDoQI+4KeuDqum72csVc="
Cancel-Lock: sha1:8Zh+5Ixse9nafuyMlm+6KKAdYxU= sha1:kSU9tceQvi7EaBerKz5dwE8IUhA=
X-BSB-Auth: 1.0387a58d4e4c0bf69267.20170531215231BST.87efv4zr3k.fsf@bsb.me.uk
Xref: csiph.com comp.lang.javascript:33458

Thomas 'PointedEars' Lahn <PointedEars@web.de> writes:

> Ben Bacarisse wrote:
>
>> Thomas 'PointedEars' Lahn <PointedEars@web.de> writes:
>>> Stefan Ram wrote:
>>>> Ben Bacarisse <ben.usenet@bsb.me.uk> writes:
>>>>> What's wrong with {} === {}?
>>>>   It depends on the context.
>>> No, it does not.  “===” *makes* the (expression) context.
>> 
>> $ nodejs -c -e '[] === [];' && echo "ok"
>> ok
>> $ nodejs -c -e '{} === {};' && echo "ok"
>> [eval]:1
>> {} === {};
>>    ^^^
>> 
>> SyntaxError: Unexpected token ===
>> <snip trace-back>
>> $ nodejs -c -e '({} === {});' && echo "ok"                        ✘
>> ok
>
> Fascinating.  I had tried in the Chrome DevTools Console, and there the 
> trailing semicolon makes the difference:
>
> | > {} === {}
> | <- false
> | {} === {};
> | (x) Uncaught SyntaxError: Unexpected token ===
>
> I assume that the explicit semicolon there is considered an /EmptyStatement/ 
> that switches back to statement context.

I can't find any reference to "statement context" or "expression
context" in the language standard.  I assume it's your own terminology,
but I don't see how it helps explain why [] === []; is OK but {} === {};
is a syntax error.  (I know why, I'm just wondering how talking about
=== making expression context and ; switching back helps to explain this
difference.)

Just

  {} === {}

both on it's own and with a semicolon, is a syntax error because it is
not a /Script/ (or a /Module/).  It's not a /Script/ because none of the
options for a /Statement/ apply.  There is an explicit exemption that
prevents an /ExpressionStatement/ (the only reasonable candidate[1])
from staring with a {.  On the other hand

  [] === []

(again, with or without a semicolon) does parse as a /Script/ because it
*is* an /ExpressionStatement/.

The trouble with testing witch a REPL is that there is no standard that
mandates what the reader in such a loop should be expecting.

[1] Unfortunately, when explaining why something isn't valid syntax, you
should, formally, say why it is not any of the possible options.  I hope
it's clear that I'm just applying a short-cut by talking about "the only
reasonable candidate".  It's obviously not a /BlockStatement/ or a
/VariableStatement/ or a ... and so on but, were it not for the
exemption, it looks most like an /ExpressionStatement/.

-- 
Ben.