Path: csiph.com!aioe.org!news.mb-net.net!open-news-network.org!.POSTED.112.228.197.178.dynamic.wless.zhbmb00p-cgnat.res.cust.swisscom.ch!not-for-mail
From: Thomas 'PointedEars' Lahn <PointedEars@web.de>
Newsgroups: comp.lang.javascript
Subject: Re: The less sign "<" in script elements
Date: Tue, 20 Mar 2018 16:45:28 +0100
Organization: PointedEars Software (PES)
Lines: 157
Message-ID: <1682634.CQOukoFCf9@PointedEars.de>
References: <less-20180318165920@ram.dialup.fu-berlin.de> <slrnpatp5k.4an.jon+usenet@sable.unequivocal.eu> <6800844.EvYhyI6sBW@PointedEars.de> <slrnpavfqp.4an.jon+usenet@sable.unequivocal.eu> <3165380.iIbC2pHGDl@PointedEars.de> <slrnpb0hk1.2gf.jon+usenet@sable.unequivocal.eu> <8233097.CDJkKcVGEf@PointedEars.de> <slrnpb1sio.2gf.jon+usenet@sable.unequivocal.eu> <3095067.QJadu78ljV@PointedEars.de> <slrnpb22mf.2gf.jon+usenet@sable.unequivocal.eu>
Reply-To: Thomas 'PointedEars' Lahn <cljs@PointedEars.de>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8Bit
Injection-Info: gwaiyur.mb-net.net; posting-host="112.228.197.178.dynamic.wless.zhbmb00p-cgnat.res.cust.swisscom.ch:178.197.228.112"; logging-data="632"; mail-complaints-to="abuse@open-news-network.org"
User-Agent: KNode/4.14.2
X-User-ID: U2FsdGVkX18DwnPoYL1y2aiohPYtpfhyq/+I4cVz6/vy05ILQiUlxg==
Cancel-Lock: sha1:FolcxuTAjyl79prto/iWTuKrACI=
X-Face: %i>XG-yXR'\"2P/C_aO%~;2o~?g0pPKmbOw^=NT`tprDEf++D.m7"}HW6.#=U:?2GGctkL,f89@H46O$ASoW&?s}.k+&.<b';Md8`dH6iqhT)6C^.Px|[=M@7=Ik[_w<%n1Up"LPQNu2m8|L!/3iby{-]A+#YE}Kl{Cw$\U!kD%K}\2jz"QQP6Uqr],./"?;=4v
Face: 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
Xref: csiph.com comp.lang.javascript:34804

Jon Ribbens wrote:

> On 2018-03-20, Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote:
>> Jon Ribbens wrote:
>>> On 2018-03-19, Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote:
>>>> Jon Ribbens wrote:
>>>>> Perhaps you could progress beyond just spouting nonsense then,
>>>>> and make some specific criticisms or corrections.
>>>> I did.  Had you not full-quoted what I have replied to you, and had
>>>> you read what I have posted in the other subthread, you would have
>>>> realized that.
>>> 
>>> I already read that post and there is nothing in it that is a specific
>>> criticism of or correction to my post. There's barely anything in it
>>> that disagrees with what I said, except that you clearly have not read
>>> either the HTML 3.2 or HTML 4.01 specifications.
>>
>> So much for “specific creticism”.

I probably wanted to say “cretinism” instead.
  
> You said:

In Usenet, one posts a follow-up to the posting with the content that one is 
referring to; one does _not_ quote statements out of context like you do.
 
>> However, in HTML 3.2 to 4.01, because of limitations of the SGML-
>> based DTD, the CDATA content of a SCRIPT element was defined to end
>> at the first occurrence of “</” (End-Tag Open delimiter, ETAGO).
> 
> In fact the HTML 3.2 specification says 'ETAGO ("</") delimiters
> followed immediately by a name character [a-zA-Z]', and the HTML 4.01
> specification says 'ETAGO ("</") delimiter followed by a name start
> character ([a-zA-Z])'.

You ought to have *cited* evidence.

It is correct that those Specifications said/say that.  HTML 3.2 said it in 
a section that was normative¹ (but back in the day the distinction between 
normative and informative was not made).  HTML 4.01, however, says it in 
Appendix B which is *informative* only.

AFAIK, SGML, with which HTML up to version 4.01 is defined –

<http://www.w3.org/TR/1999/REC-html401-19991224/sgml/sgmldecl.html>

– makes no such statements; instead one finds in the official summary:

,-<http://xml.coverpages.org/wlw14.html>
|
| […]
| As an example, a typical end tag in an SGML document "</elemname>"
| contains two delimiters and a name.  "</;" is the delimiter which
| indicates the start of an end tag; its role is called "etago".

(Given that the first sentence does not contain a semicolon after “</”, the 
one in the second sentence must be considered a typo.)

I do not know how the W3C Markup Validator handled it when it (the 
Validator’s behavior) had been *cited by people to me repeatedly* as the 
reason for wanting to comment out the “script” element contents; currently, 
it does not consider a standalone “</” within a HTML 4.01 “SCRIPT” element 
an ETAGO.

How HTML user agents actually handled it back then I do not know; but 
kangax’s test case should provide some insight; in fact, he claims that “It 
seems like de-facto standard is [the] ‘</script’ sequence, not just ‘</"’.”, 
which would refute *both* our statements.

_______
¹ HTML 3.2 is a *Superseded* Recommendation since 2018-03-15, due to
  HTML 5.2 at least.  It is now inappropriate to cite the former as
  normative:

  <https://www.w3.org/TR/2018/SPSD-html32-20180315/> p.

> You also said:
> 
>> The misconception that it would be necessary to hide the offending
>> content from *seemingly* bad validators and editors (which were
>> actually *working*) caused the misconception in many people that in
>> XHTML (and in HTML for *actually* bad validators and editors) it
>> would a good idea to comment out the entire script contents:
>> 
>>    <script …><!-- … --></script>
> 
> This was nothing to do with "seemingly bad validators and editors",
> it was to do with backwards-compatibility with browsers and tools
> that pre-dated the <script> tag and therefore could not know that
> its contents should be parsed as CDATA.

Incorrect.  First of all, apparently you have not noticed that I was 
primarily talking about *XHTML* in that paragraph.

In XHTML, the content model of the “script” element is _not_ CDATA, but 
PCDATA (parsed CDATA).  This causes those *XML* comment delimiters to 
actually comment out the content:

<https://www.w3.org/TR/2008/REC-xml-20081126/#sec-comments>

So, AISB, people who used this in XHTML simply did not know what they are 
doing.

As regards my referring to HTML, maybe you are dereferring to HTML 4.01, 
§ 18.3.2, where this is claimed, or to Mathias Bynens’ article that I 
referred to where the latter is quoted.  (Given that you did not know that 
HTML 4.01 Appendix B is informative only, it is doubtful whether you have 
actually read the HTML 4.01 Specification.)

However, the cited section is questionable in several regards.

First of all, the W3C does not have a mandate to specify, contrary to the 
statement in that section, how script engines work.  ECMAScript, for 
example, is specified by Ecma International instead.  In particular (with 
the exception of the period from 1995-12 to 1996-08; see the ECMAScript 
Support Matrix) there has not been and there is no single “the JavaScript 
engine” (as there is no single JavaScript, “Javascript” or “javascript” 
programming language; ibid.)

Second, HTML 4.01 became a W3C Recommendation on 1999-12-24.  But versions 
of HTML before 3.2, which did not specify the SCRIPT element, have been made 
*obsolete* by RFC 2854 already; that became a RFC in 2000-06, so must have 
been submitted to the IETF about 6 months prior to that, i.e. at the time of 
promoting HTML 4.01 a REC; in fact, the first draft, which already contained 
“Obsoletes: RFC 1866”, was submitted 1999-09-21 (see “draft-conolly-…”).  
Given that Conolly drafted and published this as a representative of the 
W3C, the content of that section must be considered an oversight by the HTML 
Working Group and the W3C Membership.  It should never have made it to REC 
in its current form.  At best, this section should be considered 
informative, if not entirely *historical* now.

But more importantly, this was published *long* before people *kept using* 
this.  That is, there was no software that should have been considered 
*working* at the time that still parsed the content of the SCRIPT element 
(CDATA provisions aside), including, but not limited to, in HTML 3.2, ISO 
HTML, and HTML 4.01.  There certainly is no such software now – not based on 
wishful thinking, but on *normative requirements*.

Therefore, my statement is *correct* that only “*actually* bad validators 
and editors” would (have [and actually had]) require(d) this.

> It is now your turn to provide specific criticism of my earlier post.

I see no logical reason to repeat myself.

My replying simply with “Nonsense.” to most (_not_ *all*) of your statements 
is the summary of what I wrote in my other post that you have evidently not 
read carefully enough, and to which you did not post a follow-up.

Where you have offered unfounded assumption based on wishful thinking, I 
offered historical fact because (apparently by contrast to you) not only did 
I do research on this, *I* *have* *been* *there*.  Do the comparison.

-- 
PointedEars
FAQ: <http://PointedEars.de/faq> | <http://PointedEars.de/es-matrix>
<https://github.com/PointedEars> | <http://PointedEars.de/wsvn/>
Twitter: @PointedEars2 | Please do not cc me./Bitte keine Kopien per E-Mail.