Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.javascript > #34794 > unrolled thread

Re: The less sign "<" in script elements

Back to article view | Back to comp.lang.javascript

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: The less sign "<" in script elements Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2018-03-19 00:41 +0100

#34794 — Re: The less sign "<" in script elements

Stefan Ram wrote:

>   So, is my memory correct that in HTML 2, HTML 3 and/or
>   HTML 4, one had to escape a less sign in a script element,
>   using, e.g., »&#60;«? E.g., for a comparison like »x<y«?

No.  Escaping using *other* means was only necessary if “<” was followed by 
“/”.  

The SCRIPT element was formally introduced in HTML 3.2 (there is no HTML 
3.0).  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).  And this 
substring occurred frequently in script code that generated HTML (especially 
before the W3C DOM, hence the example below).

The easiest workaround, thus recommended and used by me as well, was to 
write "<\/" in a string literal, or to move the source code for generated 
markup to an external script resource.  (Later, you could also get rid of
“</” by strictly using features introduced with the W3C DOM.)

    [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>

     (a common example of this was code generated by Macromedia Dreamweaver)

     But that merely masks the markup errors which is a problem when serving
     XHTML with the proper Content-Type “application/xhtml+xml” (those
     people then resorted to serving XHTML with Content-Type “text/html”,
     which I think contributed to the demise of XHTML – most people simply
     did not bother to understand the powerful tool that they were given). 
     The correct way in XHTML (as “application/xhtml+xml”) is, of course,
     as you indicated,

       <script …><![CDATA[ … ]]></script>

     because the CDATA section declaration makes clear where the CDATA
     begins and ends.  However “]]>” must to be escaped instead, or an
     external script resource must be used.]

I have written (this) *countless* times (about that) in several 
newsgroups/forums, including this one; so have other people [the first 
Google hit for “script etago” is an article by Mathias Bynens citing a test 
case by IIRC then-cljs regular Juriy ‘kangax’ Zaytsev (Юрий Зайцев)].

You can still verify it if you use the W3C Markup Validator at 
<http://validator.w3.org/>:

  1. <https://validator.w3.org/#validate_by_input+with_options>
  2. More Options → (·) Validate HTML fragment → Use Doctype: HTML 4.01
  3. Markup to validate (without indentation):

     <script type="text/javascript">document.write("<p>foo</p>");</script>

  4. Check →
  5. “Error: end tag for element "P" which is not open
      […]
      If this error occurred in a script section of your document, you 
      should probably read this FAQ entry
      <https://validator.w3.org/docs/help.html#faq-javascript>.”

-- 
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.

[toc] | [standalone]

Back to top | Article view | comp.lang.javascript

csiph-web