[ 14 October 1997
 The Linux HTML Validation mini-HOWTO is not being maintained by
 the author any more.  If you are interested in maintaining the
 HTML Validation mini-HOWTO, please get in touch with me at
 <[email protected]>. ]

 The HTML Validation HOWTO
 Keith M. Corbett, [email protected]
 v0.2, 29 October 1995

 This document explains how to use the nsgmls parser to validate HTML
 documents for conformance with the HTML 2.0 document type definition,
 or "DTD".  This DTD is the most commonly accepted SGML based defini-
 tion of HTML, and thus defines a subset of current practice in HTML
 markup that is likely to be portable to a wide number of HTML users
 agents (browsers).
 ______________________________________________________________________

 Table of Contents:

 1.      Introduction

 1.1.    Costs and benefits

 1.2.    Getting started

 2.      Tools

 2.1.    The

 2.2.    The

 2.3.    Download the HTML specification materials

 3.      Parsing an HTML document

 3.1.    Parser input

 3.2.    Parser output

 3.3.    Parser messages

 3.4.    Return status

 4.      Resources
 ______________________________________________________________________

 1.  Introduction

 This is a guide to using the nsgmls parser to validate and process
 HTML documents.

 1.1.  Costs and benefits

 Using the full features of SGML markup will enrich your HTML
 documents.  However, validating your documents to the HTML DTD has
 certain cost / benefit tradeoffs, basically because you are dealing
 with a more circumscribed dialect of HTML than is currently in vogue.
 The "official" HTML rules for enforcing document structure, and the
 SGML rules for data content markup, are more restrictive than current
 practice on the Web.

 The main issue you must consider is that valid HTML is restricted to a
 standard set of element tags.

 There isn't an accepted DTD that accurately reflects "browser HTML" as
 understood by many client browser programs.  For the most part, the
 HTML 2.0 DTD reflects tags and attributes that were commonly in use on
 the Web around June 1994.  Various efforts to define a more advanced
 HTML+ or HTML 3.0 DTD have gotten somewhat bogged down.  And none of
 the DTDs in circulation will recognize all of the tags that have been
 popularized recently by browser vendors such as Netscape and
 Microsoft.

 1.2.  Getting started

 Contrary to popular opinion, working with SGML does not have to cost a
 lot of time and money.  It is possible to build a robust development
 environment consisting entirely of software that is freely available
 on a wide range of platforms, including Linux, DOS, and most Unix
 workstations.  Thanks to a few very dedicated folks, all the tools you
 need to work with SGML have been made publicly available on the
 Internet.

 Setting up your environment (the parser and supporting program
 libraries) takes a bit of work but not nearly as much as one might
 think.

 You may also want to peruse an introductory SGML text such as "SGML:
 An Author's Guide to the Standard Generalized Markup Language" by
 Martin bryan, or "Practical SGML" by Eric van Herwijnen.

 2.  Tools

 2.1.  The HTML Check toolkit package

 If you want a completely self-installing / canned package, check out
 the HalSoft HTML Check Toolkit at URL: http://www.halsoft.com/html-
 tk/index.html

 The only disadvantage of using the HalSoft kit is that it uses the
 older sgmls parser, which produces error messages that are sometimes
 (even) more cryptic than those from nsgmls.

 I've used nsgmls on Linux and Windows (3.x and NT); it is supposed to
 work on many other platforms as well.

 2.2.  The nsgmls parser

 James Clark has built a software kit called sp which includes the
 validating SGML parser, nsgmls.  (This is the successor to the sgmls
 parser which has long been considered the reference parser.)

 For information on the sp kit, see URL: http://www.jclark.com/sp.html

 You can download the kit directly from: ftp://ftp.jclark.com/pub/sp/

 You may be able to pick up nsgmls executable files for your platform.
 Or, download the source kit and follow the directions in the README
 file for running make.

 Consider creating a high level public directory that will contain
 SGML-related files.  For example, on my Linux PC I have various SGML
 related directories including:

 /usr/sgml/bin

 /usr/sgml/html

 /usr/sgml/sgmls

 /usr/sgml/sp

 2.3.  Download the HTML specification materials

 The draft standard for HTML 2.0 includes SGML definition files you
 need to run the parser, namely the DTD (Document Type Definition),
 SGML Declaration, and entity catalog.  To obtain the HTML 2.0 public
 text, see URL:

 http://www.w3.org/hypertext/WWW/MarkUp/html-spec/

 Download and install the following files:

 DTD html*.dtd

 SGML declaration html.decl

 Entity catalog catalog

 You can add two entries to the HTML entity catalog for ease of use
 with nsgmls:

      ______________________________________________________________________
              -- catalog: SGML Open style entity catalog for HTML --
              -- $Id: catalog,v 1.2 1994/11/30 23:45:18 connolly Exp $ --
       :
       :
              -- Additions for ease of use with nsgmls --
      SGMLDECL        "html.decl"
      DOCTYPE HTML    "html.dtd"
      ______________________________________________________________________

 Alternatively, you can create a second catalog containing these
 entries; you will have to pass this catalog to nsgmls as an argument
 with the -m switch.

 3.  Parsing an HTML document

 Following is a "cookbook" for validating a single document.  Simply
 invoke the nsgmls parser and pass it the pathnames of the HTML catalog
 file(s) and the document:

      % nsgmls -s -m /usr/sgml/html/catalog <test.html

 The -s switch suppresses the parser's output; see below.

 3.1.  Parser input

 Your document must conform to SGML, which means, among other things,
 that the document type must be declared at the beginning of the input.
 (You can fudge this by prepending the information to the document
 instance on the nsgmls command line.)

 Here's a simple HTML document that can be parsed correctly using the
 scheme I've outlined:

      ______________________________________________________________________
      <!doctype html public "-//IETF//DTD HTML 2.0//EN">
      <html>
      <head>
      <title>Simple HTML document.</title>
      </head>
      <body>
      <h1>Test document</h1>
      <p>This is a test document.</p>
      </body>
      </html>
      ______________________________________________________________________

 3.2.  Parser output

 The standard output of nsgmls is a digested form of the SGML input
 that processing systems can use as a lexer for navigating the
 structure of the document.  For the purpose of validation, you can
 throw the standard output away and rely on the error output.

 If you do want the full output, omit the -s switch and pipe standard
 output to a file:

      % nsgmls -m /usr/sgml/html/catalog <test.html >test.out

 3.3.  Parser messages

 Error and warning messages from nsgmls can be very cryptic.  And you
 may see very many errors from illegal markup.

 To pipe messages to a file, use the -f switch:

      % nsgmls -s -m /usr/sgml/html/catalog -f test.err <test.html

 3.4.  Return status

 The parser indicates whether the input document conforms to the HTML
 DTD in two ways:

 Return code - the parser returns a 0 exit status on success, non-zero
 otherwise.

 Output - if the document conforms to the DTD, the last line of
 standard output will consist of a single C character.

 4.  Resources

 The HalSoft HTML Check Toolkit is at URL: http://www.halsoft.com/html-
 tk/index.html

 James Clark's page on sp is at URL: http://www.jclark.com/sp.html

 The W3C page on the HTML specification is at URL:
 http://www.w3.org/hypertext/WWW/MarkUp/html-spec/

 Feel free to contact me via email: [email protected].

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.