W3C home > Mailing lists > Public > www-html@w3.org > June 1999

Re: avoid visitors viewing inside pages

From: <sgambhir@web.fairfax.com.au>
Date: Wed, 30 Jun 1999 10:19:48 +1000
Message-ID: <377962A4.17924F43@web.fairfax.com.au>
To: Nicolas Lesbats <nlesbats@etu.utc.fr>
CC: Nathan Zaetta <nzaetta@brw.fairfax.com.au>, Aditya Hermawan <aditya@sby.globalinfo.net>, W3HTML <www-html@w3.org>
>From my understanding of things... there is a potential solutions to the problem.

Since you can run javascript functions in the 'parent' frame... how about doing something
like (here some pseudo-code :-)

   <script language="pseudo-code">
     if ( NOT parent.returntrue() ) {
     // the returntrue function could not be run from the parent frame
     // so this page was progably accessed directly
     location = index.html
   </script>

this will probably only work for framed sites where you don't want people to access pages
that are not via the index file (which contains the returntrue() function!).

For non-framed sites, you will probably have to do something nifty with the referer field!

:-)



Nicolas Lesbats wrote:

> On Tue, 29 Jun 1999, Nathan Zaetta wrote:
>
> | > You can add to each page of the site a
> | >
> | >  <meta http-equiv="refresh" content="0, http://.....index.html">
> |
> | Um, this solution also means that nobody can ever get to any other page, as
> | they will always be redirected back to the index
>
> In fact, this is because it *is not* a solution ! Just a part of a
> potential one...
>
> I had a similar problem (which is not solved). My site was built around a
> index.html file, which was a frames structure. To simplify, a 'left' frame
> (for a general menu), and a 'main' frame (for contents).
>
> I had, for instance, an article.html file, which was accessable from the
> index.html frameset. But it was too able to open it directly from a search
> engine.
>
> What I wanted to do yet was to 'reincorporate' (directly translated from
> french, I've some doubts about it) = 'integrate' = 'introduce' this file
> _in_ the index.html frameset.
>
> Let's take a possible processing, which will be clearer I hope.
>
>  * The user opens the article.html file, out of any frameset. An
> instruction in the article.html file tells the User Agent to re-open it in
> the index.html frameset, in the "main" frame.
>
> For example (totally invented XHTML) :
>
>         <link rev="frameset"    { or role="frameset" }
>          href="index.html?main=article.html" target="_parent"
>          xml:link="simple" actuate="auto" show="remplaced"
>         />
>
> What is important here is the possibility to specify the content of each
> frame in the URL of a frameset file. What do you think about the
> syntax above :
>
> http://www.foo.com/frameset.html?frame_name=http%3A%2F%2Fwww.foo.com%2Fframe_content.html
>
>  * When the UA parses index.html, for example :
>
>         <frameset>
>          <frame name="left" href="left_frame.html" />
>          <frame name="main" href="main_frame.html" />
>         </frameset>
>
> it replaces the hyperreference of the main frame by the processing
> instruction included in the URL, here http://www.foo.com/article.html
>
> [Of course, it doesn't solve entirely the problem because the use of the
> 'actuate="auto"' attribute in the <link> element can be processed like a
> perpetual refresh of the page, though the 'auto' value still is ambiguous.
> It would be necessary to correct that too.
>
> Furthermore, it would be necessary to refer to the active page directly,
> without name it, for instance with a
>
>         href="index.html?main=" origin="this"
> or
>         href="index.html" frame="main" frame_href="this" ]
>
> Then, a solution for one of the Aditya's problems is to use an inline
> frame in index.html (<iframe> element) to display any other page.
>
> I think javascript can be used, but incompatibiliy between
> some browsers are very dissuading.
>
> The more workable solution I have found is to never use frames anymore...
> Furthermore, frame elements are deprecated according to some W3C working
> groups (but I don't agree with it).
>
> Since this list is designated to correct problems, have anyone any
> proposal about it ?
>
>                         Nicolas
>
> --
> Nicolas Lesbats - nlesbats@etu.utc.fr
> 85 r. Carnot 60200 Compiegne - France
>  06 86 800 908
>
> Plaider <http://wwwassos.utc.fr/~plaider/>
>
> 3:-)

--
Simran Gambhir
NBD, Fairfax
201 Sussex St. Darling Harbour, NSW, 2000.
Tel: +61 2 9282-2777  Fax: +61 2 9282-2256
Received on Tuesday, 29 June 1999 21:30:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:39 GMT