W3C home > Mailing lists > Public > html-tidy@w3.org > July to September 1999

Disappearing Frameset Problem - Parsing Viewpoint

From: Gary Deschaines <garyd.deschaines@netmechanic.com>
Date: Thu, 12 Aug 1999 15:45:29 -0500
Message-ID: <37B33269.BE3F2E4@netmechanic.com>
To: html-tidy@w3.org
To Those Concerned,

The disappearing frameset problem can be traced to the parsing routines
used in HTML Tidy.  When a <BODY> tag is encountered, continued parsing
of the HTML document is performed by the ParseBody routine.  At this
point, since a <BODY> tag was encountered prior to a <FRAMESET> tag and
the ParseBody routine does not check for the occurrence of a <FRAMESET>
tag (or other frames content), the frames content is silently (a code
error?) discarded , leaving just body content.  If one assumes the
frames content is valid, just incorrectly placed within a body content,
then when the <FRAMESET> tag is encountered by the ParseBody routine,
continued parsing should be performed by the ParseFrameSet routine.
However, there would still be a content problem.  If an HTML document
contains a FRAMESET element, then body content must be placed within a
NOFRAMES element in the FRAMESET.  But a <BODY> tag was encountered
before the <FRAMESET> tag. What to do?  Keep the body content and
discard the frames content, discard the body content and keep the frames
content, or keep the frames content and move the body content within a
NOFRAMES element?

Respectfully,
Gary Deschaines
Received on Thursday, 12 August 1999 16:45:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 April 2012 06:13:42 GMT