W3C home > Mailing lists > Public > html-tidy@w3.org > October to December 2000

Re: form as a direct table child

From: Gary L Peskin <garyp@firstech.com>
Date: Thu, 23 Nov 2000 01:09:28 -0800
Message-ID: <3A1CDEC8.DC9C7F95@firstech.com>
To: Luba Roitman <l_roitman@yahoo.com>
CC: Html-Tidy <html-tidy@w3.org>
Luba Roitman wrote:
> 
> Dear Gary,
> thank you very much for your very detailed reply.
> While looking into the Tidy code trying to find the
> solution I saw a comment, apparently by Andy Quick, in
> file ParserImpl.java, class ParseTableTag, line 2120:
> -----------
>   /* AQ: TODO
>   Line 2040 of parser.c (13 Jan 2000) reads as
> follows:
>   if (!node->type == TextNode)
>   This will always evaluate to false.
>   This has been reported to Dave Raggett <dsr@w3.org>
>   */
>   //Should be?: if (!(node.type == Node.TextNode))
>   if (false)
>     parseTag(lexer, node, Lexer.IgnoreWhitespace);
> -----------
> When I did what AQ suggests (changed false for
> !(node.type == Node.TextNode) ) I got a correct Tidy
> output. The <form> in my case was moved outside of the
> <table>, which makes sense.
> 
> I just thought you might find this useful. Can you
> make this change in JTidy? Should I inform Dave
> Raggett about this? It seems that he's been informed
> about this already.

Luba --

I'm forwarding this note to the html-tidy list.  Dave, this error is now
on line 2085 of the Tidy version of 4aug2000 module parser.c.  The line
as written will always evaluate to false so the parseTag code will never
be executed.  Can you make the correction in routine ParseTableTag?

Thanks,
Gary
Received on Thursday, 23 November 2000 04:10:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 April 2012 06:13:44 GMT