Re: Semi-random Parser Failure

Unfortunately it seems the mailing list doesn't support UTF-8... I will try stripping out some php code to see if I can't isolate the issue down to a specific area when I get a chance later.  I did just use the validator again with the option to show source, although nothing seems particularly abnormal on a cursory glance.

The error i got was:
Line 69, Column 348: End of file seen and there were open elements.

However, that error occurs in the middle of a p tag surrounded by other text, and that specific p tag is also surrounded by other p tags, as it's in the middle of a paragraph.  There's no other tags being closed that would suggest I missed closing a tag, and there certainly isn't an end of file indicator (the word being highlighted is simplified chinese for "media", although both simplified and traditional chinese seem to get caught sometimes).

I also used the "clean up markup with HTML Tidy" option and then reran that code directly through the validator using the direct input option, and it successfully validated.  Interestingly, I tried to validate the code numerous times and never was able to get an error.

I've attempted to attach the html from the validator with the error mentioned above (with show source and tidy html options selected), although I'm not sure if it will go through the mailing list or not.

Cheers.


在 2010/9/20 上午1:01 時， Leif Halvard Silli 寫到：

> 黃俊銓, Sat, 18 Sep 2010 18:53:46 +0800:
>> Validation: http://validator.w3.org/check?uri=http%3A%2F%2Fwww.dak.org%2F
> 
>> [...] reporting it as an end-of-line error.  The rest of the time the site 
> is reported 
>> as valid HTML5. [...]
> 
> Exept one time (see bottom of letter) I got the same kind of reaction 
> as you: sometimes the page is reported as valid, other times it is not.
> 
> Perhaps you are confusing the Validator. I looked at the source, and it 
> seems as if some lines uses CRLF (Windows) line endings, while others 
> lines uses LF (Unix) line endings. (There is no reason this should 
> affect the Validator though.)
> 
>> NOTE: There are some dynamic elements on the page that change each 
>> time the page is loaded; however, I've looked through these and none 
>> of them seem to be able to explain the above error.
> 
> Perhaps the dynamics affects the line endings as well? Also, one of the 
> times I validated your page, I got a typical URL encoding error: 
> 
> ]]
> Line 44, Column 277: & did not start a character reference. (& 
> probably should have been escaped as &.)
> ….co.uk/ws/zh/thread.jspa?forumID=12582&start=0&zh=trad" 
> target="_blank">
> [[
> 
> May be this is an indication that the dynamic parts of the page don't 
> load fast enough, so that the Validator really doesn't see the tags 
> that you have placed there. 
> 
> I you use the Show source option in the Validator, then you can verify 
> that the validator checked the code that you expected it to check.
> 
> (PS: right now your page is unavailable - I suppose the entire 
> mailinglist is trying to help you.) :-)
> -- 
> leif halvard silli