[yunqa.de] Re: DiHtmlParser problems and questions
- From: Delphi Inspiration <delphi@xxxxxxxx>
- To: yunqa@xxxxxxxxxxxxx
- Date: Fri, 09 Mar 2012 09:51:23 +0100
On 08.03.2012 23:48, Max Terentiev wrote:
> But I found another page what produce same problem.
>
> See attachment.
>
> In AllInOne demo -> Tag Lister tags skipped between
> line 79 and 293.
>
> Problem again in parsing javascript code. But no any
> regexp at this time !
The problem is the unterminated string in line 82:
alert(\'Необходимо ввести запрос для поиска!\');
It contains two backslash-apostrophe-sequences "\'":
The 1st is outside a string. Hence the backslash does not escape the
apostrophe and the apostrophe starts a string.
The 2nd is inside a string where the backslash escapes the apostrophe.
Hence it does not terminate the string before it see the next unescaped
apostrophe in line 285.
[Btw, Firefox does not recognize the string either. Pressing the button
does not show the alert.]
Why does DIHtmlParser analyze JavaScipt? Because JavaScript may contain
strings like '</script>' which do NOT mark the end of the script.
Unfortunately, analysis may fail for invalid JavaScript syntax.
If you prefer DIHtmlParser not to analyze JavaScript, you can set a
default dummy script type for your TDIHtmlParser instance:
DIHtmlParser1.DefaultContentScriptType := 'dummy'; // <-- DUMMY!!!
This will trick DIHtmlParser not to analyze JavaScript so it will parse
your HTML correctly. It will, however, fail on HTML which contains the
unescaped </script> end marker inside JavaScript strings.
Ralf
_______________________________________________
Delphi Inspiration mailing list
yunqa@xxxxxxxxxxxxx
//www.freelists.org/list/yunqa
Other related posts:
