Hello

Is there any known way to parse html files with:
<script>
some script with tricky symbols , like "< >" that ruins the parsing </script> tag?

I've not found any better option but to re-implement THTMLReader (copy-paste + modify) and run a custom parser to skip the content of the script tag, and then go-on parsing html file.

Also, is there any reason for f parameter to be "var"?

procedure ReadHTMLFile(var ADoc: THTMLDocument; var f: TStream);
procedure ReadHTMLFragment(AParentNode: TDOMNode; var f: TStream);

thanks,
Dmitry

What about using CDATA section?

<script>
<![CDATA[

]]>
</script>

Lee Jenkins wrote

What about using CDATA section?

There's no support for CDATA in the HTML parser.
I'm trying to read arbitrary html files, so there's almost no "<![CDATA" usage.

thanks,
dmitry