Re: [Htmlparser-user] parsing bug?

[Derrick O. <der...@ro...>]

You might want to try setting the ScriptScanner STRICT memeber to false:=0A=
=0Apublic class ScriptScanner=0A    extends=0A        CompositeTagScanner=
=0A{=0A    /**=0A     * Strict parsing of CDATA flag.=0A     * If this flag=
 is set true, the parsing of script is performed without=0A     * regard to=
 quotes. This means that erroneous script such as:=0A     * <pre>=0A     * =
document.write("&lt;/script&gt");=0A     * </pre>=0A     * will be parsed i=
n strict accordance with appendix=0A     * =0A     * B.3.2 Specifying non-H=
TML data</a> of the=0A     * HTML'>http://www.w3.org/TR/html4/">HTML 4.01 S=
pecification and=0A     * hence will be split into two or more nodes. Corre=
ct javascript would=0A     * escape the ETAGO:=0A     * <pre>=0A     * docu=
ment.write("&lt;\/script&gt");=0A     * </pre>=0A     * If true, CDATA pars=
ing will stop at the first ETAGO ("&lt;/") no matter=0A     * whether it is=
 quoted or not. If false, balanced quotes (either single or=0A     * double=
) will shield an ETAGO. Beacuse of the possibility of quotes within=0A     =
* single or multiline comments, these are also parsed. In most cases,=0A   =
  * users prefer non-strict handling since there is so much broken script=
=0A     * out in the wild.=0A     */=0A    public static boolean STRICT =3D=
 true;=0A=0A=0A----- Original Message ----=0AFrom: Subramanya Sastry <sastr=
y...@cs...>=0ATo: htmlparser user list <htm...@li...urceforg=
e.net>=0ASent: Thursday, October 25, 2007 2:00:56 PM=0ASubject: [Htmlparser=
-user] parsing bug?=0A=0A=0AI am writing to check if what I am observing is=
 a parsing bug, and if =0Aso, if there are any known workarounds.=0A=0AWhen=
 javascript is being parsed, at the start, my NodeVisitor's visitTag =0Amet=
hod gets called.  As expected, all starting html tags within the =0Ajavascr=
ipt itself are being ignored since they are are part of the =0Ajavascript a=
nd not the HTML.  However, at the first closing tag that is =0Aencounted wi=
thin the javascript code (even within strings), the parser =0Atriggers a cl=
ose script tag and calls my visitor's visitEndTag method =0Agets called.=0A=
=0AFor example with "</textarea>" string within the javascript, I get 2 =0A=
successive calls to visitEndTag, first with the SCRIPT tag and next with =
=0Athe TEXTAREA tag.  Or, with a "</he" + "ad>" string within javascript, I=
 =0Aget 2 successive calls to visitEndTag, first with the SCRIPT tag and =
=0Anext with the TEXTAREA tag.=0A=0AAs an example, test the 2 urls:=0A=0A1.=
 http://youtube.com/?v=3Dqf4tdOKKWic=0A2. http://www.deccanherald.com/Conte=
nt/Oct202007/city2007102031594.asp=0A=0AAny leads to fix this are much appr=
eciated.=0A=0AThanks,=0ASubbu.=0A=0A---------------------------------------=
----------------------------------=0AThis SF.net email is sponsored by: Spl=
unk Inc.=0AStill grepping through log files to find problems?  Stop.=0ANow =
Search log events and configuration files using AJAX and a browser.=0ADownl=
oad your FREE copy of Splunk now >> http://get.splunk.com/=0A______________=
_________________________________=0AHtmlparser-user mailing list=0AHtmlpars=
er...@li...=0Ahttps://lists.sourceforge.net/lists/listinf=
o/htmlparser-user=0A=0A=0A=0A=0A