|
|
Posted by shimmyshack on 05/14/07 21:00
On May 14, 9:58 pm, shimmyshack <matt.fa...@gmail.com> wrote:
> On May 14, 7:47 pm, loretta <lorb...@optonline.net> wrote:
>
>
>
> > On May 14, 2:16 pm, shimmyshack <matt.fa...@gmail.com> wrote:
>
> > > On May 14, 6:08 pm, loretta <lorb...@optonline.net> wrote:
>
> > > > This code is just reading html and printing , eventually I want to
> > > > modify the html. However, the original html contains javascript and
> > > > the output html contains tags not in the original.
>
> > > > $url = "http://www.something.com";
> > > > $doc = new DOMDocument();
> > > > $doc->loadHTMLFile($url);
> > > > print $doc->saveHTML();
>
> > > > Original html snippet:
> > > > function exampleFunction() {
> > > > var doc = '<html><head>';
> > > > doc += '<title>Title</title>';
> > > > doc += '</head>';
> > > > doc += '<body onload="self.focus();">';
> > > > doc += '</body></html>';
> > > > }
>
> > > > Html after saveHTML:
> > > > function exampleFunction() {
> > > > ('about:blank','imagemanagerpopup',settings);
> > > > var doc = '<html><head>';
> > > > doc += '<title>Title</title>';
> > > > doc += '</script>
> > > > </head>
> > > > <body>
> > > > <p>';
> > > > doc += '</body>
> > > > </html><html><body>
> > > > <p>';
>
> > > > }
>
> > > > Extra tags to end the script, head and begin a new body are being
> > > > added before the </body> tag and after the <body onload=self.focus()>
> > > > tag in the js variable. Is there a way for the Dom to leave the
> > > > javascript as is without trying to 'fix' the html ? The changes being
> > > > made are causing a javascript error.
> > > > Thanks
>
> > > start off with xHTML, so it can be loaded with no errors, see google
> > > on how to add javascript in a way that is compliant with xml standards- Hide quoted text -
>
> > > - Show quoted text -
>
> > The html I am retrieving has a xhtml doctype. I also have no control
> > over the original webpage. The original webpage loads with no errors
> > in both IE and FF.
>
> this is what i find on google.http://developer.mozilla.org/en/docs/Properly_Using_CSS_and_JavaScrip...
> use <!CDATA or the "xhtml" document is no such thing, btw it should
> not just claim to be xhtml but should be properly validated as such,
> including the content-type text/xml+xhtml (served with as .xhtml)
> once you have obtained the webpage, and parsed it adding the right
> instructions for the xml parser, all should work, if indeed the rest
> of the doc is valid xml.
oops, application/xml+xhtml of course
[Back to original message]
|