On 2010/08/17 20:19:04, xtof wrote:
On 2010/08/17 18:39:28, jat wrote:
> On 2010/08/16 23:39:47, tbroyer wrote:
> > The HtmlSanitizer is a good idea, but the implementation is very
weak [2].
>
> Note that the API is what is important, and SimpleHtmlSanitizer is
just that,
a
> simple implementation.  A more involved implementation can be added
later.
>
> Also, we aren't trying to parse HTML with a regex here, it simply
looks for
> opening tags and allows unescaped on a small set of whitelisted tags
--
> everything else gets escaped.  If you think it fails to do its job,
can you
> supply a string which would not be propertly sanitized?

Looking at the code more closely it would merely "fail" by overly
rejecting tags that are whitelisted: i.e. "<b foo=<i>should be bold"
would be sanitized to "&lt;b foo=<i>should be bold" and the end part
would be italicized instead of bold.
But there's more:

In other words, does SimpleHtmlSanitizer adhere to the SafeHtml type
contract?
I believe it does, but of course would like arguments/hunches towards
the
contrary.

SafeHtml calls for XSS mitigation, and SimpleHtmlSanitizer doesn't
sanitize most XSS attacks, that use attributes:
"<b
style='position:absolute;z-index:2147483646;left:0px;top:0px;right:0px;bottom:0px'
onmousemove=\"alert(&quot;you've been highjacked&quot;);\">"

Having said that, I kind of agree that SimpleHtmlSanitizer is of
pretty limited
use; one of the few scenarios where I can see it used is for rendering
messages/snippets that are formatted with limited HTML markup, and are
obtained
from a complex backend that the developer of the GWT client doesn't
quite want
to trust to uphold the SafeHtml type contract.  I wouldn't be terribly
opposed
to droppping it.

I didn't want to implement a full parser, because parsing full HTML
(say in an
email app) inside the browser generally seems like a bad approach.

I agree, and I'd be OK with removing HtmlSanitizer from this patch to
better reintroduce it later.



http://gwt-code-reviews.appspot.com/771801/show

--
http://groups.google.com/group/Google-Web-Toolkit-Contributors

Reply via email to