RE: HTML Imports and CSP

Sorry, it is so, regardless of whether anyone likes it ☺ The vast majority of security vulnerability classes result from confusing code with data:

·         Buffer overflows, integer overflows, UAFs, etc., all memory corruption, happens because of confusing code with data: the CPU gets its instructions from the same memory as it gets its data. A Harvard architecture would eliminate this problem.

·         XSS results from mixing code with data: hand a user a URL that they think is a web address, and it loads and runs code in their context.

·         CSP failed to wipe out XSS at a stroke, precisely because of inline scripting that makes creating effective CSP difficult.

When you mix code with data inline, you create signaling and escaping problems that make it easy to confuse code with data. So there are exactly two choices here:

1.       Don’t mix code with data.

2.       Do mix code with data, and accept a never-ending burden of perpetual security issues.

Here’s a talk I gave at the NSF last year on mixing code with data http://www.tvworldwide.com/events/nsf/140717/globe_show/default_go_archive.cfm?gsid=2562&type=flv&test=0&live=0

That web site will ask you to “register” before you can view the video, which only means that they ask you for your e-mail address, they don’t even use passwords. So you can register yourself, or you can just enter crispin@microsoft.com<mailto:crispin@microsoft.com> and it will let you in.


From: Jim Manico [mailto:jim.manico@owasp.org]
Sent: Thursday, April 2, 2015 10:11 AM
To: Crispin Cowan
Cc: Mike West; Devdatta Akhawe; Justin Fagnani; public-webappsec@w3.org
Subject: Re: HTML Imports and CSP

This should not be so. Make it easier for coders with better franeworks. Systems like .NET LINQ allow the mixing and does the parameterization behind the hood. Also consider autoescaping templates.
--
Jim Manico
@Manicode
(808) 652-3805

On Apr 2, 2015, at 10:07 AM, Crispin Cowan <crispin@microsoft.com<mailto:crispin@microsoft.com>> wrote:
My point was that mixing code with data is always bad for security, but that doesn’t always make it the wrong choice, sometimes other considerations are more important. But I do not think we should mix code with data on the basis of “we do it elsewhere, so what the heck” ☺

From: Jim Manico [mailto:jim.manico@owasp.org]
Sent: Thursday, April 2, 2015 10:05 AM
To: Crispin Cowan
Cc: Mike West; Devdatta Akhawe; Justin Fagnani; public-webappsec@w3.org<mailto:public-webappsec@w3.org>
Subject: Re: HTML Imports and CSP

There are edge cases where inline script is necessary, like when dynamically loading JS based on certain conditions. Those edge cases are rare and only needed for extreme   performance tuning.
--
Jim Manico
@Manicode
(808) 652-3805

On Apr 2, 2015, at 9:37 AM, Crispin Cowan <crispin@microsoft.com<mailto:crispin@microsoft.com>> wrote:
Mixing code inline with data is *always* a bad idea. The web never should have allowed inline script, but it is too late to fix that now. The best we can do is avoid creating *new* opportunities to mix code with data.

So, I would block inline event handlers just on principle, unless there is some compelling compat need.

From: Mike West [mailto:mkwst@google.com]
Sent: Wednesday, April 1, 2015 11:50 PM
To: Devdatta Akhawe
Cc: Justin Fagnani; public-webappsec@w3.org<mailto:public-webappsec@w3.org>
Subject: Re: HTML Imports and CSP

On Thu, Apr 2, 2015 at 6:32 AM, Devdatta Akhawe <dev.akhawe@gmail.com<mailto:dev.akhawe@gmail.com>> wrote:
> Sure, why not? I'm not a huge fan of inline event handlers, but I don't
> think that allowing them (and inline script) in Imports actually increases
> the risk a developer exposes herself to. Do you think that's an incorrect
> analysis?

There is a huge difference in risk of XSS in inline event handlers and
inline scripts. Different types of contexts, nested contexts etc all
are issues. *cough* I think Joel wrote a paper about this :D

You're not wrong there; inline event handlers are bad and they should feel bad.

That said, is the risk really different in kind from just allowing plain old inline script that executes directly? It doesn't seem to be. Allowing one without allowing the other seems capricious.

> There's a difference between supporting a new feature which isn't currently
> covered by any directive, and changing the meaning of a directive that
> already covers a featue.

Again, because of the DOMXSS risk (even just injecting a new inline script tag
without nonce), I think not using a new directive messes up the
security invariants for any CSP adopters.

If we move imports from `script-src` to `import-src`, then sites that currently disallow imports will no longer have that protection. That seems bad.

What's your ideal solution?

-mike