Fire error events for a variety of script and style failures.

Fire error events for a variety of script and style failures.

This is a follow up to https://codereview.chromium.org/1010433007/ that
adds support for firing error events for a variety of script and style
failures. In particular, for scripts, it adds error firing for MIME type
blocking and CSP script hash failures. For styles, it add support for
general CSP style-hash failures and style-hash and style-nonce failures.

BUG=464648
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=192646

[jww, 2015-03-25 02:34:23]

Mike, can you take a look at this? I'm a bit unsure about how we should be
treating CSP failures inside SVGStyleElement (pdr@ is CC'd to get his feedback
as well). In particular, pdr@ pointed out that there are no onerror handlers for
SVG style elements, so it might not make sense to have errors there. Thus, in
this first stab, I've left out treating SVGStyleElements at all. Let me know
what you think.

[jww, 2015-03-25 02:34:45]

jww@chromium.org changed reviewers:
+ mkwst@chromium.org

[jww, 2015-03-25 02:34:46]

Whoops, forgot to actually add mkwst@ and pdr@ the first time :-)

[pdr., 2015-03-25 03:08:36]

On 2015/03/25 at 02:34:23, jww wrote:
> Mike, can you take a look at this? I'm a bit unsure about how we should be
treating CSP failures inside SVGStyleElement (pdr@ is CC'd to get his feedback
as well). In particular, pdr@ pointed out that there are no onerror handlers for
SVG style elements, so it might not make sense to have errors there. Thus, in
this first stab, I've left out treating SVGStyleElements at all. Let me know
what you think.

Sorry, I was not clear here. I think we should implement this for SVG, as SVG
does have an onerror handler. This patch gets close--both SVGStyleElement and
HTMLStyleElement inherit from StyleElement so there isn't much more work needed
to support SVG. Similarly for SVGScriptElement. Fortunately, no SVGLinkElement!

"bool childrenChanged" would be confusing to someone unfamiliar with this code.
Could we use an enum here to make this clearer?

[jww, 2015-03-25 20:20:57]

On 2015/03/25 03:08:36, pdr wrote:
> On 2015/03/25 at 02:34:23, jww wrote:
> > Mike, can you take a look at this? I'm a bit unsure about how we should be
> treating CSP failures inside SVGStyleElement (pdr@ is CC'd to get his feedback
> as well). In particular, pdr@ pointed out that there are no onerror handlers
for
> SVG style elements, so it might not make sense to have errors there. Thus, in
> this first stab, I've left out treating SVGStyleElements at all. Let me know
> what you think.
> 
> Sorry, I was not clear here. I think we should implement this for SVG, as SVG
> does have an onerror handler. This patch gets close--both SVGStyleElement and
> HTMLStyleElement inherit from StyleElement so there isn't much more work
needed
> to support SVG. Similarly for SVGScriptElement. Fortunately, no
SVGLinkElement!
Cool! You're right; I totally misunderstood :-) How would you feel if I did that
in a separate CL? It seems like it might be a fair amount of extra plumbing to
get it working.
> 
> "bool childrenChanged" would be confusing to someone unfamiliar with this
code.
> Could we use an enum here to make this clearer?
Sure, make's sense. I'll do that shortly.

[Mike West, 2015-03-26 04:41:08]

LGTM once pdr@ is happy.

https://codereview.chromium.org/1032033002/diff/20001/Source/core/html/HTMLLi...
File Source/core/html/HTMLLinkElement.cpp (right):

https://codereview.chromium.org/1032033002/diff/20001/Source/core/html/HTMLLi...
Source/core/html/HTMLLinkElement.cpp:749:
notifyLoadedSheetAndAllCriticalSubresources(true);
Nit: We should totally change this bool to an enum. It makes no sense without
checking the header. :/

[pdr., 2015-03-26 19:07:23]

LGTM for SVG as a followup.

[jww, 2015-03-26 21:04:16]

jww@chromium.org changed reviewers:
+ pdr@chromium.org

[jww, 2015-03-26 21:04:16]

The CQ bit was checked by jww@chromium.org

[jww, 2015-03-26 21:04:17]

Thanks for the reviews. Will start work on the SVG business today.

https://codereview.chromium.org/1032033002/diff/20001/Source/core/html/HTMLLi...
File Source/core/html/HTMLLinkElement.cpp (right):

https://codereview.chromium.org/1032033002/diff/20001/Source/core/html/HTMLLi...
Source/core/html/HTMLLinkElement.cpp:749:
notifyLoadedSheetAndAllCriticalSubresources(true);
On 2015/03/26 at 04:41:08, Mike West wrote:
> Nit: We should totally change this bool to an enum. It makes no sense without
checking the header. :/

Done.

[jww, 2015-03-26 21:04:17]

The patchset sent to the CQ was uploaded after l-g-t-m from pdr@chromium.org,
mkwst@chromium.org
Link to the patchset: https://codereview.chromium.org/1032033002/#ps40001
(title: "Changed bool to enum")

[commit-bot: I haz the power, 2015-03-26 21:04:38]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1032033002/40001

[jww, 2015-03-26 22:32:54]

The CQ bit was checked by jww@chromium.org

[jww, 2015-03-26 22:32:55]

The patchset sent to the CQ was uploaded after l-g-t-m from pdr@chromium.org,
mkwst@chromium.org
Link to the patchset: https://codereview.chromium.org/1032033002/#ps60001
(title: "Fixed test failures")

[commit-bot: I haz the power, 2015-03-26 22:33:14]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1032033002/60001

[commit-bot: I haz the power, 2015-03-27 03:36:22]

Committed patchset #4 (id:60001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=192646