KURL not handling NULL m_string members properly.

KURL not handling NULL m_string members properly.

Due to some recent refactorings, its possible to get a NULL
m_string inside of KURL given some bad UFT-8 input. This
corrects the checks in several places where we would formerly
segv.

BUG=275868
R=abarth@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=157922

[Tom Sepez, 2013-09-17 19:34:37]

Adam, please review.  I'm afraid you'll have to look at inline diffs, since
reitveld isn't co-operating at the moment.

[Tom Sepez, 2013-09-17 19:38:02]

> I'm afraid you'll have to look at inline diffs, since reitveld isn't
co-operating at the moment.

Yay! Base files.

[abarth-chromium, 2013-09-17 19:48:34]

https://codereview.chromium.org/24095009/diff/5001/Source/weborigin/KURL.cpp
File Source/weborigin/KURL.cpp (right):

https://codereview.chromium.org/24095009/diff/5001/Source/weborigin/KURL.cpp#...
Source/weborigin/KURL.cpp:786: if (m_string.isNull() || !m_isValid) {
Shouldn't m_string.isNull() imply !m_isValid?

[Tom Sepez, 2013-09-17 19:55:37]

https://codereview.chromium.org/24095009/diff/5001/Source/weborigin/KURL.cpp
File Source/weborigin/KURL.cpp (right):

https://codereview.chromium.org/24095009/diff/5001/Source/weborigin/KURL.cpp#...
Source/weborigin/KURL.cpp:786: if (m_string.isNull() || !m_isValid) {
On 2013/09/17 19:48:35, abarth wrote:
> Shouldn't m_string.isNull() imply !m_isValid?
Yep. Just confirmed this one worked previously.  Hence m_isValid =>
!m_string.isNull(), so the check below can just be removed.

[abarth-chromium, 2013-09-17 19:57:02]

https://codereview.chromium.org/24095009/diff/12001/Source/weborigin/KURL.cpp
File Source/weborigin/KURL.cpp (right):

https://codereview.chromium.org/24095009/diff/12001/Source/weborigin/KURL.cpp...
Source/weborigin/KURL.cpp:300: if (m_string.isNull() || !m_isValid)
I don't understand why this change is needed.  If m_string.isNull, then
necessarily !m_isValid and we'll return early without this change.

[Tom Sepez, 2013-09-17 20:05:09]

On 2013/09/17 19:57:02, abarth wrote:
> https://codereview.chromium.org/24095009/diff/12001/Source/weborigin/KURL.cpp
> File Source/weborigin/KURL.cpp (right):
> 
>
https://codereview.chromium.org/24095009/diff/12001/Source/weborigin/KURL.cpp...
> Source/weborigin/KURL.cpp:300: if (m_string.isNull() || !m_isValid)
> I don't understand why this change is needed.  If m_string.isNull, then
> necessarily !m_isValid and we'll return early without this change.
Right.  Fixed this check in several other places in Patch Set 5.

[abarth-chromium, 2013-09-17 20:07:03]

lgtm

[commit-bot: I haz the power, 2013-09-17 22:09:15]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/tsepez@chromium.org/24095009/29001

[commit-bot: I haz the power, 2013-09-17 23:15:18]

Change committed as 157922