Fix nested GuestView's container bounds calculation.

Fix nested GuestView's container bounds calculation.
This results in incorrect context menu positioning in pdf in webview or
  webview inside appview.

If a guest A is nested within an embedder B, which in turn is nested inside
  an embedder C, then we used to add offset twice to calculate the bounds of A
  because BrowserPluginGuest::GetScreenCoordinates would add:
    1. offset of C from A and
    2. offset of B from A.
This CL removes step #2.
From logs, it seems to be there from the beginning of appview implementation.

BUG=507223
Test=Load a PDF inside <webview> and check context menu inside the PDF.
Easier to test with the browser sample:
https://github.com/GoogleChrome/chrome-app-samples/tree/master/samples/webview-samples/browser
You should observe the context menu is shown in the right place instead of
it being shifted.

Committed: https://crrev.com/87a98e428afa9dff6d02ff37d1504411d40fda3d
Cr-Commit-Position: refs/heads/master@{#339635}

[lazyboy, 2015-07-21 01:12:47]

lazyboy@chromium.org changed reviewers:
+ fsamuel@chromium.org, raymes@chromium.org

[lazyboy, 2015-07-21 01:12:47]

+fsamuel for BrowserPluginGuest,
+raymes for test related changes in chrome/browser/pdf/*

[raymes, 2015-07-21 01:20:14]

Thanks :)

https://chromiumcodereview.appspot.com/1237343002/diff/80001/chrome/browser/p...
File chrome/browser/pdf/pdf_extension_util.cc (right):

https://chromiumcodereview.appspot.com/1237343002/diff/80001/chrome/browser/p...
chrome/browser/pdf/pdf_extension_util.cc:53: bool
EnsurePDFHasLoaded(content::WebContents* web_contents) {
This file is used in production code however this function should never be used
in production code and I wouldn't want someone to accidentally do that. Can we
move this function to a test-only file?

[lazyboy, 2015-07-21 03:55:06]

https://chromiumcodereview.appspot.com/1237343002/diff/80001/chrome/browser/p...
File chrome/browser/pdf/pdf_extension_util.cc (right):

https://chromiumcodereview.appspot.com/1237343002/diff/80001/chrome/browser/p...
chrome/browser/pdf/pdf_extension_util.cc:53: bool
EnsurePDFHasLoaded(content::WebContents* web_contents) {
On 2015/07/21 at 01:20:14, raymes wrote:
> This file is used in production code however this function should never be
used in production code and I wouldn't want someone to accidentally do that. Can
we move this function to a test-only file?

Oh I missed that seeing pdf_extension_util and pdf_extenion_test side by side, I
was reading it "pdf_extension_test_util" for some reason.

Moved stuff to pdf_extension_test_util.*

[raymes, 2015-07-21 04:20:36]

lgtm

https://chromiumcodereview.appspot.com/1237343002/diff/100001/chrome/browser/...
File chrome/browser/pdf/pdf_extension_test_util.h (right):

https://chromiumcodereview.appspot.com/1237343002/diff/100001/chrome/browser/...
chrome/browser/pdf/pdf_extension_test_util.h:18: // BrowserPluginGuest to ensure
the PDFScriptingAPI works correctly from there.
You can remove the last sentence. It doesn't make sense in the context here.

[Fady Samuel, 2015-07-21 04:21:58]

lgtm

[lazyboy, 2015-07-21 06:32:23]

https://chromiumcodereview.appspot.com/1237343002/diff/100001/chrome/browser/...
File chrome/browser/pdf/pdf_extension_test_util.h (right):

https://chromiumcodereview.appspot.com/1237343002/diff/100001/chrome/browser/...
chrome/browser/pdf/pdf_extension_test_util.h:18: // BrowserPluginGuest to ensure
the PDFScriptingAPI works correctly from there.
On 2015/07/21 at 04:20:36, raymes wrote:
> You can remove the last sentence. It doesn't make sense in the context here.

Done.

[lazyboy, 2015-07-21 07:33:18]

The CQ bit was checked by lazyboy@chromium.org

[lazyboy, 2015-07-21 07:33:18]

The patchset sent to the CQ was uploaded after l-g-t-m from
fsamuel@chromium.org, raymes@chromium.org
Link to the patchset:
https://chromiumcodereview.appspot.com/1237343002/#ps140001 (title: "fix type
mismatch for win_x64 compile")

[commit-bot: I haz the power, 2015-07-21 07:33:28]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1237343002/140001

[commit-bot: I haz the power, 2015-07-21 10:12:16]

Committed patchset #8 (id:140001)

[commit-bot: I haz the power, 2015-07-21 10:13:33]

Patchset 8 (id:??) landed as
https://crrev.com/87a98e428afa9dff6d02ff37d1504411d40fda3d
Cr-Commit-Position: refs/heads/master@{#339635}