| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
coming in from the renderer.
A failed HTTP navigation like a 404 response to a request is followed by two responses. The first one which is associated with the failed response. This does not send over any information about the failure to the browser and thus appears as a normal navigation.The second response is for the actual 404 page being loaded.
For network errors the browser does get notified via RenderView::DidFailProvisionalLoadWithError. However due to a prototype mismatch the corresponding function in WebContents is never invoked.
Added a new automation message AutomationMsg_NavigationFailed, which carries information about failed navigations to automation clients.The changes to the navigation controller include sending over the http status code and the URL to observers.
The ExternalTabContainer also subscribes to the FAIL_PROVISIONAL_LOAD_WITH_ERROR notification, so it can inform clients about errors. We also ignore the next NAV_ENTRY_COMMITTED notification after an error due to the reasons mentioned above.
Review URL: http://codereview.chromium.org/21495
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@10023 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ExternalTabContainer. This ensures that we don't crash the browser while trying to dereference a freed AutomationProvider pointer.
When a Chrome browser instance starts up, we attempt to locate an already running instance and defer to it to complete the navigation request. However it is quite possible for the running instance to exit while we attempt to send a WM_COPYDATA message to it. This caused a bunch
of ASSERTS to fire off in the browser.
Fixes as below:-
1. If GetWindowThreadProcessId fails, we bail out and try to launch a new chrome instance
2. If SendMessageTimeout fails, we check if the window is still valid. If not we bail out and try to launch a new chrome instance.
3. We return an error from the WM_COPYDATA handler if the chrome instance is in the process of shutting down. We handle this at the caller end, i.e. in NotifyOtherProcess.
Bug=1643310
Review URL: http://codereview.chromium.org/23016
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@9536 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
found the automation tests not hanging.
I also took a look at the logs from the hanging buildbots and the reverted change didn't touch on any of the code that appeared to be hanging.
I also put breakpoints on the modified functions while running ui_tests.exe and nothing hit, so it looks like a fluke.
I did get a hang though when running an old build of ui_tests.exe with a new version of chrome.dll.
Is there any chance that there could have been a stale version on the build bots?
Review URL: http://codereview.chromium.org/20214
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@9478 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
TBR=tommi
Review URL: http://codereview.chromium.org/20179
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@9392 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
automated tab.
Review URL: http://codereview.chromium.org/20125
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@9388 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
really painful to add more types, since lots of headers include the
notification service to derive from the notification observer. This splits that
out, so much less of the project should end up including notification_types.h
---Paths modified but not in any changelist:
Review URL: http://codereview.chromium.org/19744
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@9020 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
Improving automation interface to offer better control over
window creation of external tabs. The changes allow us
to specify a parent window, initial size and window style.
Review URL: http://codereview.chromium.org/19048
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@8828 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
| |
Review URL: http://codereview.chromium.org/18250
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@8058 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When a browser window is closed, we just terminate
all existing renderers for fast shutdown. If we have
any renderers hosted externally using automation they
are also killed during this process.
The fix is to prevent fast termination if any of the
tabs are hosted by automation.
Review URL: http://codereview.chromium.org/17453
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7901 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
Review URL: http://codereview.chromium.org/14173
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@7126 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
R=erg
review url = http://codereview.chromium.org/11348/show
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@5849 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Explicitly set_delegate(NULL) before releasing a TabContents. (I think
the order messages are getting handled is different from XP; this only
seems neccessary under Vista.)
- r4523 and someone elses change added entries to theme_resources.rc at
roughly the same time, causing happy fun crashes.
- Roll in sgk's SConscript fix.
Review URL: http://codereview.chromium.org/9373
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@4799 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
TBR=nsylvain
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@4539 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
(while leaving the HTTP authentication dialog support) and add a new type of constrained window which owns all blocked popups. Rip out dead unit tests.
Review URL: http://codereview.chromium.org/8782
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@4523 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
http://code.google.com/p/chromium/issues/detail?id=3224
Caveats:
* Did not update TabNavigation yet. Hence session restore will continue to load the tabs with empty referrer.
* Did not plumb referrer into incognito url open. (Not sure what the right thing to do is here with respect to privacy vs compatibility.)
* Did not plumb referrer throughout the automation controller. No functional impact here, but it makes the code inconsistent with the non-automation version.
Review URL: http://codereview.chromium.org/8186
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@4036 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
http://crbug.com/2186
Review URL: http://codereview.chromium.org/7468
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@3550 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
| |
http://crbug.com/2188
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@3495 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
http://crbug.com/3430
Review URL: http://codereview.chromium.org/7376
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@3441 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
I added additional information to the regular load commit notification so all
interested parties can listen for that instead.
I removed the old navigation type enum, and replaced it with the enum from
the NavigationController, so it's now public.
Review URL: http://codereview.chromium.org/3112
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@2573 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
This makes view source for some pages (for example Google Reader) not work
properly. I speculate that telling WebKit to change the encoding (which causes
a reload) right after starting the real load makes it confused.
Review URL: http://codereview.chromium.org/3022
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@2143 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
| |
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@1287 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
One item not changed yet is to rename 'receiver' in ForwardMessageToExternalHost.
The idea is to invoke receiver("message") at the other end. So for example if the
args to ForwardMessageToExternalHost("hello", "world") then we will invoke a
script hello("world") on the other side.
'receiver' doesn't really describe the first argument here so if there is a
better suggestion, I would be happy to change it :)
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@1176 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
| |
they're always ASCII. Plus, wstrings will also be 32-bit characters on Linux and Mac, which is expensive. So I make this CL by using std::string instead of std::wstring every time you have an encoding name.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@1144 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
LG by brett. I just merge them to single one for passing compilation.
The CL is to fix bug http://b/issue?id=1281734. FireFox uses encoding used by the original webpage to decode the source of the webpage, Chrome does not. We need to add a new parameter which carries the override encoding when opening a view-source URL for viewing its source.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@997 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
TBR=iyengar
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@959 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
- Slightly reduce the size of the generated .lib files ~3%.
- Reduce the number of implicit and explicit atl and windows includes. hooray!
- Help incremental build by reducing the number of unnecessary included files.
- Split some template class in two, one base class for the common code and the specialization that inherits from the base class.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@937 0039d316-1c4b-4281-b951-d872f2087c98
|
|
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15 0039d316-1c4b-4281-b951-d872f2087c98
|