summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* This is the first part of the PageAction implementation. More work is ↵finnur@chromium.org2009-05-0124-40/+701
| | | | | | | | | | | | | | | | required, but this is a good checkpoint. Design doc: http://dev.chromium.org/developers/design-documents/extensions/page-actions-api This checkin only covers Tab scoped page actions (not type "permanent"). It works end to end (if you have an extension that supplies the page action info -- I created an RSS page action that links to Google Reader). Please note that TabIndex is hard coded to 0 until the extension system can provide the tab id to the extensions (which I understand is in progress). This means that page action(s) only show up for the first tab in the tabstrip. :) BUG=None TEST=There is a unit test for the API, but apart from that it is not possible to test this manually without writing an extension that adds a PageAction. My RSS page action is not ready to be checked in but I can provide it if there is interest in a sneak preview during review/QA. Review URL: http://codereview.chromium.org/99253 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15105 0039d316-1c4b-4281-b951-d872f2087c98
* Ensure that the underlying web_contents has focus in the external tab containerananta@chromium.org2009-05-011-0/+1
| | | | | | | | | when it receives the SetInitialFocus notification from the client. Review URL: http://codereview.chromium.org/102022 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15104 0039d316-1c4b-4281-b951-d872f2087c98
* Temporarily disable two DOM checker tests to see if it has any effect onpatrick@chromium.org2009-05-011-2/+4
| | | | | | | | | | non-determinism. BUG=9394 TEST=DOM checker automated tests should still function properly. Review URL: http://codereview.chromium.org/99299 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15103 0039d316-1c4b-4281-b951-d872f2087c98
* POSIX: Add a macro for handling EINTR.agl@chromium.org2009-05-0127-212/+221
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On POSIX systems, system calls can be interrupted by signals. In this case, they'll return EINTR, indicating that the system call needs to be restarted. (The situation is a little more complicated than this with SA_RESTART, but you can read man 7 signal if you like.) The short of it is that you need to catch EINTR and restart the call for these system calls: * read, readv, write, writev, ioctl * open() when dealing with a fifo * wait* * Anything socket based (send*, recv*, connect, accept etc) * flock and lock control with fcntl * mq_ functions which can block * futex * sem_wait (and timed wait) * pause, sigsuspend, sigtimedwait, sigwaitinfo * poll, epoll_wait, select and 'p' versions of the same * msgrcv, msgsnd, semop, semtimedop * close (although, on Linux, EINTR won't happen here) * any sleep functions (careful, you need to handle this are restart with different arguments) We've been a little sloppy with this until now. This patch adds a macro for dealing with this and corrects every case of these system calls (that I found). The macro is HANDLE_EINTR in base/eintr_wrapper.h. It's safe to include on Windows and is a no-op there. On POSIX, it uses GCC magic to return the correct type based on the expression and restarts the system call if it throws EINTR. And you can use it like: HANDLE_EINTR(close(fd)); Or: ssize_t bytes_read = HANDLE_EINTR(read(fd, buffer, len)); *BEWARE* that it will evaluate the argument multiple times, so this is not safe: HANDLE_EINTR(close(FireMissiles())); http://groups.google.com/group/chromium-dev/browse_thread/thread/41a35b2a457d73a0 http://codereview.chromium.org/100225 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15102 0039d316-1c4b-4281-b951-d872f2087c98
* Don't valgrind ThirtyFourTabs, it takes too long :-(dkegel@google.com2009-05-011-0/+2
| | | | | | Review URL: http://codereview.chromium.org/99291 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15100 0039d316-1c4b-4281-b951-d872f2087c98
* Enable user_script_master_unittest for Mac since DirectoryWatcher was ported.shess@chromium.org2009-05-012-8/+10
| | | | | | | | [See http://crbug.com/10967 ] Review URL: http://codereview.chromium.org/99280 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15099 0039d316-1c4b-4281-b951-d872f2087c98
* Run workers in separate processes.jam@chromium.org2009-05-0117-78/+392
| | | | | | Review URL: http://codereview.chromium.org/99016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15098 0039d316-1c4b-4281-b951-d872f2087c98
* Pass down the opener tab when a message channel is opened to an extension.mpcomplete@google.com2009-05-0125-200/+281
| | | | | | | | | | | Also did a bunch of cleanup of ExtensionMessageService. I converted it to primarily UI-thread habitation, with one function that needs to be on the IO thread so it can handle a synchronous IPC message. TEST=N/A Review URL: http://codereview.chromium.org/99261 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15097 0039d316-1c4b-4281-b951-d872f2087c98
* Revert "POSIX: Don't allow onunload handlers to hang a renderer forever."agl@chromium.org2009-05-014-39/+0
| | | | | | | | | | | | | This reverts commit r15025. Calling exit() on another thread meant V8's destructor functions got called while another thread was still in V8. I might not have a better plan, but it's making a mess of the valgrind builders so I'm reverting for now with an eye to landing again next week. git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15095 0039d316-1c4b-4281-b951-d872f2087c98
* Remove StopAutocomplete() helper from AutocompletePopupViewMac.shess@chromium.org2009-05-013-6/+2
| | | | | | | | When the code was originally written, GetModel() was not required by AutocompletePopupView, now it's easy enough to use that instead. Review URL: http://codereview.chromium.org/99278 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15094 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a crash on the SSL logic, when a state transitionrvargas@google.com2009-05-011-5/+9
| | | | | | | | | | | | | to DoHandshakeReadComplete() is performed somewhere else than on DoHandshakeRead(). BUG=11296 TEST=navigate to an SSL page (see the bug, and crbug.com/1135 for a simpler test case). Review URL: http://codereview.chromium.org/100269 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15093 0039d316-1c4b-4281-b951-d872f2087c98
* Set endianness for ARM.thestig@chromium.org2009-05-011-2/+1
| | | | | | Review URL: http://codereview.chromium.org/100271 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15092 0039d316-1c4b-4281-b951-d872f2087c98
* Fix typo in test expectationsamanda@chromium.org2009-05-011-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15091 0039d316-1c4b-4281-b951-d872f2087c98
* Create the directory for Mac dev build archiving to use and put in an ↵thomasvl@chromium.org2009-05-011-0/+1
| | | | | | | | initial files list for what to archive. Review URL: http://codereview.chromium.org/100273 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15089 0039d316-1c4b-4281-b951-d872f2087c98
* Deprecate wstring version of PathService::Get() in net.thestig@chromium.org2009-05-018-49/+52
| | | | | | Review URL: http://codereview.chromium.org/100240 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15088 0039d316-1c4b-4281-b951-d872f2087c98
* Call WebWorkerClient on the main thread. This makes it consistent with the ↵jam@chromium.org2009-05-018-120/+200
| | | | | | | | | rest of the WebKit API, which is single threaded. Also a bunch of small fixes to make layout tests pass: the dll was being unloaded while its functions were still queued to be dispatched, and a string allocated in the dll was being GC'd in test shell. BUG=11011 Review URL: http://codereview.chromium.org/102005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15087 0039d316-1c4b-4281-b951-d872f2087c98
* Add AS to scons_import_variables and scons_propagate_variables.thestig@chromium.org2009-05-011-0/+2
| | | | | | Review URL: http://codereview.chromium.org/99292 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15086 0039d316-1c4b-4281-b951-d872f2087c98
* Test regressions for WebKit merge 43059:43114amanda@chromium.org2009-05-011-1/+9
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15085 0039d316-1c4b-4281-b951-d872f2087c98
* Roll DEPS to update webkit versionawalker@google.com2009-05-011-1/+1
| | | | git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15082 0039d316-1c4b-4281-b951-d872f2087c98
* Chromium side of merge 43059:43114awalker@google.com2009-05-012-2/+2
| | | | | | Review URL: http://codereview.chromium.org/99290 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15080 0039d316-1c4b-4281-b951-d872f2087c98
* Only unset the kWebPluginDelegateProperty property after calling the ↵jam@chromium.org2009-05-011-25/+19
| | | | | | | | plugin's windowproc, otherwise if it ends up dispatching another message to itself we'll hit the NOTREACHED at the beginning of the function. Review URL: http://codereview.chromium.org/99267 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15077 0039d316-1c4b-4281-b951-d872f2087c98
* Roll DEPS to pick up ResourceResponse unforking.dglazkov@chromium.org2009-05-011-1/+1
| | | | | | | | | TBR=eseidel BUG=8207 TEST=0 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15075 0039d316-1c4b-4281-b951-d872f2087c98
* Strings for default browser infobar.ben@chromium.org2009-05-013-0/+19
| | | | | | | http://crbug.com/9049 Review URL: http://codereview.chromium.org/100267 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15073 0039d316-1c4b-4281-b951-d872f2087c98
* Suppress valgrind warnings fordank@chromium.org2009-05-011-0/+89
| | | | | | | | | | | | | | | http://crbug.com/9561 http://crbug.com/11270 http://crbug.com/11271 http://crbug.com/11286 and a couple of things that seem clearly to be system issues rather than chromium issues. Just trying to get the mac unit_test valgrind builder green. Review URL: http://codereview.chromium.org/99259 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15071 0039d316-1c4b-4281-b951-d872f2087c98
* Add atomic operations for ARM.thestig@chromium.org2009-05-013-1/+133
| | | | | | Review URL: http://codereview.chromium.org/57031 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15070 0039d316-1c4b-4281-b951-d872f2087c98
* DevTools: Fix access violation (null pointer) on document tear down.pfeldman@chromium.org2009-05-013-5/+8
| | | | | | Review URL: http://codereview.chromium.org/100261 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15067 0039d316-1c4b-4281-b951-d872f2087c98
* Measure IPC latency for audiohclam@chromium.org2009-05-013-7/+69
| | | | | | | | | | | | | | | | | | | | | | | With this patch there's totally 5 measurements for audio IPC 1. Total round trip time browser request a packet until it gets a packet 2. Browser receive latency time for a packet to arrive browser from renderer 3. Browser process time time for browser to process the packet 4. Renderer receive latency time for a packet to arrive renderer from browser 5. Renderer process time process time on a packet in renderer They are all using low resolution timer with accuracy of ~1-15ms. Since we have a budget of ~200ms (for now) without clicks, I think the accuracy shouldn't be a big problem, unless for case 1 where we want to measure exact amount of clicks. Review URL: http://codereview.chromium.org/99213 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15063 0039d316-1c4b-4281-b951-d872f2087c98
* Release the autocomplete popup panel when not visible.shess@chromium.org2009-05-011-0/+8
| | | | | | Review URL: http://codereview.chromium.org/100260 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15061 0039d316-1c4b-4281-b951-d872f2087c98
* Increase the firstrun bubble width for Tamil.jungshik@google.com2009-05-011-1/+1
| | | | | | | | | | | | To be merged with 2.0 (172) branch. BUG=http://crbug.com/11208 TBR=mal Review URL: http://codereview.chromium.org/99260 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15060 0039d316-1c4b-4281-b951-d872f2087c98
* Fix make build by making sure .pak files are listed as outputs of grd files.tc@google.com2009-05-015-53/+130
| | | | | | | | | | | Specifically: 1) In webkit.gyp and test_shell.gyp, convert the rules to actions because of different grd outputs (localized grd files have locale in the .pak filename). 2) In chrome, make two targets each with a custom rule. One target is for localized resources, the other for non-localized resources. This is like the current windows build. Review URL: http://codereview.chromium.org/100174 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15059 0039d316-1c4b-4281-b951-d872f2087c98
* Making CloseWindow and CloseTab automation APIhuanr@chromium.org2009-05-0110-66/+240
| | | | | | | | | | | | | synchronous and robust. Adding automation APIs with corresponding IPC messages to count and find normal browser windows. Review URL: http://codereview.chromium.org/99268 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15058 0039d316-1c4b-4281-b951-d872f2087c98
* Enables key view loop autorecalculation in the BrowserWindow on Mac. This fixesrohitrao@chromium.org2009-05-011-1/+1
| | | | | | | | | | | | a bug where you would cycle through a background tab's links even after selecting a different tab. TEST=Open Chromium. Open a new tab. Press tab and verify that we are correctly cycling through the links in the new tab. Review URL: http://codereview.chromium.org/100233 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15057 0039d316-1c4b-4281-b951-d872f2087c98
* Replace window.external with a v8::extension.mbelshe@google.com2009-05-019-79/+89
| | | | | | | | | BUG=http://crbug.com/10957 TEST=Verify that the AddSearchProvider logic still works. This dates back to an old bug http://b/1069781 Review URL: http://codereview.chromium.org/99168 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15056 0039d316-1c4b-4281-b951-d872f2087c98
* More focus fixing.avi@google.com2009-05-011-1/+2
| | | | | | Review URL: http://codereview.chromium.org/99262 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15055 0039d316-1c4b-4281-b951-d872f2087c98
* Allow Frame.cpp to be unforked.levin@chromium.org2009-05-011-1/+7
| | | | | | | | | | BUG=11263 TEST=Go to any page and then click on a link to navigate to a new page. Review URL: http://codereview.chromium.org/99263 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15052 0039d316-1c4b-4281-b951-d872f2087c98
* Only send events to renderers that are listening to an event.Notify browser ↵erikkay@google.com2009-05-0110-11/+137
| | | | | | | | when a renderer is listening to an event. Review URL: http://codereview.chromium.org/100214 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15050 0039d316-1c4b-4281-b951-d872f2087c98
* chrome.gyp updates:sgk@google.com2009-05-011-11/+104
| | | | | | | | | | | | | | | | * remove chrome\browser\views\find_bar_win_unittest.cc * add Windows-only test targets: ** interactive_ui_tests ** plugin_tests ** selenium_tests * Move from unit_tests to test_support_common: ** test/in_process_browser_test.{cc,h} ** test/ui_test_utils.{cc,h} Review URL: http://codereview.chromium.org/100168 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15045 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a small undercalculation of the space available a result description.deanm@chromium.org2009-05-011-3/+3
| | | | | | | | | Calculate from the actual text width, not the allocated text width. Review URL: http://codereview.chromium.org/99222 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15042 0039d316-1c4b-4281-b951-d872f2087c98
* Glue part for the provisional WebKit merge ↵pfeldman@chromium.org2009-05-013-1/+8
| | | | | | | | (https://bugs.webkit.org/show_bug.cgi?id=25347).TBR=yurys Review URL: http://codereview.chromium.org/99274 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15041 0039d316-1c4b-4281-b951-d872f2087c98
* Undo the reference counting experiment. The crashes are stillager@chromium.org2009-05-011-30/+2
| | | | | | | | | | happening, so it seems unlikely that this is a reference counting problem. BUG=9746 Review URL: http://codereview.chromium.org/100247 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15038 0039d316-1c4b-4281-b951-d872f2087c98
* Wrappers for inline style declarations disappeared on GC. Fix theager@chromium.org2009-05-011-1/+14
| | | | | | | | issue by adding the style declaration wrappers to the object groups of the elements they belong to. Review URL: http://codereview.chromium.org/100204 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15037 0039d316-1c4b-4281-b951-d872f2087c98
* Add a title to the "download in progress" dialog.abarth@chromium.org2009-05-011-0/+4
| | | | | | | | | | | R=thestig BUG=10681 TEST=Start a long download. Try to quit the browser. Notice the spiffy title on the dialog box. Review URL: http://codereview.chromium.org/100241 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15035 0039d316-1c4b-4281-b951-d872f2087c98
* Roll DEPS to pick up Document/Element unfork.dglazkov@google.com2009-05-011-1/+1
| | | | | | | | | | TBR=eseidel BUG=10898 TEST=0 Review URL: http://codereview.chromium.org/100244 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15033 0039d316-1c4b-4281-b951-d872f2087c98
* Deprecate wstring version of PathService::Get() in webkit.thestig@chromium.org2009-05-019-42/+53
| | | | | | Review URL: http://codereview.chromium.org/99266 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15032 0039d316-1c4b-4281-b951-d872f2087c98
* Fix issue 11264fqian@google.com2009-05-011-16/+30
| | | | | | | | | | | http://code.google.com/p/chromium/issues/detail?id=11264 Integer 0 and -1 have special meaning in HashMap<int, ...>, both cannot be used as key. Specially filter out these two keys. Review URL: http://codereview.chromium.org/100242 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15031 0039d316-1c4b-4281-b951-d872f2087c98
* Second part of fix of issue ↵fqian@google.com2009-05-011-1/+12
| | | | | | | | | | | | http://code.google.com/p/chromium/issues/detail?id=3285. A NPObject can be called as a construct if it implements NPN_Construct, and NOT implements NPN_InvokeDefault. Otherwise, NPN_InvokeDefault is called even when the object is called as a constructor. Review URL: http://codereview.chromium.org/56167 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15029 0039d316-1c4b-4281-b951-d872f2087c98
* Linux: terminate clipboard handling on the UI thread.agl@chromium.org2009-05-013-27/+141
| | | | | | | | | | | | | | | | | | | | | | | After discussing with various people, I think we're going to do this in the short term at least. Currently we're calling into GTK from the IO thread and Bad Things Happen when one does that. We would like to write an Xlib, fully asynchronous system for dealing with the clipboard (and to get the clipboard stuff out of base!). That would let us avoid sending the selection over the IPC channel each time it updates too. However, that's going to be a lot of work and we have crashing browsers happening /right now/. Also, Evan thinks that maybe we don't have the deadlock situation on Linux that we do on Windows with terminating sync requests from the renderer on the UI thread. http://codereview.chromium.org/100238 BUG=9865 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15028 0039d316-1c4b-4281-b951-d872f2087c98
* Undisable the temporarily disabled test, by fixing the URL expectation to ↵ericroman@google.com2009-05-011-3/+2
| | | | | | | | | | | have %22 instead of \" (to match change in GURL canonicalization). BUG=11142 TBR=evan Review URL: http://codereview.chromium.org/100148 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15027 0039d316-1c4b-4281-b951-d872f2087c98
* Fix to test_expectations.txtasargent@chromium.org2009-05-011-3/+7
| | | | | | | | | | | | I accidentally marked mac/linux as crash when I really wanted to mark win BUG=11264 TBR=dglazkov Review URL: http://codereview.chromium.org/100239 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15026 0039d316-1c4b-4281-b951-d872f2087c98
* POSIX: Don't allow onunload handlers to hang a renderer forever.agl@chromium.org2009-05-014-0/+39
| | | | | | | | | | | | | | | | | | | | | | On POSIX one can install an unload handler which loops forever and leave behind a renderer process which eats 100% CPU forever. This is because the terminate signals (ViewMsg_ShouldClose and the error from the IPC channel) are routed to the main message loop but never processes (because that message loop is stuck in V8). One could make the browser SIGKILL the renderers, but that leaves open a large window where a browser failure (or a user, manually terminating the browser because "it's stuck") will leave behind a process eating all the CPU. On Windows we don't have this issue because all the processes are in a job so when the parent dies, all the children are killed too. http://codereview.chromium.org/100222 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@15025 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-02-11 20:06:30 +0000