summaryrefslogtreecommitdiffstats
path: root/chrome/common
Commit message (Collapse)AuthorAgeFilesLines
* Fonts and Languages Linux GTK Preparationmhm@chromium.org2009-06-221-0/+11
| | | | | | | | | | | Rearranged the windows and mac ports to use the cross platform interfaces. Updated windows fonts and languages to use those interfaces. As well, created stubs for mac platform. BUG=13524 (http://crbug.com/13524) TEST=Open a webpage with a textfield, right click and choose Spell checker options > Language Settings. Tested this on windows as well. Review URL: http://codereview.chromium.org/140043 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18900 0039d316-1c4b-4281-b951-d872f2087c98
* posix: two related changes to make plugin IPC work on POSIX.evan@chromium.org2009-06-206-35/+170
| | | | | | | | | | | | | | | | | | | | | | | | [retry, fix windows compile failure] * use a new ChannelHandle type when passing IPC channels over IPC The current POSIX code assumes that one end of a channel is always a new child process (a renderer). For plugins we need to be able to construct channels between each of the browser, plugin, and renderer. This change augments the messages related to creating channels to allow passing in a base::FileDescriptor containing the socket. The intent is that the browser process, as the initial interchange between plugin and renderer, creates the socketpair() on their behalf and hands each their respective end of the connection. * register channel endpoint names in the global pipe map The plugin code assumes it can map from a string to a channel endpoint at basically any time. So whenever we get a channel endpoint over IPC, we install it in a global map of channel endpoints. Review URL: http://codereview.chromium.org/113157 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18888 0039d316-1c4b-4281-b951-d872f2087c98
* Finish the wiring of the inactive user toast experimentcpu@google.com2009-06-191-1/+8
| | | | | | | | | | | | | | - Adds read-write to the client key of google_update - Adds the logic to track experiment outcome - Some new geric exitc codes (non error codes) BUG= b/1484308 TEST= see the bug Review URL: http://codereview.chromium.org/132058 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18871 0039d316-1c4b-4281-b951-d872f2087c98
* GTK: HTTP Auth dialogs under linux.erg@google.com2009-06-192-0/+6
| | | | | | | | | | | | - Implements a ConstrainedWindowGtk which positions itself in the center of its corresponding TabContentsViewGtk. - Implements LoginPromptGtk. HTTP Auth now works under Linux. - Renames ConstrainedWindowImpl to ConstrainedWindowWin http://crbug.com/11512 Review URL: http://codereview.chromium.org/132047 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18864 0039d316-1c4b-4281-b951-d872f2087c98
* back out r18853jeremy@chromium.org2009-06-191-5/+0
| | | | | | | | This reverts commit f2f30e9d0108f4a776672c6b29d25ed3a8ad19ad. Review URL: http://codereview.chromium.org/140013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18856 0039d316-1c4b-4281-b951-d872f2087c98
* This patch adds Cocoa editing selectors to the RenderViewWidgetHostMac class.jeremy@chromium.org2009-06-191-0/+5
| | | | | | | | | | | | | | | This makes us a better citizen in OSX text-editing land and is a first step in getting our text entry story to work more natively on OSX. The selectors are added at runtime to the Cocoa class and invoking them causes the appropriate WebKit core command to be invoked on the currently focused WebFrame. As a side-effect bug 10862 is fixed. BUG=10862 TEST=Open a new tab and type some text into an input element, change to a Hebrew keyboard layout and hit command-a, all text should be selected. Review URL: http://codereview.chromium.org/114070 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18853 0039d316-1c4b-4281-b951-d872f2087c98
* Revert "posix: two related changes to make plugin IPC work on POSIX."evan@chromium.org2009-06-196-170/+35
| | | | | | This reverts commit r18850, as it had a compile failure. git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18851 0039d316-1c4b-4281-b951-d872f2087c98
* posix: two related changes to make plugin IPC work on POSIX.evan@chromium.org2009-06-196-35/+170
| | | | | | | | | | | | | | | | | | | | | | * use a new ChannelHandle type when passing IPC channels over IPC The current POSIX code assumes that one end of a channel is always a new child process (a renderer). For plugins we need to be able to construct channels between each of the browser, plugin, and renderer. This change augments the messages related to creating channels to allow passing in a base::FileDescriptor containing the socket. The intent is that the browser process, as the initial interchange between plugin and renderer, creates the socketpair() on their behalf and hands each their respective end of the connection. * register channel endpoint names in the global pipe map The plugin code assumes it can map from a string to a channel endpoint at basically any time. So whenever we get a channel endpoint over IPC, we install it in a global map of channel endpoints. Review URL: http://codereview.chromium.org/113157 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18850 0039d316-1c4b-4281-b951-d872f2087c98
* part 3 of toolstrip dragging - persist the drag ordererikkay@google.com2009-06-191-0/+3
| | | | | | | | | | | | Also notify other browser windows of changes to the order. BUG=http://code.google.com/p/chromium/issues/detail?id=12123 TEST=browser_tests.exe --gtest_filter=ExtensionShelfModelTest.* Review URL: http://codereview.chromium.org/131114 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18845 0039d316-1c4b-4281-b951-d872f2087c98
* Enable DownloadRequestManager for Mac.paul@chromium.org2009-06-191-1/+1
| | | | | | | | | | | | | | | | DownloadRequestManager is the object that manages the number of downloads that can occur in a given tab. It throttles non-user initiated downloads to prevent carpet-bombing download attacks. Enabling this object for Mac fixes the issue where we can only download one file in a given tab. BUG=None (mac work). TEST=Can now download more than one item in a tab. Review URL: http://codereview.chromium.org/132039 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18840 0039d316-1c4b-4281-b951-d872f2087c98
* TBR: Fix Toolstrips sometimes show scrollbars as renderingaa@chromium.org2009-06-191-0/+4
| | | | | | | See: http://codereview.chromium.org/126098 Patch from Jerry Tang <tangjie@google.com>. git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18837 0039d316-1c4b-4281-b951-d872f2087c98
* Add a temporary command-line switch --auto-ssl-client-auth forwtc@chromium.org2009-06-192-0/+10
| | | | | | | | | | | | | | | | | automatically selecting a client certificate when an SSL server requests client authentication. This switch will be removed when we implement client certificate selection UI. Also fix some cpplint.py nits. R=jcampan BUG=http://crbug.com/318 TEST=none Review URL: http://codereview.chromium.org/131090 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18819 0039d316-1c4b-4281-b951-d872f2087c98
* Add Print Selection support to Chrome. This change is fairly involved since ↵sverrir@google.com2009-06-191-0/+5
| | | | | | | | | | | | | | | | this means that the printing is done async instead of the fully synchronous mode the normal full page printing is. This means we create an in memory copy of the selected text for printing. This is the next step to move to fully async printing with print frame support. This change also removes the print on demand functionality that was no longer used. BUG=http://crbug.com/1682 TEST=The print dialog on Windows now contains an option to print selection only. Test that with various pages and various selections. Review URL: http://codereview.chromium.org/125082 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18815 0039d316-1c4b-4281-b951-d872f2087c98
* Display an infobar alert when extension process crashes.phajdan.jr@chromium.org2009-06-191-0/+8
| | | | | | | | | | | | Also correctly handle crashes in task manager. The infobar allows user to restart the extension process. TEST=Install buildbot extension, see its content in the shelf. Open task manager, kill extension process. Click "restart" in the infobar that should appear. The extension content in the shelf should re-appear after the crash. http://crbug.com/14111 Review URL: http://codereview.chromium.org/126289 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18809 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a valgrind error related to Zygote's CommandLine usage.craig.schlenter@chromium.org2009-06-191-3/+3
| | | | | | | | Also touch up some include guards and a comment. Review URL: http://codereview.chromium.org/131054 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18801 0039d316-1c4b-4281-b951-d872f2087c98
* Implement window open disposition for (some) navigation buttons.estade@chromium.org2009-06-192-0/+40
| | | | | | | | | | still need to do similar for link buttons and such. BUG=14518 Review URL: http://codereview.chromium.org/131071 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18786 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 18765 to try to repair perf regressionjar@chromium.org2009-06-191-5/+0
| | | | | | | TBR=mpcomplete Review URL: http://codereview.chromium.org/131093 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18784 0039d316-1c4b-4281-b951-d872f2087c98
* First draft of web resource service; fetches data from a JSON feed mirandac@chromium.org2009-06-187-0/+131
| | | | | | | | | | and stores it in user prefs, where it can be used by the new tab page. BUG = http://crbug.com/13363 Review URL: http://codereview.chromium.org/125052 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18766 0039d316-1c4b-4281-b951-d872f2087c98
* Send port-closed notification when a frame with ports unloads.mpcomplete@google.com2009-06-181-0/+5
| | | | | | | | | | | | | Also add onLoad and onUnload chrome Event to our bindings, so we can add listeners to these events without needing a DOM. These don't hook into the window "unload" event, so we no longer prevent Chrome's sudden termination of tabs on shutdown. BUG=12686 TEST=no Review URL: http://codereview.chromium.org/125280 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18765 0039d316-1c4b-4281-b951-d872f2087c98
* Implement a max worker count of 16 per tab and 64 total. Any workers ↵jam@chromium.org2009-06-181-0/+11
| | | | | | | | created after that are queued. Review URL: http://codereview.chromium.org/125242 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18763 0039d316-1c4b-4281-b951-d872f2087c98
* Initial download shelf on OS X.thakis@chromium.org2009-06-182-26/+13
| | | | | | | | | | | | | | | | | | | This has lots of missing stuff (e.g. a custom download item view that shows download progress, the popup is the same for in-progress and completed downloads, no animation, everything looks ugly, the info bubble overlaps the shelf when it's visible, no "open download manager page" link, etc), but the basic functionality is hooked up: The shelf appears when files are downloaded, and something ugly is added to the shelf for each download. The popup's "Reveral in Finder" even works. The shelf is per-window as it should be. BUG=12500 TEST=Download something and check the shelf appears. Click the close button and make sure it disappears again. Review URL: http://codereview.chromium.org/93129 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18757 0039d316-1c4b-4281-b951-d872f2087c98
* Small piece of extensions auto-update: add an auto-update url member varasargent@chromium.org2009-06-183-6/+72
| | | | | | | | | | | to the Extension class. BUG=http://crbug.com/12117 TEST=none Review URL: http://codereview.chromium.org/131044 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18736 0039d316-1c4b-4281-b951-d872f2087c98
* Small cleanup of constness.mad@google.com2009-06-184-14/+16
| | | | | | | | | | | | | Added const to methods not changing the state, except for the lazy setting of an escaped version of a path which is now mutable. Also made some lint fixes. BUG=none TEST=none Review URL: http://codereview.chromium.org/132024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18733 0039d316-1c4b-4281-b951-d872f2087c98
* Add gtk EditKeywordController.mattm@chromium.org2009-06-181-0/+8
| | | | | | | | BUG=13326 Review URL: http://codereview.chromium.org/126305 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18727 0039d316-1c4b-4281-b951-d872f2087c98
* Abstract ExtensionPrefs into their own class.erikkay@google.com2009-06-181-0/+5
| | | | | | | | | Relands http://codereview.chromium.org/126281 which had been reverted due to a leak. The only difference is in extension_prefs.h where extension_data_ is now a scoped_ptr. Review URL: http://codereview.chromium.org/131040 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18717 0039d316-1c4b-4281-b951-d872f2087c98
* Revert r18661.willchan@chromium.org2009-06-181-5/+0
| | | | | | | | Broke purify and mac valgrind. "Pull Extension-related prefs into its own class. Also add a notification for when the extensions service has finished its initial load of extensions, separate from EXTENSIONS_LOADED.TEST=noneBUG=none" Review URL: http://codereview.chromium.org/132008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18682 0039d316-1c4b-4281-b951-d872f2087c98
* Enable off-by-default monitor color management on Windows. This is enabled viabrettw@chromium.org2009-06-182-0/+7
| | | | | | | | | a command-line switch. BUG=4938 Review URL: http://codereview.chromium.org/131002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18675 0039d316-1c4b-4281-b951-d872f2087c98
* Pull Extension-related prefs into its own class. Also add a notification ↵erikkay@google.com2009-06-171-0/+5
| | | | | | | | for when the extensions service has finished its initial load of extensions, separate from EXTENSIONS_LOADED.TEST=noneBUG=none Review URL: http://codereview.chromium.org/126281 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18661 0039d316-1c4b-4281-b951-d872f2087c98
* Add a port disconnect event for when one side of an extension message portmpcomplete@google.com2009-06-171-17/+5
| | | | | | | | | | | | | goes away. Combine the various ExtensionMessageService IPC message into a single "Invoke" message. BUG=12686 TEST=no Review URL: http://codereview.chromium.org/126234 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18645 0039d316-1c4b-4281-b951-d872f2087c98
* Implements the Page Info window on Mac.mark@chromium.org2009-06-171-19/+0
| | | | | | | | | | | | | | | | | | | | | | | - Splits out the code that was in chrome/browser/views/page_info_window.{cc,h} into a platform-independent class PageInfoWindow in chrome/browser/page_info_window.{cc,h} - The Windows implementation now lives in chrome/browser/views/page_info_window_win.{cc,h} - Created a Mac implemenation PageInfoWindowMac in chrome/browser/cocoa/page_info_window_mac.{h,mm} to bridge to a Cocoa NSWindowController - Created a new NSWindowController subclass PageInfoWindowController that is in chrome/browser/cocoa/page_info_window_controller.{h,mm} - Created a XIB for the page info window in chrome/app/nibs/en.lproj/PageInfo.xib Patch by Robert Sesek <rsesek@bluestatic.org> Review URL: http://codereview.chromium.org/115116 Review URL: http://codereview.chromium.org/125266 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18643 0039d316-1c4b-4281-b951-d872f2087c98
* Remove tooltip from page_action and just use the name as the tooltip. We ↵finnur@chromium.org2009-06-173-23/+1
| | | | | | | | don't seem to have a need for having the name and the tooltip be separate. Review URL: http://codereview.chromium.org/125264 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18642 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a crash when the ThumbnailGenerator.brettw@chromium.org2009-06-171-1/+0
| | | | | | | | | | This happens when the ThumbnailGenerator is initialized because the NotificationService doesn't exist yet. This patch adds a function that TabContentscalls to make sure it is registered to avoid this problem. Review URL: http://codereview.chromium.org/126239 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18572 0039d316-1c4b-4281-b951-d872f2087c98
* Make LocalStrings work without manipulating the DOM.arv@google.com2009-06-161-2/+5
| | | | | | | | | | | BUG=None TEST=Make sure that localized text on NTP, History and Downloads still look the same. Review URL: http://codereview.chromium.org/125076 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18551 0039d316-1c4b-4281-b951-d872f2087c98
* Fix the mac build by removing extra object namebrettw@chromium.org2009-06-161-1/+2
| | | | | | Review URL: http://codereview.chromium.org/125222 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18544 0039d316-1c4b-4281-b951-d872f2087c98
* Adds kind-of-live thumbnail generation for a potential tab switcher.brettw@chromium.org2009-06-166-1/+37
| | | | | | | | | | | | | | | | | | | | This listens to tab events and tries to keep thumbnails ready to go. See thumbnail_generator.cc for a more detailed design. This adds a painting observer to the RenderWidgetHost to enable this new behavior, as well as a notification to allow the thumbnail generator to hook its observer in. There is also a new notification that a backing store has been disabled, which required making the backing stores know about their owning widget hosts. This component is currently disabled. We just need to uncomment the member in Profile and it will start to work. Original review: http://codereview.chromium.org/118420 Review URL: http://codereview.chromium.org/126101 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18540 0039d316-1c4b-4281-b951-d872f2087c98
* Since the introduction of PushSource, there are two buffering layers in thehclam@chromium.org2009-06-161-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | browser process, the hardware buffer used in AudioOutputStream and transportation buffer in PushSource. Together with the latency in the IPC audio layer we have a serious AV sync problem. To compensate the delay and latency introduced by these three factors two parameters are added in RequestAudioPacket message that include the buffer fill level and timestamp of the request. These two parameters are used to determine the playback delay to be used by the audio renderer to update the pipeline with the time delta. So we have three parameters we need to care about: 1. Hardware buffer in AudioOutputStream 2. Buffered data in PushSource 3. IPC latency We have accurate values for 2 and 3 but not 1. We currently don't have the API in AudioOutputStream to query the remaining buffer in the hardware buffer. But usually there is a large amount of data in it, e.g. on Windows 400ms worth of data. Since we now detached the hardware buffer request of OnMoreData() from the actual packet request of IPC (by the introduction of PushSource), it is really critical to know the buffer level in the hardware. I made a guess of this buffer level by using the amount of last buffer copy. Review URL: http://codereview.chromium.org/122020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18536 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 18512 - Revert 18373 Consider a redirect following user gesture as ↵laforge@chromium.org2009-06-161-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | userinitiated in maintaining navigation entries. Also, ignore redirect or machineinitiated new subframe navigations. The current code treats all redirects as machineinitiated in processing navigation to a new page (to fix Bugs 9663 and 10531). This is not always appropriate, because some sites, e.g., www.google.com/ig, use redirect to implement userinitiated navigation (Bug 11896). This change assumes that a machineinitiated redirect happens within 300ms since the last document load was completed, while a userinitiated one happens later. This assumption is not always correct, e.g., a user may cause transition within 300ms. But I cannot think of any better ways to tell if a redirect is machine initiated or userinitiated. I believe this change works good enough, at least better than the status quo. Review URL: http://codereview.chromium.org/115919 TEST=Open http://www.hp.com and observe it redirects to http://www.hp.com/#Product . Hit Back button and observe the former URL is not visited. Open http://www.google.com/ig and click tabs inside the page, and try hitting Back and Forward to see if the navigation is right. Open http://www.google.com/codesearch, search for something, click on a result item, and try hitting Back. BUG=11896,12820 TBR=yuzo@chromium.org Review URL: http://codereview.chromium.org/125202 TBR=laforge@chromium.org Review URL: http://codereview.chromium.org/126221 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18522 0039d316-1c4b-4281-b951-d872f2087c98
* Move TableModel out of views/ and into app/.estade@chromium.org2009-06-161-17/+0
| | | | | | | | | Remove stub implementation in temp_scaffolding_stubs.h Use l10n_util collator helper function in TableModel::Compare Review URL: http://codereview.chromium.org/126184 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18518 0039d316-1c4b-4281-b951-d872f2087c98
* Extract load times from WebDataSource. Move them to NavigationState.darin@chromium.org2009-06-162-4/+4
| | | | | | | | | | | | Move PasswordForm into the webkit_glue namespace. TEST=none BUG=10041 R=brettw Review URL: http://codereview.chromium.org/126190 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18515 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 18373 - Consider a redirect following user gesture as userinitiated ↵laforge@chromium.org2009-06-161-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | in maintaining navigation entries. Also, ignore redirect or machineinitiated new subframe navigations. The current code treats all redirects as machineinitiated in processing navigation to a new page (to fix Bugs 9663 and 10531). This is not always appropriate, because some sites, e.g., www.google.com/ig, use redirect to implement userinitiated navigation (Bug 11896). This change assumes that a machineinitiated redirect happens within 300ms since the last document load was completed, while a userinitiated one happens later. This assumption is not always correct, e.g., a user may cause transition within 300ms. But I cannot think of any better ways to tell if a redirect is machine initiated or userinitiated. I believe this change works good enough, at least better than the status quo. Review URL: http://codereview.chromium.org/115919 TEST=Open http://www.hp.com and observe it redirects to http://www.hp.com/#Product . Hit Back button and observe the former URL is not visited. Open http://www.google.com/ig and click tabs inside the page, and try hitting Back and Forward to see if the navigation is right. Open http://www.google.com/codesearch, search for something, click on a result item, and try hitting Back. BUG=11896,12820 TBR=yuzo@chromium.org Review URL: http://codereview.chromium.org/125202 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18512 0039d316-1c4b-4281-b951-d872f2087c98
* Make XHR work in Workers. Creates a 'shadow page' in a worker process to ↵dimich@google.com2009-06-161-3/+3
| | | | | | | | | | proxy the loading requests through. BUG=4361 TEST=none Review URL: http://codereview.chromium.org/126070 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18465 0039d316-1c4b-4281-b951-d872f2087c98
* Add a switch for changing the fd limit on Mac/Linux.stuartmorgan@chromium.org2009-06-152-0/+7
| | | | | | | | | | Bump up the default fd limit on the Mac; stopgap fix for page cyclers, but something we want long-term to improve performance once we handle fd exhaustion with delays. BUG=none TEST=Launching with --file-descriptor-limit=20 should crash the renderer almost immediately. Review URL: http://codereview.chromium.org/125151 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18446 0039d316-1c4b-4281-b951-d872f2087c98
* Linux: Add support for chrooted renderers.agl@chromium.org2009-06-151-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | http://code.google.com/p/chromium/wiki/LinuxSandboxIPC Without filesystem access from the renderers, we need another way of dealing with fontconfig and font loading. This add support for: * An "SBX_D" environment variable in the renderers which is used to signal the end of dynamic linking so that the chroot can be enforced. * A sandbox_host process, running outside the sandbox, to deal with fontconfig requests from the renderers. See the wiki page for the reasoning behind making it a separate process. * A new, custom SkFontHost for Skia. Because this is Chrome specific, it will live outside the upstream Skia tree. This FontHost can be configured either to drive fontconfig directly (for the browser process and for any unsandboxed renderers) or to use an IPC system. Since the same SkFontHost has to be linked into both the browser and renderer (they are the same binary), this switch has to be made at run time. Sandbox IPC calls are rare (a couple of dozen at page load time) and add about 50us of overhead for each call. (Reland of r17575 which was reverted in r17577) http://codereview.chromium.org/112074 BUG=8081 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18405 0039d316-1c4b-4281-b951-d872f2087c98
* Extract form related classes from the guts of WebFrameImpl.darin@chromium.org2009-06-152-8/+8
| | | | | | | | | | | | | | | | | | | Instead of having WebFrameImpl generate SearchableFormData, PasswordForm, and AutofillForm classes, allow the embedder (RenderView) to do so. This is done to help minimize the dependencies WebFrameImpl has on other code, which will make it easier to move WebFrame and WebDataSource into the WebKit API. Most significant change: Now, RenderView always sets a NavigationState on WebDataSource instances. We used to only do so for browser initiated navigations. This is done so that we can store things like SearchableFormData and friends on the NavigationState. To facilitate this change, it was necessary to add a way through the WebKit API to refer to a HTMLFormElement. This CL introduces WebForm, which is like a RefPtr<HTMLFormElement>, so you can just copy a WebForm around by value and the right thing happens. Some of the other changes are about moving more things into the webkit_glue namespace. On hindsight, I probably should have done that as a separate CL. BUG=10041 TEST=none R=brettw Review URL: http://codereview.chromium.org/126083 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18395 0039d316-1c4b-4281-b951-d872f2087c98
* Reverting Rev 18383 since it brings down the performance of single core page ↵mad@chromium.org2009-06-151-11/+2
| | | | | | cyclers too much... git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18385 0039d316-1c4b-4281-b951-d872f2087c98
* This is to thest the page cycler performance without the resize corner, to ↵mad@chromium.org2009-06-151-2/+11
| | | | | | | | | | | | | see if these paint optimizations are worth it. I will revert these changes as needed... You may revert them yourself if they cause you trouble before I get to revert them. I have tested these changes on two different linux configuration, but there are more code paths that I couldn't verify myself, though agl gave me the OK anyway. These changes have already been reviewed here: http://codereview.chromium.org/108040 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18383 0039d316-1c4b-4281-b951-d872f2087c98
* Consider a redirect following user gesture as user-initiated in maintainingyuzo@chromium.org2009-06-151-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | navigation entries. Also, ignore redirect- or machine-initiated- new subframe navigations. The current code treats all redirects as machine-initiated in processing navigation to a new page (to fix Bugs 9663 and 10531). This is not always appropriate, because some sites, e.g., www.google.com/ig, use redirect to implement user-initiated navigation (Bug 11896). This change assumes that a machine-initiated redirect happens within 300ms since the last document load was completed, while a user-initiated one happens later. This assumption is not always correct, e.g., a user may cause transition within 300ms. But I cannot think of any better ways to tell if a redirect is machine- initiated or user-initiated. I believe this change works good enough, at least better than the status quo. Review URL: http://codereview.chromium.org/115919 TEST=Open http://www.hp.com and observe it redirects to http://www.hp.com/#Product . Hit Back button and observe the former URL is not visited. Open http://www.google.com/ig and click tabs inside the page, and try hitting Back and Forward to see if the navigation is right. Open http://www.google.com/codesearch, search for something, click on a result item, and try hitting Back. BUG=11896,12820 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18373 0039d316-1c4b-4281-b951-d872f2087c98
* Reduce header dependencies in browser/phajdan.jr@chromium.org2009-06-132-1/+4
| | | | | | | | | | | | - remove unneeded header includes - move implementation bits out of headers - more explicit header deps In my scan of headers I got up to (including) dom_ui. Review URL: http://codereview.chromium.org/126071 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18353 0039d316-1c4b-4281-b951-d872f2087c98
* Make extensions visible in the Task Manager.phajdan.jr@chromium.org2009-06-131-0/+8
| | | | | | | | | | | | | - register for notifications when extension hosts are created/destroyed - collapse extension tabs into just one table row, also for tabs opened while the task manager stays open TEST=Start chrome.exe with --enable-extensions. Install the buildbot monitor extension. Open the task manager. You should see an extension process. http://crbug.com/12127 Review URL: http://codereview.chromium.org/125047 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18352 0039d316-1c4b-4281-b951-d872f2087c98
* Change extension id format back to mp-decimal, and change theaa@chromium.org2009-06-132-14/+64
| | | | | | | | length to 32 chars (exactly 128 bits). Review URL: http://codereview.chromium.org/126074 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@18351 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-01-19 21:45:25 +0000