It fixes at least two real bugs, one in the tar generator, and one
in stream_bank.h.

Review URL: http://codereview.chromium.org/159168

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21227 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/155886

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21225 0039d316-1c4b-4281-b951-d872f2087c98

TEST=none
BUG=none

Review URL: http://codereview.chromium.org/159164

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21224 0039d316-1c4b-4281-b951-d872f2087c98

so that we get consistent results.  Also bumped up the pdiff threshold for 2d sample
on the mac to its previous number to get the test to pass until we fix it.
Review URL: http://codereview.chromium.org/155885

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21223 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/159158

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21221 0039d316-1c4b-4281-b951-d872f2087c98

New Linux baselines for dotted/dashed CSS borders, changed by http://bugs.webkit.org/show_bug.cgi?id=27388 .

BUG=http://crbug.com/474
TEST=After sheriff rolls DEPS, Linux buildbots should be greener.

Review URL: http://codereview.chromium.org/155877

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21220 0039d316-1c4b-4281-b951-d872f2087c98

https://bugs.webkit.org/show_bug.cgi?id=27388

BUG=http://crbug.com/474
TEST=Windows webkit bots should be greener after sheriff rolls DEPS.

Review URL: http://codereview.chromium.org/155880

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21219 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21217 0039d316-1c4b-4281-b951-d872f2087c98

Also included in this revert: r21165, r21180, and a couple valgrind suppression edits

TBR=huanr

Review URL: http://codereview.chromium.org/155876

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21216 0039d316-1c4b-4281-b951-d872f2087c98

fix warnings treated as errors on line 94 and 97 that are allowed on try server, but not buildbot.  doh

BUG=17304
TEST=buildbot has different warnings than try server.  this patch reenables the qtfaststart build with 2 warnings fixed.

Review URL: http://codereview.chromium.org/155834

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21215 0039d316-1c4b-4281-b951-d872f2087c98

when the section is larger.

Make the window menu look and behave more like a tooltip.

Return more recently closed windows and tabs when using the NNTP.

BUG=16539, 16835, 14963

TEST=Hide and show the tips section. More items (if there are more
items) should be shown. Hover over a closed window and the tooltip
should look like a tooltip on Vista/Win7

Review URL: http://codereview.chromium.org/155865

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21214 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21213 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21212 0039d316-1c4b-4281-b951-d872f2087c98

Fix a missing ... in a suppression for an infrequent warning
BUG=16625,17385
TEST=watch mac valgrind unit test bot go green

Review URL: http://codereview.chromium.org/159150

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21211 0039d316-1c4b-4281-b951-d872f2087c98

The previous code only support names up to 99 chars.
This one supports up to 1024. The code actually 
supports 32bit length but given that Linux, OSX and
other only support 255 I set it 1024. Why 1024?
Because Windows actually supports 255 wchar characters
which when converted to utf-8 could be up to 1024 bytes.

I'm not 100% sure the format is correct. I could not
find any docs on the format, Just reverse engineered it.

Looking at hex dumps by both 7zip and gnu tar it was
pretty clear what it does including zeroing many
of the standard fields.

Review URL: http://codereview.chromium.org/159129

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21210 0039d316-1c4b-4281-b951-d872f2087c98

If you need the tests, refractor them to follow the 
guidelines instead of having them fail intermittently 

http://big.corp.google.com/~joejoejoe/testing/2008/05/episode-90-sleeping-synchronization_27.html

Review URL: http://codereview.chromium.org/159117

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21209 0039d316-1c4b-4281-b951-d872f2087c98

http://code.google.com/p/o3d/issues/detail?id=67

Review URL: http://codereview.chromium.org/155872

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21208 0039d316-1c4b-4281-b951-d872f2087c98

application.

This works by adding the top-level GtkWindow objects, such as BrowserWindowGtk
and BookmarkManagerGtk, to their own unique window groups. Without this change,
all top-level windows are added to a default application-wide window group.

This ensures that all grabs created with gtk_grab_add(...) only affect the
window group of the grabbed widget, as opposed to the entire application.

Note that gtk_window_set_modal(...) is implemented with gtk_grab_add(...) and
therefore is only modal to a specific window group.

In order for this to work correctly, changes were also made to the info
bubble and render widget popup (<select> tag) code. Since these widgets
also call gtk_grab_add(...), they must be added to the top level window
group to work correctly.

Test 1:
- Open two new chrome window: A and B
- Open "Save file as..." dialog in window A
- Verify that window A does not respond to keyboard or mouse events.
- Verify that window B does responde to keyboard and mouse events.
- Open "Save file as..." dialog in window B
- Verify that window B does not respond to keyboard or mouse events.
- Cancel dialog on window A.
- Verify that window A starts responding to keyboard and mouse events.
- Cancel dialog on window B.
- Verify that window B starts responding to keyboard and mouse events.

Test 2:
- Verify that <select> tag allows for correct selection of items.

Test 3:
- Click bookmark star and verify that info bubble works correctly.

BUG=8727
TEST=none

patch by Mohit Muthanna Cheppudira <mohit.muthanna [at] gmail>
original review url: <http://codereview.chromium.org/155852>

Review URL: http://codereview.chromium.org/159147

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21207 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/155860

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21206 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=Install a theme, notice that the download shelf now at least somewhat fits with the new theme.
Review URL: http://codereview.chromium.org/159126

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21205 0039d316-1c4b-4281-b951-d872f2087c98

BUG=16128
TEST=watch valgrind ui bots go green

Review URL: http://codereview.chromium.org/155867

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21204 0039d316-1c4b-4281-b951-d872f2087c98

handled in original code are addressed, for example the input method in password
box case.

The most important change in this CL is the change to key event processing flow.
 In old code, a key event will first be sent to webkit then dispatched to IME
for filtering. With this CL, a key event will first be dispatched to IME for
filtering, then how to send the event to webkit is decided by the filtering
result.

This CL tries to emulate the keyboard input behavior on Windows as much as
possible. For example, if a keydown event is filtered by IME, then its virtual
key code will be changed to VK_PROCESSKEY(0xE5) to prevent webkit from
processing it again. This behavior can workaround bug 16281.

To test this CL, you may cut and save following html code into a file and open
it in chrome.
------------------8<----cut here----->8---------------------
<html><head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">

<script>
function keyEventHandler(event) {
  var props = [ "type", "charCode", "keyCode", "altKey",
    "ctrlKey", "shiftKey", "metaKey" ];
  var info = document.getElementById('info');
  var s = '';
  for (var i in props) {
    s += props[i] + ':' + event[props[i]] + ' ';
  }
  info.value += s + '\n';
}

function textEventHandler(event) {
  info.value += "type:" + event['type'] + " data:" + event['data'] + '\n';
}

function passwordChangeEventHandler(event) {
  var input2 = document.getElementById('input2');
  info.value += "password:" + input2.value + '\n';
}

function onLoad() {
  var input = document.getElementById('input');
  input.addEventListener('keydown', keyEventHandler, false);
  input.addEventListener('keyup', keyEventHandler, false);
  input.addEventListener('keypress', keyEventHandler, false);
  input.addEventListener('textInput', textEventHandler, false);
  var input2 = document.getElementById('input2');
  input2.addEventListener('change', passwordChangeEventHandler, false);
}
</script>
</head><body onload="onLoad()">
<input id="input" size="20">
<input id="input2" type="password" size="20">
<p>
<textarea id="info" rows="40" cols="150"></textarea>
</p></body></html>
------------------8<----cut here----->8---------------------

This CL was confirmed to fix following issues:
BUG=16281 "arrow keys and backspace/delete keys move/delete two characters at a
time when xim immodule is used"
BUG=16282 "Disable IMEs in a password input"
BUG=16596 "fcitx (chinese input method) not working in ubuntu 9.04"
BUG=16659 "Crash near RenderWidgetHostViewGtk::IMEUpdateStatus"
BUG=16699 "Can't move cursor to omnibox and use input method if cursor is
currently in a text input box in web page."
BUG=16796 "Input method issue: When inputting text in a text box, if there is a
composition string then pressing Backspace or Delete will cause the composition
string be cleared and the text box refuses to accept any further input.

All tests assume above html code is used.
TEST=Start scim with "scim -d" and start chrome with GTK_IM_MODULE=xim and
XMODIFIERS=@im=SCIM. Type something in input box, eg. "hello", then press
backspace, to see if only one character is deleted.
TEST=Move cursor to password input box, press ctrl-space to see if input method
is not activated. Switch keyboard layout to Canadian-French then type a'[{' key
and an 'a' key, then press enter, to see if a Latin character "U+00E2" is
inputted in password box.
TEST=Install fcitx with "sudo apt-get install fcitx" (assume you are using
Ubuntu/Debian). Open a terminal, export XMODIFIERS=@im=fcitx and
GTK_IM_MODULE=xim then run fcitx, then start chrome. Move cursor into an input
box, press ctrl-space and input something, eg. "nihao" then press space. Check
if some Chinese characters are inputted.
TEST=Start chrome with GTK_IM_MODULE=scim. Move cursor into a text input box
then move into a password box, chrome should not crash.
TEST=Move cursor into a text input box, then click omnibox, and see if the text
input box has lost focus. Press ctrl-space to activate input method, then type
something, and see of the input goes into omnibox.
TEST=Move cursor into a text input box and enable a Chinese Pinyin input method,
then type something, eg. "dajiahao", make sure a composition string is displayed
in the text input box, then press backspace and see if there is still a
composition string and backspace is handled by input method.

patch by James Su <james.su [at] gmail>
original review URL: <http://codereview.chromium.org/149755>


Review URL: http://codereview.chromium.org/155869

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21203 0039d316-1c4b-4281-b951-d872f2087c98

- NativeViewHostGtk::RemovedFromWidget crashes due to WidgetGtk::window_contents_ being NULL... basically when RemovedFromWidget is called from the RootView's dtor, the window_contents_ and widget_ properties of the containing WidgetGtk are NULL. It seems acceptable to not clean up in this case.
- TabContentsViewGtk dtor needs to call CloseNow(). This is similar to ~TabContents() calling DestroyWindow on the view's HWND. Without this, the TabContentsViewGtk object was destroyed but the corresponding GtkWidget wasn't, and so subsequent signal handlers would crash.

BUG=none
TEST=none 
Review URL: http://codereview.chromium.org/159130

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21202 0039d316-1c4b-4281-b951-d872f2087c98

BUG=14659,17100
TEST=Download multiple things. Items should now be inserted from the left. When a download is in progress, the time to completion should be displayed, when it's done the time should fade out and the filename should move down. Download items should appear in a sweep animation. Clicking the download directly should open it; clicking the arrow on the right of the download item should show the correct context menu. If a download filename is long, it should be elided in the middle.

Review URL: http://codereview.chromium.org/159060

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21201 0039d316-1c4b-4281-b951-d872f2087c98

GTK Themes can set colors on a per widget basis, and many do. Previously we were getting
the text color from a GtkWindow object, which sometimes is wrong--the theme just sets the
text color on labels...

Review URL: http://codereview.chromium.org/159143

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21200 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=go to a bookmarked page; the outline of the yellow star should be themed

Review URL: http://codereview.chromium.org/155851

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21199 0039d316-1c4b-4281-b951-d872f2087c98

BUG=17289
TEST=no

Review URL: http://codereview.chromium.org/159095

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21197 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21196 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21194 0039d316-1c4b-4281-b951-d872f2087c98

http://crbug.com/10934
TEST=Type 'something', and press Control.  Autocomplete should show www.something.com at the top.  If that's what you'd get anyhow, try a different word.  Control-enter should take you there.

Review URL: http://codereview.chromium.org/159105

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21192 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/155816

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21190 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=Make sure the Options dialog still works as expected (tab selection, focus traversal...)
Review URL: http://codereview.chromium.org/155668

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21189 0039d316-1c4b-4281-b951-d872f2087c98

Prevents case where init callback is called before NPP_SetWindow (which initializes the renderer).

Review URL: http://codereview.chromium.org/155716

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21188 0039d316-1c4b-4281-b951-d872f2087c98

is swapped out while a drag is happening.

Since the tab contents is swapped out, our RVH that the drag
started on is no longer valid.  Simply null out the pointer
and no longer try to send feedback to the drag source.  This
allows the user to continue the drag even though the source
RVH is gone.

BUG=16073

Review URL: http://codereview.chromium.org/159040

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21187 0039d316-1c4b-4281-b951-d872f2087c98

BUG=17370
TEST=FirefoxImporterTest.Firefox3NSS3Decryptor isn't run by the purify bots
TBR=wtc

Review URL: http://codereview.chromium.org/159141

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21186 0039d316-1c4b-4281-b951-d872f2087c98

BUG=14662 (http://crbug.com/14662)
TEST=Download something so that the shelf is visible and make sure
     that hovering over links on the page produce a status bubble
     above the download shelf.
Review URL: http://codereview.chromium.org/155706

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21185 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/155859

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21184 0039d316-1c4b-4281-b951-d872f2087c98

Disable all possible things a theme can do to make this false on the GtkToolbar widget.

Review URL: http://codereview.chromium.org/155821

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21183 0039d316-1c4b-4281-b951-d872f2087c98

The renderer doesn't give the settings to Skia yet -- that'll
go in after agl's work on hinting and subpixel rendering in Skia.

http://codereview.chromium.org/155787

(Patch by Dan Erat)


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21182 0039d316-1c4b-4281-b951-d872f2087c98

TBR=dkegel
Review URL: http://codereview.chromium.org/155836

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21181 0039d316-1c4b-4281-b951-d872f2087c98