OSDN Git Service
Ben Murdoch [Mon, 29 Nov 2010 13:25:42 +0000 (13:25 +0000)]
AutoFill no longer needs a unique profile id native side.
AutoFill profiles now use GUIDs that they generate themselves
rather than injecting a unique int ID so we no longer need to
reach into the java class to grab the ID.
Change-Id: I26492aeddd130831e941ec79ef1f5e36271903c8
Steve Block [Fri, 26 Nov 2010 17:18:30 +0000 (09:18 -0800)]
Merge "Minor code cleanup in WebCookieJar"
Steve Block [Fri, 26 Nov 2010 14:31:37 +0000 (14:31 +0000)]
Minor code cleanup in WebCookieJar
Change-Id: Ia68bac6bd93b0afeff049a9ac0f056191d92d2a0
Steve Block [Fri, 26 Nov 2010 13:31:28 +0000 (05:31 -0800)]
Merge "Factor out CookieManager from WebRequestContext"
Kristian Monsen [Fri, 5 Nov 2010 14:43:45 +0000 (14:43 +0000)]
Factor out CookieManager from WebRequestContext
This is the first step in allowing us to have a WebRequestContext per
WebView, which is required to allow the useragent string to be
configured per WebView.
Bug:
3113804
Change-Id: I9ccf07d4a277147e74e2f3e701bd6d2166954280
Steve Block [Fri, 26 Nov 2010 12:28:14 +0000 (04:28 -0800)]
Merge "No longer need to pass storage paths to WebRequestContext.cleanupPrivateBrowsingFiles()"
Steve Block [Fri, 26 Nov 2010 11:27:42 +0000 (11:27 +0000)]
No longer need to pass storage paths to WebRequestContext.cleanupPrivateBrowsingFiles()
Requires the following changes to frameworks/base and packages/apps/Browser
https://android-git.corp.google.com/g/81782
https://android-git.corp.google.com/g/81780
Bug:
3232569
Change-Id: I73cabb561e206beee2bc1716f04aed8764797cc1
Steve Block [Fri, 26 Nov 2010 10:06:54 +0000 (02:06 -0800)]
Merge "Fixes WebResourceRequest to correctly set load flags"
Steve Block [Thu, 25 Nov 2010 12:11:15 +0000 (12:11 +0000)]
Fixes WebResourceRequest to correctly set load flags
We now correctly set the cookie load flags and the cache control HTTP
headers.
Bug:
2889880
Change-Id: I759a3131b73a8fc82432a067a31754fcc566c136
Iain Merrick [Thu, 25 Nov 2010 15:26:17 +0000 (15:26 +0000)]
Session cookies are now persistent.
We previously checked !IsPersistent() to determine whether a
cookie is a session cookie, but now all cookies are persistent,
so we need to call the new method IsSessionCookie().
Depends on https://android-git.corp.google.com/g/81755
Bug:
2864791
Change-Id: Ia44cd0cdc5e68a4b0a61bdeebd70cdccfc8e4e1b
Iain Merrick [Wed, 24 Nov 2010 10:38:33 +0000 (10:38 +0000)]
Set useCachedCredentials to true on first auth request.
This flag previously defaulted to false, but the Android CTS
specifies that a "sufficient" implementation should return true
on the first authentication request. (The second and subsequent
auth requests aren't tested.)
This CL sets the flag to true on the first authentication request
for a given WebRequest, but false for subsequent requests (i.e.,
if the first attempt to authenticate failed).
Using a counter rather than a flag in WebRequest for readability.
"isFirstAuthRequest" would be a bit ambiguous as a member variable,
but "authRequestCount" seems pretty clear.
Change-Id: Id9baae4c30111f743e079f90940078b07498b168
Cary Clark [Wed, 24 Nov 2010 19:58:36 +0000 (11:58 -0800)]
Merge "use style ignoring pending style sheets"
Cary Clark [Wed, 24 Nov 2010 19:50:46 +0000 (14:50 -0500)]
use style ignoring pending style sheets
If the style is not yet computed, attempting to
compute the style may blow up (e.g., the font points
to zero and is dereferenced). The webkit way around
this appears to be to ignore pending style sheets,
even though six lines down from this change, the
regular webkit makes the original styleForElement
call.
This does fix sfgate.com, so it's not a bad thing to
do, and the real style change should retrigger this
code path later on.
bug:
3194102
Change-Id: Ia558e3d49c94ed817d69d528b9e914f906f62a66
Cary Clark [Wed, 24 Nov 2010 18:34:30 +0000 (10:34 -0800)]
Merge "allow animated gifs to skip frames"
Ben Murdoch [Wed, 24 Nov 2010 18:23:00 +0000 (10:23 -0800)]
Merge "Fix crash in JavaClassV8"
Cary Clark [Wed, 24 Nov 2010 18:17:30 +0000 (13:17 -0500)]
allow animated gifs to skip frames
Removing this patch causes gifs to skip frames rather than
play all intermediate frames in an attempt to catch up.
bug:
2621902
Change-Id: Idc19b11991c37c685d1dc61bb5daddc6d12d30e0
Ben Murdoch [Wed, 24 Nov 2010 12:16:03 +0000 (12:16 +0000)]
Fix crash in JavaClassV8
It is possible that the m_name member of JavaClass is uninitialized as
we possibly early out from the constructor if we could not load the Java
class from the JVM. In this case when we call the dtor we try to free
uninitialized memory and crash. m_name is not actually used anywhere
so we just remove it.
Will upstream to webkit.
Bug:
3222518
Change-Id: I9354e9914157d2c4afd72dcce9122329c2486c89
Kristian Monsen [Wed, 24 Nov 2010 17:54:55 +0000 (09:54 -0800)]
Merge "Removing CRASH() from asserts in WebRequest.cpp"
Steve Block [Wed, 24 Nov 2010 16:25:11 +0000 (08:25 -0800)]
Merge "Add LayoutTests http/tests/cookies at r71558"
Cary Clark [Wed, 24 Nov 2010 15:45:12 +0000 (10:45 -0500)]
fix radio, checkbox hit testing
Radio and checkbox input types draw only a bitmap in
their navigation bounds, and require some special treatment.
Never treat them as transparent.
Use the slop box instead of the exact box when detecting
them, since our asset may be larger than the space webkit
provides.
Also, fix a debug test if the layer is null.
bug:
3215174
Change-Id: I3b573e3aa215268390db2d58435ac47115804ed7
Steve Block [Wed, 24 Nov 2010 13:46:35 +0000 (13:46 +0000)]
Add LayoutTests http/tests/cookies at r71558
All tests pass or have failing test expectations.
Bug:
3227269
Change-Id: Ia0d58eb29d312c57c2f687e5431bb997b9433560
Kristian Monsen [Wed, 24 Nov 2010 10:34:19 +0000 (10:34 +0000)]
Removing CRASH() from asserts in WebRequest.cpp
These asserts have not been triggered in months, so removing the
crash (but keeping logging), and closing bug
2951985.
Change-Id: I5bb31c6e23c18e83301428582ad895b0b679ff86
Cary Clark [Fri, 19 Nov 2010 19:23:02 +0000 (14:23 -0500)]
reenable draw extras when GL is turned on
This captures the drawing in the UI thread, then passes
the drawing to the tile imaging thread.
The draw extras interface now takes an additional rectangle,
and each draw extra fills in the inval area formed as that
part is drawn.
The old extra implementation in GLWebViewState has been
removed. The inval portion of the setBaseLayer call has been split
out so it can be called directly.
bug:
3161294
Change-Id: I28d3e6879059770b973e7c0f7c0796909f7359aa
Ben Murdoch [Tue, 23 Nov 2010 14:42:50 +0000 (14:42 +0000)]
Revert "Merge Chromium at r65505: Work around regression introduced by 64761"
This reverts commit
0c918ae77b5edfe59b726ff40a7c86dbc4a94307.
This was actually not really a regression as we were still broken
previously it's just that this change makes that breakage more visible.
The problem reproduces on tip of tree Chromium, so we can make the fix
upstream and then cherry pick into Android.
Change-Id: I4d65544b4ea137d77c00a39fcd29edb28b40aa1c
Ben Murdoch [Tue, 23 Nov 2010 12:20:38 +0000 (04:20 -0800)]
Merge "By default, provide native AutoFill with an empty profile."
Ben Murdoch [Wed, 17 Nov 2010 11:03:51 +0000 (11:03 +0000)]
By default, provide native AutoFill with an empty profile.
If no profile is set then AutoFill will not inform Java that it has
found a form that could be autofilled. We want to be able to offer
the user the option Java side to set up a profile in this case, so
we set by default an empty profile that is active until the user
syncs their real profile. This ensures that the AutoFill detection
code will not early out.
Change-Id: I474b25c5b31dae59e4ddb23c7f398d013e9d3beb
Nicolas Roard [Tue, 23 Nov 2010 03:25:45 +0000 (19:25 -0800)]
Fix the Zoom out / dirty tiles bug
Bug:
3212531 Bug:
3212527
Change-Id: Ie038f5b42cd34653f0a3e8fbe8c1fb1c0c3aceca
Derek Sollenberger [Mon, 22 Nov 2010 21:13:02 +0000 (16:13 -0500)]
Fix case where the Texture class held a reference to a deleted Tile.
bug:
3217947
Change-Id: If75673ece459d6539f1fd990d538347e13d60d68
Shimeng (Simon) Wang [Sat, 20 Nov 2010 01:50:21 +0000 (17:50 -0800)]
Clear GL inside clear view.
issue:
3215000
Change-Id: I8668a857d840084c363b13c95efffc740a1451db
Nicolas Roard [Mon, 22 Nov 2010 21:01:53 +0000 (13:01 -0800)]
Fix browser ANR when using Hardware Acceleration
Bug:
3217947
Change-Id: Ic0da5c4c131026a88afff2acc254a23a4d6a42ff
Iain Merrick [Mon, 22 Nov 2010 17:34:46 +0000 (09:34 -0800)]
Merge "Call maybeSavePassword() from Chrome HTTP stack."
Iain Merrick [Mon, 22 Nov 2010 17:34:39 +0000 (09:34 -0800)]
Merge "Add maybeSavePassword() method to WebFrame."
John Reck [Mon, 22 Nov 2010 17:29:44 +0000 (09:29 -0800)]
Merge "Fixes the handling of unspecified mime-type"
John Reck [Fri, 19 Nov 2010 23:01:47 +0000 (15:01 -0800)]
Fixes the handling of unspecified mime-type
Bug:
3211038 and
3211569
If the server doesn't specify a Content-Type header in the response,
there was no mime-type set when using the Chrome HTTP stack. This change
fixes it so that it makes a guess based off of the URL extension.
Change-Id: I33ef90c1db54bdccb4e143795656f4f9b56572ca
Iain Merrick [Fri, 19 Nov 2010 12:07:55 +0000 (12:07 +0000)]
Call maybeSavePassword() from Chrome HTTP stack.
This will prompt the user to save the username and password
when posting a form. Same behaviour as the Android HTTP stack.
Bug:
3060688
Test: Logged into amazon.co.uk, checked that password is saved.
Change-Id: I4185b7ef2fbb299e4441b18b50f0018f0208d264
Iain Merrick [Thu, 18 Nov 2010 18:18:30 +0000 (18:18 +0000)]
Add maybeSavePassword() method to WebFrame.
This calls through to the new entrypoint on the Java size
(see change I3f3f84a1). I've done some modest refactoring
to minimize code duplication. The way we obtain a pointer
to the frame is still a bit messy, but I couldn't find a
really clean solution to that.
Change-Id: Ia5c12f615b8c41f1c846bbf6fd6dcfe56f3e2a51
Leon Scroggins [Thu, 18 Nov 2010 18:58:43 +0000 (13:58 -0500)]
Provide the line spacing attribute for textareas.
Bug:
3085564
In order to line up the WebTextView text with the page's text,
we need to know the line spacing attribute stored in the DOM.
Requires a change in frameworks/base:
https://android-git.corp.google.com/g/#change,80646
Change-Id: I539454df318826233496ffcd140ba1d6e396a4a0
Ben Murdoch [Mon, 22 Nov 2010 10:12:24 +0000 (10:12 +0000)]
Merge Chromium at r65505: Work around regression introduced by 64761
This is a temporary workaround to fix the regression introduced on
Android by http://src.chromium.org/viewvc/chrome?view=rev&revision=64761
It should be reverted when we understand better the root cause of the
regression.
Change-Id: I6652d817254f1b8c3857e2f14b53d5d034d2f12f
Ben Murdoch [Fri, 19 Nov 2010 18:41:33 +0000 (18:41 +0000)]
Merge Chromium at r65505: Merge AutoFill change 64761
Merge in http://src.chromium.org/viewvc/chrome?view=rev&revision=64761
This is not part of the bigger autofill merge as this introdcues
a regression that we need to understand.
Change-Id: I08cfcefb93181e70a8a48d0c6da2217d3a5bbcf9
Ben Murdoch [Fri, 19 Nov 2010 18:26:32 +0000 (18:26 +0000)]
Merge Chromium at r65505: Update AutoFill files.
Bring our forked AutoFill files in line with r65505.
http://src.chromium.org/viewvc/chrome?view=rev&revision=63527
http://src.chromium.org/viewvc/chrome?view=rev&revision=63797
http://src.chromium.org/viewvc/chrome?view=rev&revision=64470
http://src.chromium.org/viewvc/chrome?view=rev&revision=64572
http://src.chromium.org/viewvc/chrome?view=rev&revision=64612
http://src.chromium.org/viewvc/chrome?view=rev&revision=64691
http://src.chromium.org/viewvc/chrome?view=rev&revision=64696
Change-Id: I34755e649d1093a79ad50de53dfd1d1f9e92a584
Ben Murdoch [Fri, 19 Nov 2010 13:40:28 +0000 (13:40 +0000)]
Merge Chromium at r65505: Initiliase the OpenSSL library.
See https://chromereview.prom.corp.google.com/
2010002/
Change-Id: I16754f6f18a54350671a2226d4c8c69615c63812
Ben Murdoch [Thu, 18 Nov 2010 19:56:26 +0000 (19:56 +0000)]
Merge Chromium at r65505: Update use of AutoFillProfile.
With this new Chromium merge, unique profile ids have been
deprecated and profile previews are obtained through the
Label function rather that PreviewSummary.
WebKit/android/WebCoreSupport/autofill/WebAutoFill.cpp:
WebKit/android/jni/WebSettings.cpp:
WebKit/android/WebCoreSupport/autofill/WebAutoFill.h:
http://src.chromium.org/viewvc/chrome?view=rev&revision=63806
http://src.chromium.org/viewvc/chrome?view=rev&revision=65100
Change-Id: I8af86c8be8627e17b13edf66349d9df8f85638a1
Ben Murdoch [Thu, 18 Nov 2010 19:54:50 +0000 (19:54 +0000)]
Merge Chromium at r65505: Update log message handler
Add new parameters to our error logging handler.
WebKit/android/WebCoreSupport/ChromiumLogging.cpp
http://src.chromium.org/viewvc/chrome?view=rev&revision=65062
Change-Id: I6f34f2c5170b218c81894da541c923782e150e2e
Ben Murdoch [Thu, 18 Nov 2010 19:50:49 +0000 (19:50 +0000)]
Merge Chromium at r65505: Update call to CreateSystemHostResolver.
Pass NULL, which uses getaddrinfo() as the host resolver.
WebKit/android/WebCoreSupport/WebRequestContext.cpp:
http://src.chromium.org/viewvc/chrome?view=rev&revision=64271
Change-Id: I85d9371d568911ab925307bfbeb2e984bd947e2b
Cary Clark [Fri, 19 Nov 2010 17:19:21 +0000 (09:19 -0800)]
Merge "do a better job of finding closest selectable text"
Cary Clark [Fri, 19 Nov 2010 16:28:17 +0000 (11:28 -0500)]
do a better job of finding closest selectable text
Measure the closest to the edge, instead of closest to the center.
Identify vertical overlaps, and use them to discount characters
which are vertically closer.
Add more debugging.
bug:
3183434
Change-Id: I80a14be25260113220e446b876868566440be252
Cary Clark [Thu, 18 Nov 2010 19:32:25 +0000 (11:32 -0800)]
Merge "fix indices in cached frame"
Cary Clark [Thu, 18 Nov 2010 19:10:13 +0000 (14:10 -0500)]
call ReverseBidi as needed in FindCanvas
bug:
3184921
Change-Id: I2f05db656ab467a361fd395d264be3281f69b78c
Cary Clark [Thu, 18 Nov 2010 14:39:19 +0000 (09:39 -0500)]
fix indices in cached frame
After the CachedFrame is built, it may replace the previous
node. Replace the indices stored in the CachedFrame and in the
CachedNode with the index from the overwritten node.
bug:
3201246
Change-Id: Ia70f4f2c1f8009db2370e723bc8631e4a3873e75
Teng-Hui Zhu [Tue, 16 Nov 2010 01:22:49 +0000 (17:22 -0800)]
Merge WebKit at r71558: Update WebKit revision number.
Change-Id: I4839c51a42fb175f7e13934b637c2dc669953726
Teng-Hui Zhu [Tue, 16 Nov 2010 22:07:38 +0000 (14:07 -0800)]
Merge Webkit at r71558: Functions generated from .idl file should be guarded.
This is a cherry pick from the webkit fix.
http://trac.webkit.org/changeset/72216
The change cause this compile error is
http://trac.webkit.org/changeset/71515
Change-Id: Ibf7f689b1463116bb856739af7bbab8457498d05
Teng-Hui Zhu [Fri, 12 Nov 2010 18:27:14 +0000 (10:27 -0800)]
Merge Webkit at r71558: Add computeDefaultLanguage
Because navigotor.language no longer hardcoded.
Note that on android HTTP stack, this function will
naivelly return English in order to avoid code dependency
on chrome stack specific code.
http://trac.webkit.org/changeset/71484
Change-Id: I151105bb1478fa2da12ecf50c45102c3afceeb82
Teng-Hui Zhu [Fri, 12 Nov 2010 01:35:28 +0000 (17:35 -0800)]
Merge Webkit at r71558: fix code generation script
This is a cherry pick of webkit change.
http://trac.webkit.org/changeset/71984
Change-Id: I7bc944b6e9105a3cbfa52be305b7bd4755205244
Teng-Hui Zhu [Thu, 11 Nov 2010 23:23:48 +0000 (15:23 -0800)]
Merge Webkit at r71558: Define ImageInterpolationQuality in Android
http://trac.webkit.org/changeset/71337
Change-Id: I2c800902231b676f6b9ec73fe43c6621939b9fb9
Teng-Hui Zhu [Mon, 15 Nov 2010 22:04:10 +0000 (14:04 -0800)]
Merge Webkit at r71558: Resolve conflit with INSPECTOR
This is a cherry pick from one latest webkit fix.
http://trac.webkit.org/changeset/71981
Change-Id: I46d45af50654d836c7bfcf4c7d2d58a827cd366b
Teng-Hui Zhu [Thu, 11 Nov 2010 19:39:36 +0000 (11:39 -0800)]
Merge Webkit at r71558: resolve BackForwardList refactor conflict
http://trac.webkit.org/changeset/70960
Change-Id: I9a423e7069bd90a37edfdfcc4bd66fdc55600bcb
Teng-Hui Zhu [Tue, 16 Nov 2010 19:10:55 +0000 (11:10 -0800)]
Merge Webkit at r71558: Add function focusedFrameChanged
http://trac.webkit.org/changeset/71541
Change-Id: Ic068c75dec986e8761c6d57030ce7cfba2e8eb01
Teng-Hui Zhu [Thu, 11 Nov 2010 19:36:28 +0000 (11:36 -0800)]
Merge Webkit at r71558: Vitual function implementation in ChromeClientAndroid.*
http://trac.webkit.org/changeset/71041
Change-Id: Ifc88920a1cce9bf483d06a68a5ab4d23b63228d0
Teng-Hui Zhu [Thu, 11 Nov 2010 18:20:19 +0000 (10:20 -0800)]
Merge Webkit at r71558: Remove clipOutEllipseInRect
http://trac.webkit.org/changeset/70996
Change-Id: I0e12e9f1b94fc1a9798ab2faa61d4fd2aeb8f076
Teng-Hui Zhu [Thu, 11 Nov 2010 18:19:50 +0000 (10:19 -0800)]
Merge Webkit at r71558: Fix compiling error after a renaming
http://trac.webkit.org/changeset/71503
Change-Id: I12f477f18630b364b704b28fd63aaf46abb1be9a
Teng-Hui Zhu [Thu, 11 Nov 2010 18:14:40 +0000 (10:14 -0800)]
Merge Webkit at r71558: Cache.h relocated as MemoryCache.h
http://trac.webkit.org/changeset/71492
http://trac.webkit.org/changeset/71499
Change-Id: Iacd8a22d0c1caece7bec2273f1d4e69dd3e4eb2c
Teng-Hui Zhu [Thu, 11 Nov 2010 18:10:30 +0000 (10:10 -0800)]
Merge Webkit at r71558: Fix Makefiles
Change-Id: I6e437c61405ca117da435b32a8a933a3aa53ba32
Teng-Hui Zhu [Thu, 11 Nov 2010 02:17:16 +0000 (18:17 -0800)]
Merge Webkit at r71558: Fix conflicts.
WebCore/loader/FrameLoader.cpp
http://trac.webkit.org/changeset/70960
http://trac.webkit.org/changeset/70971
WebCore/css/CSSStyleSelector.cpp
http://trac.webkit.org/changeset/71218
WebCore/css/CSSParser.cpp
http://trac.webkit.org/changeset/71218
WebCore/html/HTMLLinkElement.cpp
http://trac.webkit.org/changeset/71159
WebCore/platform/android/PlatformBridge.h
http://trac.webkit.org/changeset/71484
WebCore/platform/android/TemporaryLinkStubs.cpp
http://trac.webkit.org/changeset/71484
WebCore/rendering/RenderTableSection.cpp
http://trac.webkit.org/changeset/71382
http://trac.webkit.org/changeset/71251
WebCore/rendering/RenderLayer.cpp
http://trac.webkit.org/changeset/71536
WebCore/rendering/RenderTable.cpp
http://trac.webkit.org/changeset/71382
WebCore/Android.v8bindings.mk
http://trac.webkit.org/changeset/71515
WebCore/bindings/scripts/CodeGeneratorV8.pm
http://trac.webkit.org/changeset/71227
http://trac.webkit.org/changeset/71230
Change-Id: I3ad8c6bc6e36c90aedebedf915a8614b2c786550
Teng-Hui Zhu [Wed, 10 Nov 2010 23:31:59 +0000 (15:31 -0800)]
Merge WebKit at r71558: Initial merge by git.
Change-Id: Ib345578fa29df7e4bc72b4f00e4a6fddcb754c4c
Kristian Monsen [Wed, 17 Nov 2010 19:48:10 +0000 (11:48 -0800)]
Merge "Removing tests that pass after a v8 merge"
Kristian Monsen [Wed, 17 Nov 2010 17:53:59 +0000 (17:53 +0000)]
Removing tests that pass after a v8 merge
Change-Id: Ia69c678a3098cc39c1e4687bd00f8ebba41cf732
Cary Clark [Wed, 17 Nov 2010 16:23:36 +0000 (08:23 -0800)]
Merge "improve select text anchor tracking"
Cary Clark [Wed, 17 Nov 2010 16:00:36 +0000 (11:00 -0500)]
improve select text anchor tracking
The initial point for the touch relative to the anchor
was erroneously calculated, causing the select anchor
to jump to the left or right each time a drag was
initiated. With this fix, the anchor no longer jumps.
bug:
3089212
Change-Id: I995afa97ef9851fc5f15d642d7975797d8eb2022
Cary Clark [Wed, 17 Nov 2010 15:35:32 +0000 (07:35 -0800)]
Merge "make helper class private"
Cary Clark [Wed, 17 Nov 2010 14:47:19 +0000 (09:47 -0500)]
make helper class private
Confusion on my part about nested classes prevented
me from getting this right the first time. Thanks to
Pat for setting me straight.
bug:
3114609
Change-Id: If6551ec98c08c45e0bbd84a51a3a48bda0e55619
Kristian Monsen [Wed, 17 Nov 2010 11:30:31 +0000 (03:30 -0800)]
Merge "Updating class used for getDatabaseDirectory and getCacheDirectory"
Ben Murdoch [Wed, 17 Nov 2010 11:06:00 +0000 (03:06 -0800)]
Merge "Merge autofill files after last Chromium merge"
Ben Murdoch [Tue, 16 Nov 2010 15:33:28 +0000 (15:33 +0000)]
Merge autofill files after last Chromium merge
Merge by hand the files that have been gotked implementing
AutoFill on Android. This classes use WebCore types rather
than the Chrome WebKit API types that the code would
normally use.
This brings the files up to date with the last external/chromium
merge to r63472 and updates their usages in line with that so
that the feature continues to work as intended.
Change-Id: Ic110873dc48c59555f0db76e7ad0c138c2fe96f2
Kristian Monsen [Tue, 16 Nov 2010 18:22:23 +0000 (18:22 +0000)]
Updating class used for getDatabaseDirectory and getCacheDirectory
Required by this CL in frameworks/base:
https://android-git.corp.google.com/g/#change,80051
Change-Id: I673e8a5a112d96721fe2e06d84653d1c75fd667a
Cary Clark [Tue, 16 Nov 2010 14:52:37 +0000 (09:52 -0500)]
improve left edge tap detection
Sites like Gizmodo.com layout paragraphs with extra space
between lines. Increase the slop allowed when looking for
the left edge of a paragraph to accommodate this.
Add #define to disable sideways scrolling if the line is
already entirely visible.
Make various slop consts more clearly named and commented.
A companion fix in frameworks/base is not required, but
makes scrolling as a result of taps more reliable.
bug:
1331125
bug:
3099569
Change-Id: I0ed89c1f53de99831e835e417b5409c1176fc841
Patrick Scott [Mon, 15 Nov 2010 19:20:19 +0000 (14:20 -0500)]
Fix a crash due to null java strings.
Change-Id: I29a229da168181ba3d5bc5fb84365f275eb42869
Cary Clark [Mon, 15 Nov 2010 17:05:59 +0000 (12:05 -0500)]
increase the select text anchor hit slop
bug:
3195900
Change-Id: I55317b6c2d1f2dd48e2522df9ba45653a479793a
Cary Clark [Fri, 12 Nov 2010 21:18:31 +0000 (16:18 -0500)]
make the button state revert to normal
This moves the logic used to ensure that cursor rings
eventually go away after being tapped to buttons, so
that they too go back to their normal state after a short
period.
bug:
2127333
Change-Id: Ia2e1d4b7f735581dbd88bf32b35fe5b24922ea11
Patrick Scott [Thu, 11 Nov 2010 18:14:26 +0000 (13:14 -0500)]
Allow applications to intercept requests.
The api returns a response containing the mime type, encoding, and an optional
InputStream for reading data. Move the asset and content url logic into the
new api to consolidate some code.
Requires a change in frameworks/base.
Bug:
2905943
Change-Id: Ic7af410308872042c412aedf62e589f6d2095782
Iain Merrick [Fri, 12 Nov 2010 14:24:49 +0000 (14:24 +0000)]
Don't set error description strings in ResourceError.
These strings may be displayed to the user, so they need to be
localized. In a separate change (I7dec8dff) I updated the Java
framework to use a standard localized message if the description
is empty, so here we just use the empty string (same as Chrome).
Bug:
3172265
Test: Load http://google.com:81 to force a timeout error, check
that the correct message is displayed.
Change-Id: I9042a2ee99cd78619cc0392cdc2cbb82ebd51743
Patrick Scott [Fri, 12 Nov 2010 20:53:20 +0000 (15:53 -0500)]
Use LayerAndroid::find with the original x and y.
If the picture is not provided, use the findInner case to find the layer at the
given position.
Change-Id: I811c1c4d631086832a3a32038834117a1c4e9601
Leon Scroggins [Fri, 12 Nov 2010 19:39:06 +0000 (14:39 -0500)]
Return a float for text size.
Bug:
3085564
Requires a change in frameworks/base: https://android-git.corp.google.com/g/#change,79591
Change-Id: Iaf3ebc405fca349e23bebb113228c4b550e6b88e
Shimeng (Simon) Wang [Fri, 12 Nov 2010 18:36:13 +0000 (10:36 -0800)]
Merge "Fix mobile google.com flickering issue."
Shimeng (Simon) Wang [Fri, 12 Nov 2010 01:22:57 +0000 (17:22 -0800)]
Fix mobile google.com flickering issue.
Implementing request cache policy in chrome http integration.
issue:
3008478
Change-Id: I3d7c32912e72e7639885056fb70f9f627d075c4e
Derek Sollenberger [Fri, 12 Nov 2010 17:53:51 +0000 (09:53 -0800)]
Merge "Track all tiles in the viewport instead of only the upper left tile."
Derek Sollenberger [Thu, 11 Nov 2010 20:22:47 +0000 (15:22 -0500)]
Track all tiles in the viewport instead of only the upper left tile.
We currently store only the coordinates of the upper left tile and
leave it to the caller to compute the actual number of tiles in the
viewport. This CL computes that information upfront and then passes
a rect containing the tiles in the viewport to the necessary parties.
This also allows us to intelligently prepare tiles that are currently
off-screen but given the user's scroll direction will be on-screen
shortly.
bug:
3190926
Change-Id: I50a5922ed8bc7e8fd8e4cacd4986a9f275cd37b0
Jeff Brown [Fri, 12 Nov 2010 02:57:31 +0000 (18:57 -0800)]
Merge "Add new keycodes initially used for GoogleTV devices."
Cary Clark [Thu, 11 Nov 2010 18:02:24 +0000 (13:02 -0500)]
compute select text distance biasing x over y
Rather than using a regular distance function, first
find the closest match in Y then look for the closest
match in X. This allows the touch point to be well to
the right or left of the paragraph and still move the
selection to the next or preceeding line while the
selection is extended.
bug:
3183434
Change-Id: I6abf14b9dbca4b7bafe2cb9a96a2915732038ce1
Cary Clark [Thu, 11 Nov 2010 15:14:14 +0000 (10:14 -0500)]
use layer id to maintain selection
The pictures used to draw the page content can change
in response to refreshes, as the selection is moved or
as the screen is scrolled or zoomed.
Track the layer id, instead of the picture pointer,
to find the picture that contains the current selection.
Before, the selection would disappear as the screen
was pinched, or the highlight might disappear even
though the anchors draw, or the highlight and anchors
might be drawn offset from the screen data. This change
eliminates this class of bugs.
Change-Id: Ifc440b8aa0c9d6d232e298e42c4c544a41629442
http://b/
3183492
Steve Block [Thu, 11 Nov 2010 16:56:18 +0000 (08:56 -0800)]
Merge "Remove the failing expectation for fast/encoding/char-encoding.html for DRT2"
Derek Sollenberger [Thu, 11 Nov 2010 14:59:44 +0000 (09:59 -0500)]
Only check to see if the page is ready when necessary.
This CL reorders the contents of an if statement to ensure that
we do not pointlessly check to see if the tiledPage is ready.
Change-Id: I7155c4beb5a6dcbcb195e2c80c317a599a9553af
Derek Sollenberger [Thu, 11 Nov 2010 14:56:03 +0000 (06:56 -0800)]
Merge "Use a fixed number of BaseTiles per TiledPage."
Steve Block [Thu, 11 Nov 2010 14:25:11 +0000 (14:25 +0000)]
Remove the failing expectation for fast/encoding/char-encoding.html for DRT2
This test now passes with both HTTP stacks.
Bug:
3047156
Change-Id: I44e1834782acc3d0674beb70f2a4ac78343cb2c5
Derek Sollenberger [Fri, 5 Nov 2010 12:12:51 +0000 (08:12 -0400)]
Use a fixed number of BaseTiles per TiledPage.
This CL ensures that the TiledPage only keeps track of BaseTiles
that are backed by a texture. In order to avoid unecessary
allocations on the stack we pre-allocate the memory for the tiles
and reuse tiles.
Change-Id: I894cf4d5c6b564169ccf7a967629f5d670a5426d
Bart Sears [Thu, 11 Nov 2010 06:05:43 +0000 (22:05 -0800)]
Revert "Merge "Fix text-shadow style when blur is 0px""
Bug:
3185661
This reverts commit
ccbb8a238ca0ba5fc0780884e281e6b6a601a675, reversing
changes made to
fcb462de639304deaa2da761823fbfeef88b36d4.
Steve Block [Wed, 10 Nov 2010 20:42:24 +0000 (12:42 -0800)]
Merge "Set failing expectations for DRT2 for some pixel LayoutTests"
Cary Clark [Wed, 10 Nov 2010 20:13:23 +0000 (12:13 -0800)]
Merge "build select region for one character"
Cary Clark [Wed, 10 Nov 2010 20:07:49 +0000 (15:07 -0500)]
build select region for one character
bug:
3183119
Change-Id: Iebd070feeb9c9a52e0d6214563c378647d97b977
Steve Block [Wed, 10 Nov 2010 19:49:25 +0000 (19:49 +0000)]
Set failing expectations for DRT2 for some pixel LayoutTests
When these tests were added to the triaged set they were observed to
pass with DRT1 because they have no generic expected result. DRT1
considers this to imply a non-text test and hence ignores the result.
Change-Id: Id3e6edb8e1e3927e8441e2b88761cfa348688155
Cary Clark [Wed, 10 Nov 2010 15:26:31 +0000 (10:26 -0500)]
Ratchet back style changes that rebuild nav caches
Many style changes do not visibly affect the way
links are detected -- so limit the rebuild to
style changes that either move the size the style,
or changes that edit the style's display property.
Styles appearing or disappearing (i.e., one is 0)
is not treated as a style version change since
these events accompany dom edits which are detected
outside of this change.
bug:
3181683
Change-Id: Ie62ef524a956fcd9298ce5878b632645a22e461a