On 2013/06/04 09:01:51, tbroyer wrote:
Can someone move that patch over to Gerrit? Would you prefer that I do
it?
(BTW, any reason this hadn't been reviewed yet?)
No particular reason other than it dropping off the radar with other
stuff going on. Of course we will want unit tests for this.
Please go ahead and move it, but with a couple of team members away we
probably won't get to it this week.
http://gwt-code-reviews.appspot.com/1839803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
---
You received this message because you are subscribed to the Google
On 2013/05/06 08:05:00, John A. wrote:
On 2013/04/30 23:05:08, skybrian wrote:
Hello, could you avoid reformatting code that you're not working on?
Or if you
want to just fix the javadoc, do that in a separate CL.
Appears the eclipse auto formatting had changed Element JavaDocs
without
Hello, could you avoid reformatting code that you're not working on? Or
if you want to just fix the javadoc, do that in a separate CL.
http://gwt-code-reviews.appspot.com/1815803/
--
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
---
You received this message because you
Why protected? Also, shouldn't the unit tests be updated?
http://gwt-code-reviews.appspot.com/1891804/
--
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
---
You received this message because you are subscribed to the Google Groups Google Web Toolkit Contributors group.
To
LGTM
http://gwt-code-reviews.appspot.com/1890803/
--
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
---
You received this message because you are subscribed to the Google Groups Google Web Toolkit Contributors group.
To unsubscribe from this group and stop receiving emails
Seems fine. (I'll take care of these nits.)
http://gwt-code-reviews.appspot.com/1587803/diff/10006/user/src/com/google/gwt/editor/client/adapters/HasDataEditor.java
File user/src/com/google/gwt/editor/client/adapters/HasDataEditor.java
(right):
I'm not all that familiar with the editor framework but perhaps there's
a cleaner way to do this? (See comments.)
http://gwt-code-reviews.appspot.com/1587803/diff/1/user/src/com/google/gwt/editor/client/adapters/EditorSource.java
File
http://gwt-code-reviews.appspot.com/1587803/diff/5002/user/src/com/google/gwt/editor/client/adapters/EditorSource.java
File user/src/com/google/gwt/editor/client/adapters/EditorSource.java
(right):
LGTM. Thanks!
http://gwt-code-reviews.appspot.com/1881803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM
http://gwt-code-reviews.appspot.com/1859804/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Reviewers: cromwellian,
Description:
Fix a GWT-RPC issue where covariant array types for lists of raw types
were
being generated inconsistently, causing serialization policy generation
to
be nondeterministic. Includes tests.
Fixes issue 7791.
Please review this at
On 2012/12/05 23:41:03, cromwellian wrote:
LGTM. General potentially unrelated question, what about the case
static class Covariant extends Itemint[]
Is this handled properly?
It looks like it. In that case, isClassOrInterface() will return null,
so it won't look at covariant types, just
Reviewers: mdempsky,
Description:
Update devmode Firefox plugin to support Firefox 17 on the Mac:
- started compiling with C++11, using clang (from XCode 4.5.2)
- dropped PPC from build
Includes release candidate.
Review by: mdemp...@google.com
Please review this at
On 2012/11/29 05:45:46, cromwellian wrote:
On 2012/11/29 03:35:49, skybrian wrote:
LGTM. No API/Source breakages this time around?
There were but I already fixed them when I did Linux. For the Mac it was
upgrading XCode and figuring out how much I need to mirror Firefox's
compiler options
On 2012/11/23 08:51:26, nfm wrote:
ping?
Sorry about that. It's hard to say which is faster without testing, but
I think having one cache is simpler than having two. So let's go ahead
with merging these variables.
http://gwt-code-reviews.appspot.com/1859804/
--
Reviewers: mdempsky,
Description:
Add Windows DLL for Firefox 17 to dev mode plugin.
Please review this at http://gwt-code-reviews.appspot.com/1871803/
Affected files:
A plugins/xpcom/VisualStudio/ff170-xpcom.vcxproj
M plugins/xpcom/prebuilt/extension/chrome.manifest
--
http://gwt-code-reviews.appspot.com/1869804/diff/1/user/test/com/google/gwt/user/rebind/rpc/SerializableTypeOracleBuilderTest.java
File
user/test/com/google/gwt/user/rebind/rpc/SerializableTypeOracleBuilderTest.java
(right):
Reviewers: mdempsky,
Description:
Update devmode Firefox plugin to support Firefox 17 on 64-bit Linux.
(Other platforms coming soon.)
Please review this at http://gwt-code-reviews.appspot.com/1870803/
Affected files:
M plugins/xpcom/ExternalWrapper.cpp
M plugins/xpcom/FFSessionHandler.cpp
Reviewers: mdempsky,
Description:
Move non-GWTTestCase tests in RpcSuite to RpcSuiteNoBrowser,
so they don't unnecessarily get run multiple times in different browser
environments.
Also fix the build to handle running *SuiteNoBrowser suites.
Review by: mdemp...@google.com
Please review this at
Reviewers: mdempsky,
Description:
Minor cleanup in SerializableTypeOracleBuilder:
- remove unused fields
- TypeInfoComputed can be static
- tighten visibility
- remove some indirection
- fix some warnings
Review by: mdemp...@google.com
Please review this at
Reviewers: cromwellian,
Description:
Fix a bug where GWT-RPC would not generate some covariant array types,
possibly causing inconsistent serialization policies. Includes a test.
Fixes issue 7791
Please review this at http://gwt-code-reviews.appspot.com/1869804/
Affected files:
M
http://gwt-code-reviews.appspot.com/1859804/diff/7001/user/src/com/google/gwt/user/client/Window.java
File user/src/com/google/gwt/user/client/Window.java (right):
http://gwt-code-reviews.appspot.com/1859804/diff/7001/user/src/com/google/gwt/user/client/Window.java#newcode322
LGTM
http://gwt-code-reviews.appspot.com/1866803/diff/1/user/src/com/google/gwt/logging/server/StackTraceDeobfuscator.java
File user/src/com/google/gwt/logging/server/StackTraceDeobfuscator.java
(right):
http://gwt-code-reviews.appspot.com/1866803/diff/1/user/src/com/google/gwt/logging/server/StackTraceDeobfuscator.java
File user/src/com/google/gwt/logging/server/StackTraceDeobfuscator.java
(right):
On 2012/11/05 16:34:46, perneto wrote:
FYI: I added http://code.google.com/p/rietveld/issues/detail?id=402 to
track this change, as requested by Brian.
Oops, that's actually the rietveld issue tracker, not the GWT issue
tracker. The one you want is at:
Context: the goal is to put a footer in the middle section so it will
scroll into view below all the data rows.
http://gwt-code-reviews.appspot.com/1864803/diff/1/user/src/com/google/gwt/user/cellview/client/DataGrid.java
File user/src/com/google/gwt/user/cellview/client/DataGrid.java (right):
http://gwt-code-reviews.appspot.com/1859804/diff/1/user/src/com/google/gwt/user/client/Window.java
File user/src/com/google/gwt/user/client/Window.java (right):
http://gwt-code-reviews.appspot.com/1859804/diff/1/user/src/com/google/gwt/user/client/Window.java#newcode238
LGTM
http://gwt-code-reviews.appspot.com/1863803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Reviewers: mdempsky,
Description:
Chrome devmode plugin: recompile Windows DLL to be statically linked.
Other fixes:
- Check in zip file as uploaded to Chrome web store; remove crx file.
- Makefile zip rule didn't include the manifest.
Please review this at
Reviewers: mdempsky,
Description:
Chrome plugin: don't use an iterator after the underlying entry has been
freed.
Rebuilt Windows DLL and created a release.
(This patch might help on Linux and Mac as well, but for safety I'm not
updating them yet.)
Please review this at
http://gwt-code-reviews.appspot.com/1859803/diff/2001/user/src/com/google/web/bindery/requestfactory/apt/ValidationTool.java
File
user/src/com/google/web/bindery/requestfactory/apt/ValidationTool.java
(right):
On 2012/10/24 21:24:02, mdempsky wrote:
Alternatively, take a look at Patch Set #1, which simply added an
extra
parameter to exec() to pass in a custom JavaCompiler instance. Would
that patch
(updated to make the exec() methods public) be preferable?
Yeah, seems simpler.
LGTM
http://gwt-code-reviews.appspot.com/1857803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Reviewers: mdempsky,
Description:
Update the DevMode Chrome plugin's Makefile to build a zip file to
upload to the Chrome web store.
Instead of building a crx file, we build a zip file containing the files
to upload. Also, the Chrome web store doesn't allow the key or update
URL to be set in
We have no plans to support reflection in client-side code. What problem
are you trying to solve?
http://gwt-code-reviews.appspot.com/1855803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM
I don't quite understand why it's capturing click events when it's 10x10
and it's 20 units offscreen, but setting visibility seems like the right
thing to do anyhow.
I'd guess that having it explicitly offscreen makes it less likely to be
in the way somehow when using developer tools that
LGTM
http://gwt-code-reviews.appspot.com/1850804/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Hi, one-line typo fixes like this aren't worth bothering with,
particularly since the constructor isn't public. If you want to help, I
would suggest either doing a more extensive spell-checking pass, or
figure out how to say something more useful.
In this case, it looks like all
11305)
+++ plugins/npapi/npapi/npapi.h (working copy)
@@ -258,6 +258,16 @@
} NPDrawingModel;
#endif
+// copied from newer npapi.h - skybrian
+#ifdef XP_MACOSX
+typedef enum {
+#ifndef NP_NO_CARBON
+ NPEventModelCarbon = 0,
+#endif
+ NPEventModelCocoa = 1
+} NPEventModel;
+#endif
+
typedef
http://gwt-code-reviews.appspot.com/1844803/diff/1/plugins/npapi/Makefile
File plugins/npapi/Makefile (right):
http://gwt-code-reviews.appspot.com/1844803/diff/1/plugins/npapi/Makefile#newcode17
plugins/npapi/Makefile:17: ARCH=x86
On 2012/09/29 02:20:02, jtamplin wrote:
I assume this means
http://gwt-code-reviews.appspot.com/1840803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/09/26 18:51:33, jtamplin wrote:
LGTM, though I still wonder if you want -style PRETTY to be the
default.
It only increases the size of the zipped extension by about 5% (from
349k to 365k). I think I'd rather leave it on since it makes debugging
easier.
For comparison, the Firefox extension is now a 7M download, due to
having a separate binary for each version of Firefox * three platforms.
http://gwt-code-reviews.appspot.com/1840803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Reviewers: cromwellian,
Description:
Upgraded the devmode Chrome extension to manifest version 2.
To do this, we need to replace all inline javascript with code in a
separate js file. Also, changed the options page to use the
DirectInstall linker so that GWT bootstrap doesn't use eval() or
Could you write a test for this change?
http://gwt-code-reviews.appspot.com/1836803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/09/17 20:43:56, tbroyer wrote:
Do you think it would be possible to share the BUILD file(s), or a
stripped-down
version of it, even privately?
There's a high-level overview here:
http://google-engtools.blogspot.com/2011/08/build-in-cloud-how-build-system-works.html
It doesn't
http://gwt-code-reviews.appspot.com/1834803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM. Thanks!
http://gwt-code-reviews.appspot.com/1831803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/09/13 09:57:02, manolo.carrasco wrote:
RF fails in the case of @ProxyFor pointing to a domain-interface
instead of a
domain-implementation, could you consider to include the patch
https://gwt-code-reviews.appspot.com/1764804/ as well in 2.5?
I think that counts as a new feature
I don't see any usages of XmlHttpRequest.ResponseType in Google code. We
could have constants instead of the enum.
https://gwt-code-reviews.appspot.com/1830803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/09/13 20:22:43, tbroyer wrote:
I kind of like the enums as in com.google.gwt.dom.client.Style (also
used in
SafeStyles), but I'll make the setResponseType(String) public, and add
a
getResponseTypeString() to the enum.
Oops, I guess I should have responded. I'm fine with it either
LGTM
https://gwt-code-reviews.appspot.com/1830803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM
https://gwt-code-reviews.appspot.com/1829804/diff/1/user/src/com/google/web/bindery/autobean/shared/impl/AutoBeanCodexImpl.java
File
user/src/com/google/web/bindery/autobean/shared/impl/AutoBeanCodexImpl.java
(right):
We've decided against this change for now (increases the size of the API
too much). A HasElement interface sounds somewhat reasonable but we can
add that later if needed.
I suppose if Element implemented HasElement, returning itself, we could
avoid doubling the number of methods, but this seems
Hmm, I'm inclined to skip this patch for GWT 2.5. The typo fix at least
makes it work on some browsers...
https://gwt-code-reviews.appspot.com/1830803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Hmm. If it were named HasAriaElement, then it would make sense that
not all widgets implement it. Only those that need to be customized for
proper Aria support would have it.
http://gwt-code-reviews.appspot.com/1828804/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
There's a usage in Google code as well, but it looks like it can't pass
in a null.
I'd rather fix iterableFromConstrantViolations [sic] but I don't care
much, so submitting.
https://gwt-code-reviews.appspot.com/1826803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM. I'll commit this soon.
https://codereview.appspot.com/6492092/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
I don't know this code, but since pushViolations() and
iterableFromConstraintViolations() are both public methods, under the
principle of being conservative about what you send and liberal about
what you receive, it seems like iterableFromConstraintViolations()
should return an empty iterable
LGTM. Committing...
https://gwt-code-reviews.appspot.com/1712803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/09/07 19:57:58, skybrian wrote:
LGTM. Committing...
I'm getting 500 errors in MultipleFactoriesTest.
java.lang.RuntimeException: The RequestFactory ValidationTool must be
run for the
com.google.web.bindery.requestfactory.shared.MultipleFactoriesTest.Factory1
RequestFactory type
Never mind, looks like a hard-coded list in our build file.
https://gwt-code-reviews.appspot.com/1712803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM, but I'm not going to be able to submit this right away. (Working
on Firefox plugin.)
https://gwt-code-reviews.appspot.com/1818803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
http://gwt-code-reviews.appspot.com/1817803/diff/1/plugins/xpcom/JavaObject.cpp
File plugins/xpcom/JavaObject.cpp (right):
http://gwt-code-reviews.appspot.com/1817803/diff/1/plugins/xpcom/JavaObject.cpp#newcode385
plugins/xpcom/JavaObject.cpp:385: (JSVAL_IS_PRIMITIVE(argv[1])
Reviewers: acleung,
Description:
Got the development mode plugin to compile for Firefox 15 using
xulrunner-15.0b6 on 64-bit Linux.
(This is not ready to commit yet, since it's not tested, but here's a
preview.)
Please review this at http://gwt-code-reviews.appspot.com/1816803/
Affected
Thanks! In that case, I'll commit this and follow up in another CL if
needed.
http://gwt-code-reviews.appspot.com/1816803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/08/24 20:41:13, jtamplin wrote:
#ifndef JSVAL_IS_OBJECT
#define JSVAL_IS_OBJECT(x) (!JSVAL_IS_PRIMITIVE(x))
#endif
They're actually subtly different because the removed JSVAL_IS_OBJECT
would return true for a null, and so does JSVAL_IS_PRIMITIVE. However, I
didn't see any code that
Strangely, this isn't working for me. The plugin seems to install and I
can bring up preferences, but the __gwt_HostedModePlugin javascript
variable doesn't show up, so it goes to the missing plugin page. Any
ideas?
Firefox version:
Mozilla/5.0 (X11; Linux i686 on x86_64; rv:15.0) Gecko/20100101
Turns out I needed to update chrome.manifest. After fixing this, I'm
getting an error on reloading the page. I'm surprised it was working for
jat, unless that was with Firefox 14.
Connect(url=http://localhost:8080/Showcase.html?gwt.codesvr=xx.xx.xx.xx:9997#!CwCheckBox,
sessionKey=XXX,
Reviewers: acleung,
Description:
Fix development mode Firefox plugin to work on with Firefox 15 on Linux.
Tested with 64-bit linux. Now builds 32-bit Linux for 15, but that's not
tested yet.
Please review this at http://gwt-code-reviews.appspot.com/1817803/
Affected files:
M
Reviewers: mdempsky,
Description:
Attempt to fix flakiness in testGetClientSize by bumping up the timeout.
(I didn't track down the root cause.)
Please review this at http://gwt-code-reviews.appspot.com/1814803/
Affected files:
M user/test/com/google/gwt/user/client/WindowTest.java
Index:
LGTM
http://gwt-code-reviews.appspot.com/1813803/diff/1/user/test/com/google/gwt/dev/js/client/CoverageTest.java
File user/test/com/google/gwt/dev/js/client/CoverageTest.java (right):
http://gwt-code-reviews.appspot.com/1813803/diff/1/user/test/com/google/gwt/dev/js/client/CoverageTest.java
File user/test/com/google/gwt/dev/js/client/CoverageTest.java (right):
LGTM
http://gwt-code-reviews.appspot.com/1800805/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Okay, you fixed autohide. They also mentioned that DialogBox can't be
dragged. Does this change fix that problem too?
http://gwt-code-reviews.appspot.com/1810803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
I'm not familiar with FocusPanel, but it looks like it's used by a
variety of subclasses that need to accept keyboard input for various
reasons. So I don't think this is a good idea in all cases. Perhaps
FocusPanel needs a way to set the ARIA role and you should do this only
for your app?
LGTM
http://gwt-code-reviews.appspot.com/1810803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM
http://gwt-code-reviews.appspot.com/1803804/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
It makes sense, considering that I don't know anything about iPads.
Could you describe the bug in more detail? Is there an issue for it in
the GWT issue tracker?
http://gwt-code-reviews.appspot.com/1810803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/08/14 01:01:20, skybrian wrote:
It makes sense, considering that I don't know anything about iPads.
Could you
describe the bug in more detail? Is there an issue for it in the GWT
issue
tracker?
It looks like this issue is related:
http://code.google.com/p/google-web-toolkit/issues
The idea is that any SingleSelectionModel can be thought of as a set
that contains either zero or one items. So it seems appropriate that it
should have the new methods - and after all, the whole point is to be
able to treat them uniformly.
SetSelectionModel isn't intended to imply that the
Yes, by editing the description in Critique. (I just did that for you.)
http://gwt-code-reviews.appspot.com/1798806/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
http://gwt-code-reviews.appspot.com/1799804/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Reviewers: mdempsky,
Description:
Rename aria.Role to aria.RoleImpl and make private. (The next step will
be to
rename aria.RoletypeRole to aria.Role.)
Rationale: this hides the implemention so that it's not in the public
API,
and we can change the implementation more freely in future GWT
Reviewers: atincheva,
Description:
Rename RoletypeRole to Role and improve the javadoc. Subtypes of Role
now
link to the ARIA specification.
Please review this at http://gwt-code-reviews.appspot.com/1799804/
Affected files:
M user/src/com/google/gwt/aria/client/AlertRole.java
M
LGTM
http://gwt-code-reviews.appspot.com/1793806/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM
http://gwt-code-reviews.appspot.com/1800803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Reviewers: cromwellian,
Description:
Increase a timeout in JsonpRequestSuite to see if it fixes a flaky test.
(It seems to help when running the tests by hand, but it's hard to
tell.)
Please review this at http://gwt-code-reviews.appspot.com/1798804/
Affected files:
M
http://gwt-code-reviews.appspot.com/1800803/diff/1/user/test/com/google/gwt/core/public/script_injector_test_absolute.js
File
user/test/com/google/gwt/core/public/script_injector_test_absolute.js
(right):
On 2012/07/27 07:48:14, cromwellian wrote:
LGTM. Is this the only case? I would imagine there's probably many
tests that
were unknowingly dependent on the DOM being modified and left with
persistent
state between tests.
I don't know. My guess is that most tests don't care that there's
LGTM
http://gwt-code-reviews.appspot.com/1797803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2012/07/26 07:02:08, tbroyer wrote:
I'm not against this kind of change but IMO any deviation from the
JSON and/or
ECMAScript specs should be documented, or we risk removing them at a
later time
and break things again (that being said, I don't understand what's
broken here
and how this
Thanks, that's gotten me reading the right code; before I couldn't even
find the SelectionChangeListener and now I see that it's in
HasDataPresenter.
So for this CL, I think we just assume rendering happens when it needs
to and update aria-selected (and other attributes) whenever a node gets
Reviewers: cromwellian,
Description:
Fix UiSuite to pass when test methods are run in a different order.
The symptom was that many tests would fail, starting with
RootPanelTest.testGetById(). I don't know the root cause, but it seems
that
leaving a RichTextArea with focus in the RootPanel
Thanks John.
On 2012/07/25 22:42:13, john.labanca wrote:
As far as issue 7480, it sounds like a bug. Updating the
SelectionModel should
update the CellTree's keyboard selection.
Well, the question is how to do this. I assume we install a
SelectionChangeHandler. However, we don't know how
Reviewers: cromwellian,
Description:
Fix TestOracleMediatorFromByteCodeTest to work when run on JDK 7.
The issue is that java.lang.Throwable has nested classes in JDK 7 but
not JDK 6.
So, change the assertion to allow any nested classes.
Please review this at
Reviewers: acleung,
Description:
Fix JSON escaping of unicode characters to work in JDK 7.
JDK 7 supports Unicode 6 and some characters changed:
- zero-width-space is no longer a whitespace character
- invisible-plus is new
This caused JSON encoding tests to fail in HTMLUnit somehow, so
escape
LGTM
http://gwt-code-reviews.appspot.com/1749803/diff/12001/dev/core/src/com/google/gwt/dev/javac/ProgressLogger.java
File dev/core/src/com/google/gwt/dev/javac/ProgressLogger.java (right):
Thanks Alan.
http://gwt-code-reviews.appspot.com/1792803/diff/1/plugins/xpcom/JavaObject.h
File plugins/xpcom/JavaObject.h (right):
http://gwt-code-reviews.appspot.com/1792803/diff/1/plugins/xpcom/JavaObject.h#newcode55
plugins/xpcom/JavaObject.h:55: static void finalize(JSFreeOp* fop,
1 - 100 of 238 matches
Mail list logo
