Revision: 7834
Author: j...@google.com
Date: Wed Mar 31 20:30:31 2010
Log: Initial work on adding pluggable transports on the Java side, adding
prebuilt binaries for Linux JNI code, build changes.
http://code.google.com/p/google-web-toolkit/source/detail?r=7834
Added:
/changes/jat/csproto/dev
Revision: 7832
Author: j...@google.com
Date: Wed Mar 31 17:07:40 2010
Log: Add timings, separate detach from destructor.
http://code.google.com/p/google-web-toolkit/source/detail?r=7832
Modified:
/changes/jat/csproto/plugins/platform/Linux/Futex.cpp
/changes/jat/csproto/plugins/platform/Linux/
What happened with the files? It seems the patch that was uploaded got all
messed up. Git5 mis-identified renames, perhaps?
2010/3/31
>
> http://gwt-code-reviews.appspot.com/291801/diff/4001/5003
> File
> bikeshed/src/com/google/gwt/collections/ImmutableArrayEmptyImpl.java
> (right):
>
> http://
http://gwt-code-reviews.appspot.com/291801/diff/4001/5003
File
bikeshed/src/com/google/gwt/collections/ImmutableArrayEmptyImpl.java
(right):
http://gwt-code-reviews.appspot.com/291801/diff/4001/5003#newcode27
bikeshed/src/com/google/gwt/collections/ImmutableArrayEmptyImpl.java:27:
Assertions.ass
I have uploaded another patch set to http://gwt-code-reviews.appspot.com/226801
to address the concerns raised. See inline messages below.
This latest version has an ASCII only option for the is*() methods
witch has an overhead of a couple hundred bytes. See below for the
size penalties for the ta
http://gwt-code-reviews.appspot.com/291801/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
To unsubscribe, reply using "remove me" as the subject.
LGTM
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
To unsubscribe, reply using "remove me" as the subject.
Reviewers: Ray Ryan,
Description:
Use PagingTableListView in TableEntityListView
Work on Mail sample app, remove index param from Column.getValue()
Review by: rj...@google.com
Please review this at http://gwt-code-reviews.appspot.com/298801/show
Affected files:
M bikeshed/src/com/google/gwt/
The strings look correct when running the sample. I made sure the files
were UTF8 encoded before editing them.
http://gwt-code-reviews.appspot.com/297801/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
To unsubscribe, reply using "remove me" as the subject.
LGTM with one caveat.
http://gwt-code-reviews.appspot.com/297801/diff/1/3
File
samples/showcase/src/com/google/gwt/sample/showcase/client/ShowcaseConstants_fr.properties
(right):
http://gwt-code-reviews.appspot.com/297801/diff/1/3#newcode29
samples/showcase/src/com/google/gwt/sample/showcase/cl
Reviewers: jat,
Description:
We got some updated translations for Showcase.
I verified that Showcase runs correctly, and that comma-delimited lists
have the correct English comma.
Please review this at http://gwt-code-reviews.appspot.com/297801/show
Affected files:
samples/showcase/src/com
Revision: 7831
Author: j...@google.com
Date: Wed Mar 31 12:05:41 2010
Log: Measure and tweak spinlock times, add some of the work to support
kernel
cleanup of futexes, restructure lock loop for better performance.
http://code.google.com/p/google-web-toolkit/source/detail?r=7831
Modified:
/ch
Well, it looks like the bar for getting code into that package is pretty
low, based on the existing classes there. :)
But more seriously, this class appears to do three useful things, so let me
comment on them in turn:
1) Remember a package name, genCtx, and logger so you don't have to keep
track
Reviewers: rdayal,
Description:
Selenium.start() can block indefinitely if the Selenium RC cannot
successfully start an instance of the browser, such as when an
unexpected popup appears. This has annoying results for
RunStyleSelenium.
First, we were calling Selenium.start() inside of the synchro
LGTM
http://gwt-code-reviews.appspot.com/294801/diff/1/2
File
user/src/com/google/gwt/user/client/rpc/core/java/lang/Void_CustomFieldSerializer.java
(right):
http://gwt-code-reviews.appspot.com/294801/diff/1/2#newcode29
user/src/com/google/gwt/user/client/rpc/core/java/lang/Void_CustomFieldSeri
Revision: 7830
Author: j...@google.com
Date: Wed Mar 31 10:18:10 2010
Log: Add timeout support.
http://code.google.com/p/google-web-toolkit/source/detail?r=7830
Modified:
/changes/jat/csproto/plugins/platform/Linux/Futex.cpp
/changes/jat/csproto/plugins/platform/Linux/Futex.h
/changes/jat/csp
I found com.google.gwt.uibinder.rebind.PrintWriterManager very handy
for UiBinderGenerator, where one template leads to the creation of a
whole family of source files. And I'm about to find it handy again as
I write RequestFactoryGenerator in bikeshed.
Shall I move it to com.google.gwt.user.rebind
Oh, I get your point now. Sorry.
On Wed, Mar 31, 2010 at 12:42 PM, Ray Ryan wrote:
> I don't follow your assertion, scott. I'm pretty sure that null is the
> only legal value for a Void field, so what's the big deal about
> serializing it?
>
> On Wed, Mar 31, 2010 at 9:40 AM, Scott Blum wrote:
On Wed, Mar 31, 2010 at 12:42 PM, Ray Ryan wrote:
> I don't follow your assertion, scott. I'm pretty sure that null is the
> only legal value for a Void field, so what's the big deal about
> serializing it?
null is the only value. This supports a command-style use case where
you have an RPC inte
Revision: 7829
Author: j...@google.com
Date: Wed Mar 31 09:44:49 2010
Log: Run more cycles to get more consistent timing in test, spin longer
while
waiting for a lock (need to revisit when it is hooked into the real protocol
where more work is being done between messages), move spin loop into t
I don't follow your assertion, scott. I'm pretty sure that null is the
only legal value for a Void field, so what's the big deal about
serializing it?
On Wed, Mar 31, 2010 at 9:40 AM, Scott Blum wrote:
> Could you please explain this? I would assert that GenericType should
> be serializable if a
Could you please explain this? I would assert that GenericType should
be serializable if and only if it contains no fields of type T, where T is
substituted for void. Do you disagree with this assertion, or is there a
shortcoming in the current implementation that makes the assertion fail to
be a
Revision: 7828
Author: j...@google.com
Date: Wed Mar 31 09:29:51 2010
Log: Initial implementation of Linux-specific Shm+Futex implementation in
C++.
TODO: register Futexes for kernel cleanup on unclean shutdown.
http://code.google.com/p/google-web-toolkit/source/detail?r=7828
Added:
/change
Reviewers: Dan Rice,
Message:
Review requested.
Description:
GWT RPC doesn't conside GenericType to be serializable because it
doesn't think that Void is serializable.
Please review this at http://gwt-code-reviews.appspot.com/294801/show
Affected files:
A
user/src/com/google/gwt/user/clien
Reviewers: ruce_google.com,
Description:
Fixes a list subset error in AbstractListModel that was breaking paging.
Changes ListListModel's deferred command to be more careful about
clearing
its pending state (so that an exception won't break it permanently).
Please review this at http://gwt-code
On 2010/03/30 21:40:48, Lex wrote:
LGTM. I didn't realize Charset was 1.6-only.
It has Charset but not String.getBytes(Charset).
http://gwt-code-reviews.appspot.com/281801/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
To unsubscribe, reply using "remove me" as the
Exactly. Thanks.
On Tue, Mar 30, 2010 at 10:51 PM, wrote:
> LGTM
>
> Just to make sure I understand, you added the following:
> 14=\#
> 15=
>
> which specifies the import order for all other packages not mentioned in
> 0-13? So "foo.bar.baz" now appears at the end of the imports.
>
>
> http://
On 31 Mar 2010, at 18:59, Joel Webber wrote:
John,
Sorry for the slow response -- I've had a browser window open for
several days to respond, but I keep getting pulled off onto other
things.
Ha ha! I know exactly how it is.
It sounds like this is indeed an ordering bug. It's hard to be
On Wed, Mar 24, 2010 at 1:01 PM, dflorey wrote:
> I've been missing that one as well.
> I can provide patches if desired.
>
> --
On Wed, Mar 24, 2010 at 11:40 AM, dflorey wrote:
>
> I just needed a TabLayoutPanel with Tabs on the bottom instead of tabs
> on the top. I copied the TabLayoutPana
John,
Sorry for the slow response -- I've had a browser window open for several
days to respond, but I keep getting pulled off onto other things.
It sounds like this is indeed an ordering bug. It's hard to be sure
precisely when IE's going to call onresize() (onmove() and onresize() are
obscure n
I was hoping that because most people (not using the Maps API) wouldn't need
this, it wouldn't be necessary to provide a special case for "layout
complete" outside of the existing animation callbacks. Does it not work to
put the Map in a RequiresResize container and call its layout method when
onRe
Comment by j...@google.com:
I'm not aware of any way on most browsers to force a GC. On IE, there's the
global CollectGarbage() function, but IE already runs GC too frequently
most of the time anyway, so it's not much help. I'm less familiar with
Firefox and Safari, but Chrome's GC is gener
Comment by atul.dambalkar:
We also have quite a big GWT based application. In our application, we have
also seen similar memory issues showing up in FireFox3.6, IE and Chrome4.1
as well. Browsers eventually stopped responding (with CPU more than 70%)
after memory reaching up to 1GB (happene
Does anyone think this is a bug? Or is it expected behaviour?
On Mar 23, 10:40 pm, jd wrote:
> Something else related to this problem: when a DockLayoutPanel is
> animated to resize its children, onResize() is called *before* the
> resize but my child component was expecting to examine its new
34 matches
Mail list logo