[gwt-contrib] Move more ie6 gis into resource space in Mail sample
Reviewers: bobv, Message: Code review, please Description: Move more IE6 urls out of war and into resource space Review by bobv Manually tested on Safari, Ie6, Ie7, Mac Chrome Please review this at http://gwt-code-reviews.appspot.com/126808 Affected files: M samples/mail/src/com/google/gwt/sample/mail/client/Shortcuts.ui.xml A samples/mail/src/com/google/gwt/sample/mail/client/contactsgroup_ie6.gif A samples/mail/src/com/google/gwt/sample/mail/client/mailboxesgroup_ie6.gif A samples/mail/src/com/google/gwt/sample/mail/client/tasksgroup_ie6.gif M samples/mail/war/contactsgroup_ie6.gif M samples/mail/war/mailboxesgroup_ie6.gif M samples/mail/war/tasksgroup_ie6.gif -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] [google-web-toolkit] r7325 committed - Fix Mail sample to use a data url for IE6, baking in the image...
Revision: 7325 Author: rj...@google.com Date: Wed Dec 16 21:09:31 2009 Log: Fix Mail sample to use a data url for IE6, baking in the image instead of calling back to the server for it. Desk review by bobv Manually tested on Chrome, Safari, IE6 http://code.google.com/p/google-web-toolkit/source/detail?r=7325 Added: /trunk/samples/mail/src/com/google/gwt/sample/mail/client/logo_ie6.gif Deleted: /trunk/samples/mail/war/logo_ie6.gif Modified: /trunk/samples/mail/src/com/google/gwt/sample/mail/client/TopPanel.ui.xml === --- /dev/null +++ /trunk/samples/mail/src/com/google/gwt/sample/mail/client/logo_ie6.gif Wed Dec 16 21:09:31 2009 Binary file, no diff available. === --- /trunk/samples/mail/war/logo_ie6.gifTue Nov 17 14:23:51 2009 +++ /dev/null Binary file, no diff available. === --- /trunk/samples/mail/src/com/google/gwt/sample/mail/client/TopPanel.ui.xml Tue Nov 17 19:34:50 2009 +++ /trunk/samples/mail/src/com/google/gwt/sample/mail/client/TopPanel.ui.xml Wed Dec 16 21:09:31 2009 @@ -8,7 +8,11 @@ + + + + .statusDiv { text-align: right; margin: 1em; @@ -19,8 +23,9 @@ } @if user.agent ie6 { +@url logoIe6 logoIe6Data; .logo { - background-image: url(logo_ie6.gif); + background-image: logoIe6; width: 140px; height: 75px; position: absolute; -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] [google-web-toolkit] r7324 committed - Edited wiki page through web user interface.
Revision: 7324 Author: b...@google.com Date: Wed Dec 16 20:42:07 2009 Log: Edited wiki page through web user interface. http://code.google.com/p/google-web-toolkit/source/detail?r=7324 Modified: /wiki/ClientBundle.wiki === --- /wiki/ClientBundle.wiki Mon Apr 6 08:51:55 2009 +++ /wiki/ClientBundle.wiki Wed Dec 16 20:42:07 2009 @@ -1,5 +1,7 @@ #summary Compile-time resource optimizations +* Documentation for the [http://code.google.com/webtoolkit/doc/latest/DevGuideClientBundle.html GWT 2.0 release of ClientBundle] can be found on the GWT Developer's Guide website. * + = Introduction = -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] [google-web-toolkit] r7323 committed - Edited wiki page through web user interface.
Revision: 7323 Author: b...@google.com Date: Wed Dec 16 20:41:16 2009 Log: Edited wiki page through web user interface. http://code.google.com/p/google-web-toolkit/source/detail?r=7323 Modified: /wiki/CssResource.wiki === --- /wiki/CssResource.wiki Wed Nov 18 06:09:49 2009 +++ /wiki/CssResource.wiki Wed Dec 16 20:41:16 2009 @@ -1,6 +1,8 @@ #summary Compile-time CSS processing #labels Phase-Implementation +* Documentation for the [http://code.google.com/webtoolkit/doc/latest/DevGuideClientBundle.html GWT 2.0 release of CssResource] can be found on the GWT Developer's Guide website. * + See also the CssResourceCookbook and StyleInjector. -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] now.. afetr GWT 2.0?
I'd like to see SuggestBox get a little love. For example: http://code.google.com/p/google-web-toolkit/issues/detail?id=2311 http://code.google.com/p/google-web-toolkit/issues/detail?id=2739 http://code.google.com/p/google-web-toolkit/issues/detail?id=3409 And one of those issues links to this thread where I had some other thoughts on SuggestBox improvements: http://groups.google.com/group/Google-Web-Toolkit-Contributors/browse_thread/thread/83db777f0e0602d0 And if I could include a wish for future GPE development it would be wizards, automated refactorings, and reusable code templates. Things like: http://code.google.com/p/google-web-toolkit/issues/detail?id=3914 I already have at least one SuggestBox patch floating around out there. I'd be happy to update that and work on other SuggestBox features, too, in my "spare" time. If somebody with commit privs cares to buddy up for design and review, please feel free to contact me on or off list. :) Oh, and woohoo GWT 2.0! Go team! - Isaac On Wed, Dec 16, 2009 at 12:01 PM, Bruce Johnson wrote: > Working on a draft one. > What do folks here think is important? > > On Wed, Dec 16, 2009 at 7:42 AM, tfreitas wrote: >> >> What about roadmap? >> >> -- >> http://groups.google.com/group/Google-Web-Toolkit-Contributors > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] [google-web-toolkit] r7322 committed - Add class and method resymbolization to JUnitShell to improve producti...
Revision: 7322 Author: b...@google.com Date: Wed Dec 16 17:16:05 2009 Log: Add class and method resymbolization to JUnitShell to improve production-mode stack traces for continuous builds. Turns on emulated stack traces across all browsers to ensure consistent behavior when running GWTTestCases. Patch by: bobv Review by: jat, scottb, rjrjr http://code.google.com/p/google-web-toolkit/source/detail?r=7322 Added: /trunk/user/src/com/google/gwt/junit/linker /trunk/user/src/com/google/gwt/junit/linker/JUnitSymbolMapsLinker.java Deleted: /trunk/user/test/com/google/gwt/core/client/impl/EmulatedStackTraceTest.java Modified: /trunk/dev/core/src/com/google/gwt/core/linker/IFrameLinker.java /trunk/user/src/com/google/gwt/junit/JUnit.gwt.xml /trunk/user/src/com/google/gwt/junit/server/JUnitHostImpl.java /trunk/user/test/com/google/gwt/core/CoreSuite.java /trunk/user/test/com/google/gwt/core/client/impl/StackTraceCreatorTest.java === --- /dev/null +++ /trunk/user/src/com/google/gwt/junit/linker/JUnitSymbolMapsLinker.java Wed Dec 16 17:16:05 2009 @@ -0,0 +1,46 @@ +/* + * Copyright 2009 Google Inc. + * + * Licensed under the Apache License, Version 2.0 (the "License"); you may not + * use this file except in compliance with the License. You may obtain a copy of + * the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT + * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the + * License for the specific language governing permissions and limitations under + * the License. + */ +package com.google.gwt.junit.linker; + +import com.google.gwt.core.ext.TreeLogger; +import com.google.gwt.core.ext.UnableToCompleteException; +import com.google.gwt.core.ext.linker.ArtifactSet; +import com.google.gwt.core.ext.linker.CompilationResult; +import com.google.gwt.core.ext.linker.EmittedArtifact; +import com.google.gwt.core.linker.SymbolMapsLinker; + +import java.io.ByteArrayOutputStream; + +/** + * Emits the symbol maps into the application output directory so that the + * JUnitHostImpl servlet can read them. + */ +public class JUnitSymbolMapsLinker extends SymbolMapsLinker { + @Override + protected void doEmitSymbolMap(TreeLogger logger, ArtifactSet artifacts, + CompilationResult result, ByteArrayOutputStream out) + throws UnableToCompleteException { +// Collaborate with JUnitHostImpl.loadSymbolMap +String partialPath = ".junit_symbolMaps/" + result.getStrongName() ++ STRONG_NAME_SUFFIX; + +EmittedArtifact symbolMapArtifact = emitBytes(logger, out.toByteArray(), +partialPath); + +artifacts.add(symbolMapArtifact); + } + +} === --- /trunk/user/test/com/google/gwt/core/client/impl/EmulatedStackTraceTest.java Tue Jul 28 09:27:08 2009 +++ /dev/null @@ -1,29 +0,0 @@ -/* - * Copyright 2009 Google Inc. - * - * Licensed under the Apache License, Version 2.0 (the "License"); you may not - * use this file except in compliance with the License. You may obtain a copy of - * the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT - * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the - * License for the specific language governing permissions and limitations under - * the License. - */ -package com.google.gwt.core.client.impl; - -/** - * Reuses the existing StackTraceCreatorTest, but forces emulated JS stack trace - * code to be emitted.. - */ -public class EmulatedStackTraceTest extends StackTraceCreatorTest { - - @Override - public String getModuleName() { -return "com.google.gwt.core.EmulateJsStack"; - } - -} === --- /trunk/dev/core/src/com/google/gwt/core/linker/IFrameLinker.javaWed Oct 21 13:20:55 2009 +++ /trunk/dev/core/src/com/google/gwt/core/linker/IFrameLinker.javaWed Dec 16 17:16:05 2009 @@ -259,7 +259,7 @@ out.newlineOpt(); // Setup the well-known variables. -out.print("
[gwt-contrib] Re: RR : Improve web-mode stack trace presentation for continuous builds
JUnit fixes LGTM http://gwt-code-reviews.appspot.com/126806 -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: now.. afetr GWT 2.0?
On 16 déc, 21:06, Thomas Broyer wrote: > On 16 déc, 18:01, Bruce Johnson wrote: > > > Working on a draft one. > > > What do folks here think is important? > > - remove listeners (wasn't this targeted to 2.0 ?) > - "port" all widgets to c.g.g.dom.client.Element and deprecate both > c.g.g.user.client.Element and c.g.g.user.client.DOM (eventually > enhancing c.g.g.d.c.Element with, e.g. insertChild(Element,int)) > - further optimize code, both what the compiler produces (I'm told > this is being worked on while we talk) and the client code (widgets > and JRE) Hmm, I'd add (with high priority) further rework and improvements of events: - to make it easier/possible to add support for other DOM events in third-party libs (see issue 2562, focusIn/focusOut would be cool too, see issue 1431), by moving the sinking into the DomEvent itself (at least decoupling it from the bit-mask used in Widget::sinkEvents) - to finally have a *good* handling of keyboard/text input events (already improved in 1.6, many nits left, see suggestions given on the issue tracker) - to eventually allow adding handlers in "capture" mode (you know, the last parameter in W3C DOM's addEventListener) -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Improve NumberFormat
Reviewers: jat, Description: This passes all tests but could probably be made a bit clearer. Please review this at http://gwt-code-reviews.appspot.com/126807 Affected files: user/src/com/google/gwt/i18n/client/NumberFormat.java user/test/com/google/gwt/i18n/client/NumberFormat_en_Test.java -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] Re: Problems with UiBinder Internals
Jarrod, Try toggling display: none/block on the element after the param values are setup and see what happens. On the Flash plugin, this causes reinitialization in IE. -Ray On Wed, Dec 16, 2009 at 2:33 PM, Ray Cromwell wrote: > I'm not sure about this particular issue, but this is a well known > problem when injecting Flash into IE DOM. IE will not only > restart/reinitialize ActiveX plugins when you move nodes around the > DOM, it will also do it when you set display: none|block, which causes > havoc if you put a widget into something like a stack panel or tab > panel. The only solution I've found is to have the plugin serialize > state to GWT, and then restore it whenever it boots up. > > See the attach() function here: > http://code.google.com/p/gwt-chronoscope/source/browse/trunk/chronoscope/src/main/java/org/timepedia/chronoscope/client/browser/flashcanvas/FlashCanvas.java > > When Flash starts up, it calls into my GWT app. If it is the first > time, then the app continues initialization. If it is the second time, > it means IE has restarted the SWF, so resyncLayers() is called, which > causes my app to push all of the state back to the plugin. > > I tried ways of getting around this, like creating a proxy widget is > attached and controls the Flash widget which lives as an absolutely > positioned element of , and then trying to look for the proxy > being hidden and hiding the Flash layer by moving it offscreen, but it > is laggy and you still get into situations where IE can spontaneously > restart your SWF sometimes. > > -Ray > > On Wed, Dec 16, 2009 at 4:27 AM, djd wrote: >> Jarrod, you could try to code 2 versions of that class, say >> UploaderWidget and UploaderWidgetIE, and dinamically switch from one >> to another with (almost) the same rules >> >> On Dec 15, 8:13 pm, jarrod wrote: >>> Actually, I was just struggling with this a bit... I have my two >>> ui.xml templates setup, but I can't seem to get GWT.create() to select >>> the right one. Here's what I tried: >>> >>> @UiTemplate("JavaApplet.ui.xml") >>> interface JavaApplet extends UiBinder { >>> >>> } >>> >>> @UiTemplate("JavaAppletIE.ui.xml") >>> interface JavaAppletIE extends JavaApplet { >>> >>> } >>> >>> JavaApplet binder = GWT.create(JavaApplet.class); >>> >>> The following rules were defined in my module: >>> >> class="com.playonsports.web.uploader.client.JavaAppletIE"> >>> >> class="com.playonsports.web.uploader.client.JavaApplet" /> >>> >>> >>> >>> >>> >>> >>> But unfortunately, I just get compiler errors: >>> 12/15/09 12:45:14 PM EST: [INFO] Compiling module >>> com.playonsports.web.uploader.Uploader >>> 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Errors in 'file:/C:/Users/ >>> jcarlson/Projects/com.playonsports.web.uploader/src/main/java/com/ >>> playonsports/web/uploader/client/UploaderWidget.java' >>> 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Line 16: Rebind result >>> 'com.playonsports.web.uploader.client.JavaAppletIE' must be a class >>> 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Cannot proceed due to >>> previous errors >>> >>> I guess that makes sense, but how does one get around this? >>> >>> On Dec 15, 11:44 am, John Tamplin wrote: >>> >>> >>> >>> > On Tue, Dec 15, 2009 at 11:29 AM, jarrod wrote: >>> > > Combined with the ability to use different ui.xml templates documented >>> > > here: >>> >>> > >http://code.google.com/webtoolkit/doc/latest/DevGuideUiBinder.html#Ap... >>> > > I should be able to declare one ui.xml for IE and another for >>> > > everything else. I'll try that next and that would really simplify >>> > > things. >>> >>> > Speaking of that example, shouldn't the GWT.create calls refer to >>> > RedBinder/BlueBinder? >>> >>> > -- >>> > John A. Tamplin >>> > Software Engineer (GWT), Google >> >> -- >> http://groups.google.com/group/Google-Web-Toolkit-Contributors > -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] Re: Problems with UiBinder Internals
I'm not sure about this particular issue, but this is a well known problem when injecting Flash into IE DOM. IE will not only restart/reinitialize ActiveX plugins when you move nodes around the DOM, it will also do it when you set display: none|block, which causes havoc if you put a widget into something like a stack panel or tab panel. The only solution I've found is to have the plugin serialize state to GWT, and then restore it whenever it boots up. See the attach() function here: http://code.google.com/p/gwt-chronoscope/source/browse/trunk/chronoscope/src/main/java/org/timepedia/chronoscope/client/browser/flashcanvas/FlashCanvas.java When Flash starts up, it calls into my GWT app. If it is the first time, then the app continues initialization. If it is the second time, it means IE has restarted the SWF, so resyncLayers() is called, which causes my app to push all of the state back to the plugin. I tried ways of getting around this, like creating a proxy widget is attached and controls the Flash widget which lives as an absolutely positioned element of , and then trying to look for the proxy being hidden and hiding the Flash layer by moving it offscreen, but it is laggy and you still get into situations where IE can spontaneously restart your SWF sometimes. -Ray On Wed, Dec 16, 2009 at 4:27 AM, djd wrote: > Jarrod, you could try to code 2 versions of that class, say > UploaderWidget and UploaderWidgetIE, and dinamically switch from one > to another with (almost) the same rules > > On Dec 15, 8:13 pm, jarrod wrote: >> Actually, I was just struggling with this a bit... I have my two >> ui.xml templates setup, but I can't seem to get GWT.create() to select >> the right one. Here's what I tried: >> >> @UiTemplate("JavaApplet.ui.xml") >> interface JavaApplet extends UiBinder { >> >> } >> >> @UiTemplate("JavaAppletIE.ui.xml") >> interface JavaAppletIE extends JavaApplet { >> >> } >> >> JavaApplet binder = GWT.create(JavaApplet.class); >> >> The following rules were defined in my module: >> > class="com.playonsports.web.uploader.client.JavaAppletIE"> >> > class="com.playonsports.web.uploader.client.JavaApplet" /> >> >> >> >> >> >> >> But unfortunately, I just get compiler errors: >> 12/15/09 12:45:14 PM EST: [INFO] Compiling module >> com.playonsports.web.uploader.Uploader >> 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Errors in 'file:/C:/Users/ >> jcarlson/Projects/com.playonsports.web.uploader/src/main/java/com/ >> playonsports/web/uploader/client/UploaderWidget.java' >> 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Line 16: Rebind result >> 'com.playonsports.web.uploader.client.JavaAppletIE' must be a class >> 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Cannot proceed due to >> previous errors >> >> I guess that makes sense, but how does one get around this? >> >> On Dec 15, 11:44 am, John Tamplin wrote: >> >> >> >> > On Tue, Dec 15, 2009 at 11:29 AM, jarrod wrote: >> > > Combined with the ability to use different ui.xml templates documented >> > > here: >> >> > >http://code.google.com/webtoolkit/doc/latest/DevGuideUiBinder.html#Ap... >> > > I should be able to declare one ui.xml for IE and another for >> > > everything else. I'll try that next and that would really simplify >> > > things. >> >> > Speaking of that example, shouldn't the GWT.create calls refer to >> > RedBinder/BlueBinder? >> >> > -- >> > John A. Tamplin >> > Software Engineer (GWT), Google > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: now.. afetr GWT 2.0?
1) Improve change-compile-refresh development experience. This, IMO, is GWT's one and only pain point and its a big one. DevMode refresh, while much improved, still is like watching paint dry compared to other web development platforms. Something like an incremental, non- optimized web mode compile, synced with eclipse compiles, would be friggin amazing. Distant 2nd) Formalize data binding Brad -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: RR : Improve web-mode stack trace presentation for continuous builds
+1000, one potential problem tho http://gwt-code-reviews.appspot.com/126806/diff/1002/1005 File user/src/com/google/gwt/junit/server/JUnitHostImpl.java (right): http://gwt-code-reviews.appspot.com/126806/diff/1002/1005#newcode293 Line 293: private Map loadSymbolMap(StrongName strongName) { This method should synchronize, right? I think the rest of the code is fine unsynchronized. http://gwt-code-reviews.appspot.com/126806 -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] now.. afetr GWT 2.0?
> 3) DataBinding & Validation (btw see this post from Ray announcing something > for Q1 2009 ;-) For me, it's the main feature GWT should have. Next would be "pretty" widgets, Drag'n Drop (in the framework, GWT DnD is very good, but DnD deserve to be in the framework by itself). And on top of it, yes, the WYSIWYG Eclipse plugin (with databing and validation, it could make people more easily start with GWT!) Thank you for the usefull features of 2.0 (just devMode in Firefox is worth it)! -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: now.. afetr GWT 2.0?
On 16 déc, 18:01, Bruce Johnson wrote: > Working on a draft one. > > What do folks here think is important? - remove listeners (wasn't this targeted to 2.0 ?) - "port" all widgets to c.g.g.dom.client.Element and deprecate both c.g.g.user.client.Element and c.g.g.user.client.DOM (eventually enhancing c.g.g.d.c.Element with, e.g. insertChild(Element,int)) - further optimize code, both what the compiler produces (I'm told this is being worked on while we talk) and the client code (widgets and JRE) -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] now.. afetr GWT 2.0?
GWT 2.0 was so awesome, it'll be hard to top any of the new stuff with my feature wishlist. A few things I'd like: - moving as many compiler properties as possible into configuration properties so we can build an instrumented release (with type cast checking, assertions, emulated stack traces) at the same time as release that can be turned on via - A DOM object to represent the window - Less of a hit on first load in development mode - New linker that uses iframes with dynamic scripts and a more generic, more easily reusable hosted mode script Matt. On 2009-12-16, at 10:01 AM, Bruce Johnson wrote: > Working on a draft one. > > What do folks here think is important? > > On Wed, Dec 16, 2009 at 7:42 AM, tfreitas wrote: > What about roadmap? > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors > > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] now.. afetr GWT 2.0?
Hey Bruce, GWT 2 has provided great features focusing on the hidden parts of the iceberg (devmode, codesplitting, clientbundle, ...). It would be great to focus 2.1 on the user experience and the visible part of the user framework. Here's my wish list : 0) More widgets/containers compliant with the new CSS Layout API to get rid of the RootPanel.get() 1) Enrich the existing widgets and provide a good user experience (not the 3 predefined awful themes ;-)). Copy/Paste Wave/adWords apps themes is a good starting point :-D. With a multiple windows management framework (docking, maximize, etc ...) 2) A real DataGrid (sorting, paging, filtering, async...). Maybe by Improving the design (and the look) of the current incubator ScrollTable 3) DataBinding & Validation (btw see this post from Ray announcing something for Q1 2009 ;-) -> http://groups.google.com/group/Google-Web-Toolkit-Contributors/browse_thread/thread/8c611ab8bb076ead?pli=1 4) Replace RemoteService and RemoteServiceServlet by deRPC "for real" 5) Provide a UiBinder WYSIWYG Eclipse plugin (this would be the killer app) 6) provide a framework for UiBinder custom parsers long term - 7) Rewrite entirely the compiler ;-) instant dev mode (to benefit from incremental compilation) and generate the permutations from .class, not source code (to facilitate other language integration) I didn't congratulate all the GWT team for this fantastic 2.0 release. Let me thank you once more (some of you deserve a vacation ;-)). Sami On Wed, Dec 16, 2009 at 6:01 PM, Bruce Johnson wrote: > Working on a draft one. > > What do folks here think is important? > > > On Wed, Dec 16, 2009 at 7:42 AM, tfreitas wrote: > >> What about roadmap? >> >> -- >> http://groups.google.com/group/Google-Web-Toolkit-Contributors >> > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors > -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] now.. afetr GWT 2.0?
Working on a draft one. What do folks here think is important? On Wed, Dec 16, 2009 at 7:42 AM, tfreitas wrote: > What about roadmap? > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors > -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: RR : Improve web-mode stack trace presentation for continuous builds
Much better, thanks. LGTM http://gwt-code-reviews.appspot.com/126806 -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
Re: [gwt-contrib] Using cellIndex and rowIndex/sectionRowIndex to get cells' position in tables
Thomas, Your assessment that this is an accidental legacy is basically correct. This code should definitely be optimized -- if you could file an issue, that would be helpful. Hopefully it makes little difference in practice, because of the infrequent nature of click events, but it would still be good to get it fixed. I'm also sure that it will ultimately be more efficient to build up HTML strings than to build a table programatically (no matter how silly that may be). I'm starting to do a lot of experiments on DOM and HTML parsing speed that tentatively support this view. We will ultimately need to find a way to support this mechanism in the core GWT widgets, which we'll be looking into this year. Cheers, joel. On Tue, Dec 15, 2009 at 6:45 AM, Thomas Broyer wrote: > Hi all, > > [long intro, sorry, you can skip to "interesting bits start here" if > you want] > > SpeedTracer tells me that in rendering a list of results > (JsArray returned by a RequestBuilder as JSON, displayed in > a Grid widget; no other widgets involved), Chrome is spending 57.2% of > its time, i.e. 437ms out of 764ms in DOM Events, most (if not all) of > them being mutation events which I don't care about. > > We're targetting IE6 and although I haven't checked with dynaTrace or > similar tools, I'm confident that the results would be approx. the > same (in proportion, IE6 being of course way slower, > unfortunately). > > So, given that I'm only interested in populating the grid once and > then listening the click events (clicking on a row goes from list to > details, much like Google Code issues list), I'm investigating into > building the table as an HTML string and then using innerHTML to parse > it (similar to how BulkTableLoader works in the GWT-Incubator). > And because I'm not interested in the Grid APIs to manipulate the > grid, I was thinking about creating a widget from scratch and only > adding the click-handling logic. Which lead me to look at how it's > done in HTMLTable (most notably getCellForEvent/getEventTargetCell). > > [interesting bits start here] > > ...and I then wondered why calculating the cell and row indices of the > clicked cell was done by walking the tbody's children (not even the > rows[] collection) and the tr's children (not even the cells[] > collection), instead of just calling the cell's cellIndex and row's > rowIndex or sectionRowIndex, which according to PPK are supported in > all GWT-supported browsers (rowIndex is incorrect in Opera but only in > the presence of a which HTMLTable does not have): > http://www.quirksmode.org/dom/w3c_html.html#t31 > http://www.quirksmode.org/dom/w3c_html.html#t317 > http://www.quirksmode.org/dom/w3c_html.html#t319 > > Is there any reason not to do it? or is it just because > DOM.getChildIndex() existed so it was used as it "just worked"? > > > (for those who read the intro, I think I might improve perfs without > this bulkloading technique, for instance building the table off the > DOM and attach it later, whereas I'm just hiding/showing it using > setVisible(); but I believe the bulk load is the way to go for lists > like mine, and I don't like the API of the tables in the GWT- > Incubator, I want something simpler, even if its more "close to metal") > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors > -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] now.. afetr GWT 2.0?
What about roadmap? -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: Problems with UiBinder Internals
Jarrod, you could try to code 2 versions of that class, say UploaderWidget and UploaderWidgetIE, and dinamically switch from one to another with (almost) the same rules On Dec 15, 8:13 pm, jarrod wrote: > Actually, I was just struggling with this a bit... I have my two > ui.xml templates setup, but I can't seem to get GWT.create() to select > the right one. Here's what I tried: > > @UiTemplate("JavaApplet.ui.xml") > interface JavaApplet extends UiBinder { > > } > > @UiTemplate("JavaAppletIE.ui.xml") > interface JavaAppletIE extends JavaApplet { > > } > > JavaApplet binder = GWT.create(JavaApplet.class); > > The following rules were defined in my module: > class="com.playonsports.web.uploader.client.JavaAppletIE"> > class="com.playonsports.web.uploader.client.JavaApplet" /> > > > > > > > But unfortunately, I just get compiler errors: > 12/15/09 12:45:14 PM EST: [INFO] Compiling module > com.playonsports.web.uploader.Uploader > 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Errors in 'file:/C:/Users/ > jcarlson/Projects/com.playonsports.web.uploader/src/main/java/com/ > playonsports/web/uploader/client/UploaderWidget.java' > 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Line 16: Rebind result > 'com.playonsports.web.uploader.client.JavaAppletIE' must be a class > 12/15/09 12:45:21 PM EST: [INFO] [ERROR] Cannot proceed due to > previous errors > > I guess that makes sense, but how does one get around this? > > On Dec 15, 11:44 am, John Tamplin wrote: > > > > > On Tue, Dec 15, 2009 at 11:29 AM, jarrod wrote: > > > Combined with the ability to use different ui.xml templates documented > > > here: > > > >http://code.google.com/webtoolkit/doc/latest/DevGuideUiBinder.html#Ap... > > > I should be able to declare one ui.xml for IE and another for > > > everything else. I'll try that next and that would really simplify > > > things. > > > Speaking of that example, shouldn't the GWT.create calls refer to > > RedBinder/BlueBinder? > > > -- > > John A. Tamplin > > Software Engineer (GWT), Google -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: RR : Improve web-mode stack trace presentation for continuous builds
> This is pretty ugly -- is there some other way to handle this? I have updated the patch to have JUnit.gwt.xml override the symbolMaps Linker definition. http://gwt-code-reviews.appspot.com/126806 -- http://groups.google.com/group/Google-Web-Toolkit-Contributors