Last 7 days

• Dec 26, 2009

  issue 4413 (Developer plugin issues on OSX 10.4) changed by jat@google.com   -  

  Status: Duplicate

  Status: Duplicate
• Dec 26, 2009

  issue 4381 (GWT 2 Dev mode does not work in OS X 10.4, PowerPC (don't kn...) changed by jat@google.com   -   Not just PPC

  Status: Accepted

  Owner: j...@google.com

  Cc: b...@google.com

  Labels: Category-DevPlugin Type-Defect

  Not just PPC

  Status: Accepted

  Owner: j...@google.com

  Cc: b...@google.com

  Labels: Category-DevPlugin Type-Defect
• Dec 26, 2009

  issue 4413 (Developer plugin issues on OSX 10.4) commented on by jdoug...@basis.com   -   http://code.google.com/p/google-web-toolkit/issues/detail?id=4381

  http://code.google.com/p/google-web-toolkit/issues/detail?id=4381
• Dec 26, 2009

  issue 4228 ([patch] I18N: use browsers' locale at startup) commented on by jat@google.com   -   Basically, I think it needs to be in such a way that apps which want the default behavior should be able to get it without copy/pasting the locale property provider. Since the patch as proposed would basically never leave the locale as the fallback, I think there needs to be opt-in to get this behavior. Probably the easiest way to do that is to allow a special token for the locale fallback value which invokes this behavior if locale hasn't been set in some other way.

  Basically, I think it needs to be in such a way that apps which want the default behavior should be able to get it without copy/pasting the locale property provider. Since the patch as proposed would basically never leave the locale as the fallback, I think there needs to be opt-in to get this behavior. Probably the easiest way to do that is to allow a special token for the locale fallback value which invokes this behavior if locale hasn't been set in some other way.
• Dec 26, 2009

  issue 4413 (Developer plugin issues on OSX 10.4) reported by jat@google.com   -   Found in GWT Release (e.g. 1.5.3, 1.6 RC): 2.0.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): MacOSX 10.4, Safari/FF Detailed description (please be as specific as possible): The developer plugins built on 10.6 do not work on 10.4. Workaround if you have one: Upgrade to 10.5 or later. Links to relevant GWT Developer Forum posts: https://groups.google.com/group/google-web- toolkit/browse_thread/thread/5e4f82a9069cf965?hl=en We need to either find a way to compile the plugins that also works on 10.4, or we need to build separate versions and provide a way to load the right one. Given how the FF plugin is setup, it wouldn't be too hard to load an alternate .dylib on 10.4, but I am not sure how that would work on Safari.

  Found in GWT Release (e.g. 1.5.3, 1.6 RC): 2.0.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): MacOSX 10.4, Safari/FF Detailed description (please be as specific as possible): The developer plugins built on 10.6 do not work on 10.4. Workaround if you have one: Upgrade to 10.5 or later. Links to relevant GWT Developer Forum posts: https://groups.google.com/group/google-web- toolkit/browse_thread/thread/5e4f82a9069cf965?hl=en We need to either find a way to compile the plugins that also works on 10.4, or we need to build separate versions and provide a way to load the right one. Given how the FF plugin is setup, it wouldn't be too hard to load an alternate .dylib on 10.4, but I am not sure how that would work on Safari.
• Dec 26, 2009

  r7349 (Revert r7347, which caused problems in DevMode. svn merge -...) committed by jat@google.com   -   Revert r7347, which caused problems in DevMode. svn merge -c-7347 https://google-web-toolkit.googlecode.com/svn/trunk . Patch by: jat Review by: rice (email)

  Revert r7347, which caused problems in DevMode. svn merge -c-7347 https://google-web-toolkit.googlecode.com/svn/trunk . Patch by: jat Review by: rice (email)
• Dec 26, 2009

  issue 4408 (UBinder with <ui:with> not styling with clientBundle and css...) commented on by TPetricek   -   I came across the same problem: I am referencing a ClientBundle with ui:with element in a UiBuilder template. There is a CssResource interface defined in the ClientBundle linked to an external stylesheet by the @Source annotation (relative path, the file is in the same directory). There are no warnings or errors on compile but these styles are not "in effect". When investigated in Firebug I can see that a separate class name ("G1fktep7B") was generated for the CSS rule from the external stylesheet (or from the CssResource interface) and this class name is assigned to the element as specified in the UiBinder template. But that rule does not make it to the client - no such CSS rule is sent to the client. There are two other rules but not that one from external stylesheets: .G1fktep7C{font-weight:bold;}.G1fktep7A{font-weight:bold;}

  I came across the same problem: I am referencing a ClientBundle with ui:with element in a UiBuilder template. There is a CssResource interface defined in the ClientBundle linked to an external stylesheet by the @Source annotation (relative path, the file is in the same directory). There are no warnings or errors on compile but these styles are not "in effect". When investigated in Firebug I can see that a separate class name ("G1fktep7B") was generated for the CSS rule from the external stylesheet (or from the CssResource interface) and this class name is assigned to the element as specified in the UiBinder template. But that rule does not make it to the client - no such CSS rule is sent to the client. There are two other rules but not that one from external stylesheets: .G1fktep7C{font-weight:bold;}.G1fktep7A{font-weight:bold;}
• Dec 26, 2009

  issue 4236 (Provide support for NavigableMap/NavigableSet implementation) commented on by mmaterzok   -   I consider this issue important.

  I consider this issue important.
• Dec 26, 2009

  issue 4412 (Code Split and impossible Superclass dependencies preventing...) reported by Tristan....@gmail.com   -   Found in GWT Release (e.g. 1.5.3, 1.6 RC): 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): all Detailed description (please be as specific as possible): It appears that the compiler is not capable of properly determining dependencies when a superclass calls a subclass method. It would be beneficial for code reuse if the compiler, when determining dependencies, would keep track which subclass called the superclass method, so that when the superclass method calls an abstract method (assumed to be implemented by the subclass), it only depends on the currently calling subclass, and not all subclasses in existence. See code below. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): public class Superclass { public abstract void doSpecific(); public void doGeneric(){ doSpecific(); } } public class SubclassA extends Superclass{ @Override public void doSpecific(){ // do something for A } } public class SubclassB extends Superclass{ @Override public void doSpecific(){ // do something for B } } public class GenericAsyncSplit { HashMap providers; // providers that provide SubclassA and SubclassB public void onDoStuff(final object AorBselector){ GWT.runAsync(new RunAsyncCallback(){ ... onSuccess(){ Superclass myObject = providers.get(AorBselector).get(); myObject.doGeneric(); } } } } When compiled, the SOYC indicates the following dependencies: SubclassA.doGeneric() -> Superclass.doGeneric() -> SubclassB.doSpecific() SubclassB.doGeneric() -> Superclass.doGeneric() -> SubclassA.doSpecific() This chain of dependency is clearly impossible. Workaround if you have one: I'm not familiar with the innards of the compiler, but it would probably require an extra state to maintain track of what subclass called the superclass? Links to relevant GWT Developer Forum posts: (I did a poor job describing the issue here) http://groups.google.com/group/google-web- toolkit/browse_thread/thread/fd2b2a27a4481aa3/0d3124a35c347a92#0d3124a35c34 7a92

  Found in GWT Release (e.g. 1.5.3, 1.6 RC): 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): all Detailed description (please be as specific as possible): It appears that the compiler is not capable of properly determining dependencies when a superclass calls a subclass method. It would be beneficial for code reuse if the compiler, when determining dependencies, would keep track which subclass called the superclass method, so that when the superclass method calls an abstract method (assumed to be implemented by the subclass), it only depends on the currently calling subclass, and not all subclasses in existence. See code below. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): public class Superclass { public abstract void doSpecific(); public void doGeneric(){ doSpecific(); } } public class SubclassA extends Superclass{ @Override public void doSpecific(){ // do something for A } } public class SubclassB extends Superclass{ @Override public void doSpecific(){ // do something for B } } public class GenericAsyncSplit { HashMap providers; // providers that provide SubclassA and SubclassB public void onDoStuff(final object AorBselector){ GWT.runAsync(new RunAsyncCallback(){ ... onSuccess(){ Superclass myObject = providers.get(AorBselector).get(); myObject.doGeneric(); } } } } When compiled, the SOYC indicates the following dependencies: SubclassA.doGeneric() -> Superclass.doGeneric() -> SubclassB.doSpecific() SubclassB.doGeneric() -> Superclass.doGeneric() -> SubclassA.doSpecific() This chain of dependency is clearly impossible. Workaround if you have one: I'm not familiar with the innards of the compiler, but it would probably require an extra state to maintain track of what subclass called the superclass? Links to relevant GWT Developer Forum posts: (I did a poor job describing the issue here) http://groups.google.com/group/google-web- toolkit/browse_thread/thread/fd2b2a27a4481aa3/0d3124a35c347a92#0d3124a35c34 7a92
• Dec 26, 2009

  issue 4411 (When overriding the buttons (or for that sake any other widg...) reported by KRdata   -   Found in GWT Release 2.0.0 Encountered on Kubuntu with Firefox 3.5.6 Detailed description (please be as specific as possible): When KDE has been set with a dark theme Firefox adapts to the theme (I find it safe to assume that other browsers do the same in some cases), so eg. buttons have the same background and foreground color properties as the theme. For me that means a light gray text at almost the same color as the one used on buttons in the Standard theme making them close to unuasable. So basically, please set "color: black;" in the standard and chrome thems and i general, whenever the default background is changed to somegthing specifik, the foreground should be changed accordingly. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): public void onModuleLoad() { VerticalPanel contentPanel = new VerticalPanel(); // code omitted contentPanel.add(new Button("Hello world!"); RootPanel.get("stockList").add(contentPanel); Workaround if you have one: in the css-file for your project you can (and should) add: .gwt-Button { color: black; }

  Found in GWT Release 2.0.0 Encountered on Kubuntu with Firefox 3.5.6 Detailed description (please be as specific as possible): When KDE has been set with a dark theme Firefox adapts to the theme (I find it safe to assume that other browsers do the same in some cases), so eg. buttons have the same background and foreground color properties as the theme. For me that means a light gray text at almost the same color as the one used on buttons in the Standard theme making them close to unuasable. So basically, please set "color: black;" in the standard and chrome thems and i general, whenever the default background is changed to somegthing specifik, the foreground should be changed accordingly. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): public void onModuleLoad() { VerticalPanel contentPanel = new VerticalPanel(); // code omitted contentPanel.add(new Button("Hello world!"); RootPanel.get("stockList").add(contentPanel); Workaround if you have one: in the css-file for your project you can (and should) add: .gwt-Button { color: black; }
• Dec 25, 2009

  issue 3742 (Make Goole Waves UI Look/Feel the default for GWT.) commented on by jprichardson   -   I definitely support this. The default GWT themes are ugly. Plus, I feel teased seeing the themes on the GWT screenshots on the GWT docs. http://code.google.com/webtoolkit/doc/latest/DevGuideUiPanels.html Maybe I'm missing this theme somewhere?

  I definitely support this. The default GWT themes are ugly. Plus, I feel teased seeing the themes on the GWT screenshots on the GWT docs. http://code.google.com/webtoolkit/doc/latest/DevGuideUiPanels.html Maybe I'm missing this theme somewhere?
• Dec 25, 2009

  issue 4325 (GWT Developer Plugin for Chrome on Linux) commented on by jat@google.com   -   We can't do it until the necessary support is there, but if you are expecting DevMode in Chrome to be fast it will be longer before it is nearly as fast as the other browsers.

  We can't do it until the necessary support is there, but if you are expecting DevMode in Chrome to be fast it will be longer before it is nearly as fast as the other browsers.
• Dec 25, 2009

  issue 4325 (GWT Developer Plugin for Chrome on Linux) commented on by mayank.r.jain   -   Must have! We need it. The only alternative we have is FF and that is dead-slow and a CPU hog! Eats up 98+% of CPU and takes ages to render the app.

  Must have! We need it. The only alternative we have is FF and that is dead-slow and a CPU hog! Eats up 98+% of CPU and takes ages to render the app.
• Dec 25, 2009

  issue 4410 (devmode without GUI support) commented on by m.bnovc   -   http://paste.pocoo.org/raw/158535/ is selckin's patch I mentioned.

  http://paste.pocoo.org/raw/158535/ is selckin's patch I mentioned.
• Dec 25, 2009

  issue 4410 (devmode without GUI support) reported by m.bnovc   -   Found in GWT Release (e.g. 1.5.3, 1.6 RC): Not available in r7348. Detailed description (please be as specific as possible): I believe that GWT's devmode should be allow to operate without a GUI. I went to try out GWT, but I was unable to use it because the server I put the code on didn't have a GUI (this causes `ant devmode` to exit immediately). I know a lot of others do their work over SSH as well, so I think this would be a helpful enhancement. selckin from freenode/##gwt provided a patch for "-headless" as a parameter, but the HeadlessUI doesn't give any information about what URL to open/etc.

  Found in GWT Release (e.g. 1.5.3, 1.6 RC): Not available in r7348. Detailed description (please be as specific as possible): I believe that GWT's devmode should be allow to operate without a GUI. I went to try out GWT, but I was unable to use it because the server I put the code on didn't have a GUI (this causes `ant devmode` to exit immediately). I know a lot of others do their work over SSH as well, so I think this would be a helpful enhancement. selckin from freenode/##gwt provided a patch for "-headless" as a parameter, but the HeadlessUI doesn't give any information about what URL to open/etc.
• Dec 24, 2009

  issue 4409 ("JavaScriptException: (null): null" in Development Mode) reported by jaroslav.zaruba   -   Found in GWT Release (e.g. 1.5.3, 1.6 RC): 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): Vista64, FF3.5, Development Mode Detailed description (please be as specific as possible): This is an exception I get in Development Mode when trying to validate SmartGWT's DynamicForm having FormItemValueParser on one of its fields: com.google.gwt.core.client.JavaScriptException: (null): null at com.google.gwt.dev.shell.BrowserChannelServer.invokeJavascript (BrowserChannelServer.java:195) at com.google.gwt.dev.shell.ModuleSpaceOOPHM.doInvoke (ModuleSpaceOOPHM.java:120) at com.google.gwt.dev.shell.ModuleSpace.invokeNative(ModuleSpace.java: 507) at com.google.gwt.dev.shell.ModuleSpace.invokeNativeBoolean (ModuleSpace.java:179) at com.google.gwt.dev.shell.JavaScriptHost.invokeNativeBoolean (JavaScriptHost.java:35) at com.smartgwt.client.widgets.form.DynamicForm.validate (DynamicForm.java) at cz.acolyte.caesar.client.Caesar$2.onClick(Caesar.java:72) ...when compiled it works fine.

  Found in GWT Release (e.g. 1.5.3, 1.6 RC): 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): Vista64, FF3.5, Development Mode Detailed description (please be as specific as possible): This is an exception I get in Development Mode when trying to validate SmartGWT's DynamicForm having FormItemValueParser on one of its fields: com.google.gwt.core.client.JavaScriptException: (null): null at com.google.gwt.dev.shell.BrowserChannelServer.invokeJavascript (BrowserChannelServer.java:195) at com.google.gwt.dev.shell.ModuleSpaceOOPHM.doInvoke (ModuleSpaceOOPHM.java:120) at com.google.gwt.dev.shell.ModuleSpace.invokeNative(ModuleSpace.java: 507) at com.google.gwt.dev.shell.ModuleSpace.invokeNativeBoolean (ModuleSpace.java:179) at com.google.gwt.dev.shell.JavaScriptHost.invokeNativeBoolean (JavaScriptHost.java:35) at com.smartgwt.client.widgets.form.DynamicForm.validate (DynamicForm.java) at cz.acolyte.caesar.client.Caesar$2.onClick(Caesar.java:72) ...when compiled it works fine.
• Dec 24, 2009

  issue 4408 (UBinder with <ui:with> not styling with clientBundle and css...) reported by coderina...@gmail.com   -   (If this is your first time submitting an issue report to the GWT Issue Tracker, please see the sample issue report linked below for an example of a well reported issue) Sample issue report: http://code.google.com/p/google-web-toolkit/wiki/IssueReportSample Found in GWT Release 2: Encountered on OS / Browser (firefox 3.5.6, chrome, ie 8): Detailed description (please be as specific as possible): The use case is below...styling lost unless hard code css in uiBinder with Style source...also image sprite style comes out blank in the div, however other styles are there in the widgets/divs (I have set the CSSResource.Style=pretty). No Styling works and get styleless output. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): 1)----myView.ui.xml-------- <!DOCTYPE ui:UiBinder SYSTEM "http://dl.google.com/gwt/DTD/xhtml.ent"> <ui:UiBinder xmlns:ui="urn:ui:com.google.gwt.uibinder" xmlns:g="urn:import:com.google.gwt.user.client.ui"> <ui:with type="co.client.config.myResources" field="myres"></ui:with> <g:VerticalPanel width="100%"> <g:HTMLPanel> Hello, <g:Button styleName="{myres.mystyle.important}" ui:field="button" /> </g:HTMLPanel> <g:FlowPanel styleName="{myres.mystyle.panelback}"> <g:HTML>This is the heading</g:HTML> </g:FlowPanel> <g:FlowPanel styleName="{myres.mystyle.panelback2}" height="150px"> <g:Button styleName="{myres.mystyle.important}" ui:field="button2" / > </g:FlowPanel> </g:VerticalPanel> </ui:UiBinder> 2)-------------myView.java----------------- package co.client.views; import co.client.config.myResources; import com.google.gwt.core.client.GWT; import com.google.gwt.event.dom.client.ClickEvent; import com.google.gwt.resources.client.ClientBundle; import com.google.gwt.uibinder.client.UiBinder; import com.google.gwt.uibinder.client.UiField; import com.google.gwt.uibinder.client.UiHandler; import com.google.gwt.user.client.Window; import com.google.gwt.user.client.ui.Button; import com.google.gwt.user.client.ui.Composite; import com.google.gwt.user.client.ui.Widget; public class PeopleView extends Composite { private static PeopleViewUiBinder uiBinder = GWT .create(PeopleViewUiBinder.class); interface PeopleViewUiBinder extends UiBinder<Widget, PeopleView> { } @UiField Button button; public PeopleView(String firstName) { initWidget(uiBinder.createAndBindUi(this)); button.setText(firstName); } @UiHandler("button") void onClick(ClickEvent e) { Window.alert("Hello!"); } } 3)----MyResources.java-------- package co.client.config; import com.google.gwt.core.client.GWT; import com.google.gwt.resources.client.ClientBundle; import com.google.gwt.resources.client.CssResource; import com.google.gwt.resources.client.DataResource; import com.google.gwt.resources.client.ImageResource; import com.google.gwt.resources.client.TextResource; public interface MyResources extends ClientBundle { public static final BillResources INSTANCE = GWT.create (MyResources.class); @Source("myapp.css") public Style mystyle(); @Source("mylogo.png") public ImageResource logoImage(); public interface Style extends CssResource { String important(); String panelback(); String panelback2(); String mysprite(); } } 4)----myapp.css--------------- .important { font-weight: bold; } .panelback { background-color: #174c5d; color: white; width: 80%; padding: 20px; margin-left: 20px; } .panelback2 { border-color: #174c5d; border-style: solid; color: white; padding-top: 0; width: 80%; margin-left: 20px; } @sprite .mysprite { gwt-image: "logoImage"; background-color: #174c5d; } Workaround if you have one: Links to relevant GWT Developer Forum posts:

  (If this is your first time submitting an issue report to the GWT Issue Tracker, please see the sample issue report linked below for an example of a well reported issue) Sample issue report: http://code.google.com/p/google-web-toolkit/wiki/IssueReportSample Found in GWT Release 2: Encountered on OS / Browser (firefox 3.5.6, chrome, ie 8): Detailed description (please be as specific as possible): The use case is below...styling lost unless hard code css in uiBinder with Style source...also image sprite style comes out blank in the div, however other styles are there in the widgets/divs (I have set the CSSResource.Style=pretty). No Styling works and get styleless output. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): 1)----myView.ui.xml-------- <!DOCTYPE ui:UiBinder SYSTEM "http://dl.google.com/gwt/DTD/xhtml.ent"> <ui:UiBinder xmlns:ui="urn:ui:com.google.gwt.uibinder" xmlns:g="urn:import:com.google.gwt.user.client.ui"> <ui:with type="co.client.config.myResources" field="myres"></ui:with> <g:VerticalPanel width="100%"> <g:HTMLPanel> Hello, <g:Button styleName="{myres.mystyle.important}" ui:field="button" /> </g:HTMLPanel> <g:FlowPanel styleName="{myres.mystyle.panelback}"> <g:HTML>This is the heading</g:HTML> </g:FlowPanel> <g:FlowPanel styleName="{myres.mystyle.panelback2}" height="150px"> <g:Button styleName="{myres.mystyle.important}" ui:field="button2" / > </g:FlowPanel> </g:VerticalPanel> </ui:UiBinder> 2)-------------myView.java----------------- package co.client.views; import co.client.config.myResources; import com.google.gwt.core.client.GWT; import com.google.gwt.event.dom.client.ClickEvent; import com.google.gwt.resources.client.ClientBundle; import com.google.gwt.uibinder.client.UiBinder; import com.google.gwt.uibinder.client.UiField; import com.google.gwt.uibinder.client.UiHandler; import com.google.gwt.user.client.Window; import com.google.gwt.user.client.ui.Button; import com.google.gwt.user.client.ui.Composite; import com.google.gwt.user.client.ui.Widget; public class PeopleView extends Composite { private static PeopleViewUiBinder uiBinder = GWT .create(PeopleViewUiBinder.class); interface PeopleViewUiBinder extends UiBinder<Widget, PeopleView> { } @UiField Button button; public PeopleView(String firstName) { initWidget(uiBinder.createAndBindUi(this)); button.setText(firstName); } @UiHandler("button") void onClick(ClickEvent e) { Window.alert("Hello!"); } } 3)----MyResources.java-------- package co.client.config; import com.google.gwt.core.client.GWT; import com.google.gwt.resources.client.ClientBundle; import com.google.gwt.resources.client.CssResource; import com.google.gwt.resources.client.DataResource; import com.google.gwt.resources.client.ImageResource; import com.google.gwt.resources.client.TextResource; public interface MyResources extends ClientBundle { public static final BillResources INSTANCE = GWT.create (MyResources.class); @Source("myapp.css") public Style mystyle(); @Source("mylogo.png") public ImageResource logoImage(); public interface Style extends CssResource { String important(); String panelback(); String panelback2(); String mysprite(); } } 4)----myapp.css--------------- .important { font-weight: bold; } .panelback { background-color: #174c5d; color: white; width: 80%; padding: 20px; margin-left: 20px; } .panelback2 { border-color: #174c5d; border-style: solid; color: white; padding-top: 0; width: 80%; margin-left: 20px; } @sprite .mysprite { gwt-image: "logoImage"; background-color: #174c5d; } Workaround if you have one: Links to relevant GWT Developer Forum posts:

Last 30 days

• Dec 23, 2009

  issue 4372 (GWT Random.nextInt() always generates even integers on Googl...) Status changed by r...@google.com   -   I've filed a bug against V8 (Chrome's Javascript engine) (http://code.google.com/p/v8/issues/detail?id=558). We can also look at making changes to GWT code to use multiple calls to Math.random to generate the bits we need if V8 doesn't supply a fix. Decreasing the constants won't get away from the fact that only 30 bits or randomness are coming from Math.random(). While it will alter the set of values that are never seen, we would still see only about 1/4 of all possible 32-bit integer values. While I think we should aim for all 32 bits of Random.nextInt to be meaningful, in general many random number generators have the issue that the lower bits tend to be less random than the upper ones. For example, they may do things like alternate between even and odd, or have other repeatable patterns. So, as a practical matter, if the quality of randomness is significant to you, you might want to discard some of the low bits in your app anyway.

  Status: Accepted

  I've filed a bug against V8 (Chrome's Javascript engine) (http://code.google.com/p/v8/issues/detail?id=558). We can also look at making changes to GWT code to use multiple calls to Math.random to generate the bits we need if V8 doesn't supply a fix. Decreasing the constants won't get away from the fact that only 30 bits or randomness are coming from Math.random(). While it will alter the set of values that are never seen, we would still see only about 1/4 of all possible 32-bit integer values. While I think we should aim for all 32 bits of Random.nextInt to be meaningful, in general many random number generators have the issue that the lower bits tend to be less random than the upper ones. For example, they may do things like alternate between even and odd, or have other repeatable patterns. So, as a practical matter, if the quality of randomness is significant to you, you might want to discard some of the low bits in your app anyway.

  Status: Accepted
• Dec 23, 2009

  issue 3917 (ImageBundle cant resize images from bundle pls help !!!) commented on by fenster.ben   -   this cannot be done dou to the way that the image is taken from the main connected image this will not work for you for the near future anyway but u can use repeat instead althugh it will not work on ie 6

  this cannot be done dou to the way that the image is taken from the main connected image this will not work for you for the near future anyway but u can use repeat instead althugh it will not work on ie 6
• Dec 23, 2009

  issue 4407 (ListBox is not properly resizing in IE8 when zoom is used) reported by jackiedpeacock   -   This problem can be replicated using the ListBox in the GWT showcase: http://gwt.google.com/samples/Showcase/Showcase.html#CwListBox If you use the zoom of the browser to increase the zoom level (e.g. by pressing Ctrl and +) - and then decrease the zoom level back to 100% - the size of the ListBox is not set back to its original size. This problem appears to only happen in IE8 - and not in IE7.

  This problem can be replicated using the ListBox in the GWT showcase: http://gwt.google.com/samples/Showcase/Showcase.html#CwListBox If you use the zoom of the browser to increase the zoom level (e.g. by pressing Ctrl and +) - and then decrease the zoom level back to 100% - the size of the ListBox is not set back to its original size. This problem appears to only happen in IE8 - and not in IE7.
• Dec 23, 2009

  issue 4383 (Serilization exception in GWT 2.0) Status changed by r...@google.com   -   What errors are you seeing?

  Status: NeedsInfo

  What errors are you seeing?

  Status: NeedsInfo
• Dec 23, 2009

  issue 4403 (NPE is not thrown when calling a method on null object in De...) Owner changed by r...@google.com   -   Looks like we are not checking that 'this' is non-null (and the method being called is non-static) when calling native methods in dev mode.

  Owner: j...@google.com

  Looks like we are not checking that 'this' is non-null (and the method being called is non-static) when calling native methods in dev mode.

  Owner: j...@google.com
• Dec 23, 2009

  issue 4372 (GWT Random.nextInt() always generates even integers on Googl...) Owner changed by r...@google.com   -  

  Owner: r...@google.com

  Owner: r...@google.com
• Dec 23, 2009

  issue 4346 (Class.getCanonicalName() is not implemented) commented on by Konstantin.Scheglov   -   Right now I don't need Class.getCanonicalName(). I found this problem during adding support for GWT 2.0 widgets into GWT Designer and wondered that may be this is bug in GWT 2.0. But now I found that this method does not exist in GWT 1.7, I just never hit it before. I've added my own implementation for this method, so it is not problem for me anymore.

  Right now I don't need Class.getCanonicalName(). I found this problem during adding support for GWT 2.0 widgets into GWT Designer and wondered that may be this is bug in GWT 2.0. But now I found that this method does not exist in GWT 1.7, I just never hit it before. I've added my own implementation for this method, so it is not problem for me anymore.
• Dec 23, 2009

  issue 3861 (when autoopen is set to false submenus stay open when the mo...) commented on by ro...@sierrasky.net   -   Issue 3860 was a request for new behavior for autoopen menus. This issue is a bug that exists for non-autoopen menus. I do not believe these two issues should have been merged.

  Issue 3860 was a request for new behavior for autoopen menus. This issue is a bug that exists for non-autoopen menus. I do not believe these two issues should have been merged.
• Dec 23, 2009

  issue 4297 (Mathematical operation including / and + not being converted...) Status changed by r...@google.com   -   Could you verify whether this is still not working in GWT 2.0.0?

  Status: NeedsInfo

  Could you verify whether this is still not working in GWT 2.0.0?

  Status: NeedsInfo
• Dec 23, 2009

  issue 4305 (JDO Serialization CRITICAL Issue in 2.0 RC) commented on by r...@google.com   -   It would be tough for GWT to do the "right thing" on the client since the enhanced version of the class that contains the jdoDetachedState instance field and the enhanced versions of the various setters is: 1) not necessarily available at GWT compile time 2) enhanced in bytecode form -- GWT compiles from Java source So I think that solutions that involve the client having deep knowledge of the enhancement are not very realistic.

  It would be tough for GWT to do the "right thing" on the client since the enhanced version of the class that contains the jdoDetachedState instance field and the enhanced versions of the various setters is: 1) not necessarily available at GWT compile time 2) enhanced in bytecode form -- GWT compiles from Java source So I think that solutions that involve the client having deep knowledge of the enhancement are not very realistic.
• Dec 23, 2009

  issue 4358 (IE Developer Plugin does not install correctly) commented on by allurefx   -   Yes, this worked for me too. 32 bit IE8 on Vista 64 bit. Thanks!

  Yes, this worked for me too. 32 bit IE8 on Vista 64 bit. Thanks!
• Dec 23, 2009

  issue 3811 (KeyCodes class is not final, but cannot be extended to inclu...) Status changed by r...@google.com   -   Can you elaborate on why it is not suitable to define the additional constants in a separate class? Are there particular key codes you need support for?

  Status: NeedsInfo

  Can you elaborate on why it is not suitable to define the additional constants in a separate class? Are there particular key codes you need support for?

  Status: NeedsInfo
• Dec 23, 2009

  issue 214 (Support for Cross Domain Browsing) commented on by vishal.santoshi   -   Thank u. I tried this out I have an html page on a loaded on xxx.somedomain:301/some.html This html has <script type="text/javascript" language="javascript" src="http:// yyy.somedomain:301/g-control/gwtc/z/z.nocache.js"> document.domain = 'somedomain:301'; </script> and thus has to load up a GWT generated nocache.js from http:// yyy.somedomain:301/g-control/gwtc/z/z.nocache.js. I have the <add-linker name="xs"/> added to the GWT.xml to allow Cross Domain Scripting. I am able to execute the nocache.js , but any RPC fail with com.google.gwt.http.client.RequestPermissionException: The URL http://yyy.somedomain:301/g-control/actionservice.foundation is invalid or violates the same-origin security restrictionCaused by: com.google.gwt.http.client.RequestException: the url being , where my RemoteServlet is mounted. So me think that this might not work ( or may be it is not supposed to work :)) What am I doing wrong here ?

  Thank u. I tried this out I have an html page on a loaded on xxx.somedomain:301/some.html This html has <script type="text/javascript" language="javascript" src="http:// yyy.somedomain:301/g-control/gwtc/z/z.nocache.js"> document.domain = 'somedomain:301'; </script> and thus has to load up a GWT generated nocache.js from http:// yyy.somedomain:301/g-control/gwtc/z/z.nocache.js. I have the <add-linker name="xs"/> added to the GWT.xml to allow Cross Domain Scripting. I am able to execute the nocache.js , but any RPC fail with com.google.gwt.http.client.RequestPermissionException: The URL http://yyy.somedomain:301/g-control/actionservice.foundation is invalid or violates the same-origin security restrictionCaused by: com.google.gwt.http.client.RequestException: the url being , where my RemoteServlet is mounted. So me think that this might not work ( or may be it is not supposed to work :)) What am I doing wrong here ?
• Dec 23, 2009

  issue 4368 (Wrong result when formatting a double) commented on by heniaymen   -   Yes I think it's the same bug.

  Yes I think it's the same bug.
• Dec 23, 2009

  issue 4368 (Wrong result when formatting a double) commented on by jaroslav.zaruba   -   Isn't this one a duplicate of #4173? http://code.google.com/p/google-web-toolkit/issues/detail?id=4173&q=numberformat

  Isn't this one a duplicate of #4173? http://code.google.com/p/google-web-toolkit/issues/detail?id=4173&q=numberformat
• Dec 23, 2009

  issue 3917 (ImageBundle cant resize images from bundle pls help !!!) commented on by kondalar9   -   Still not able to re size my Images with ClientBundle. ImageResource myImageBundle; Image top = new Image(myImageBundle.top()); top.setPixelSize(topWidth, 6); panel.add(top); Pls Help

  Still not able to re size my Images with ClientBundle. ImageResource myImageBundle; Image top = new Image(myImageBundle.top()); top.setPixelSize(topWidth, 6); panel.add(top); Pls Help
• Dec 23, 2009

  issue 4406 (UiBinder dimensions and percentages) reported by ivan.rdz83   -   Found in GWT Release (GWT 2.0): Encountered on OS / Browser: Mac Snow Leopard, Pc WinXP, IE8, FF3 Detailed description: I ´m triying to desing a basic structure with percentages and there is an unexpected behaviour of the layout because of the child widgets don´t adopt the correct dimension. The code is as follows and I attach two files, the unexpected behaviour and the correct behaviour modified with firebug. <ui:UiBinder xmlns:ui="urn:ui:com.google.gwt.uibinder" xmlns:g="urn:import:com.google.gwt.user.client.ui" xmlns:e="urn:import:eu.example.autowork.client.ui.modules.employers"> <ui:style> .grid { width: 100%; height: 250px; } .scrollpanel { height: 250px; } .verticalpanel { width: 100%; height: 100%; padding: 20px; } .horizontalpanel { height: 30px; vertical-align: top; } .employer { width: 100%; margin-top: 20px; vertical-align: top; height: 100%; width: 100%; background: #e3e8f3; font-family:Arial Unicode MS,Arial,sans-serif; font-weight: bold; color: #666666; } .wrapper { height: 100%; width: 100%; } </ui:style> <g:VerticalPanel styleName="{style.verticalpanel}" > <g:ScrollPanel styleName="{style.scrollpanel}"> <e:GridEmployers ui:field="grid" styleName="{style.grid}"> </e:GridEmployers> </g:ScrollPanel> <g:HorizontalPanel styleName="{style.horizontalpanel}"> <g:Button>Nuevo</g:Button> <g:Button>Modificar</g:Button> <g:Button>Eliminar</g:Button> </g:HorizontalPanel> <g:VerticalPanel styleName="{style.wrapper}" > <e:Employer styleName="{style.employer}"> </e:Employer> </g:VerticalPanel> </g:VerticalPanel> </ui:UiBinder>

  Found in GWT Release (GWT 2.0): Encountered on OS / Browser: Mac Snow Leopard, Pc WinXP, IE8, FF3 Detailed description: I ´m triying to desing a basic structure with percentages and there is an unexpected behaviour of the layout because of the child widgets don´t adopt the correct dimension. The code is as follows and I attach two files, the unexpected behaviour and the correct behaviour modified with firebug. <ui:UiBinder xmlns:ui="urn:ui:com.google.gwt.uibinder" xmlns:g="urn:import:com.google.gwt.user.client.ui" xmlns:e="urn:import:eu.example.autowork.client.ui.modules.employers"> <ui:style> .grid { width: 100%; height: 250px; } .scrollpanel { height: 250px; } .verticalpanel { width: 100%; height: 100%; padding: 20px; } .horizontalpanel { height: 30px; vertical-align: top; } .employer { width: 100%; margin-top: 20px; vertical-align: top; height: 100%; width: 100%; background: #e3e8f3; font-family:Arial Unicode MS,Arial,sans-serif; font-weight: bold; color: #666666; } .wrapper { height: 100%; width: 100%; } </ui:style> <g:VerticalPanel styleName="{style.verticalpanel}" > <g:ScrollPanel styleName="{style.scrollpanel}"> <e:GridEmployers ui:field="grid" styleName="{style.grid}"> </e:GridEmployers> </g:ScrollPanel> <g:HorizontalPanel styleName="{style.horizontalpanel}"> <g:Button>Nuevo</g:Button> <g:Button>Modificar</g:Button> <g:Button>Eliminar</g:Button> </g:HorizontalPanel> <g:VerticalPanel styleName="{style.wrapper}" > <e:Employer styleName="{style.employer}"> </e:Employer> </g:VerticalPanel> </g:VerticalPanel> </ui:UiBinder>
• Dec 22, 2009

  issue 3742 (Make Goole Waves UI Look/Feel the default for GWT.) commented on by srdrucker   -   Would it be possible just to have the Google Wave Theme as an option?

  Would it be possible just to have the Google Wave Theme as an option?
• Dec 22, 2009

  issue 3964 (history.html does not exists) commented on by bess.mah   -   same is true for gwt 2.0

  same is true for gwt 2.0
• Dec 22, 2009

  issue 1438 (Stack overflow at line: 0) commented on by sandor.n...@erinors.com   -   We had the same problem, but it seems that switching from Gilead to hibernate-pojo-bridge (http://code.google.com/p/hibernate-pojo-bridge/) solved the problem.

  We had the same problem, but it seems that switching from Gilead to hibernate-pojo-bridge (http://code.google.com/p/hibernate-pojo-bridge/) solved the problem.
• Dec 22, 2009

  issue 49 (ListBox addItem of 100 or more has performance problems.) commented on by sandor.n...@erinors.com   -   Please take a look at my workaround, which may be used in specific cases: http://jvminside.blogspot.com/2009/12/adding-many-items-to-gwt-listbox-faster.html

  Please take a look at my workaround, which may be used in specific cases: http://jvminside.blogspot.com/2009/12/adding-many-items-to-gwt-listbox-faster.html
• Dec 22, 2009

  issue 214 (Support for Cross Domain Browsing) commented on by dobesv   -   Hi vishal, Actually I didn't try it yet, because the cross-domain linker isn't supported for a few things, like hosted mode.

  Hi vishal, Actually I didn't try it yet, because the cross-domain linker isn't supported for a few things, like hosted mode.
• Dec 22, 2009

  issue 4405 (IE8 event handling takes 100% CPU when using a FlowPanel wit...) reported by arseny.bogomolov   -   Found in GWT Release (e.g. 1.5.3, 1.6 RC): 1.7, 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): IE8 Detailed description (please be as specific as possible): The below code builds 250 small tables, each of which contains a short label and a checkbox. Each table receives onmouseover and onmouseout events, and sets or unsets a red background style element. These tables are then added to a DIV (FlowPanel), and the FlowPanel is placed within a cell of another table (VerticalPanel). The performance on IE8 is terrible. The CPU goes to 100% and the browser really starts lagging. If you try running this code, you'll see that the red background barely switches from item to item. I did not observe the issue in IE6, IE7, or even IE8 Compatibility Mode, and of course everything works OK in FF, etc. The interesting thing is that if I use a SimplePanel instead of a VerticalPanel as the overall container, the performance issue goes away. I initially thought this was a browser issue, but so far have been unable to reproduce using just native HTML and Javascript. The code snippet below is a very trivial example (I'm actually running into this issue using Incubator's FastTree), but it should hopefully be sufficient to reproduce the problem. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): FlowPanel fp = new FlowPanel(); for ( int i = 0; i < 250; ++i ) { HorizontalPanel row = new HorizontalPanel() { @Override public void onBrowserEvent( Event e ) { super.onBrowserEvent( e ); switch ( DOM.eventGetType(e) ) { case Event.ONMOUSEOVER: getElement().getStyle().setBackgroundColor( "red" ); break; case Event.ONMOUSEOUT: getElement().getStyle().setBackgroundColor( "" ); break; } } }; row.setWidth( "100%" ); row.sinkEvents( Event.ONMOUSEOVER | Event.ONMOUSEOUT ); row.add( new InlineLabel("Item " + i) ); row.add( new CheckBox() ); fp.add( row ); } VerticalPanel layout = new VerticalPanel(); layout.setWidth( "100%" ); layout.add( fp ); // If using SimplePanel as below, everything works fine // SimplePanel layout = new SimplePanel(); // layout.setWidget( fp ); RootPanel.get().add( layout ); Workaround if you have one: None that I could find. The problem actually happens when a large FastTree is placed inside a DisclosurePanel (results in a similar combination of widgets as the sample above) Links to relevant GWT Developer Forum posts:

  Found in GWT Release (e.g. 1.5.3, 1.6 RC): 1.7, 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): IE8 Detailed description (please be as specific as possible): The below code builds 250 small tables, each of which contains a short label and a checkbox. Each table receives onmouseover and onmouseout events, and sets or unsets a red background style element. These tables are then added to a DIV (FlowPanel), and the FlowPanel is placed within a cell of another table (VerticalPanel). The performance on IE8 is terrible. The CPU goes to 100% and the browser really starts lagging. If you try running this code, you'll see that the red background barely switches from item to item. I did not observe the issue in IE6, IE7, or even IE8 Compatibility Mode, and of course everything works OK in FF, etc. The interesting thing is that if I use a SimplePanel instead of a VerticalPanel as the overall container, the performance issue goes away. I initially thought this was a browser issue, but so far have been unable to reproduce using just native HTML and Javascript. The code snippet below is a very trivial example (I'm actually running into this issue using Incubator's FastTree), but it should hopefully be sufficient to reproduce the problem. Shortest code snippet which demonstrates issue (please indicate where actual result differs from expected result): FlowPanel fp = new FlowPanel(); for ( int i = 0; i < 250; ++i ) { HorizontalPanel row = new HorizontalPanel() { @Override public void onBrowserEvent( Event e ) { super.onBrowserEvent( e ); switch ( DOM.eventGetType(e) ) { case Event.ONMOUSEOVER: getElement().getStyle().setBackgroundColor( "red" ); break; case Event.ONMOUSEOUT: getElement().getStyle().setBackgroundColor( "" ); break; } } }; row.setWidth( "100%" ); row.sinkEvents( Event.ONMOUSEOVER | Event.ONMOUSEOUT ); row.add( new InlineLabel("Item " + i) ); row.add( new CheckBox() ); fp.add( row ); } VerticalPanel layout = new VerticalPanel(); layout.setWidth( "100%" ); layout.add( fp ); // If using SimplePanel as below, everything works fine // SimplePanel layout = new SimplePanel(); // layout.setWidget( fp ); RootPanel.get().add( layout ); Workaround if you have one: None that I could find. The problem actually happens when a large FastTree is placed inside a DisclosurePanel (results in a similar combination of widgets as the sample above) Links to relevant GWT Developer Forum posts:
• Dec 22, 2009

  issue 214 (Support for Cross Domain Browsing) commented on by vishal.santoshi   -   dobesv, Did the 2 steps work for u?

  dobesv, Did the 2 steps work for u?
• Dec 22, 2009

  issue 4057 (String.split(String) does not return the same value when you...) Status changed by r...@google.com   -   Fixed in r7348. Now "".split(",") will return { "" }. Although Sun's javadoc is ambiguous, my interpretation is that the sentence "If the expression does not match any part of the input then the resulting array has just one element, namely this string" takes precedence over what comes after it.

  Status: FixedNotReleased

  Fixed in r7348. Now "".split(",") will return { "" }. Although Sun's javadoc is ambiguous, my interpretation is that the sentence "If the expression does not match any part of the input then the resulting array has just one element, namely this string" takes precedence over what comes after it.

  Status: FixedNotReleased
• Dec 22, 2009

  issue 4048 (error of java.sql.Timestamp.valueOf in Hosted mode) Status changed by r...@google.com   -   Timestamp is being too picky -- it expects the time to have a nanoseconds field, like so: "2009-09-27 20:2:23.123456789". That format is what the JRE javadoc claims to expect, although in practice it will accept timestamps without the nanoseconds field.

  Status: ReviewPending

  Timestamp is being too picky -- it expects the time to have a nanoseconds field, like so: "2009-09-27 20:2:23.123456789". That format is what the JRE javadoc claims to expect, although in practice it will accept timestamps without the nanoseconds field.

  Status: ReviewPending
• Dec 22, 2009

  r7348 (Ensure "".split(",") returns { "" }. This is the behavior o...) committed by r...@google.com   -   Ensure "".split(",") returns { "" }. This is the behavior of Sun's JRE, although the JRE javadoc is somewhat ambiguous. Review by: jlabanca

  Ensure "".split(",") returns { "" }. This is the behavior of Sun's JRE, although the JRE javadoc is somewhat ambiguous. Review by: jlabanca
• Dec 22, 2009

  issue 4404 (Add "Launch Default Browser" to Development Mode view ) reported by johnthuss   -   Found in GWT Release (e.g. 1.5.3, 1.6 RC): GWT 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): Mac OS X 10.6.2 Eclipse 3.4.2 Google Plugin 1.2.0 Detailed description (please be as specific as possible): Please add a "Launch Default Browser" button to the Development Mode view from the eclipse plugin, that works exactly like the same button that appears if you use the Swing UI for development mode. Workaround if you have one: Copy the url and paste into the browser

  Found in GWT Release (e.g. 1.5.3, 1.6 RC): GWT 2.0 Encountered on OS / Browser (e.g. WinXP, IE6-7, FF3): Mac OS X 10.6.2 Eclipse 3.4.2 Google Plugin 1.2.0 Detailed description (please be as specific as possible): Please add a "Launch Default Browser" button to the Development Mode view from the eclipse plugin, that works exactly like the same button that appears if you use the Swing UI for development mode. Workaround if you have one: Copy the url and paste into the browser
• Dec 22, 2009

  issue 4249 (module rename-to causing classpath issue) Status changed by r...@google.com   -   Please retry with GWT 2.0 and let me know if the problem still exists.

  Status: NeedsInfo

  Please retry with GWT 2.0 and let me know if the problem still exists.

  Status: NeedsInfo
• Dec 22, 2009

  issue 4054 (Optimization: Add String.equals(String) method to emul libra...) Status changed by r...@google.com   -   Added in r7347.

  Status: FixedNotReleased

  Added in r7347.

  Status: FixedNotReleased
• Dec 22, 2009

  issue 1879 (GWT serialisation should stream its result) commented on by david.nouls   -   Are you assuming that deRPC solves all the issues we have seen with RPC ? I will test the new RPC mechanism after my christmas holidays.

  Are you assuming that deRPC solves all the issues we have seen with RPC ? I will test the new RPC mechanism after my christmas holidays.
• Dec 22, 2009

  r7347 (Add a more inlinable path for String.equals(String) Review ...) committed by r...@google.com   -   Add a more inlinable path for String.equals(String) Review by: jat

  Add a more inlinable path for String.equals(String) Review by: jat
• Dec 22, 2009

  issue 488 (Functionality to add custom HTTP headers to RPC requests) commented on by burke.eric   -   @t.broyer I had a chance to play around with RpcRequestBuilder and that works great! So this issue is definitely resolved in my mind. Thanks for the help.

  @t.broyer I had a chance to play around with RpcRequestBuilder and that works great! So this issue is definitely resolved in my mind. Thanks for the help.