_id
stringlengths 36
36
| bug_id
int64 1
425k
| description
stringlengths 1
373k
⌀ | bug_severity
stringclasses 7
values | bug_status
stringclasses 3
values | resolution
stringclasses 8
values |
---|---|---|---|---|---|
{'$oid': '52e9b45554dc1c25ebdb2011'} | 301 | - create three new java projects in a workspace
- select one, right-click, and note that the "Convert to PDE Project" menu item is enabled since it's not a PDE project
- select all three of the projects, right-click -> Convert to PDE Project
- now select the first project that was selected above, right-click, and note that the item is disabled since it's now a
PDE project (ok)
- but select any of the others that should have been converted in step 3 and note that their Convert to PDE Project
menu items are still enabled, indicating that they were never converted
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2012'} | 302 | - in a workspace containing all the platform plugins, including examples (43 plugins in all),
- open the new component project wizard
- on a 1024x768 display, the wizard is too tall to show the next button
- the wizard can't be resized (see 1GHBN33: ITPUI:WIN2000 - Wizards should be resizable)
- enter a name and hit enter to go to the next page
- enter an id and version stamp and hit enter to go to the next page
- on the third page, you'll see a checkbox table for all the plugins
- it tries to fit all of them on one page, making the page too large
It should give this table a fixed height and use a scroll bar instead.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2013'} | 303 | Component builder does not handle cleanly NL for install.xml (eg. label="%name"). It also does not handle correctly NLed plugins
(eg. %name 1.0.0 appears on the content page)
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2014'} | 304 | - select a plugin.xml, right-click -> Create Plug-in JARs
- the title in the subsequent wizard reads " Configure...build an Eclipse plug-in libraries" (should say "library")
NOTES: | normal | RESOLVED | INVALID |
{'$oid': '52e9b45654dc1c25ebdb2015'} | 306 | When setting up a self-hosting Eclipse environment for JDI, I had to add the following
line to <requires> in plugin.xml of project org.eclipse.jdt.debug:
<import plugin="org.eclipse.core.runtime"/>
NOTES:
DW (7/30/01 12:01:11 PM)
Sent e-mail to JM.
JM (8/7/2001 11:01:54 AM)
There is no need to prereq the runtime plugin. Technically you can do it if you want but
the runtime plugin is added automatically and your import statement will be ignored.
DW (8/7/01 10:34:21 AM)
Closed.
MM (8/8/2001 10:10:30 AM)
Moved to ITPPDE.
When converting a Java project to a PDE project, the org.eclipse.core.runtime project
should always be added to the list of dependend projects. | normal | RESOLVED | INVALID |
{'$oid': '52e9b45654dc1c25ebdb2016'} | 308 | I found these exceptions about half an hour after they were thrown.
I might've been using self-hosting when they occured. ( not sure though because it was around the borderline where
I switched from non-self hosting to self hosting.
Log: Thu Jul 26 20:03:52 EDT 2001
2 org.eclipse.core.resources 2 Problems occurred when invoking code from plug-in: org.eclipse.core.resources.
org.eclipse.swt.SWTException: Invalid thread access
at org.eclipse.swt.SWT.error(SWT.java:1708)
at org.eclipse.swt.SWT.error(SWT.java:1642)
at org.eclipse.swt.widgets.Widget.error(Widget.java:368)
at org.eclipse.swt.widgets.Widget.checkWidget(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Composite.getChildren(Composite.java:145)
at org.eclipse.pde.internal.editor.manifest.AlertSection.update(AlertSection.java:258)
at org.eclipse.pde.internal.editor.manifest.AlertSection.resourceChanged(AlertSection.java:250)
at org.eclipse.core.internal.events.NotificationManager$1.run(NotificationManager.java:122)
at org.eclipse.core.internal.runtime.InternalPlatform.run(InternalPlatform.java:758)
at org.eclipse.core.runtime.Platform.run(Platform.java:393)
at org.eclipse.core.internal.events.NotificationManager.notify(NotificationManager.java:137)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:42)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:62)
at org.eclipse.core.internal.resources.Workspace.broadcastChanges(Workspace.java:108)
at org.eclipse.core.internal.resources.Workspace.endOperation(Workspace.java:669)
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1199)
at org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:78)
at org.eclipse.jface.operation.ModalContext$ModalContextThread.run(ModalContext.java:98)
Log: Thu Jul 26 20:03:52 EDT 2001
2 org.eclipse.core.resources 2 Problems occurred when invoking code from plug-in: org.eclipse.core.resources.
org.eclipse.swt.SWTException: Invalid thread access
at org.eclipse.swt.SWT.error(SWT.java:1708)
at org.eclipse.swt.SWT.error(SWT.java:1642)
at org.eclipse.swt.widgets.Widget.error(Widget.java:368)
at org.eclipse.swt.widgets.Widget.checkWidget(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Composite.getChildren(Composite.java:145)
at org.eclipse.pde.internal.editor.manifest.AlertSection.update(AlertSection.java:258)
at org.eclipse.pde.internal.editor.manifest.AlertSection.resourceChanged(AlertSection.java:250)
at org.eclipse.core.internal.events.NotificationManager$1.run(NotificationManager.java:122)
at org.eclipse.core.internal.runtime.InternalPlatform.run(InternalPlatform.java:758)
at org.eclipse.core.runtime.Platform.run(Platform.java:393)
at org.eclipse.core.internal.events.NotificationManager.notify(NotificationManager.java:137)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:42)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:62)
at org.eclipse.core.internal.resources.Workspace.broadcastChanges(Workspace.java:108)
at org.eclipse.core.internal.resources.Workspace.endOperation(Workspace.java:669)
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1199)
at org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:78)
at org.eclipse.jface.operation.ModalContext$ModalContextThread.run(ModalContext.java:98)
Log: Thu Jul 26 20:03:57 EDT 2001
2 org.eclipse.core.resources 2 Problems occurred when invoking code from plug-in: org.eclipse.core.resources.
org.eclipse.swt.SWTException: Widget is disposed
at org.eclipse.swt.SWT.error(SWT.java:1708)
at org.eclipse.swt.SWT.error(SWT.java:1642)
at org.eclipse.swt.widgets.Widget.error(Widget.java:368)
at org.eclipse.swt.widgets.Control.getDisplay(Control.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.isValidThread(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.checkWidget(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Composite.getChildren(Composite.java:145)
at org.eclipse.pde.internal.editor.manifest.AlertSection.update(AlertSection.java:258)
at org.eclipse.pde.internal.editor.manifest.AlertSection.resourceChanged(AlertSection.java:250)
at org.eclipse.core.internal.events.NotificationManager$1.run(NotificationManager.java:122)
at org.eclipse.core.internal.runtime.InternalPlatform.run(InternalPlatform.java:758)
at org.eclipse.core.runtime.Platform.run(Platform.java:393)
at org.eclipse.core.internal.events.NotificationManager.notify(NotificationManager.java:137)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:42)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:62)
at org.eclipse.core.internal.resources.Workspace.broadcastChanges(Workspace.java:108)
at org.eclipse.core.internal.resources.Workspace.endOperation(Workspace.java:669)
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1199)
at org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:78)
at org.eclipse.jface.operation.ModalContext$ModalContextThread.run(ModalContext.java:98)
Log: Thu Jul 26 20:04:00 EDT 2001
2 org.eclipse.core.resources 2 Problems occurred when invoking code from plug-in: org.eclipse.core.resources.
org.eclipse.swt.SWTException: Widget is disposed
at org.eclipse.swt.SWT.error(SWT.java:1708)
at org.eclipse.swt.SWT.error(SWT.java:1642)
at org.eclipse.swt.widgets.Widget.error(Widget.java:368)
at org.eclipse.swt.widgets.Control.getDisplay(Control.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.isValidThread(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.checkWidget(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Composite.getChildren(Composite.java:145)
at org.eclipse.pde.internal.editor.manifest.AlertSection.update(AlertSection.java:258)
at org.eclipse.pde.internal.editor.manifest.AlertSection.resourceChanged(AlertSection.java:250)
at org.eclipse.core.internal.events.NotificationManager$1.run(NotificationManager.java:122)
at org.eclipse.core.internal.runtime.InternalPlatform.run(InternalPlatform.java:758)
at org.eclipse.core.runtime.Platform.run(Platform.java:393)
at org.eclipse.core.internal.events.NotificationManager.notify(NotificationManager.java:137)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:42)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:62)
at org.eclipse.core.internal.resources.Workspace.broadcastChanges(Workspace.java:108)
at org.eclipse.core.internal.resources.Workspace.endOperation(Workspace.java:669)
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1199)
at org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:78)
at org.eclipse.jface.operation.ModalContext$ModalContextThread.run(ModalContext.java:98)
NOTES:
JohnA (7/30/2001 10:19:36 AM)
Appears to be a problem with the PDE editor's notification listener. Updates should be async'ed to run in
UI thread. Moving to ITPPDE | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2017'} | 305 | - turn off autobuilding
- retrieve a plugin from a repository
- select its plugin.xml, right-click -> Create plugin-in JARs
- in the subsequent wizard only check the first checkbox at the top, Finish
- a build occurs, even though Ant script generation should only be dependent
upon the available plugin.xml and build.properties files
NOTES: | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b45654dc1c25ebdb2018'} | 307 | From EC newsgroup ...
Subject:
'Name Collision' error occurs during building of PDE Project
Date:
Mon, 16 Jul 2001 23:13:19 -0400
From:
Sandy Minocha <minocha@ca.ibm.com>
Organization:
EclipseCorner
Newsgroups:
eclipse.tools
I'm having problems with PDE calculating my build path correctly. I've
created a new PDE Project and than imported my plugin source. This
plugin depends on several external plugins. In the "Dependencies" Page
of my plugin.xml, i see the required external plugins checked. However,
when I explicitly select "Compute Build Path" in the popup menu on the
"Dependencies" Page, a pop-up window appears saying "Name collision".
If i look at my stack trace, it shows the following. Note that this
plugin works fine in VAJ and i'm trying this out in R0.9 of Eclipse.
4 org.eclipse.pde 0 Name collision.
Java Model Exception: Java Model Status [Name collision.]
at
org.eclipse.jdt.internal.core.JavaProject.setRawClasspath(JavaProject.java:1507)
at
org.eclipse.jdt.internal.core.JavaProject.setRawClasspath(JavaProject.java:1491)
at
org.eclipse.pde.internal.base.BuildPathUtil.setBuildPath(BuildPathUtil.java:146)
at
org.eclipse.pde.internal.editor.manifest.PluginListSection$6.run(PluginListSection.java:220)
at org.eclipse.jface.operation.ModalContext.run(ModalContext.java:249)
at
org.eclipse.jface.dialogs.ProgressMonitorDialog.run(ProgressMonitorDialog.java:335)
at
org.eclipse.pde.internal.editor.manifest.PluginListSection.computeBuildPath(PluginListSection.java:234)
at
org.eclipse.pde.internal.editor.manifest.PluginListSection.access$4(PluginListSection.java:212)
at
org.eclipse.pde.internal.editor.manifest.PluginListSection$7.run(PluginListSection.java:295)
at
org.eclipse.jface.action.ActionContributionItem.handleWidgetSelection(ActionContributionItem.java:408)
at
org.eclipse.jface.action.ActionContributionItem.handleWidgetEvent(ActionContributionItem.java:362)
at
org.eclipse.jface.action.ActionContributionItem.access$0(ActionContributionItem.java:353)
at
org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java:47)
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled
Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled
Code))
at
org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled
Code))
at
org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled
Code))
at
org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled
Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:622)
at
org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:119)
at org.eclipse.core.launcher.Main.run(Main.java:401)
at org.eclipse.core.launcher.Main.main(Main.java:274)
4 org.eclipse.jdt.core 977 Name collision.
Thanks.
Sandy Minocha
NOTES: | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45654dc1c25ebdb2019'} | 309 | First line of PDE preference page does not display japanese characters.
PDE uses DialogPage.setDescription to display this text.
NOTES:
KK (2001/07/31 15:37:06)
Moving to ITPUI for comment
TC (9/5/2001 2:29:45 PM) - related to
1GJJLBK: SWT:WINNT - German: Text.setText() is ignored if a font is set
The problem here is that the font we use for the banner is not a valid font for
Japanese. In
build 135 or later there will be an option to change the banner font to the
system default which
should be correct for Japanese.
KH (9/10/2001 12:04:41 PM)
NL Fragement should fix the default behavior. Please verify in build
135. | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb201a'} | 310 | EC post "PDE plugin name validation" by MP on 30/07/2001:
I have a plugin that contains an underscore on its name ("my_plugin" for
example). Although this is a valid classpath and a working plugin, the PDE
plugin project creation wizard doesn't allow me to create it. Why?
NOTES:
NE (7/30/01 6:55:43 PM)
This seems like a bug in PDE, since Core and underlying file systems allow a project name to contain underscore.
I notice that you can create a Java project with an underscore, then "Convert to PDE Project" in the Navigator's popup menu. | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb201b'} | 311 | from the newsgroup
Subject:
PDE needs a decent default for the run time instance of the workbench
Date:
Wed, 22 Aug 2001 14:16:38 -0400
From:
Pat McCarthy <patmc@us.ibm.com>
Organization:
OTI-RTP - Eclipse Jumpstart Team
Newsgroups:
eclipse.tools
When we develop plug-ins we use the development instance of the workbench, the first instance started.
When we test our plug-ins we launch the run time instance from the development instance.
The PDE preference settings tell the run time instance where it should look for a workspace.
The default is blank, which triggers the run-time instance to see the current development instance workspace
directory as the run time workspace directory.
This seems to be causing us to corrupt our development workspace when we shutdown the run time instance.
I think we can all agree that we should not be using the development workspace as the run time workspace. Given
that, shouldn't we expect that:
The PDE preferences have a default value that does not allow this to occur (seems bad when the default
corrupts).
The PDE preferences dialog should allow for a directory to be identified that does not exist - and create it.
Very painful to have to go create the directory and then use the PDE preference to point to it as the run time
workspace location. I'd vote for install-dir\runtime or \PDE_runtime (install-dir\workspace is the default
development workspace).
Seems like something that corrupts by default should also be fixed for 1.0.
Comments?
Pat Mc.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb201c'} | 312 | from www.eclipse.org news group
Subject:
Porting workspaces
Date:
31 Jul 2001 16:00:33 GMT
From:
danjou@us.ibm.com (Jim D\'Anjou)
Organization:
http://www.eclipse.org
Newsgroups:
eclipse.tools
For instructional purposes we want to be able to provide to students an
image of Eclipse and a workspace that has a set of "starter" PDE projects
and solutions that students use during the course.
We found that supplying only a workspace (provided as a zip file) and
loading the unzipped workspace from an existing image of Eclipse did not
work. Possibly metadata differences. We are not sure.
We tried zipped up a working version of Eclipse and its workspace (defined
separately. We are not using the default workspace) containing the student
projects. It nearly worked. It is necessary to update the ECLIPSE_HOME
variable (all impacts are nicely handled!). Existing PDE projects are
testable. Java projects work. New Java projects can be instantiated. The
only problem, and it is a big one, is that new PDE projects cannot be
created. The PDE Project wizard fails with an null pointer exception.
Below is the stack trace. Any suggestions? Our only other recourse would
be trying to ship a CVS repository and we don't know what issues may be
involved with that. We have to have a working solution for a class in
August. Thanks.
java.lang.NullPointerException
at org.eclipse.pde.internal.PDEPlugin.findPlugin(PDEPlugin.java:113)
at org.eclipse.pde.internal.PDEPlugin.findPlugin(PDEPlugin.java:125)
at org.eclipse.pde.internal.PDEPlugin.findPlugin(PDEPlugin.java:121)
at
org.eclipse.pde.internal.wizards.PluginPathUpdater.addImplicitLibraries(PluginPathUpdater.java:98)
at
org.eclipse.pde.internal.base.BuildPathUtil.setBuildPath(BuildPathUtil.java:117)
at
org.eclipse.pde.internal.wizards.project.ProjectCodeGeneratorsPage.setJavaSettings(ProjectCodeGeneratorsPage.java:220)
at
org.eclipse.pde.internal.wizards.project.ProjectCodeGeneratorsPage.access$7(ProjectCodeGeneratorsPage.java:201)
at
org.eclipse.pde.internal.wizards.project.ProjectCodeGeneratorsPage$3.execute(ProjectCodeGeneratorsPage.java:184)
at
org.eclipse.ui.actions.WorkspaceModifyOperation$1.run(WorkspaceModifyOperation.java:64)
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1194)
at
org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:78)
at org.eclipse.jface.operation.ModalContext.run(ModalContext.java:249)
at org.eclipse.jface.wizard.WizardDialog.run(WizardDialog.java:715)
at
org.eclipse.pde.internal.wizards.project.ProjectCodeGeneratorsPage.runJavaSettingsOperation(ProjectCodeGeneratorsPage.java:195)
at
org.eclipse.pde.internal.wizards.project.ProjectCodeGeneratorsPage.finish(ProjectCodeGeneratorsPage.java:163)
at
org.eclipse.pde.internal.wizards.project.NewProjectWizard.performFinish(NewProjectWizard.java:141)
at
org.eclipse.jface.wizard.WizardDialog.finishPressed(WizardDialog.java:572)
at
org.eclipse.jface.wizard.WizardDialog.buttonPressed(WizardDialog.java:313)
at org.eclipse.jface.dialogs.Dialog$1.widgetSelected(Dialog.java:211)
at
org.eclipse.swt.widgets.TypedListener.handleEvent(TypedListener.java:83)
at org.eclipse.swt.widgets.EventTable.sendEvent(EventTable.java(Compiled
Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java:622)
at
org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled
Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled
Code))
at org.eclipse.jface.window.Window.runEventLoop(Window.java(Compiled
Code))
at org.eclipse.jface.window.Window.open(Window.java:517)
at org.eclipse.ui.actions.NewWizardAction.run(NewWizardAction.java:74)
at
org.eclipse.jface.action.ActionContributionItem.handleWidgetSelection(ActionContributionItem.java:408)
at
org.eclipse.jface.action.ActionContributionItem.handleWidgetEvent(ActionContributionItem.java:362)
at
org.eclipse.jface.action.ActionContributionItem.access$0(ActionContributionItem.java:353)
at
org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java:47)
at org.eclipse.swt.widgets.EventTable.sendEvent(EventTable.java(Compiled
Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java:622)
at
org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled
Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled
Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java:639)
at org.eclipse.ui.internal.Workbench.run(Workbench.java:622)
at
org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:119)
at org.eclipse.core.launcher.Main.run(Main.java:401)
at org.eclipse.core.launcher.Main.main(Main.java:274)
Jim D'Anjou
Eclipse Jumpstart Team
NOTES:
ARCH (8/20/01 3:54:42 PM)
Fix for 1.0
DG (9/5/2001 12:41:57 PM)
Using the build 134, I tried setting up a workspace with couple of PDE projects, zipping it up, unzipping in a different
location and updating ECLIPSE_HOME variable to point to the new location. Everything seems to be working fine,
except that plugin.xml editor shows some funny errors that vanish when Eclipse is restarted.
At this point, my suggestion is to document this requirement in the installation procedures as follows:
1) Unzip the packaged Eclipse
2) Update path-sensitive variables to point to the new location (Java variables, run-time workspace location etc.)
3) Restart Eclipse | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45654dc1c25ebdb201d'} | 313 | Create a fragement project
Edit fragement.xml
In the "Library Content" area press the new button
Add an item.
Notice that a scroll bar is added to the list area and you
can no longer interact with the list.
NOTES: | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45654dc1c25ebdb201e'} | 314 | Create a fragement project
Edit fragement.xml
In the "Run-time libaries" area create some new libraries. Give each library a unique name
Notice that "Up and Down" have no effect.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb201f'} | 316 | You can create build scripts (build.xml) for a component and its related plugins by using the PDE wizards in the workbench, or using the api specified as in
http://www.eclipse.org/downloads/drops/sdk/R0.9/online-doc/html/plugins/org.eclipse.pde.doc.user/guide/pde_component_generating_antcommandline.htm
Then you can run the build.xml in the component project to build the component and the plugins, and to create jars.
I am trying to call the build.xml for a component from another ant script that I have written. But it doesn't seem to work this way. I always get an "AssertionFailedException".
I wonder if this is a problem because of the "InitTemplate" or if I am doing something wrong.
STEPS:
1. start eclipse
2. Create a sample plugin project, say myTestPlugin
3. Create a sample component project, say myTestComponent, and choose the myTestPlugin to be a plugin in this component
4. Open the install.xml for the component
5. Click on "Create jars" , then OK
6. " Build.xml "s for both the component and the plugin are now created.
7.Create a simple project called "ant"
8. create a "test.xml" file in this project with the following content:
---cut here---
<?xml version="1.0"?>
<project default = "main" basedir="../">
<taskdef name="initTemplate" classpath = "../plugins/org.eclipse.core.boot/boot.jar;../plugins/org.eclipse.core.runtime/runtime.jar;../plugins/org.eclipse.ant.core/antsupport.jar;../plugins/org.apache.xerces/xerces.jar;../plugins/org.eclipse.pde.core/pdecore.jar" classname="org.eclipse.pde.internal.core.InitTemplate" />
<target name="main" >
<property name ="componentProject" value = "myTestComponent"/>
<property name="componentId" value="org.myTestComponent" />
<property name ="tag" value = "1.0.0"/>
<property name="buildScriptPath" value = "${componentProject}/install/components/${componentId}_${tag}" />
<property name ="buildScript" value="build.xml" />
<ant antfile="${buildScript}" dir="${buildScriptPath}"/>
</target>
</project>
---cut here---
9. If you run this ant file (test.xml), you will get the following error:
BuildException: org.eclipse.core.internal.runtime.AssertionFailedException: assertion failed: meta.appNotInit.
BuildException: org.eclipse.core.internal.runtime.AssertionFailedException: assertion failed: meta.appNotInit.
BuildException: org.eclipse.core.internal.runtime.AssertionFailedException: assertion failed: meta.appNotInit.
10. But if you try to run the build.xml for the component manually, it works.
NOTES:
GWG (8/28/2001 9:20:55 AM)
I made the invocation work in test.xml by removing the taskdef and adding an initTemplate invocation, as is shown below.
<?xml version="1.0"?>
<project default = "main" basedir="../">
<target name="main" >
<initTemplate/>
<property name ="componentProject" value = "myTestComponent"/>
<property name="componentId" value="org.myTestComponent" />
<property name ="tag" value = "1.0.0"/>
<property name="buildScriptPath" value = "${componentProject}/install/components/${componentId}_${tag}" />
<property name ="buildScript" value="build.xml" />
<ant antfile="${buildScript}" dir="${buildScriptPath}"/>
</target>
</project>
I'm not sure if this is the fix that's being sought though, if the original test.xml is indeed valid and should work as is. | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45654dc1c25ebdb2020'} | 317 | This is a posting from the eclipse newsgroup by Birgit Guder
("DEFECT REPORT: Internal Error when defining an Editors Extension"
posted on 8/22/2001 6:21 PM)
Hi,
I just started using the plugin development tool and wanted to recreate the
readme-plugin step by step.
So I created a plugin project and added an Editor Extension. But whenever I
added a new editor-Element to the Editors-Extension using the Plugin
Manifest Editor I got an Internal Error.
The resulting log-file looked like this:
2 org.eclipse.ui 2 Problems occurred when invoking code from plug-in:
org.eclipse.ui.
java.lang.NullPointerException
at
org.eclipse.pde.internal.editor.manifest.ExtensionElementPropertySource.getP
ropertyValue(ExtensionElementPropertySource.java:222)
at
org.eclipse.ui.views.properties.PropertySheetEntry.refreshValues(PropertyShe
etEntry.java:484)
at
org.eclipse.ui.views.properties.PropertySheetEntry.refreshChildEntries(Prope
rtySheetEntry.java:450)
at
org.eclipse.ui.views.properties.PropertySheetEntry.setValues(PropertySheetEn
try.java:606)
at
org.eclipse.ui.views.properties.PropertySheetViewer.setInput(PropertySheetVi
ewer.java:700)
at
org.eclipse.ui.views.properties.PropertySheetPage.selectionChanged(PropertyS
heetPage.java:219)
....
After a bit of debugging the IDE itself (which is pretty straight forward by
the way, good job) I stumbled over the following code snippet in
ExtensionElementPropertySource.getPropertyValue(Object):
...
if (att == null) {
if (isEditable() && attInfo != null) {
if (attInfo.getKind() == ISchemaAttribute.JAVA) {
IPluginModelBase model = att.getModel(); // NullPointerException
!!
.....
Since att is null at this point (explicitly tested) this will cause the
above NullPointerException.
The statement should probably look like this:
IPluginModelBase model = getElement().getModel();
This error occurs both in wsw-beta and eclipse build 0.131.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2021'} | 315 | Build 0.129
Fail sniff test on Win-XP (reported by Sonia Dimitrov).
PDE - running new extension
java.lang.reflect.InvocationTargetException: java.lang.reflect.InvocationTargetException: java.lang.NullPointerException
at org.eclipse.core.internal.runtime.InternalPlatform.activateDefaultPlugins(InternalPlatform.java:100)
at org.eclipse.core.internal.runtime.InternalPlatform.loaderStartup(InternalPlatform.java:530)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.internal.boot.InternalBootLoader.startup(InternalBootLoader.java:912)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:803)
Exception launching the Eclipse Platform UI:
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:119)
at org.eclipse.core.launcher.Main.run(Main.java:401)
at org.eclipse.core.launcher.UIMain.main(UIMain.java:52)
NOTES: | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b45654dc1c25ebdb2022'} | 318 | The interface AntRunnerListener will become IAntRunnerListener for naming
consistency purpose.
This will affect the class
org.eclipse.pde.internal.component.ComponentJarWizardPage, and maybe also other
ones.
NOTES: | normal | RESOLVED | INVALID |
{'$oid': '52e9b45654dc1c25ebdb2023'} | 320 | Need to make sure we are writing generated Java using the encoding expected by
Java tooling. Currently we write UTF-8, they expect native.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2024'} | 319 | I got the following in the log when I opened a plugin.xml and moved the mouse over the
"Required plugins". Note that it only happens when the required plugin name in the list
is fully qualified, (i.e. org.eclipse.core.resources) and not the user-readable name.
(i.e. Core Resource Management)
LOG:
Log: Wed Aug 29 11:11:08 EDT 2001
1 org.eclipse.core.resources 4 Unhandled exception caught in event loop.
Log: Wed Aug 29 11:11:08 EDT 2001
4 org.eclipse.ui 0 java.lang.NullPointerException
java.lang.NullPointerException
at org.eclipse.pde.internal.editor.manifest.RequiresSection.linkEntered(RequiresSection.java:110)
at org.eclipse.pde.internal.forms.HyperlinkHandler.mouseEnter(HyperlinkHandler.java:86)
at org.eclipse.swt.widgets.TypedListener.handleEvent(TypedListener.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:622)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:119)
at org.eclipse.core.launcher.Main.run(Main.java:401)
at org.eclipse.core.launcher.Main.main(Main.java:274)
PLUGIN:
Here is the plugin.xml which caused the problems:
<?xml version="1.0"?>
<plugin
name="%pluginName"
id="org.eclipse.core.examples.site.tests"
version="1.0"
class="org.eclipse.core.examples.site.tests.SiteManagementTestsPlugin"
provider-name="Object Technology International, Inc.">
<requires>
<import plugin="org.eclipse.core.resources"/>
<import plugin="org.eclipse.core.tests.harness"/>
<import plugin="org.junit"/>
<import plugin="org.eclipse.core.examples.site"/>
</requires>
<runtime>
<library name="siteexampletests.jar"/>
</runtime>
<extension point="org.eclipse.core.tests.harness.tests">
<test id="site.AllTests">
<run class="org.eclipse.core.examples.site.tests.AllTests"/>
</test>
<test id="site.SiteObjectTest">
<run class="org.eclipse.core.examples.site.tests.SiteObjectTest"/>
</test>
<test id="site.SiteObjectFactoryTest">
<run class="org.eclipse.core.examples.site.tests.SiteObjectFactoryTest"/>
</test>
</extension>
</plugin>
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2025'} | 321 | When I tried to add an extension point to a plugin I was defining I got
"Internal Error - Widget is disposed"
STEPS
1) Create a PDE project
2) Open the plugin.xml
3) Select the extension points tab
4) add an extension point for org.eclipse.ui.preferencePages
5) Hit OK
Exception
org.eclipse.swt.SWTException: Invalid thread access
at org.eclipse.swt.SWT.error(SWT.java:1708)
at org.eclipse.swt.SWT.error(SWT.java:1642)
at org.eclipse.swt.widgets.Widget.error(Widget.java:368)
at org.eclipse.swt.widgets.Widget.checkWidget(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Composite.getChildren(Composite.java(Compiled Code))
at org.eclipse.pde.internal.editor.manifest.AlertSection.update(AlertSection.java:259)
at org.eclipse.pde.internal.editor.manifest.AlertSection.resourceChanged(AlertSection.java:251)
at org.eclipse.core.internal.events.NotificationManager$1.run(NotificationManager.java:122)
at org.eclipse.core.internal.runtime.InternalPlatform.run(InternalPlatform.java:758)
at org.eclipse.core.runtime.Platform.run(Platform.java:393)
at org.eclipse.core.internal.events.NotificationManager.notify(NotificationManager.java:137)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:42)
at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:62)
at org.eclipse.core.internal.resources.Workspace.broadcastChanges(Workspace.java:111)
at org.eclipse.core.internal.resources.Workspace.endOperation(Workspace.java(Compiled Code))
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1202)
at org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:78)
at org.eclipse.jface.operation.ModalContext$ModalContextThread.run(ModalContext.java:98)
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2026'} | 322 | from EC
Subject:
PDE Bugs reports
Date:
Thu, 6 Sep 2001 20:13:36 -0400
From:
"Olivier Refalo" <orefalo@yahoo.com>
Organization:
FPL
Newsgroups:
eclipse.tools
Version: Eclipse pre 1.0 R 134
From the PDE, if you run the plugin.xml file, in other word if you test your
plugin in a new instance of Eclipse. The plugin.xml file becomes suddently
READ-ONLY ! The file is not locked by the OS because I can change it with an
editor. But it becomes... Un-editable. Restarting Eclipse magically solves
the problem.
I got more or less the same behavior with the file build.properties...
One more question where can we check PR and bug status ?
Regards,
OR.
-*=
Olivier Refalo - Software engineer -
Java/DistibutedSystems/UnixSysop/WebSpecialist/Security
http://www.crionics.com/orefalo / mailto:orefalo@yahoo.com
Java: A simple, object oriented, distributed, interpreted, robust, secure,
architecture neutral, portable, high-performance, multi-threaded, and
dynamic language.
-*=
NOTES: | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45654dc1c25ebdb2027'} | 323 | PDE allows NL characters in plugin project names (and I presume fragment, component).
Since the project name == plugin id, plugin ids are spec'ed as a-z,A-Z,0-9 and .
See help docs ... plugin developers guide, Reference, Other reference information,
Plugin manifest.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45654dc1c25ebdb2028'} | 324 | 1) Open the Window -> Preferences dialog.
2) Select the Plug-in Development item
3) Click on the "Select All" button
4) Unselect a few items.
5) Click on the "Apply" button.
Get the following exception:
java.lang.NullPointerException
at org.eclipse.scripting.ScriptAbstractPreferencePage.getScriptNamesFromManifest(ScriptAbstractPreferencePage.java:364)
at org.eclipse.scripting.ScriptAbstractPreferencePage.populateScriptList(ScriptAbstractPreferencePage.java:539)
at org.eclipse.scripting.ScriptAbstractPreferencePage.createContents(ScriptAbstractPreferencePage.java:177)
at org.eclipse.jface.preference.PreferencePage.createControl(PreferencePage.java:194)
at org.eclipse.jface.preference.PreferenceDialog.showPage(PreferenceDialog.java:685)
at org.eclipse.jface.preference.PreferenceDialog$6.run(PreferenceDialog.java:393)
at org.eclipse.swt.custom.BusyIndicator.showWhile(BusyIndicator.java(Compiled Code))
at org.eclipse.jface.preference.PreferenceDialog$5.widgetSelected(PreferenceDialog.java:385)
at org.eclipse.swt.widgets.TypedListener.handleEvent(TypedListener.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.jface.window.Window.runEventLoop(Window.java(Compiled Code))
at org.eclipse.jface.window.Window.open(Window.java:523)
at org.eclipse.ui.internal.OpenPreferencesAction.run(OpenPreferencesAction.java:47)
at org.eclipse.jface.action.Action.runWithEvent(Action.java:451)
at org.eclipse.jface.action.ActionContributionItem.handleWidgetSelection(ActionContributionItem.java:407)
at org.eclipse.jface.action.ActionContributionItem.handleWidgetEvent(ActionContributionItem.java(Compiled Code))
at org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java(Compiled Code))
at org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:658)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:820)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:151)
at org.eclipse.core.launcher.Main.run(Main.java:502)
at org.eclipse.core.launcher.Main.main(Main.java:362)
NOTES: | normal | RESOLVED | INVALID |
{'$oid': '52e9b45754dc1c25ebdb2029'} | 325 | DS (9/20/01 3:52:39 PM)
Edit the source of of your plugin.xml to
include
<requires>
<import plugin="a.b.c"/>
</requires>
Go to the overview view, and put the mouse of the Require plugin link for a.b.c--->
4 org.eclipse.ui 0 java.lang.NullPointerException
java.lang.NullPointerException
at org.eclipse.pde.internal.editor.manifest.RequiresSection.linkEntered(RequiresSection.java:111)
at org.eclipse.pde.internal.forms.HyperlinkHandler.mouseEnter(HyperlinkHandler.java:86)
at org.eclipse.swt.widgets.TypedListener.handleEvent(TypedListener.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:656)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:151)
at org.eclipse.core.launcher.Main.run(Main.java:433)
at org.eclipse.core.launcher.Main.main(Main.java:306)
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb202a'} | 326 | 1) Install eclipse.
2) Create a new PDE plugin using the wizard called com.examples.test.
3) Run eclipse from with eclipse using the Run-> Run-time Workbench
4) Create a project in the launched eclipse.
5) Quit.
The Window -> Preferences -> Plug-in Development dialog said Workspace Location was /home/torres/Eclipse/ide/eclipse/runtime-workspace
The actual Workspace Location on the file system was /home/torres/Eclipse/test/"/home/torres/Eclipse/ide/eclipse/runtime-workspace"
Note, I launched eclipse from a script located in /home/torres/Eclipse/test/
Note, there is a directory created called " (that is the directory name is just the quote mark) and the final directory is
called runtime-workspace" (including the quote at the end).
Now edit the Workspace Location and set the new location to be 'junk' (without the single quotes) and repeat steps 3 through 4.
This time a directory is created called /home/torres/Eclipse/test/"junk" (including the double quotes)
NOTES:
DG (9/21/2001 4:36:17 PM)
The problem is due to two related problems in Core on Linux:
1) Double quotes are not stripped from the command line argument values for workspace
2) Because of quotes, Core thinks that the path is not absolute and prepends user.home
RTP is currently looking into the original PR that was opened for this problem:
1GJM757: ITPCORE:Linux - Bogus state location returned on Linux
Providing that the original core problem is fixed, PDE will need to remove 'linux' conditional code
that was written to work around the double quote problem. The conditional code is localized in
WorkbenchLauncher. java class and after the removal, the same code will execute on both Linux and
Windows.
DG (9/24/2001 12:08:25 PM)
The bug is in the core and the core cannot fix it in 1.0 stream. We have no choice but to readme
this problem. We must indicate that PDE will not be able to start the run-time workbench
on Linux if the run-time platform location and/or run-time workspace location contain spaces.
DG (9/24/2001 3:15:53 PM)
readme | normal | RESOLVED | DUPLICATE |
{'$oid': '52e9b45754dc1c25ebdb202b'} | 328 | Add the ability to pass arguments to the VM for the Target VM.
NOTES:
KJD (9/25/2001 1:41:11 PM)
It makes sense to keep the workspace arguments and the vm arguments together (and perhaps
the workspace location and the platform home).
KJD (9/25/2001 1:46:01 PM)
Code exists for this feature.... request it from me | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb202c'} | 327 | Create a PDE Plugin-project using the PDE Default Plugin Generator.
Edit the plugin.xml file.
Click on the "Runtime" tab.
Do NOT select anything in the "runtime" libraries list.
(the problem happens regardless of the number of items in the list.)
Click on the Up or Down button beside the Runtime libraries list.
Note you get the following exception:
Unhandled exception caught in event loop.
Reason: -1
Log: Mon Sep 24 14:49:06 EDT 2001
1 org.eclipse.core.resources 4 Unhandled exception caught in event loop.
Log: Mon Sep 24 14:49:06 EDT 2001
4 org.eclipse.ui 0 -1
java.lang.ArrayIndexOutOfBoundsException: -1
at org.eclipse.pde.internal.editor.manifest.LibrarySection.handleUp(LibrarySection.java:288)
at org.eclipse.pde.internal.editor.manifest.LibrarySection.access$5(LibrarySection.java:283)
at org.eclipse.pde.internal.editor.manifest.LibrarySection$5.widgetSelected(LibrarySection.java:193)
at org.eclipse.swt.widgets.TypedListener.handleEvent(TypedListener.java:85)
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:658)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:151)
at org.eclipse.core.launcher.Main.run(Main.java:433)
at org.eclipse.core.launcher.UIMain.main(UIMain.java:52)
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb202d'} | 330 | EG (26.09.2001 15:43:02)
PDE puts the startup.jar on the class path automagically. The problem I see
is that the startup.jar comes from the runtime/development workspace.
For example, I'm running 136 and have code from build 200 loaded in my
workspace (I don't have 200 installed). When I run my workspace with PDE I'm running it
with the startup.jar from 136 and this could result in surprises.
NOTES:
from DG
It's a tough question. The class that has 'main' I need to run is UIMain
and it is in startup.jar. Therefore, it belongs to the 'non-plugin' code
that is somewhat hard to handle.
I agree that 1.0 has an issue with the respect to startup.jar, but this
issue will probably not show itself as a critical problem any time soon. It
may become a problem for people that work on 2.0 version of Eclipse while
using 1.0 as a development tool (as you have observed).
I would definitely recommend to open a PR against PDE re this problem. A
possible solution would be for PDE to first try to locate UIMain in the
workspace. If found, it should run it, if not it should run the version
from the development tool space. You are free to comment on possible
problems of this solution.
EG (26.09.2001 15:45:35)
I agree that the problem will not show up soon. Looking for UIMain
in the workspace or asking the user once for the startup.jar would make
the current behavior less magic. | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb202e'} | 329 | I have tried many times to launch the Target workbench resulting in a silent failure due to having
a .java (class) file selected. It would be a lot more useable to have the Project the class that is selected be
the Project that is used to launch the workbench.
i.e.
/myProj/src/foo.java
is selected when launching ('Run-time Workbench').
Look up the project ('myProj')
Launch using 'myProj'
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb202f'} | 331 | There are instances where
plug-ins might want to include a web enabled runtime to get
javax.servlet, etc. J2EE uses it too. There's no reason to include runtime
libraries:
<library name="lib"/>
<library name="lib/ext"/>
<library name="web/help"/>
<library name="properties"/>
as they don't contain any loadable .class files. So getting rid of these
wouldn't change anything other than to fix the name collision problem. It
shouldn't
actually be doing anything with the above libraries since they don't contribute
anything to the classpath. They should just be ignored.
The problem is with the following elements in the runtime element:
<library name="lib"/>
<library name="lib/ext"/>
<library name="web/help"/>
<library name="properties"/>
Somehow these are causing a name collision. It might be because "lib" is
included along with lots of references to jar
files in the lib folder. But I'm not sure why.
I have no specific idea why either, except for my explanation above -- we need
a directory called "lib" and it contains
only jars and no classes and it must be defined for our internal use.
These directories don't contain any .class files, so do they need to be in the
runtime?
Yes, for the reasons stated -- we need to configure it this way to fire it up
properly.
Vlad, what does including these libraries mean if they only contain .jar files,
not .class files.
Are all the lib/*.jar files automatically included? The doc says only .classes.
This would cause
collisions because there are other places where specific .jar files are
referenced:
<library name="lib/vaprt.jar">
<export name="*"/>
</library>
NOTES: | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45754dc1c25ebdb2030'} | 333 | Image form_banner.gif is missing in the 'basic' version of PDE icons. Since PDE tries to load it nonetheless, the editor
ends up displaying a red dot where the image should be.
PDE should not try to load the image or supply a small transparent image that will not be seen instead.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2031'} | 332 | From Jim Amsden:
Degan,
Found another potential problem. You can put dependencies in the
requires element of the manifest file with PDE, but you can't specify that these are exported.
This is preventing classes from being loaded at runtime. Need a toggle next to each dependency to set export="true".
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2032'} | 334 | From VK:
Dejan, as part of the 2.0 development stream we should transfer the responsibility
for shipping/ defining/ maintaining the PDE schema files back to the teams. Can you
please put together simple note with instructions for the teams to follow. Thanks.
NOTES: | normal | CLOSED | WONTFIX |
{'$oid': '52e9b45754dc1c25ebdb2033'} | 335 | From Carolyn MacLeod:
It's kind of hard to tell, but I think (?) most of these are PDE feature
requests?
Car
"Marcelo Paternostro" <marcelop@ca.ibm.com> wrote in message
news:<9omcuh$6aa$1@rogue.oti.com>...
> Hi,
>
> Some simple features I am missing on Eclipse:
>
> 1. Export/Import Plugins Preference
>
> I have multiple Eclipse folders on my machine and it is really painful -
and
> non-productive - to synchronize all the settings on the preferences. It
you
> be very good if the UI allows us to Export and Import the plugins'
> preferences (maybe serializing IPreferenceStore...).
>
> The best interface should allow the user to select which preferences
he/she
> wants to Export or Import - without throwing exceptions or presenting
popups
> if someone tries to import the preferences of a missing plugin.
>
>
> 2. PDE Preference page
>
> It would be very good if the number of "unselected plugin" was presented
-
> not only the number of "selected plugins".
>
>
> 3. Path Variable (Java IDE)
>
> 3.1 The "new variable dialog" should allow us to define the source
> attachment of a jar or zip file
> 3.2 We should be able to create a variable pointing to more then one file
> and/or folder
>
>
> 4. After release or update the stream the repository view should be
updated
> automatically
>
> Maybe this should be an option on the Team's preferences.
>
>
> Thx.
>
>
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2034'} | 336 | When plugin.xml exists locally and the java project is converted to PDE project
an error: plugin.xml exists locally appears and the project is not converted to
PDE and it cannot be converted even after a "Refresh from Local is performed"
NOTES:
KJD (10/3/01 1:55:58 PM)
Here is a .log snippet:
4 org.eclipse.pde 0
A resource already exists on disk /home/kdriedge/C_Development/eclipse/workspace/com.ibm.dstore.miners/plugin.xml.
org.eclipse.core.internal.resources.ResourceException: A resource already exists on disk /home/kdriedge/C_Development/eclipse/workspace/com.ibm.dstore.miners/plugin.xml.
at org.eclipse.core.internal.resources.File.create(File.java:104)
at org.eclipse.pde.internal.model.WorkspacePluginModelBase.save(WorkspacePluginModelBase.java:114)
at org.eclipse.pde.internal.wizards.project.ConvertedProjectsPage.createManifestFile(ConvertedProjectsPage.java:249)
at org.eclipse.pde.internal.wizards.project.ConvertedProjectsPage.convertProject(ConvertedProjectsPage.java:105)
at org.eclipse.pde.internal.wizards.project.ConvertJavaToPDEProjectAction$1.execute(ConvertJavaToPDEProjectAction.java:37)
at org.eclipse.ui.actions.WorkspaceModifyOperation$1.run(WorkspaceModifyOperation.java:59)
at org.eclipse.core.internal.resources.Workspace.run(Workspace.java:1187)
at org.eclipse.ui.actions.WorkspaceModifyOperation.run(WorkspaceModifyOperation.java:73)
at org.eclipse.jface.operation.ModalContext.runInCurrentThread(ModalContext.java:229)
at org.eclipse.jface.operation.ModalContext.run(ModalContext.java:274)
at org.eclipse.jface.dialogs.ProgressMonitorDialog.run(ProgressMonitorDialog.java:335)
at org.eclipse.pde.internal.wizards.project.ConvertJavaToPDEProjectAction.run(ConvertJavaToPDEProjectAction.java:49)
at org.eclipse.ui.internal.PluginAction.run(PluginAction.java:125)
at org.eclipse.jface.action.ActionContributionItem.handleWidgetSelection(ActionContributionItem.java:397)
at org.eclipse.jface.action.ActionContributionItem.handleWidgetEvent(ActionContributionItem.java:351) at org.eclipse.jface.action.ActionContributionItem.access$0(ActionContributionItem.java:342)
at org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java:47)
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:676)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:810)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:280)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:151)
at org.eclipse.core.launcher.Main.run(Main.java:433)
at org.eclipse.core.launcher.Main.main(Main.java:306)
4 org.eclipse.core.resources
272 A resource already exists on disk /home/kdriedge/C_Development/eclipse/workspace/com.ibm.dstore.miners/plugin.xml. | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2035'} | 338 | STEPS:
1. Load a plugin project, click on a class in the project in the Packages View
2. Click Run>>"Run-time Workbench"
This fails silently, it should run the target eclipse using the project for this class.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2036'} | 337 | STEPS:
1. Load a plugin project, click on the project in the Packages View
2. Click on the running man (not the drop down menu)
A dialog with the title "Run" appears but the Run-time workbench launcher is not available.
NOTES: | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b45754dc1c25ebdb2037'} | 339 | DDW (03/10/2001 10:34:33 AM)
In 202a, some timings were done to get a feel for how caching the plugin registry would affect startup times.
The following data shows how long it took to "parse and resolve" the plugin registry and the corresponding
full startup times vs. how long it took to read the plugin registry from a cache file and the corresponding full
startup times.
Coincidentally,these tests were run over 2 environments. The first environment contained one java project and
one plugin project (with one plugin with errors). The plugin project was open. The second environment
contained only one java project with 3 simple class files all of which were open. It was noticed that when
the environment with the plugin project was used, initial startup contained an extra call to parse a
registry.
Comparing groups 1 and 3, the full startup times seem larger with the plugin project while the
times for parsing and resolving the plugin registry seem comparable. The times for
reading the plugin registry from a cache file in groups 2 and 4 are similar but the startup times
are quite different.
Group 1:
Parse and Resolve (with 2 projects: 1 java project + 1 plugin project with errors)
6679 ms f ull startup = 27036 ms
6709 ms 26936
6649 27066
6789 28271
7124 30799
------------------------------------------
6790 28021.6 (AVERAGE)
Group 2:
Cached Registry (with same 2 projects)
421 25069
391 24537
401 24948
391 24547
390 24978
------------------------------------------
398.8 24815.8 (AVERAGE)
Group 3:
Parse and Resolve (with 1 java project)
6660 22764
6740 22724
6650 22674
6640 22674
6670 22635
6581 23337
------------------------------------------
6656.8 22801.3 (AVERAGE)
Group 4:
Cached Registry (with 1 project - the java project)
391 17014
401 16853
401 16944
391 16834
391 16913
------------------------------------------
395 16911.6 (AVERAGE)
NOTE: In the first two groups (the ones with the 2 projects), there was some other xml parsing done
(the RegistryLoader gave a time for a 'Parsed Registry'). In the first group (when we were doing the
Parse and Resolve of the 'real' registry) the times for this 2nd parse were about 801 ms. When we
cached the 'real' registry, the times for the 2nd parse were up significantly to over 4000 ms!! This
2nd parse was missing from the 3rd and 4th group of data (the ones without the plugin project).
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2038'} | 340 | The generated build.properties source entry needs a trailing / when specifying a source folder
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb2039'} | 341 | When dependent plugins specify $ws$/*.jar on their classpath, the generated build properties
cannot be resolved.
Same problem probably exists for handling $nl$ and $os$
NOTE: not sure if PDE or JDT should be handling this
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb203a'} | 342 | Plug-in : When you create the Plugin development project for the first time, the <require>
section is in the source code. But, the Extension point selection panel is empty.
Any Plugin developemnt project created thereafter is missing the <require> section from the code.
And the extension point selection panel is also empty.
This problem only occured in driver 136a. In the driver 135, this is not a problem for Korean.
This functionality broke in driver 136a.
Steps to recreate:
Step A: Missing <require> section.
-Create a plug-in project :File->New->Project... and selecting Plug-in Development, then Plug-in Project, press next.
-Name the project with some SBCS name "aaa", press next, then press next.
-Select "create plug-in using a template wizard" + Default plug-in Generator, press next, then press finish.
-From panel "aaa1" Plugin, select the "Source" tab at the bottom of screen
-The plugin source code is displayed as follow:
<?xml version="1.0" encoding="UTF-8"?>
<plugin
name="aaa Plugin"
id="aaa"
version="1.0.0"
class="aaa.aaaPlugin">
<requires>
<import plugin="org.eclipse.core.runtime"/>
<import plugin="org.eclipse.core.ressources"/>
<import plugin="org.eclipse.ui"/>
</requires>
<runtime>
<library name="aaa.jar"/>
</runtime>
=> The section <requires> ...</requires> is missing from for the source.
StepB: Empty Extension point selection panel displayed
- At the bottom of the screen, turn to the "Extension" page
and push the new button to start the Extension wizard.
-Select "Generic Wizard" and "schema-based Extension", then press Next
=> An empty Extension point selection is displayed.
NOTES:
DESCRIPTION:
NOTES: | critical | RESOLVED | WONTFIX |
{'$oid': '52e9b45754dc1c25ebdb203b'} | 343 | DBCS-PDE:Source Page for the Plug-in Manifest editor does not display correct DBCS char.
What was displayed look like correct DBCS char but in reality they are corrupted DBCs char.
Steps to recreate:
(Test case 2: Plug-in Development: Adding New Action Set extension point and New Top level menu)
-Create a plug-in project :File->New->Project... and selecting Plug-in Development, then Plug-in Project, press next.
- create Action Sets and Menu with DBCS in its label field.
- After changing the label to DBCS using properties view, check the source page and make sure that the characters are not corrupted.
Characters in editor (source page) was corrupted DBCS. Due to the corrupted characters, it couldn't be compiled.
Verify that the menu label shows up correctly in the menu bar in run-time workspace.
-
<?xml version="1.0" encoding="UTF-8"?>
<actionSet
label="(dbcs)"
visible="true"
id="j.actionSet2">
<menu
label="(dbcs)"
id="j.menu2"
...
=> In the above editor, where (dbcs) is used, DBCS characters should be displayed correctly in the editor as well as correct DBCS chars as menu label in the run-time window.
NOTES:
RTP Test: (E. Nair) Verified with 136a and 137. Fixed for all DBCS languages.
Project name in DBCS is also displayed properly. However Project name in Plugin is a limitation and will not be resolved when compiled. | critical | RESOLVED | FIXED |
{'$oid': '52e9b45754dc1c25ebdb203c'} | 344 | Plug-in : the <require> section, which is comprised of the basic user interfaces
and core services, is missing from the plug-in code for the Chinese languages.
Also, the an empty Extension point selection panel is being displayed.
This problem happened on W2K T. Chineses and W2K S. Chinese only.
W2K Japanese and W2K Korean worked fine.
Steps to recreate:
Step A: Missing <require> section.
-Create a plug-in project :File->New->Project... and selecting Plug-in Development, then Plug-in Project, press next.
-Name the project with some SBCS name "aaa", press next, then press next.
-Select "create plug-in using a template wizard" + Default plug-in Generator, press next, then press finish.
-From panel "aaa1" Plugin, select the "Source" tab at the bottom of screen
-The plugin source code is displayed as follow:
<?xml version="1.0" encoding="UTF-8"?>
<plugin
name="aaa Plugin"
id="aaa"
version="1.0.0"
class="aaa.aaaPlugin">
<requires>
<import plugin="org.eclipse.core.runtime"/>
<import plugin="org.eclipse.core.ressources"/>
<import plugin="org.eclipse.ui"/>
</requires>
<runtime>
<library name="aaa.jar"/>
</runtime>
=> The section <requires> ...</requires> is missing from for the both S.Chinese + T.Chinese stations.
It does display correctly on the Japanese, Korean and English stations.
StepB: Empty Extension point selection panel for S.Chinese + T.Chinese.
- At the bottom of the screen, turn to the "Extension" page
and push the new button to start the Extension wizard.
-Select "Generic Wizard" and "schema-based Extension", then press Next
=> An empty Extension point selection is displayed for the Chinese stations.
NOTES:
RTP TEST: (E. Nair) Fix Verified with 137 build. Closed. | critical | RESOLVED | WONTFIX |
{'$oid': '52e9b45854dc1c25ebdb203d'} | 345 | DBCS: Got "Exception in thread" error message when launch a Plug-in Project code
at "Runtime Workbench instance".
This problem happened on the S. Chinese and T.Chinese only.
Korean and Japanese stations worked fine.
Steps to recreate:
- Take any Plug-in project code,
- Launch it at "Runtime Workbench instance"
=> get the following error message on the console:
java.lang.NoClass DefFound Error:
NOTES:
RTP TEST: (E. Nair) Fix Verified with 137 build. Closed. (related defect, 1GKBEOU)
org/eclipse/core/launcher/UIMain
Exception in thread "main". | critical | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb203e'} | 816 | VK (5/17/01 4:56:51 PM)
PR placeholder to investigate performance issues reported by Greg Adams on 98 running disconnected
NOTES: | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b46c54dc1c25ebdb203f'} | 818 | KK (5/23/2001 4:42:37 PM)
If help is a fast view, and is hidden, pressing F1 from the Resource Navigator, and then choosing one of the link shows and then HIDES help fast view (Windows 2k).
If help is a fast view, and is shown, pressing F1 from the Topics Tree, and then choosing one of the link HIDES help fast view (Linux and Windows 2k).
NOTES:
KK (2001-05-31 13:08:00)
In driver 0.115:
If help is a fast view, and is hidden, pressing F1 from the Resource Navigator,
and then choosing one of the links shows and then HIDES help fast view (Windows 2k and Linux).
DB (6/1/2001 12:01:15 PM)
We have made a change (closing the F1 infopop BEFORE launching the fast view) and that seems
to solve the above problem.
We'll do more testing to make sure it covers all the scenarios.
KH (9/19/2001 11:53:39 AM)
Moving to Help for comment. Did your testing uncover a workbench PR? | normal | RESOLVED | INVALID |
{'$oid': '52e9b46c54dc1c25ebdb2040'} | 817 | Print topic tree in the left panel does not let you choose the printer.
Clicking the print button on the local toolbar however does let you choose a printer.
Both should have choice.
NOTES:
MF (5/22/01 12:47:23 PM)
Absolutely, but we are "somewhat" forced into not being able to support this.
Here is why: Print topic tree in the left panel prints the whole topic tree
(ie: the current topic and all it's children). In the current implementation,
these are actually printed each as separate print transactions (through
the IE embedded control). Poping up the "Printer Dialog" (for the first print
job for example) was not feasable because some versions (configuration)
of IE would not persist the dialog settings for the next invocation, and so
user selection would not be carried over for further print jobs.
But I will double check this issue again, maybe it can be fixed.
ps: we may, at some point, rely on platform printing APIs to do the nested printing
instead of the IE control. But we did not look into this since at the time, I believe, that
functionality was not there.
MF (6/4/01 2:25:14 PM)
This will be very confusing for the user.
Currently Nested printing will print to the last printer used by IE, and not the system printer.
This is the behavior of IDM_PRINT command. According to MS web pages, you can not
programatically set the printer to be used by the IE control.
We can pop-up the Printer dialog for the first transaction, and following transaction will
re-use the latest printer, but the problem is two-fold:
1) not all printer dialog settings are persistent.
2) some of the options do not make sense/ are confusing for nested printing.
eg: what does it mean to specify a range of pages for nested printing?
ARCH (7/11/01 12:21:50 PM)
Investigate but do not fix in 1.0 | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46c54dc1c25ebdb2041'} | 820 | - from Welcome view, hit Workbench User Guide
- it comes up, but nothing interesting is shown
- likewise for Java Development User Guide
NOTES:
RG (6/5/01 2:51:27 PM)
The current help api does not allow us to request anything finer than an info set.
We have requested better api.
KH (9/19/2001 2:30:17 PM)
Pls adise when/if better API is available. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb2042'} | 819 | I can select text in the contents area of the help view, but I cannot copy it.
Edit>Copy is not enabled and Ctrl-C or Ctrl-Insert does not work.
Given that example code is an important part of help, the fact that it
cannot be copied is a rather significant limitation.
NOTES:
KK (2001-06-06 11:33:38)
Ctrl-C used to work, and is broken again. Our IE control
does not receive this event. Moving PR to ITPUI.
Right click -> Copy still works.
There are issues in supporting Edit>Copy, so we do not plan on
supporting this way for the help view.
KH (6/6/2001 12:53:52 PM)
Ctrl-C is hooked as an accerator by the Workbench. The workbench should
be forwarding this to the active view. When your view is active it is your responsbility
to forward the event to the widgets you created. This includes active X controls that you
create. Pls verify that your view is getting the event. If it is not, then move the PR back
to the UI folder.
DB (6/8/2001 2:17:51 PM)
This is not a stop-ship and will be deferred.
The workaround is to use the IE5 context menu: select the text, and right click then "Copy". | normal | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb2043'} | 821 | - Perspective / Show View / Other... / Team / Resource History
- hit F1
- click on the down arrow
- it says "There are no more links."
Then why did it bother showing me the arrow?
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb2044'} | 822 | Open the help perpective.
Select a text
- If you popup menu, you see a copy option enabled.
- if you go to menu Edit/Copy, it is not enabled.
NOTES:
DB (6/12/2001 1:07:39 PM)
The help view contains two controls, the navigation tree and the IE5 browser (ActiveX web browser control).
Since we don't control the browser, we don't know when something is selected so copy can be enabled/disabled,
so the simplest solution was to just use the context menu.
If this is a bad practice, we can revisit it in the next release. | normal | RESOLVED | INVALID |
{'$oid': '52e9b46c54dc1c25ebdb2045'} | 823 | It looks like because of the emedded ActiveX web browser component in the help view, one can
get into some odd behavior when detaching the help view.
Basically, if you open the help view into the Resource perspective and detach it (i.e. separate window),
you can then cannot scroll the Getting Started view, or click on its links.
Sometimes the Eclipse menus themselves are not selectable.
I did not notice the problem when detaching other views.
NOTES:
NE (6/15/01 1:43:47 PM)
Dup of 1GFA3GO: ITPUI:ALL - Cannot restore focus to window | normal | RESOLVED | INVALID |
{'$oid': '52e9b46c54dc1c25ebdb2046'} | 825 | - create e second workbench window
- open the help perspective in it.
- go back to the other window
- request help.
observe: a help perspective opens in the current workbench window. The other help window is not reused.
Furthermore help doesn't honor the setting for perspective opening set in preferences. It always creates a new perspective
in the current active workbench window.
NOTES:
KH (9/14/2001 5:03:35 PM)
Moving to Help for comment. | normal | RESOLVED | INVALID |
{'$oid': '52e9b46c54dc1c25ebdb2047'} | 824 | This may be a duplicate. However ..
1. Invoke Help | Help Contents.
2. Drag Help View off window into detached window.
3. Click on window.
Result: Focus is not returned to window !!! None of the menus in the window are accessable,
and I need to close the help view before doing anything else.
NOTES:
KH (6/14/2001 12:35:44 PM)
SWT.Activate not being sent from ie control.
NE (6/15/01 1:38:59 PM)
On Win98, it comes back on the 3rd click.
NE (6/15/01 1:43:36 PM)
Readme doc:
99.1.1 Detaching help view can leave Workbench window unresponsive
When viewing help, if the help view is detached from the main Workbench window, it can leave the main window in an unresponsive state.
Closing the detached view makes the main window responsive again. | normal | RESOLVED | INVALID |
{'$oid': '52e9b46c54dc1c25ebdb2048'} | 826 | - select a section with several topics
- choose print from the section pane's pop-up menu
- they print out in a different order than listed
NOTES:
NE (7/16/01 9:51:02 AM)
Suspect it's iterating over a hashtable, which doesn't preserve order.
DB (8/17/2001 1:21:31 PM)
No, the code iterates correctly, but there is an issue with the print commands. I remeber Mazen answering
a similar PR somewhere. | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46c54dc1c25ebdb2049'} | 827 | FC (7/19/01 10:18:02)
Under linux, the outline of the help is shown, but not the contents of
the help files
NOTES:
DB (8/17/2001 1:20:52 PM)
The content is shown in the external Netscape browser. | normal | RESOLVED | INVALID |
{'$oid': '52e9b46c54dc1c25ebdb204a'} | 828 | In client using remote help, documents contain CSS link embedded twice.
Same problem with search results, having javascript inserted twice.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb204b'} | 829 | Adjustable font size for the help page will be a good feature, making Eclipse more accesible.
Honestly, I have a hard time reading the current font size...
NOTES:
DB: (8/17/2001 1:17:40 PM)
Unfortunately (fortnunately, for other cases:-) we have a dependency on display content using IE5.
The help system uses a CSS file to control the size of the fonts and other presention.
The current workaround is to change the properties in your Internet Explorer to ignore style sheets and
to use your desired fonts.
In the future, we will likely provide a couple of style sheets, with different font sizes. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb204c'} | 830 | MF (8/14/01 3:19:45 PM)
Just logging for future reference...
NOTES:
"Stephen Cooper" <Stephen@removethiswordtoreply.thecoopers.ws> wrote in message news:9l96r2$1v2i$1@news.boulder.ibm.com...
> 1) Good question, but no. Closing netscape & clicking on the same topic
> brings up Nestcape with the expected content.
> 2) I'm running Netscape 4.77. This isn't the version that ships with RH
> 7.1, but I ran their update utility and got it.
> 3) Since I generally use Konqueror instead of Netscape, Workbench almost
> always starts Netscape.
>
> Question back to you - can I change it so that help uses Konqueror instead?
> /Stephen
> Mazen Faraj wrote:
>
> > Stephen,
> > here is what comes to mind:
> >
> > 1) did you click on a Topic that has no content associated with it?
> > (I have to ask :-) ). For example, a the topic Concepts might not have any
> > html file
> > associated with it (ie: it is a container) and so clicking on it has no
> > effect.
> >
> > 2) what Netscape version do you have? Maybe we have an issue with
> > a specific release.
> >
> > 3) was Netscape running before you launched the Workbench? Try launching
> > the Workbench with Netscape not running. Do you see the same problem?
> >
> > hope this helps...
> > Mazen.
> >
> >
> >
> >
> >
> > "Stephen Cooper" <Stephen@removethiswordtoreply.thecoopers.ws> wrote in
> > message news:9l1bdk$1bki$1@news.boulder.ibm.com...
> >> I have the Workbench window open, the help window open and the netscape
> >> window open.
> >> I then select a different topic in the help window and nothing happens in
> >> netscape. I have to close netscape and let help launch a new netscape
> >> process to get the new help topic.
> >>
> >>
> >> I figured that the help was still in progress, I just wanted to put my
> >> two cents in - I like the up/next/prev idea from info.
> >> /Stephen
> >> Mazen Faraj wrote:
> >>
> >> > Stephen,
> >> > can you give us more info.
> >> > When you say "next time I select help", do you mean that
> >> > you had help perspective already open and then you selected Help ->
> >> > Help Content again? or do you mean that you selected a different topic
> >> > in the curernt help view and notyhing happened?
> >> >
> >> > more info please :-)
> >> >
> >> > As for TOC in the HTML file, the Help system on Linux is still under
> >> > design/re-design, and so things will certainly change.
> >> > Having a more rich HTML file to display the help is certainly an
> >> > option. Embedded browser (just like windows) is another.
> >> >
> >> > Hope this helps...
> >> > Mazen.
> >> >
> >> >
> >> >
> >> >
> >> >
> >> >
> >> > "Stephen Cooper" <Stephen@removethis.thecoopers.ws> wrote in message
> >> > news:9kt9q9$eiq$1@news.boulder.ibm.com...
> >> >> I'm running RedHat 7.1, Netscape 4.77, Workbench 0.9 beta.
> >> >> If I select a help topic, it brings Netscape up and displays the help.
> >> >> However, next time I select help, it doesn't change the page Netscape
> > is
> >> >> showing.
> >> >> Also, it would be nice in the HTML Help if you had links like TOC, Up,
> >> >> Previous topic & Next topic.
> >> >>
> >> >> /Coop | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b46c54dc1c25ebdb204d'} | 832 | | To: Brian Young/OTT/OTI@OTI, Dave Thomson/OTT/OTI@OTI,
| Erich Gamma/ZRH/OTI@OTI, Greg Adams/OTT/OTI@OTI, Jeff
| McAffer/OTT/OTI@OTI, John Wiegand/MIN/OTI@OTI, Kevin
| Haaland/OTT/OTI@OTI, Kevin McGuire/OTT/OTI@OTI, Mike
| Wilson/OTT/OTI@OTI, klicnik@ca.ibm.com
| cc: Jim des Rivieres/OTT/OTI@OTI
Date: 08:54:43 PM Yesterday
Subject: 2.0 - Problems with Javadoc - JDT (only)
What is it
Here is part of the information I mentioned in today's call. It is flagged
by an NL checking tool (CHKPII)
but it is general problem issues with our javadoc and not NL specific. It
generated wads of other
info which I have tried to filter out and make it easier to spot your
pieces.
Some web browsers (e.g. IE) are tolerant of these errors, others (i.e.
netscape 6 on linux) are less
tolerant but I can't see us stop shipping on these.
What you need to
1) Create pr for your pieces
2) Fix it -- we need to fix these problems in the 2.0 stream although I
suspect some teams may want
to address this sooner if its easier.
Where can you find the tool
Misc -> NL Tools -> chkpii & run it on *.htm
Ask BrianY if you need a hand running it.
When fixed
Please let releng know when you have fixed your problems.
Note: The above does not include pde or jdt. They will be in separate
files.
/Greg
NOTES:
VK (8/24/01 12:05:30 PM) result of CHKPII run. Ignore the CHARSET problems (should be handled in doclet)
ORG\ECLIPSE\HELP\
ICONTEXT.HTML HTML-40 833 Missing META tag. Required CHARSET value must be defined in this file.
IHELP.HTML HTML-40 813 Invalid HTML symbol: <. Line: 73
813 Invalid HTML symbol: >. Line: 73
833 Missing META tag. Required CHARSET value must be defined in this file.
IHELPTOPIC.HTML HTML-40 833 Missing META tag. Required CHARSET value must be defined in this file.
PACKAGE-FRAME.HTML HTML-40 833 Missing META tag. Required CHARSET value must be defined in this file.
PACKAGE-SUMMARY.HTML HTML-40 833 Missing META tag. Required CHARSET value must be defined in this file.
PACKAGE-TREE.HTML HTML-40 833 Missing META tag. Required CHARSET value must be defined in this file.
PACKAGE-USE.HTML HTML-40 833 Missing META tag. Required CHARSET value must be defined in this file. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46c54dc1c25ebdb204e'} | 831 | from newsgroup
From: "Dale King" <KingD@TCE.com>
Newsgroups: ibm.software.websphere.studio.workbench
Sent: Wednesday, August 15, 2001 7:26 PM
Subject: Another issue related to help system
> While we are on the subject of help and the internal web server, let me
> relate one of my annoyances with the use of a web server for the help. I
am
> using Eclipse on a laptop and I also use Outlook Express as my mail reader
> on that laptop to read newsgroups including this one. I typically do most
of
> my news reading and replying while disconnected from the network and
> synchronize when I get to work. Therefore when I am reading news I have it
> set to work off-line. Unfortunately the on-line vs. off-line setting for
> outlook express and IE are connected.
>
> So I have a conflict if I happen to be reading Eclipse help and reading
news
> at the same time, which I frequently do. To read Eclipse help I have to
> "connect" to be able to get to what it thinks is some server on the
network.
> But this takes outlook express out of off-line mode and it gives me errors
> when I try to post as it tries to contact the server. If I put outlook
> express in off-line mode then Eclipse is going to ask me to connect which
> puts outlook express back in on-line mode.
>
> In case you are wondering I have seen no correlation between this and the
> lock-ups, so I don't think this is related.
> --
> Dale King
NOTES:
DB (8/29/2001 3:48:34 PM)
If you install the loopback adapter in Win 2000 you can access your local
machine from the network even without a network cable. It is a software
device driver on the install CD from Microsoft.
"dorian" <birsan@ca.ibm.com> wrote in message
news:9lgl5g$tu0$1@news.boulder.ibm.com...
> Dale,
>
> Yes, it is unfortunate to have to alternate between off-line/on-line as
you
> swtich from Outlook to Eclipse, and I think we should try to find a better
> solution in the help system for this kind of scenario.
>
> Is there a way to have Outlook/Internet Explorer/Winsock be able to
> recognize local addresses and allow communication with them even in
off-line
> mode?
>
> Dorian
>
> "Dale King" <KingD@TCE.com> wrote in message
> news:9lf1om$170s$2@news.boulder.ibm.com...
> > While we are on the subject of help and the internal web server, let me
> > relate one of my annoyances with the use of a web server for the help. I
> am
> > using Eclipse on a laptop and I also use Outlook Express as my mail
reader
> > on that laptop to read newsgroups including this one. I typically do
most
> of
> > my news reading and replying while disconnected from the network and
> > synchronize when I get to work. Therefore when I am reading news I have
it
> > set to work off-line. Unfortunately the on-line vs. off-line setting for
> > outlook express and IE are connected.
> >
> > So I have a conflict if I happen to be reading Eclipse help and reading
> news
> > at the same time, which I frequently do. To read Eclipse help I have to
> > "connect" to be able to get to what it thinks is some server on the
> network.
> > But this takes outlook express out of off-line mode and it gives me
errors
> > when I try to post as it tries to contact the server. If I put outlook
> > express in off-line mode then Eclipse is going to ask me to connect
which
> > puts outlook express back in on-line mode.
> >
> > In case you are wondering I have seen no correlation between this and
the
> > lock-ups, so I don't think this is related.
> > --
> > Dale King
> >
>
> | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46c54dc1c25ebdb204f'} | 833 | Kevin Haaland mentioned a way of reusing NS on Linux that was used in VA/Java. Involves using a front-end script.
Should investigate as a possible short-term hack/ post to Eclipsecorner
NOTES: | enhancement | RESOLVED | WONTFIX |
{'$oid': '52e9b46c54dc1c25ebdb2050'} | 834 | - Keyboard navigation of the help system is clumsy. Say you have a topic
highlighted and the page on the right is larger than the screen. How do you
scroll it down without the mouse. The answer is you have to tab to higlight
that part of the window then hit page down. Now that you finished that
page, how do you get to the next? Answer tab around to higlight the tree of
topics and then hit the down arrow. That is a pain to read the help
documents.
NOTES: | enhancement | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb2051'} | 835 | We are missing an icon for the help view. When you select Perspectives->Customize->Other you will see there
is no icon for the Help Contents.
NOTES:
KK (9/10/2001 12:01:34 PM)
This is help action that is in the Help menu, not the help view
(help view is in Perspectives->Customize->Views and has an icon).
This help action does not have an associated icon. Do we need one?
DB (9/10/2001 1:21:48 PM)
If you decide an icon is needed, I can reuse the one use for the help view (a yellow bubble with ?) | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb2052'} | 836 | People using the Help perspective find it misleading that the print icon
and menu item are not enabled. Help should hook into the global print
action provided by the UI.
NOTES:
EJP(9/24/2001 5:35:57 PM)
Must add the following lines to EmbeddedHelpView.makeActions()
IActionBars actionBars = getViewSite().getActionBars();
if (printAction.isEnabled())
actionBars.setGlobalActionHandler(IWorkbenchActionConstants.PRINT, printAction);
PS: the copy action could be added as well:
if (copyAction.isEnabled())
actionBars.setGlobalActionHandler(IWorkbenchActionConstants.COPY, copyAction); | normal | RESOLVED | INVALID |
{'$oid': '52e9b46d54dc1c25ebdb2053'} | 837 | "It would be nice to ... "check" chapters that I have learned" - from EC.
This is in addition to needing help bookmarks.
NOTES: | enhancement | RESOLVED | WONTFIX |
{'$oid': '52e9b46d54dc1c25ebdb2054'} | 838 | If users follows link in help document, and then synchronizes with navigation,
it will be required to click "back" twice to return to the original document
NOTES:
This is caused by reloading the second document after synchronize
is pressed and corresponding topic selected in the navigation.
The event should not be sent to IE in that case. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb2055'} | 839 | A help api which takes an array of help contexts rather than a single
context adds a degree of complication.
A help context includes a description and any number of related
links. When passed more than one context for which help should be
shown, the help support system currently uses the additional contexts to
supplement its related links list. It does not show the additional descriptions.
Since a help context can be either an id or an IContext, clients can compose their
own IContext consisting of nested related links if they want to.
Thus an api taking a single context is simpler without sacrificing functionality.
NOTES:
DB (5/1/2001 4:48:34 PM)
This is a good idea. We'll need to discuss this for the next release. At this point is a bit too late to make API changes.
ARCH (7/11/01 12:25:44 PM)
Defer | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb2056'} | 840 | Help sniff test from
http://home.ott.oti.com/hubba/teams/releng/docs/codeDrops/SanityChecksForEclipse/index.html
1.1 Choose "Help"/"Help Contents" from the menu.
"Help" Perspective" should be created, with one view titled "Help Content Example"
containing navigation tree on the left, and initial HTML document on the right.
* View title is "Help" not "Help Context Example"
1.3 Press "F1" key.
Context Help should pop up, with description and related links.
* By clicking on the little down arrow icon in the info pop you see more related links.
There are two references to "embeddedHelpView Related topic 1".
When you click on the second link you get "This is embeddedHelpView context
related topic 2" in the contents area.
NOTES:
KK (2001-05-16 19:14:01)
The title of Help view is inconsistent.
It is called "Help" if help is launched from the menu,
or named after InfoSet name, if Help view created by chosing link in F1 help.
We should make it consistent (and update sniff test cases if necessary).
KK (2001-05-24 12:05:26)
one of "embeddedHelpView Related topic 1" links was renamed to
"embeddedHelpView Related topic 2" | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46d54dc1c25ebdb2057'} | 841 | In its current form, the help view is quite large and monolithic. The combination of a tree
and browser make it almost impossible to use this view in the same page as another view.
Unfortunately, there are many situations where you may want to browse help while editing a
document or navigating the workspace tree.
Suggestion:
Break the help view into two:
1. Help Topics.
2. Help Editor.
If the user selects an item within the Help Topics view open an editor. If they select a
second item in the topics just change the input for the open editor. I believe this
approach is taken with the java debugger to avoid the proliferation of too many editors
as you step through a program.
DB: (5/1/2001 5:38:20 PM)
This was discussed at some point, but it was felt that the help navigation and the html browser go together.
To address the large view area occupied, there is Show/Hide navigation button on the internal toolbar of the current help view.
You can toggle between just showing the documentation or showing both documentation and navigation.
If you feel strongly about the separation of the two views, please let me know and I'll raise the issue with
the UCD folks (actually, I will do it anyway, as previous UI designs can change).
Thanks!
Cool Feature:
"Lock Page": When invoked, the current help editor is locked to the current topic. If you
click in the Help Topics again a new editor is opened. Now you have the ability to look at
more than one page in the help doc.
Benefits:
Consistent mechanism for layout manipulation and persistance across desktop, including help.
Better integration between help view and others.
You can look at more than one page in the help content.
On platforms like Linux, where a browser component may not be available, the
html browser will open as an external editor. Result: No ugly empty
browser area in the help view.
Estimated Coding Time: 1 day.
NOTES:
DB (5/2/2001 9:40:57 PM)
There is one issue left, dealing with fast views. When help is a fast view, we cannot have both navigation
and browser inside the same "pane", as one will always go away.
I do, however, like the idea of having separate views for navigation and browser, and yes, it
should not take more than a day or so to code. Testing all the possible scenarios may be a bit
more difficult. | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46d54dc1c25ebdb2058'} | 842 | VK (5/14/01 10:50:44 AM)
The help handling of NL'ed doc webs needs to be changed to use the fragment APIs. Cannot rely on having all docs as part of
the plugins directory.
NOTES:
VK (7/11/01 9:45:35 AM)
Need to verify if this is still the case ... the help support should be calling the platform methods to locate NL-variants
of the doc files. The core api supports should now (R0.9) be searching the plugin and fragment directories.
DB (7/11/2001 11:08:56 AM)
We do correct plugin and fragment directories, but only for the current locale.
I opened a PR on core to have support for passing locale info to the Plugin.find() method.
1GGPF9N: ITPCORE:ALL - Need support for specifying the locale info in Plugin.find() | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb2059'} | 843 | PRODUCT VERSION:
105 jdk
DESCRIPTION:
(minor)
If help system cannot find the help it will give an error dialog indicating
"It appears that documentation is not installed on this machine. If it is installed on
a remote server, then from Windows -> Preferences...."
Should this give the user the option of jumping directly to the preference page.
NOTES:
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb205a'} | 844 | If help is shown (based on local install) and a remote help is set, switching to the remote
help should refresh the current opened help views.
NOTES: | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46d54dc1c25ebdb205b'} | 845 | Search should be completely contained in the search plugin.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb205c'} | 846 | When doc.properties is available in the .nl\ja and .nl\ja_JP, strings should be
looked-up in both. Currently, once a doc.properties has been found, the related ones (lang and default)
are no longer looked at.
NOTES: | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb205d'} | 848 | jkca (6/1/2001 12:10:45 PM)
jre-motif-sdk 116
The Javascript used to write the name of title pages
does not work on linux.
To Duplicate:
1. Start Eclipse
2. Switch to help perspective
3. The title page for JDT doc comes up with just an image across the top (no
text)
NOTES:
DB (6/1/2001 4:32:49 PM)
What browser do you have on the Linux machine?
It worked ok with Netscape 4.72, the version that comes by default in Red Hat
6.2
| normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b46d54dc1c25ebdb205e'} | 847 | Occassionally when printing a Topic Tree, some of the topics are reversed.
1) Right click on a tree
2) Print Topic Tree
3) Sometimes two adjacent pages will print in flipped order
NOTES:
MM (6/13/2001 3:31:52 PM)
The first time I tried this (Platform Plug-In Developers Guide -> Samples) it printed
Workbench - Property Sheet before Workbench - Multi-page Editor. Possibly others.
The second time, those two were correct, but other (seemingly random) examples
were printed in the wrong order.
MM (6/13/2001 3:36:20 PM)
Tried printing the Workbench topic tree, and it printed Property Sheet (third item)
even before the Table Of Contents. | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46d54dc1c25ebdb205f'} | 849 | The forward/back buttons should be disabled when there is no next/previous page in the history.
NOTES:
DB (6/7/2001 2:24:07 PM)
Unless you feel this PR is very important,we would like to defer it to a futher release. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46d54dc1c25ebdb2060'} | 850 | If help reuses Netscape Window, and 2 or more help perpectives are open,
the browser is shared between these perspectives.
If I select topic a in perspective A, then topic b in perspective B,
then switch back to A perspective, the browser displays topic b.
NOTES: | normal | RESOLVED | INVALID |
{'$oid': '52e9b46d54dc1c25ebdb2061'} | 851 | -In Help Perspective select PDE ISV guide
(Perspective-> Open-> Other-> Help; select PDE ISV)
-Click on "Introduction to PDE" page
-Hit Next Page button on local toolbar
BUG: nothing happens, next page is not displayed
NOTES:
KLM (6/4/2001 5:00:44 PM)
Looks like this is working as intended (more like a web browser)
I was expecting it to take me to the next page in the guide.
DB (6/7/2001 2:25:18 PM)
I changed the tooltip to say "Back" and "Forward", as this is the standard browser convention.
A next/previous type of behavior as you describe would be nice, I would treat that as a suggestion
for a further release of the code. | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b46e54dc1c25ebdb2062'} | 852 | -In Help Perspective select PDE ISV guide
(Perspective-> Open-> Other-> Help; select PDE ISV)
-Click on "Creating a plug-in" page
-Hit Previous Page button on local toolbar
The previous page is selected, but
BUG: the incorrect page is highlighted in the contents list
"Creating a plugin" is still highlighted, when "Setting up the Workbench"
is the page that should be highlighted
NOTES:
DB (6/4/2001 12:39:54 PM)
This is working as designed.
It was felt that navigation in the browser should not change the topics tree.
One can use the synchronize button on the help view toolbar to synchronize with navigation.
The behavior you suggest is very valid, and perhaps we will need to re-evaluate our decisions.
Similarly for the other PR (1GET07T) | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46e54dc1c25ebdb2063'} | 853 | Especially when Netscape is up, double clicking in the help navigation view
does not expand/collpase the tree.
NOTES: | normal | RESOLVED | INVALID |
{'$oid': '52e9b46e54dc1c25ebdb2064'} | 854 | When selecting remote install with no url (or incorrect URL ) specified
the browser return its own error message for browsing a topic.
The page should do a sanity test to the url specified before applying the changes.
NOTES: | normal | RESOLVED | INVALID |
{'$oid': '52e9b46e54dc1c25ebdb2065'} | 855 | MF (6/5/01 10:11:33 AM)
when we switch between local/ remote scenarios, we do not refresh the doc.properties files.
(and maybe context.properties should be looked at as well)
To reproduce, have same plugin, local and remote.
load remote, then swtich to local.
close perspective.
open same plugin again.
contents are picked up locally, but tree viewer still displays remote doc.properties.
NOTES:
KK (2001-06-05 10:22:32)
context.properties are always read from local first | normal | RESOLVED | FIXED |
{'$oid': '52e9b46e54dc1c25ebdb2066'} | 858 | Please respond to "Susan Griffin/OTT/OTI" <Susan_Griffin@oti.com>
To: Vlad Klicnik/Toronto/IBM@IBMCA, "Greg Adams/OTT/OTI" <Greg_Adams@oti.com>
cc:
Subject: bug in help server!
The help server could not resolve a path that was:
http://127.0.0.1:3293/org.eclipse.jdt.doc.isv/questions-index/index.html
it could resolve a path that was:
http://127.0.0.1:3293/org.eclipse.jdt.doc.isv/questions/index.html
It would seem the presence of the '-' character messes it up.
I can work around, but it took me awhile to figure it out.
susan
NOTES:
KK (2001-06-08 11:02:22)
I could not reproduce the problem. There are files in documentation
that contain "-" in the names and paths, that work fine.
I have created the file that you had problem with (included it in
manifest xml file). I was able navigate and view the file.
If it still does not work for you, send me your complete plugin,
and I will take a look. | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b46e54dc1c25ebdb2067'} | 859 | - open help perspective
- click on Hide Navigation button
- shutdown
- restart
- it comes back showing the Navigation pane
I did this while the help view was detached, but that should not matter.
NOTES:
DB (6/11/2001 1:42:01 PM)
If this is not a must fix (we don't think it is) we will defer it to a future release. | enhancement | RESOLVED | INVALID |
{'$oid': '52e9b46e54dc1c25ebdb2068'} | 857 | AK (6/5/01 7:07:04 PM)
note: i use internet explorer 2.0
1. open the getting started page
2. hit the Workbench User Guide link
3. note: no busy cursor ! <<BUG 1
4. an error message show up saying that i don't have a correct version of ie installed.
but it does not tell me which version _is_ correct. << BUG 2
5. i get the following walkback in the console << BUG 3
Check to see if the correct version of Internet Explorer is installed on your sy
stem.
org.eclipse.swt.SWTException: Class ID not found in registry result = -214722100
5
at org.eclipse.swt.ole.win32.OLE.error(OLE.java:304)
at org.eclipse.swt.ole.win32.OleClientSite.getClassID(OleClientSite.java
:631)
at org.eclipse.swt.ole.win32.OleControlSite.<init>(OleControlSite.java:8
0)
at org.eclipse.help.internal.ui.win32.HelpControlSite.<init>(HelpControl
Site.java:28)
at org.eclipse.help.internal.ui.win32.WebBrowser.<init>(WebBrowser.java:
102)
at org.eclipse.help.internal.ui.win32.BrowserFactory.createBrowser(Brows
erFactory.java:27)
at org.eclipse.help.internal.ui.HTMLHelpViewer.createControl(HTMLHelpVie
wer.java:56)
at org.eclipse.help.internal.ui.HTMLHelpViewer.<init>(HTMLHelpViewer.jav
a:35)
at org.eclipse.help.internal.ui.EmbeddedHelpView.createPartControl(Embed
dedHelpView.java:107)
at org.eclipse.ui.internal.PartPane$2.run(PartPane.java:54)
at org.eclipse.core.internal.runtime.InternalPlatform.run(InternalPlatfo
rm.java(Compiled Code))
at org.eclipse.core.runtime.Platform.run(Platform.java(Compiled Code))
at org.eclipse.ui.internal.PartPane.createChildControl(PartPane.java:52)
at org.eclipse.ui.internal.PartPane.createControl(PartPane.java:96)
at org.eclipse.ui.internal.ViewPane.createControl(ViewPane.java:108)
at org.eclipse.ui.internal.PartSashContainer.createControl(PartSashConta
iner.java:163)
at org.eclipse.ui.internal.PerspectivePresentation.activate(PerspectiveP
resentation.java:78)
at org.eclipse.ui.internal.Perspective.onActivate(Perspective.java:499)
at org.eclipse.ui.internal.WorkbenchPage.onActivate(WorkbenchPage.java:8
27)
at org.eclipse.ui.internal.WorkbenchWindow$5.run(WorkbenchWindow.java:87
2)
at org.eclipse.swt.custom.BusyIndicator.showWhile(BusyIndicator.java(Com
piled Code))
at org.eclipse.ui.internal.WorkbenchWindow.setActivePage(WorkbenchWindow
.java:860)
at org.eclipse.ui.internal.WorkbenchWindow.busyOpenPage(WorkbenchWindow.
java:281)
at org.eclipse.ui.internal.WorkbenchWindow.access$6(WorkbenchWindow.java
:269)
at org.eclipse.ui.internal.WorkbenchWindow$4.run(WorkbenchWindow.java:64
1)
at org.eclipse.swt.custom.BusyIndicator.showWhile(BusyIndicator.java(Com
piled Code))
at org.eclipse.ui.internal.WorkbenchWindow.openPage(WorkbenchWindow.java
:638)
at org.eclipse.help.internal.ui.DefaultHelp.getHelpViewInHelpPerspective
(DefaultHelp.java:223)
at org.eclipse.help.internal.ui.DefaultHelp.getHelpViewInCurrentPerpecti
ve(DefaultHelp.java:167)
at org.eclipse.help.internal.ui.DefaultHelp.displayHelp(DefaultHelp.java
:132)
at org.eclipse.ui.internal.dialogs.WelcomeItem.openHelpTopic(WelcomeItem
.java:91)
at org.eclipse.ui.internal.dialogs.WelcomeItem.access$0(WelcomeItem.java
:88)
at org.eclipse.ui.internal.dialogs.WelcomeItem$1.run(WelcomeItem.java:13
1)
at org.eclipse.swt.custom.BusyIndicator.showWhile(BusyIndicator.java(Com
piled Code))
at org.eclipse.ui.internal.dialogs.WelcomeItem.triggerLinkAt(WelcomeItem
.java:129)
at org.eclipse.ui.internal.dialogs.WelcomeEditor$1.mouseUp(WelcomeEditor
.java:71)
at org.eclipse.swt.widgets.TypedListener.handleEvent(TypedListener.java(
Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled C
ode))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled C
ode))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compil
ed Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled
Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compile
d Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:620)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoa
der.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:69)
at org.eclipse.core.launcher.Main.run(Main.java:311)
at org.eclipse.core.launcher.Main.main(Main.java:198)
NOTES:
EG (6/5/2001 9:18:48 AM)
moving to ITPUI
KH (6/5/2001 3:25:25 PM)
Moving to TOR
MF (6/7/01 3:47:46 PM)
OK.
For bug 1, I *do* get the hour glass. So can you please be more specific if you are doing some
special steps?
For bug 2: I have added the IE version to the pop-up message.
For bug 3: yes, we do stack dump, but we stack dump to our own .log file, not console.
Note that we are using the Platform's "log" object, and the platform by default has a global
log listener, that will listen to us.
So the only way that I can see where you will see us stack dumping to the console, it would be if
you have started the platform with logging directed to standard out.
And even then, it is *not* us stack dumping, we would still be logging to our .log file in this
scenario. It is the platform that is listening to us, and deciding to use standard out instead.
So the point is, can you please let us know if you had platform logging directed to standard
out ?
MF (6/7/01 4:52:12 PM)
Bug2: fixed in build > 0.119 | normal | RESOLVED | FIXED |
{'$oid': '52e9b46e54dc1c25ebdb2069'} | 856 | MF (6/5/01 12:10:21 PM)
we silently ignore two context objects that have the same id.
This should be at least logged at the info debug level.
NOTES:
MF (6/5/01 12:11:49 PM)
to reproduce, in same contexts file, cut and paste same context contribution,
and test. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46e54dc1c25ebdb206a'} | 860 | Navigating to url /pluginid/ causes Null Pointer Exception to be thrown
inside help server, and reported on command line. Eclipse and Help
continue to work.
NOTES:
KK (2001-06-11 13:45:20)
The url passed is not valid help topic id, hence should not appear
in the navigation nor links inside documents.
It is low severity defect. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46e54dc1c25ebdb206b'} | 861 | When the workbench help is set to be off the web and its URL is invalid, UnknownHostException is not caught.
Log: Wed Jun 13 13:34:05 EDT 2001
4 org.eclipse.help.ui 0 Could not copy the infoset data from server
java.net.UnknownHostException: wwwwwwwwwwwwwwwwww.ca
at java.net.InetAddress.getAllByName0(InetAddress.java:585)
at java.net.InetAddress.getAllByName0(InetAddress.java:554)
at java.net.InetAddress.getByName(InetAddress.java:463)
at java.net.Socket.<init>(Socket.java:112)
at sun.net.NetworkClient.doConnect(NetworkClient.java:56)
at sun.net.www.http.HttpClient.openServer(HttpClient.java:347)
at sun.net.www.http.HttpClient.openServer(HttpClient.java:534)
at sun.net.www.http.HttpClient.<init>(HttpClient.java:282)
at sun.net.www.http.HttpClient.<init>(HttpClient.java:292)
at sun.net.www.http.HttpClient.New(HttpClient.java:304)
at sun.net.www.protocol.http.HttpURLConnection.connect(HttpURLConnection.java:393)
at sun.net.www.protocol.http.HttpURLConnection.getInputStream(HttpURLConnection.java:487)
at java.net.URL.openStream(URL.java(Compiled Code))
at org.eclipse.help.internal.navigation.InfosetsMap.restore(InfosetsMap.java:60)
at org.eclipse.help.internal.navigation.HelpNavigationManager.getInfoSetIds(HelpNavigationManager.java:190)
at org.eclipse.help.internal.navigation.HelpNavigationManager.<init>(HelpNavigationManager.java:65)
at org.eclipse.help.internal.HelpSystem.getNavigationManager(HelpSystem.java:110)
at org.eclipse.help.internal.ui.DefaultHelp.displayHelp(DefaultHelp.java:111)
at org.eclipse.help.internal.ui.DefaultHelp.displayHelp(DefaultHelp.java:104)
at org.eclipse.help.internal.Help.displayHelp(Help.java:87)
at org.eclipse.help.internal.ui.ShowHelp$1.run(ShowHelp.java:67)
at org.eclipse.swt.custom.BusyIndicator.showWhile(BusyIndicator.java(Compiled Code))
at org.eclipse.help.internal.ui.ShowHelp.run(ShowHelp.java:61)
at org.eclipse.ui.internal.PluginAction.run(PluginAction.java:125)
at org.eclipse.jface.action.ActionContributionItem.handleWidgetSelection(ActionContributionItem.java:408)
at org.eclipse.jface.action.ActionContributionItem.handleWidgetEvent(ActionContributionItem.java(Compiled Code))
at org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java(Compiled Code))
at org.eclipse.jface.action.ActionContributionItem$ActionListener.handleEvent(ActionContributionItem.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Widget.notifyListeners(Widget.java(Compiled Code))
at org.eclipse.swt.widgets.Display.runDeferredEvents(Display.java(Compiled Code))
at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java(Compiled Code))
at org.eclipse.ui.internal.Workbench.run(Workbench.java:620)
at org.eclipse.core.internal.boot.InternalBootLoader.run(InternalBootLoader.java:815)
at org.eclipse.core.boot.BootLoader.run(BootLoader.java:285)
at java.lang.reflect.Method.invoke(Native Method)
at org.eclipse.core.launcher.Main.basicRun(Main.java:126)
at org.eclipse.core.launcher.Main.run(Main.java:408)
at org.eclipse.core.launcher.Main.main(Main.java:281)
Log: Wed Jun 13 13:34:11 EDT 2001
4 org.eclipse.help.ui 0 It appears that documentation is not installed on this machine. If it is installed on a remote server, then from Window -> Preferences, select the Help page and configure the required parameters.
NOTES:
KK (2001-06-14 11:02:50)
I cannot reproduce uncought exception.
This exception is caught by our help code, and is logged as part of the error handling/reporting.
MF (6/14/01 4:46:57 PM)
We are assuming that platform logging was directed to console for this stack dump to be seen.
If this is not the case please let us know.
WC(14/06/2001 7:28:49 PM)
The stack trace was logged to .log and it still occurs in 123 SDK. I didn't try it in VAJ though. | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46e54dc1c25ebdb206c'} | 862 | Right clicking on a topic under Contents and choosing "Print Topic Tree"
doesn't use the default Windows printer (and doesn't prompt)
1) Open Help perspective
2) choose a topic to print
3) Right click and Print Topic Tree
4) Click "OK" to confirm printing
5) Walk across the building to the first printer configured in Windows and
pick up your document.
NOTES:
MM (6/13/2001 3:34:14 PM)
After selecting the printer from the "Print page" button, it seems to print to
that printer all the time, even between restarts of Eclipse.
MF (6/14/01 12:48:55 PM)
This is duplicate to:
1GE4V5U: ITPHELP:ALL - print has inconsisent behavior
Currently Nested printing will print to the last printer used by IE, and not the system printer.
This is the behavior of IDM_PRINT command. According to MS web pages, you can not
programatically set the printer to be used by the IE control.
We can pop-up the Printer dialog for the first transaction, and following transaction will
re-use the latest printer, but the problem is two-fold:
1) not all printer dialog settings are persistent.
2) some of the options do not make sense/ are confusing for nested printing.
eg: what does it mean to specify a range of pages for nested printing? | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46e54dc1c25ebdb206d'} | 863 | The name displayed in the Eclipse Help perspective for a given topic
isn't shown anywhere on the printed page.
1) Print a topic
2) The name displayed when browsing the tree doesn't show up anywhere on the page.
NOTES:
MM (6/13/2001 3:40:04 PM)
A good example of this is Platform Plug-In Developers Guide -> Workbench ->
Property View. The title of the HTML page is "Desktop - Property Sheet Example"
(which appears at the top of the printed page), and has a large title in the body
of the HTML reading "Example - Property Sheet". Nowhere on the paper does it
say "Property View".
KK (2001-06-14 10:01:27)
It is up to the ID authors to label the topic appropriately in the tree.
The label might match the document title, which should be contained
within a body document (help does not expose the title),
or might be completely different, and make sense only in the tree
contexts
Ex. 3 consequtive topics labeled: "Exporting Project", "on Windows",
and "on Linux". It does not make sense to use these labels when
topic does not appear in the context. We use them hovewer,
in the table of contents, when printing topic sub tree.
The printing solution is not perfect, but we are not going to redesign
it for this version.
MF (6/14/01 11:13:18 AM)
Just to make things clear here. The table of contents printed is the tree that you see
in the left navigation view. Since the help system code does not rely on nor display the title
of a page, the title of the page often should be the actual title in the tree.
If we revisit printing, then this might change. | normal | RESOLVED | FIXED |
{'$oid': '52e9b46e54dc1c25ebdb206e'} | 864 | MF (6/14/01 4:32:36 PM)
Two defects have poped up against the Help System reporting that exceptions are not being caught
and are being stack dumped to standard out. I have referenced the two defects below, and I am
assuming that when these defects where opened, the platform logging was directed to the console.
The help system uses the Eclipse "log" concept, and we do stack dump the trace to our own log file.
and the errors *are* being handled. But it is scaring people when they see these stack dumps (with
logging directed to console).
So I guess my question is: is it ok to see these stack dumps (of course only if logging is directed to console)?
or is there some standard being followed that no stack dumps should happen if things are being properly
handled?
We stack dump only for debuggnig purposes. Since we do not use error codes, the same error can be reached
from two code paths, and it's better to save the stack dump.
Please move this PR to the Help system inbox if action needs to be taken.
NOTES:
MF
1GF9SDU: ITPUI:WIN2000 - (minor)UnknownHostException: Help
1GEUXRR: ITPJUI:ALL - exception in 'getting started' page
JM (6/15/2001 9:34:17 AM)
In general, exceptions, handled or otherwise, should not be
happening. If we hide the fact that they are, the responsible parties will not be held responsible and the
underlying problems will be fixed.
Note that people do not normally run with -consolelog and the console log is just a mirror of the real log.
Not much the core can do for this. Moving ot Help. | enhancement | RESOLVED | WORKSFORME |
{'$oid': '52e9b46e54dc1c25ebdb206f'} | 866 | MF (7/12/01 10:38:10 AM)
while looking at
1GE4V5U: ITPHELP:ALL - print has inconsisent behavior
I noticed a potential for a UI deadlock when doing Nested printing. While it *is* very unlikely that
it would happen, that error checking code should have been there.
NOTES:
The fix is not at all major. Simply more escape code to prevent deadlock. | normal | RESOLVED | WONTFIX |
{'$oid': '52e9b46e54dc1c25ebdb2070'} | 865 | 1) Open Help perspective
2) From the topic combobox select "Java Development User Guide"
3) Drill down to "Tasks > Using the Scrapbook > Creating a Scrapbook Page"
4) move mouse over link "See Running & Debugging"
Observe: link changes feedback; so it seems to be possible to follow it
5) Click on the link "See Running & Debugging"
Observe: nothing happens
6) Click again
Observe: now the link is followed and a new page is opened
NOTES: | normal | RESOLVED | DUPLICATE |
{'$oid': '52e9b48854dc1c25ebdb2071'} | 1,518 | JGS (8/8/01 5:20:19 PM)
We need enabled, disabled & hover icons for the following actions:
CopyToClipboardActionDelegate
RelaunchActionDelegate
TerminateAndRemoveActionDelegate
DW (9/24/2001 2:22:48 PM)
Use the standard "copy" icon for copy to clipboard (desktop likely exposes it).
DW (9/24/2001 2:23:05 PM)
Made requests for
Relaunch
Terminate All
Terminate & Remove | enhancement | VERIFIED | FIXED |
{'$oid': '52e9b48854dc1c25ebdb2072'} | 1,520 | DW (2/26/01 5:56:36 PM)
Could allow users to use styled text in console, with preference support. This would
allow console input to be shown in bold, etc. | normal | CLOSED | WONTFIX |
{'$oid': '52e9b48854dc1c25ebdb2073'} | 1,519 | JGS (6/4/01 11:43:47 AM)
Set a breakpoint in InfiniteLoop with a hit count. When suspended,
right click breakpoint in Breakpoints viewer and enable. Hit count
stays at previous value. Should it be reset to zero?
DW (6/4/01 11:50:23 AM)
This should just re-enable the breakpoint with the previous hit-count.
DS (6/4/2001 11:50:31 AM)
To change a hit count breakpoint into a normal breakpoint, you set its hit count to zero.
This should likely be a readme item.
DW (6/4/01 6:33:43 PM)
Make into a README.
DW (6/15/01 9:45:55 AM)
Added to README. | normal | RESOLVED | WORKSFORME |
{'$oid': '52e9b48854dc1c25ebdb2074'} | 1,521 | JGS (5/21/01 2:09:10 PM)
To be consistent, should we represent StringBuffer objects in the variables view like
we do Strings, with their values shown on the top node, so the user doesn't have to
expand the object and look at the char array?
DW (5/21/01 6:33:02 PM)
Deferred - requires an evaluation. | normal | VERIFIED | FIXED |