Kepler: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362014-03-04T12:24:32ZEcoinformatics Redmine
Redmine Bug #6439 (Closed): Double rounding fails in some cases while evaluating Expressionshttps://projects.ecoinformatics.org/ecoinfo/issues/64392014-03-04T12:24:32ZOwsiak Michalmichal.owsiak@man.poznan.pl
<p>It seems that addition of doubles can produce values slightly different than they should to be.</p>
<p>Please take a look at attached workflow (simple_error.xml).</p>
<p>Condition that should be satisfied to escape the loop is: 1.7 > 1.5 + 0.1</p>
<p>However, loop is interrupted sooner, because of incorrect calculation of doubles. Value of "p" is set to: 1.6000000000000003</p>
<p>This, of course, makes it impossible to use doubles as check points for the loops.</p>
<p>However, it seems that casting to string and back works fine (take a look at second workflow - simple.xml)</p>
<p>Cheers</p>
<p>Michal</p> Bug #5564 (Resolved): dragging sdf director to the canvas causes exceptionhttps://projects.ecoinformatics.org/ecoinfo/issues/55642011-12-07T00:08:19ZJing Taotao@nceas.ucsb.edu
<p>I searched sdf director and got it on the left panel. Then I tried to drag it to the canvas, but i got the exception:<br />com.microstar.xml.XmlException: Cannot create entity without a class name. in [external stream] at line 2 and column 22<br /> at ptolemy.moml.MoMLParser._checkForNull(MoMLParser.java:3992)<br /> at ptolemy.moml.MoMLParser._createEntity(MoMLParser.java:4173)<br /> at ptolemy.moml.MoMLParser.startElement(MoMLParser.java:2694)<br /> at com.microstar.xml.XmlParser.parseElement(XmlParser.java:922)<br /> at com.microstar.xml.XmlParser.parseContent(XmlParser.java:1105)<br /> at com.microstar.xml.XmlParser.parseElement(XmlParser.java:925)<br /> at com.microstar.xml.XmlParser.parseContent(XmlParser.java:1105)<br /> at com.microstar.xml.XmlParser.parseElement(XmlParser.java:925)<br /> at com.microstar.xml.XmlParser.parseDocument(XmlParser.java:481)<br /> at com.microstar.xml.XmlParser.doParse(XmlParser.java:159)<br /> at com.microstar.xml.XmlParser.parse(XmlParser.java:132)<br /> at ptolemy.moml.MoMLParser.parse(MoMLParser.java:1538)<br /> at ptolemy.moml.MoMLParser.parse(MoMLParser.java:1510)<br /> at ptolemy.moml.MoMLParser.parse(MoMLParser.java:1666)<br /> at ptolemy.moml.MoMLChangeRequest._execute(MoMLChangeRequest.java:289)<br /> at ptolemy.kernel.util.ChangeRequest.execute(ChangeRequest.java:171)<br /> at ptolemy.kernel.util.NamedObj.executeChangeRequests(NamedObj.java:738)<br /> at ptolemy.kernel.util.NamedObj.requestChange(NamedObj.java:1778)<br /> at ptolemy.actor.CompositeActor.requestChange(CompositeActor.java:1922)<br /> at ptolemy.vergil.basic.EditorDropTargetListener.drop(EditorDropTargetListener.java:369)<br /> at java.awt.dnd.DropTarget.drop(DropTarget.java:430)<br /> at sun.awt.dnd.SunDropTargetContextPeer.processDropMessage(SunDropTargetContextPeer.java:500)<br /> at sun.awt.X11.XDropTargetContextPeer.processDropMessage(XDropTargetContextPeer.java:165)<br /> at sun.awt.dnd.SunDropTargetContextPeer$EventDispatcher.dispatchDropEvent(SunDropTargetContextPeer.java:812)<br /> at sun.awt.dnd.SunDropTargetContextPeer$EventDispatcher.dispatchEvent(SunDropTargetContextPeer.java:736)<br /> at sun.awt.dnd.SunDropTargetEvent.dispatch(SunDropTargetEvent.java:30)<br /> at java.awt.Component.dispatchEventImpl(Component.java:4270)<br /> at java.awt.Container.dispatchEventImpl(Container.java:2116)<br /> at java.awt.Component.dispatchEvent(Component.java:4243)<br /> at java.awt.LightweightDispatcher.retargetMouseEvent(Container.java:4322)<br /> at java.awt.LightweightDispatcher.processDropTargetEvent(Container.java:4057)<br /> at java.awt.LightweightDispatcher.dispatchEvent(Container.java:3911)<br /> at java.awt.Container.dispatchEventImpl(Container.java:2102)<br /> at java.awt.Window.dispatchEventImpl(Window.java:2440)<br /> at java.awt.Component.dispatchEvent(Component.java:4243)<br /> at java.awt.EventQueue.dispatchEvent(EventQueue.java:599)<br /> at java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:273)<br /> at java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:183)<br /> at java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:173)<br /> at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:168)<br /> at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:160)<br /> at java.awt.EventDispatchThread.run(EventDispatchThread.java:121)</p> Bug #5549 (Resolved): Help=>Modules Documentation not working in ubuntuhttps://projects.ecoinformatics.org/ecoinfo/issues/55492011-11-21T23:30:42ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>If you click on e.g. reporting from Modules Documentation, you get an error that isn't helpful.<br />It's coming from module-manager ActiveModulesDocumentationPanel, which is not finding a browser to use to open the file.</p> Bug #5507 (Resolved): Windows Installer: The R choice is not unselectablehttps://projects.ecoinformatics.org/ecoinfo/issues/55072011-10-06T01:39:19ZChristopher Brookscxh@eecs.berkeley.edu
<p>When installing under Windows, I'm given a choice of which components<br />to install: Kepler and the R interface.</p>
<p>However, the R interface checkbox is not selectable, so I have to install<br />it.</p>
<p>Why do we offer the user the illusion of choice if they can't make<br />a choice?</p>
<p>This is minor and in no way a blocker, so feel free to retarget.</p> Bug #5506 (Resolved): Windows Installer Page does not mention 2.3?https://projects.ecoinformatics.org/ecoinfo/issues/55062011-10-06T01:36:30ZChristopher Brookscxh@eecs.berkeley.edu
<p>When I start up the Windows Installer, the page says Kepler-2.3, but<br />the date is August 12, 2011.<br />Also, the page says <br />"The 2.0 release contains a major architectural overhaul from previous releases"</p>
<p>I would remove that text, it is rather obvious.</p>
<p>Could there be text about what is new in 2.3?</p>
<p>These are minor suggestions.</p> Bug #5505 (Resolved): Kepler-2.3 installer should create directory named "Kepler-2.3", not "Kepler"https://projects.ecoinformatics.org/ecoinfo/issues/55052011-10-06T00:34:59ZChristopher Brookscxh@eecs.berkeley.edu
<p>When installing <a class="external" href="https://code.kepler-project.org/code/kepler/releases/test-installers/Kepler-2.3.dmg">https://code.kepler-project.org/code/kepler/releases/test-installers/Kepler-2.3.dmg</a><br />I get a window that prompts me to drag the Kepler icon to the applications folder.</p>
<p>The folder should be named "Kepler-2.3", not "Kepler", so that it<br />is obvious which version is installed and so that multiple versions of <br />Kepler may be installed.</p>
<p>If the user already has a Kepler folder, what happens?</p>
<p>This is a feature request, not a bug, but I find it odd to have applications<br />that have no notion of version number in the directory name.</p> Bug #5502 (Resolved): use https to svn check out moduleshttps://projects.ecoinformatics.org/ecoinfo/issues/55022011-10-03T22:48:50ZDaniel Crawldanielcrawl@gmail.com
<p>It would be nice if the build system used https when getting modules.</p> Bug #5500 (Resolved): unable to open kar in relaxed mode saved in patched 2.2.0 in 2.3.0https://projects.ecoinformatics.org/ecoinfo/issues/55002011-10-01T05:03:31ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>There was a bug w/ the version checking logic whereby the kar would show as being cleanly-openable in a suite with version greater than what it was saved by.<br />The old logic required the major, minor and macro to <strong>all</strong> be greater than or equal to the compared numbers.</p> Bug #5494 (Resolved): existence of remote 2.2 patches can cause 2.3 to fail to start 2.2 on first...https://projects.ecoinformatics.org/ecoinfo/issues/54942011-09-20T00:32:14ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>Now two patches to 2.2 exist: loader-2.1.1 and module-manager-gui-2.2.1.<br />If you launch 2.3.0 from the 2.3 app and attempt to use it to switch back to kepler-2.2, if you have not previously downloaded the patches mentioned above from within 2.2, 2.2 will fail to start on this initial attempt. The error on console is:<br />9/19/11 4:34:20 PM [0x0-0xfdafda].org.kepler.build.runner.Kepler<sup><a href="#fn38570">38570</a></sup> [null] Error: The following modules are missing: <br />9/19/11 4:34:20 PM [0x0-0xfdafda].org.kepler.build.runner.Kepler<sup><a href="#fn38570">38570</a></sup> [null] loader-2.1.1<br />9/19/11 4:34:20 PM [0x0-0xfdafda].org.kepler.build.runner.Kepler<sup><a href="#fn38570">38570</a></sup> [null] module-manager-gui-2.2.1</p>
<p>A subsequent start up of the 2.3 app does what was supposed to happen the first time: start 2.2, immediately prompting you to download the two available patches.</p>
<p>During the Module Manager restart process something seems to have knowledge of remote patches and mistakenly assumes they've been downloaded.</p> Bug #5488 (Resolved): domain/username/password in command line hadn't been parsed in kepler when ...https://projects.ecoinformatics.org/ecoinfo/issues/54882011-09-08T23:56:39ZJing Taotao@nceas.ucsb.edu
<p>When I used the command line to run a workflow which would access some private data, i found the domain/username/password arguments in command line were not used. So the execution failed since kepler thought the public user wanted to access those data.</p>
<p>The reason is the authentication path was done earlier than the path which parsed the domain/username/password in KeplerConfigurationApplication class. I changed the order and the bug was fixed.</p> Bug #5455 (Resolved): Documentation for 2.3.0https://projects.ecoinformatics.org/ecoinfo/issues/54552011-08-12T01:38:38ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>Documentation needs to be done for the 2.3.0 release. <br />The getting-started-guide needs updating, at the very least to mention that Java 1.6 is now required. This should have been done for 2.2.0.<br />The UserManual, at the very least, needs a new date on title page.<br />I think the Actor Reference and Guide to ENM can again go unchanged as we've done for the last two releases.</p> Bug #5430 (Resolved): problems with multiple instances of keplerhttps://projects.ecoinformatics.org/ecoinfo/issues/54302011-06-25T00:51:54ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>If you start two instances of kepler running at the same time on one machine, and quit the first instance, the second instance will begin having problems like being unable to search the Components library, unable to instantiate new components, etc. It appears the connection to the database is closed.</p> Bug #5254 (Resolved): provenance database connections keep growing when using Reporting suite.https://projects.ecoinformatics.org/ecoinfo/issues/52542010-12-10T01:02:15Zjianwu jianwujianwu@sdsc.edu
<p>Every time a window is open, or a workflow is saved, running, a separate db connection is gotten. but they are not released when a window is closed. It will cause memory usage goes up.</p>
<p>We guess WorkflowRunManager.disconnect() should call SQLQueryV8.disconnect() to close the connection. Currently, SQLQueryV8.disconnect() is not called by any classes.</p>
<p>If we can reuse existing connections. The performance will be better than creating and closing a new one.</p>
<p>This is one stack trace for db connection when a window is open:</p>
<p>DatabaseFactory.getConnectedDatabaseType(Map<String,String>, String) line: 154 <br />SQLQueryV8.<init>(Map<String,String>) line: 77 <br />WorkflowRunManager.connect() line: 135 <br />WorkflowRunManagerPanel.reconnect() line: 960 <br />WorkflowRunManagerPanel.resetProvenanceInfo(ProvenanceStore) line: 937 <br />WorkflowRunManagerPanel.initializeTab() line: 231 <br />TabPaneFactory.createTabPaneTabs(TableauFrame) line: 110 <br />TabManager.initializeTabs(TableauFrame) line: 87 <br />KeplerGraphFrame._initBasicGraphFrame() line: 377 <br />KeplerGraphFrame(BasicGraphFrame).<init>(NamedObj, Tableau, LibraryAttribute) line: 224 <br />KeplerGraphFrame(ExtendedGraphFrame).<init>(CompositeEntity, Tableau, LibraryAttribute) line: 95 <br />KeplerGraphFrame(ActorGraphFrame).<init>(CompositeEntity, Tableau, LibraryAttribute) line: 135 <br />KeplerGraphFrame.<init>(CompositeEntity, Tableau, LibraryAttribute) line: 180 <br />KeplerGraphTableau.<init>(PtolemyEffigy, String, LibraryAttribute) line: 164 <br />KeplerGraphTableau$Factory.createTableau(Effigy) line: 227 <br />PtolemyTableauFactory.createTableau(Effigy) line: 100 <br />TableauFactory.createTableau(Effigy) line: 127 <br />Configuration.createPrimaryTableau(Effigy) line: 569 <br />TableauFrame$1.actionPerformed(ActionEvent) line: 519</p> Bug #5186 (Resolved): Publish example KARs to repositoryhttps://projects.ecoinformatics.org/ecoinfo/issues/51862010-09-16T21:10:39ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>The two KARs on the kepler library were published inadvertently, and may and should be removed, because they are not good examples.</p>
<p>We should also publish some good examples (using 2.1, so they are KAR 2.1) for users to see.<br />Matt suggests:<br />REAP examples<br />SANPark examples<br />some useful components</p>
<p>Publishing a run kar that includes a nice report (sanparks?) would be good too.</p> Bug #4483 (Resolved): Module dependencies in MoML fileshttps://projects.ecoinformatics.org/ecoinfo/issues/44832009-10-21T23:15:30ZAaron Aaronaschultz@nceas.ucsb.edu
<p>To reproduce: <br />Save a KAR from the WRP suite<br />change to the kepler suite<br />on startup you see the error: <br />Error encountered in: <property name="TOP Provenance Recorder" class="org.kepler.provenance.ProvenanceRecorder"><br />etc....</p>
<p>I used to think this was an issue with the startup building of the library and that dependencies in the KARs was the problem. But it turns out to be much more malicious. It is actually a dependency from the MoML file itself. Because the WRP suite has the Provenance recorder included in the moml, when the provenance module is not around the error is displayed and the workflow can not be used. So this means that any moml file generated by the WRP suite cannot be run in the Kepler suite. Unlike in Reporting where the reporting files are separate from the MoML itself and therefore the MoML can still run in the Kepler suite.</p>
<p>So the task here is to figure out how to add a dependency on the provenance module to all the workflows that are created from the WRP suite. The ability to do this I think is not built into the KAR design and I'll have to figure out how to do it.</p>