Kepler: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362009-01-22T21:36:55ZEcoinformatics Redmine
Redmine Bug #3770 (Resolved): CompositeActor "Open Actor" menu item is missinghttps://projects.ecoinformatics.org/ecoinfo/issues/37702009-01-22T21:36:55ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>On my linux command line build, and my mac eclipse build, the 'Open Actor' context menu is missing for the CompositeActor. kepler r16420, ptII r52162. Debi's also seeing this problem, but Ben is not. I suspect this may be related to the build system.</p> Bug #3735 (Resolved): R actor produces 'null' message when R not found on PATHhttps://projects.ecoinformatics.org/ecoinfo/issues/37352009-01-08T19:14:09ZMatt Jonesjones@nceas.ucsb.edu
<p>Edward Lee reports thatthe R actor produces a 'null' error message when one of the demo workflows that requires R is run on a system lacking R.</p>
<p>The R actor (and other external execution actors like Matlab) should check to see if the R executable is available and the R system functioning before it tries to execute. Maybe this can be done on init() so that it happens before a workflow is run.</p>
<p>When R is not found, a sensible error message should be displayed, one that preferably points the user to an easy solution of the problem.</p>
<p>An alternative (and probably better) approach is to have the R actor formally specify a dependency on the R system executable for the local platform, so that when the R actor is installed, the module installer checks that its dependencies are all satisfied. This would be possible under the new modularization system that includes the ability to formally specify dependency graphs in module metadata.</p> Bug #3696 (Resolved): Core Build System Taskshttps://projects.ecoinformatics.org/ecoinfo/issues/36962008-12-09T18:17:47ZChad Berkleyberkley@nceas.ucsb.edu
<p>Tasks with an X are done.</p>
<p>X-get rid of env vars<br />X-get checkout working<br />X-nuke kepler/src<br />X-mv build file to kepler/trunk<br />X-search for module in multiple locations<br />X-allow for shortcuts in targets (i.e. preconfigured build)<br />X-move actors/directors<br />X-create targets to create a new module<br />X-clean up root directory<br />-move installer targets from old build<br />-create targets to create a new actor<br />-re-enable JNI<br />-get eclipse build working<br />-make SVN checkout more verbose<br />-fix typo in checkout-if-not-exists: [checkout-if-not-exists] set property moduleName =ptII<br />-update the build-ptolemy.xml file to be more optimized (see christopher's notes)<br />-set the default task to show the other targets<br />-update version on the spash screen<br />-merge exp files to ptII</p> Bug #3530 (Resolved): Locking assertion failurehttps://projects.ecoinformatics.org/ecoinfo/issues/35302008-10-15T22:44:39ZAaron Aaronaschultz@nceas.ucsb.edu
<p>Specifically, I encountered this issue on Ubuntu version 8.04.1 (Hardy Heron) with dual displays using java 1.5</p>
<p>I get several backtrace messages to the console that begin with "Locking assertion failure. Backtrace:" <br />The following site describes the solution in more detail<br /><a class="external" href="http://ubuntuforums.org/showthread.php?t=610457">http://ubuntuforums.org/showthread.php?t=610457</a></p>
<p>The command I used to solve this issue is<br />sudo sed -i 's/XINERAMA/FAKEEXTN/g' /usr/lib/jvm/java-1.5.0-sun-1.5.0.15/jre/lib/i386/xawt/libmawt.so</p> Bug #3471 (Resolved): custom input boxes in actor configuration dialoghttps://projects.ecoinformatics.org/ecoinfo/issues/34712008-08-15T16:06:19ZMatt Jonesjones@nceas.ucsb.edu
<p>Report from Peter Reutemann:</p>
<p>I've encountered a similar problem as mentioned in the following thread:<br /> <a class="external" href="http://groups.google.com/group/comp.soft-sys.ptolemy/browse_thread/thread/dc4c136eb6870d10/75e79a2afd7cc9ab">http://groups.google.com/group/comp.soft-sys.ptolemy/browse_thread/thread/dc4c136eb6870d10/75e79a2afd7cc9ab</a></p>
<p>Though using Kepler, I needed to modify the underlying Ptolemy<br />framework in order to allow custom input boxes in actor dialogs. But<br />instead of hacking the ptolemy.gui.Query and<br />ptolemy.actor.gui.PtolemyQuery classes whenever I'd need another<br />custom query box, I modified those two classes (and added<br />classes/interfaces) to make it a bit more flexible.</p>
<p>In order to get a new query box into Ptolemy, you need to modify the<br />following places:<br />- ptolemy.gui.Query<br /> ~ set(String,String)<br /> ~ getStringValue(String)</p>
<pre><code>in those two methods the type of custom query chooser, e.g.,<br /> QueryColorChooser is used to determine how to obtain/set the<br /> string representation.</code></pre>
<p>- ptolemy.actor.gui.PtolemyQuery<br /> ~ addStyledEntry(Settable)</p>
<pre><code>here the type of Parameter defines what query chooser is added<br /> for the attribute, e.g., a QueryColorChooser in case of a<br /> ColorAttribute.</code></pre>
<p>To decouple this a bit, I've come up with an interface called<br />"CustomQueryBoxParameter", to be implemented by types derived from<br />Parameter that need a custom query box. The interface consists of only<br />one method:<br /> javax.swing.Box createQueryBox(PtolemyQuery query, Settable att)<br />This method returns a custom query box (fully configured), based on<br />the current query and attribute this query belongs to. This swing<br />element contains all the necessary logic to manipulate the data, e.g.,<br />a button for popping up another dialog, etc. The addStyledEntry method<br />of the PtolemyQuery class obtains a box from an attribute implementing<br />CustomQueryBoxParameter and attaches it (via the _addPair method).</p>
<p>The query box being returned, is a custom class derived from<br />javax.swing.Box, which implements another interface called<br />"SettableQueryChooser". This interface is used in the<br />set/getStringValue methods of ptolemy.gui.Query to set/get the current<br />string value. The actual query chooser class turns that into whatever<br />internal representation it uses. This interface avoids having to add<br />new logic to the Query class, whenever a new custom QueryChooser is<br />implemented.</p>
<p>Since I'm apparently not the only user who encountered this problem,<br />I'd like to contribute my code (which will hopefully trickle down to<br />Kepler at one stage, in case it gets accepted). Not exactly sure how<br />contributions work around here.</p>
<p>Note: my original post <strong>with</strong> attachments didn't come through<br />apparently (never showed up in the Google group), you can find my<br />original post including the attachments on my homepage:<br /> <a class="external" href="http://www.scms.waikato.ac.nz/~fracpete/downloads/ptolemy/custom_query_boxes/">http://www.scms.waikato.ac.nz/~fracpete/downloads/ptolemy/custom_query_boxes/</a></p>
<p>There are several files available for download from this web page:<br />- ptolemy-patch.zip<br /> contains patches for the Query/PtolemyQuery classes<br /> (run against subversion repository August 14, 2008)<br />- ptolemy-additional_classes.zip<br /> additional classes/interfaces necessary for my proposed changes<br />- examples-classes.zip<br /> contains a simple actor that uses a custom query box with a button<br /> that pops up a JOptionPane for the user to enter some text; this is<br /> just to demonstrate how the slightly modified framework would work<br />- examples.kar<br /> a kepler archive - I know, this is not exactly the right place, but<br /> I thought some Kepler people/developers read this mailing list as well<br /> and might be interested in this</p>
<p>Hopefully, my ramblings made some sense, otherwise the code hopefully<br />speaks for itself. ;-)</p> Bug #3300 (Resolved): CompositeActor does not support multiporthttps://projects.ecoinformatics.org/ecoinfo/issues/33002008-05-16T23:06:25Zben leinfelderleinfelder@nceas.ucsb.edu
<p>You can create a CompositeActor with input and output ports that are set as Multiports, but only the first value on the port is consumed.<br />Kevin ran into this when creating the river flow workflow for kruger.</p> Bug #3236 (Resolved): remove strict requirements for PTII and KEPLER env variableshttps://projects.ecoinformatics.org/ecoinfo/issues/32362008-04-21T17:09:34ZChad Berkleyberkley@nceas.ucsb.edu
<p>From an email from matt on kepler-dev:</p>
<p>I agree, it would be better if we could build without requiring that <br />JAVA_HOME is set. For that matter, I would prefer if we didn't need to <br />set KEPLER or PTII either. For KEPLER, we can use the ant basedir <br />property which is always set to the directory where the build is <br />launched -- this would be a simple change. Its not even clear to me <br />that env.KEPLER is used anywhere except in the javadoc target, and it <br />would be easily replaced with either a relative path or ${basedir}.</p>
<p>For PTII, I think we should check if PTII is set, and if so, check it <br />and use that dir, but if not then see if ptii is found in a sibling dir <br />of $KEPLER, and if so use that. If both of those fail, then abort with <br />an error saying we can't find ptii. By checking for the environment <br />variable first, we allow people to use different versions of PTII while <br />still making the default case where it is checked out in a sibling <br />directory the default.</p>
<p>These changes would significantly simplify our build and eliminate the <br />most frequently reported build problem that is reported to us.</p>
<p>Matt</p> Bug #3215 (Resolved): Spaces in KEPLER path causes build problemshttps://projects.ecoinformatics.org/ecoinfo/issues/32152008-04-09T00:59:04ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>Aaron Schultz reports - when trying to build under windows (in eclipse) with spaces in the -DKEPLER variable, Config.XML.saveDOM gives an error that lib/config.xml does not exist. Spaces are replaced with %20 and that's not working for windows. When -DKEPLER and -DPTII are changed to not have spaces kepler builds.</p> Bug #3150 (Resolved): Need to add PTII Python actors back into Keplerhttps://projects.ecoinformatics.org/ecoinfo/issues/31502008-02-14T19:39:17ZDan Higginshiggins@nceas.ucsb.edu
<p>Ptolemy has 2 different actors that allow one to use Python scripts. These were removed from Kepler when all the actors were instantiated at startup because they greatlyt slowed down startup (by scanning all the jar files).</p>
<p>Several Kepler users have asked about using Python scripts in Kepler. Now that all actors are not instantiated at startup, the Python actors can be put back into Kepler, giving an additional scripting capability.</p>
<p>Dan Higgins - Feb 2008</p> Bug #3130 (Resolved): Add an accelerator (keyboard shortcut) for opening a new blank workflowhttps://projects.ecoinformatics.org/ecoinfo/issues/31302008-02-05T21:28:19ZDerik Barseghianbarseghian@nceas.ucsb.edu
<p>Add an accelerator for opening a new blank workflow canvas (File => New Workflow => Blank).</p>
<p>Unless 'n' is taken I will use it. e.g. apple-n will be the accelerator on a mac for this action.</p> Bug #3061 (Resolved): Need IPCC data by monthhttps://projects.ecoinformatics.org/ecoinfo/issues/30612008-01-11T20:06:58ZDan Higginshiggins@nceas.ucsb.edu
<p>The ClimateFileProcessor and ClimateChangeFileProcessor actors are custom actors for reading IPCC climate data. The data is stored in a single file with 12 grids, one for each month. Currently these actors return either annual or seasonal values by averaging the monthly data.</p>
<p>The ability to return the data for a specific month has been requested, rather than averaging over several months or a year.</p> Bug #3056 (Resolved): unable to print output of Display actorhttps://projects.ecoinformatics.org/ecoinfo/issues/30562008-01-09T19:04:59ZKevin Drurydrury@nceas.ucsb.edu
<p>The Display actor receives summary output from an RExpression actor. The summary displays perfectly, but does not print. The summary also appears normally in the text editor available from the Display Tools menu, but also does not print from there.</p> Bug #3002 (Resolved): Copy/Paste of a relation (diamond)https://projects.ecoinformatics.org/ecoinfo/issues/30022007-11-13T22:56:44ZNorbert Podhorszkipnorbert@cs.ucdavis.edu
<p>The pasted diamond symbol completely overlaps the original one. <br />In contrast, other pasted elements are relocated a bit, so that you are aware <br />of the fact that there are two of them.</p>
<p>Great possibility for novice users to screw up workflows at tutorial hands-ons (e.g. SC07 last Sunday), believing they are connecting new actors to an existing relation while they connect them to a new relation (without input link). It is hard to find the error afterwards (SDF would complain about disconnected graphs but not the other models).</p> Bug #2941 (Resolved): Command Line Exec Actor throws exception when given input on "arguments" porthttps://projects.ecoinformatics.org/ecoinfo/issues/29412007-09-07T05:17:50ZDan Homerickdanhomerick@gmail.com
<p>Providing a constant (string constant or constant) to arguments port generates the exception:</p>
<p>"Attempt to get token from an empty QueueReceiver in model.Command Line Exec.arguements"</p>
<p>Additionally, none of the test workflows in kepler/workflow/test/spa test the arguments port.</p> Bug #2870 (Resolved): Copy/Paste of composite actorshttps://projects.ecoinformatics.org/ecoinfo/issues/28702007-06-07T21:52:47ZNorbert Podhorszkipnorbert@cs.ucdavis.edu
<p>There is a composite actor A, in which a parameter P is referred to.<br />A Copy/Paste into another workflow (or some other subparts in the same workflow) where parameter P does not exists (we get an error message) puts<br />all elements of the composite into the same 0 coordinate. <br />This is very annoying and makes reuse of workflows/actors much harder than it should be.</p>
<p>I am used to drag the elements back into their place, but I am impatient to do that above 20 elements. Also I used to create parameters (if I am aware of them) before copying, but this is also cumbersome.<br />Just think about text-code programmming and copy-paste of complete loops, and then renaming the variables according to the context.</p>