Kepler: Issueshttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362005-12-09T18:54:54ZEcoinformatics Redmine
Redmine Bug #2314 (Resolved): composites show up as classes not instanceshttps://projects.ecoinformatics.org/ecoinfo/issues/23142005-12-09T18:54:54ZChad Berkleyberkley@nceas.ucsb.edu
<p>when you drag a composite to the canvas, a purple box shows up surrounding the<br />actor indicating that the actor is a class that needs to be instantiated. This<br />is happening because when the actor is added to the library dynamically, the<br />moml parser is serializing the actor as a class instead of an entity. Even<br />though the composite is, in fact, a class (in the moml def), it should be<br />declared as an entity in the actor library if you want it to show up as an<br />instance on the canvas. I can't figure out how to convince the moml parser to<br />create an entity instead of a class. I've written to pthackers and kepler-dev<br />about this. hopefully someone will have an idea.</p> Bug #2300 (Resolved): migrate to new web services-only ecogrid client in Keplerhttps://projects.ecoinformatics.org/ecoinfo/issues/23002005-11-29T20:37:40ZMatt Jonesjones@nceas.ucsb.edu
<p>Kepler currently uses the version of the EcoGrid client based on Grid service<br />APIs. THat client will be changed to utilize web service only APIs (see bug<br />2299). When that is complete we need to change Kepler so that it utilizes this<br />new Web Services-only client jar.</p> Bug #2290 (Resolved): EML2 Data Source problem with higgins.5087.3 - 'BaselineLayers'https://projects.ecoinformatics.org/ecoinfo/issues/22902005-11-28T21:09:34ZDan Higginshiggins@nceas.ucsb.edu
<p>To reproduce this bug, do a datasearch for 'BaselineLayers'. This shold give the<br />knb package 'higgins.5087.3'. The data for this package is a zip file stored on<br />the SRB; the data is treated as a single attribute table (i.e. not really a<br />table). When this datasource is dragged to the work area, a NullPointerException<br />is thrown 'at org.kepler.objectmanager.cache.DataCacheObject.getFileName'.</p> Bug #2277 (Resolved): Simplify Kar loadinghttps://projects.ecoinformatics.org/ecoinfo/issues/22772005-11-16T18:18:23ZKevin Rulandkruland@ku.edu
<p>For first release, simplify the kar loading process to operate on moml documents<br />only. Each moml file will describe a single actor.</p> Bug #2269 (Resolved): SVG - Small Icon (Actor Library Thumbnail)https://projects.ecoinformatics.org/ecoinfo/issues/22692005-11-11T20:05:32ZMatthew Brookebrooke@nceas.ucsb.edu
<p>getIcon() method in XMLIcon works, but its success seems to be linked to the odd<br />tree refreshing behavior described in bug <a class="issue tracker-1 status-3 priority-2 priority-default closed" title="Bug: unexpected workflow components tree behavior (Resolved)" href="https://projects.ecoinformatics.org/ecoinfo/issues/1843">#1843</a>, which sends multiple calls to<br />getIcon().</p>
<p>When 1843 gets fixed, we may be able to make getIcon() more efficient/intuitive,<br />and this may in turn save some memory usage (see bug <a class="issue tracker-1 status-3 priority-5 priority-highest closed" title="Bug: SVG - Memory Usage (Resolved)" href="https://projects.ecoinformatics.org/ecoinfo/issues/2267">#2267</a>). Another way to save<br />memory and improve app response time is to assign gif/png thumbnail icons to the<br />actors, since these are already available in docs/dev/usability/graphics/ - it<br />seems like an unnecessary use of resources to use Batik to render these thumbnails.</p> Bug #2268 (Resolved): SVG - Backward Compatibilityhttps://projects.ecoinformatics.org/ecoinfo/issues/22682005-11-11T19:59:38ZMatthew Brookebrooke@nceas.ucsb.edu
<p>- Need to add code so old SVG icons are translated to have their origin at the<br />top-left corner. They are currently cropped when displayed in batik, because<br />Ptolemy's default origin for svg components is at the center of each actor icon,<br />whereas the svg/batik standard is to have the origin at top-left.</p>
<p>Another possibility (and one which would temporarily help with the memory usage<br />problems - see bug <a class="issue tracker-1 status-3 priority-5 priority-highest closed" title="Bug: SVG - Memory Usage (Resolved)" href="https://projects.ecoinformatics.org/ecoinfo/issues/2267">#2267</a>) would be to continue using PTII's SVG rendering for<br />the old "simple" svg icons, and use batik only for the new icons.</p> Bug #2266 (Resolved): SVG - Assigning Iconshttps://projects.ecoinformatics.org/ecoinfo/issues/22662005-11-11T19:49:48ZMatthew Brookebrooke@nceas.ucsb.edu
<p>Icons can be assigned in actor moml, or can be assigned in the actor's java<br />code. Assigning in moml means icons will show up for newly-created workflows,<br />but <strong>not</strong> for old workflows. [Note that some actors can't be set via moml (eg<br />EML200 datasource), so would need to add some to java code anyway]</p>
<p>Adding icon paths directly to the actors' java code (note that existing svg<br />icons are currently defined in java code) means icons show up for all actors, in<br />new <strong>and</strong> old workflows. Could move actual actor->svg icon mappings into a<br />separate settings file? Or would this cause too much indirection and add<br />complication?</p>
<hr />
<p>conf call of 11/10/05 - decided to put assignments in actor code, with actual<br />mappings in a separate config file to allow for easy changes</p> Bug #2255 (Resolved): Simple Composite Actor cannot be dragged onto displayhttps://projects.ecoinformatics.org/ecoinfo/issues/22552005-11-08T17:29:37ZDan Higginshiggins@nceas.ucsb.edu
<p>The simple, empty 'Composite Actor' does not appear in the actor tree because it<br />has been 'commented out' in the actorList.moml file. Uncommenting it makes it<br />appear in the actor tree, but then dragging it to the right to place it in a<br />workflow creates an error! ("Property is not an instance of Attribute...").</p>
<p>Being able to simply create a Composite Actor by dragging it to the work area<br />seems to be fairly important.</p> Bug #2253 (Resolved): ENM Workflows errors on newer, faster computerhttps://projects.ecoinformatics.org/ecoinfo/issues/22532005-11-07T17:32:51ZDan Higginshiggins@nceas.ucsb.edu
<p>Errors are occurring in 'SingleSpecies_BestRuleSet_Aug2005 ENM workflow on new,<br />faster computer (Optiiplex GX620 - 3GHz dual core - Win XP) that do not occur<br />when same workflow is run on older, slower machine (null pointer error most of<br />time). Error seems to occur at end of GARP run; perhaps a file is not ready?</p>
<p>Possible threading problem?</p> Bug #2243 (Resolved): bug in inverse distance interpolation in Grid Rescaling actorhttps://projects.ecoinformatics.org/ecoinfo/issues/22432005-11-04T17:22:58ZDan Higginshiggins@nceas.ucsb.edu
<p>An attempt to smooth the IDW interpolation near NO_DATA edges of a grid has<br />resulted in errors in interpolation (that are clearly visible as a 'waffle' type<br />background in images).</p>
<p>Reason for this bug is currenly unclear. Needs to be fixed for use in ENM workflow.</p> Bug #2235 (Resolved): Search text-field only 1 character wide on Mac OS Xhttps://projects.ecoinformatics.org/ecoinfo/issues/22352005-10-25T22:41:14ZMatthew Brookebrooke@nceas.ucsb.edu
<p>On the left-hand side of the Kelper UI, the text field in which the user types<br />search terms is only 1 character wide when displayed on Mac OS X - this is<br />because the buttons are much wider than those on windows/linux</p> Bug #2233 (Resolved): Name duplication exception with KAR fileshttps://projects.ecoinformatics.org/ecoinfo/issues/22332005-10-19T22:07:28ZEfrat Jaegerjaeger@ecoinformatics.org
<p>The new kar system throws a name duplication exception when two kars <br />implemented by (perhaps completely) different actors share the same name. We'd <br />like to have both actors, so we probably need some kar's naming conventions.</p> Bug #2141 (Resolved): resultset cache expiration problemhttps://projects.ecoinformatics.org/ecoinfo/issues/21412005-07-05T21:31:30ZChad Berkleyberkley@nceas.ucsb.edu
<p>The cache needs to be fixed so that resultsets expire and are purged according<br />to a set policy. This was discussed at the 05 june kepler meeting. efrat and<br />jing agreed to fix it.</p> Bug #2108 (Resolved): changes to simple searchhttps://projects.ecoinformatics.org/ecoinfo/issues/21082005-06-10T17:57:54ZLaura Downeyldowney@lternet.edu
<p>1) remove "text" and "concept" checkboxes<br />2) change "go" to "search" <br />3) make "enter" key execute "search" not "reset"</p> Bug #2025 (Resolved): provide workflow and results in some publishable formathttps://projects.ecoinformatics.org/ecoinfo/issues/20252005-03-11T20:45:26ZLaura Downeyldowney@lternet.edu
<p>provide a workflow "package" of the workflow and its results in a publishable <br />format -- this would be more extensive than just the natural language summary <br />of a workflow referred to in bug 2024</p>