<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent changes to bugs</title><link href="https://sourceforge.net/p/openjgraph/bugs/" rel="alternate"/><link href="https://sourceforge.net/p/openjgraph/bugs/feed.atom" rel="self"/><id>https://sourceforge.net/p/openjgraph/bugs/</id><updated>2005-01-24T06:26:00Z</updated><subtitle>Recent changes to bugs</subtitle><entry><title>Reference to Queue class is ambiguous</title><link href="https://sourceforge.net/p/openjgraph/bugs/15/" rel="alternate"/><published>2005-01-24T06:26:00Z</published><updated>2005-01-24T06:26:00Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net3a994fe3d4f7e3820088697070a5d062d01fe82a</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;I was using Java 1.5. It has java.utill.Queue which&lt;br /&gt;
conflicts with salvo.jesus.util.Queue. The error I got is &lt;/p&gt;
&lt;p&gt;.\salvo\jesus\graph\algorithm\BreadthFirstTraversal.java:19:&lt;br /&gt;
reference to Queue is ambiguous, both class&lt;br /&gt;
salvo.jesus.util.Queue in salvo.jesus.util and class&lt;br /&gt;
java.util.Queue in java.util match&lt;br /&gt;
private Queue queue;&lt;/p&gt;
&lt;p&gt;when I tried with jdk 1.4 it worked fine&lt;/p&gt;
&lt;p&gt;Regards,&lt;br /&gt;
Jayesh Jain&lt;/p&gt;
&lt;p&gt;jjjain@buffalo.edu&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Class Cast Exception in LayeredTreeLayout</title><link href="https://sourceforge.net/p/openjgraph/bugs/14/" rel="alternate"/><published>2003-08-06T13:15:04Z</published><updated>2003-08-06T13:15:04Z</updated><author><name>Bruno C</name><uri>https://sourceforge.net/u/userid-838944/</uri></author><id>https://sourceforge.net96c498372fca0057953d97e52d2500998f1fd790</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;on Linux (redHat 8):&lt;/p&gt;
&lt;p&gt;This bug happens when you push the Layout button with &lt;br /&gt;
a tree loaded from a File.&lt;/p&gt;
&lt;p&gt;How to reproduce the bug:&lt;br /&gt;
Build a Tree through a java class (i.e let's consider&lt;br /&gt;
the SampleTreeLayout class). &lt;br /&gt;
Save it as &amp;amp;quot;tmp.xgmml&amp;amp;quot;. &lt;br /&gt;
Push the Layout button&amp;amp;gt;&amp;amp;gt;everything is ok&lt;/p&gt;
&lt;p&gt;Now, open the file you just saved: &amp;amp;quot;tmp.xgmml&amp;amp;quot;.&lt;br /&gt;
Push the Layout button&lt;br /&gt;
You'll get a Class Cast exception:&lt;/p&gt;
&lt;p&gt;This happens 'cause there's a casting from Graph to Tree:&lt;/p&gt;
&lt;p&gt;&amp;amp;quot;tree=(Tree) this.vgraph.getGraph();&amp;amp;quot;&lt;/p&gt;
&lt;p&gt;This instruction is totally safe when the Tree is&lt;br /&gt;
builded from an external class (i.e SampleTreeLayout),&lt;br /&gt;
but it gives a systematic Class Cast Exception if the&lt;br /&gt;
Tree is builded from an xgmml file...&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Loading edge labels from XGMML</title><link href="https://sourceforge.net/p/openjgraph/bugs/13/" rel="alternate"/><published>2003-05-14T16:44:00Z</published><updated>2003-05-14T16:44:00Z</updated><author><name>Andrew Fairley</name><uri>https://sourceforge.net/u/afairley/</uri></author><id>https://sourceforge.net04ee21b52fdccc3188afc21f7b015f72d221c005</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Using 0.9.2, it looks like edge labels aren't being loaded &lt;br /&gt;
from XGMML files.&lt;/p&gt;
&lt;p&gt;Changing XGMMLContentHandler::instantiateEdge() to &lt;br /&gt;
check for a label attribute, and if defined, call both &lt;br /&gt;
setLabel() and setFollowVertextLabel( false ) correctly &lt;br /&gt;
loads the label.&lt;/p&gt;
&lt;p&gt;I've attached a tarball that includes both an example &lt;br /&gt;
XGMML file that isn't loaded correctly in 0.9.2, and the &lt;br /&gt;
modified XGMMLContentHandler.java which will load the &lt;br /&gt;
edge labels.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>compileall.sh doesn't work on tcsh</title><link href="https://sourceforge.net/p/openjgraph/bugs/12/" rel="alternate"/><published>2003-02-19T13:09:09Z</published><updated>2003-02-19T13:09:09Z</updated><author><name>Michael Behrisch</name><uri>https://sourceforge.net/u/behrisch/</uri></author><id>https://sourceforge.net811d1424e7e75c905b4769b934f81a8a05f9af06</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;There is a small typo in the first line of &lt;br /&gt;
compileall.sh preventing it from running correctly. &lt;br /&gt;
Correction is appended &lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>wrong usage of == for object comparison</title><link href="https://sourceforge.net/p/openjgraph/bugs/11/" rel="alternate"/><published>2002-06-28T01:06:29Z</published><updated>2002-06-28T01:06:29Z</updated><author><name>Ramesh Pidikiti</name><uri>https://sourceforge.net/u/pidix/</uri></author><id>https://sourceforge.net5eaf0b6849e450c716551e5cb61ebf799d2cc918</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;In salvo.jesus.graph.EdgeImpl class getOppositeVertex( &lt;br /&gt;
Vertex v ) method uses the == for object comparison. I &lt;br /&gt;
think we should use &amp;amp;quot;equals&amp;amp;quot; method for correct results.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>parsing of xgmml element &amp;lt;graph&amp;gt;</title><link href="https://sourceforge.net/p/openjgraph/bugs/10/" rel="alternate"/><published>2002-02-11T13:43:49Z</published><updated>2002-02-11T13:43:49Z</updated><author><name>Knut Wannheden</name><uri>https://sourceforge.net/u/d96knut/</uri></author><id>https://sourceforge.net50f92d0438e4d303efeacdac0aa69d5dc312f9e3</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;When trying to load a simple XGMML document from &lt;br /&gt;
within the editor a NullPointerException is raised in &lt;br /&gt;
salvo.jesus.graph.xml.XGMMLContentHandler.startGraphEle&lt;br /&gt;
ment(XGMMLContentHandler.java:153).&lt;/p&gt;
&lt;p&gt;The problem seems to be that XGMMLContentHandler &lt;br /&gt;
expects a &amp;amp;quot;Graphic&amp;amp;quot; attribute in the &amp;amp;lt;graph&amp;amp;gt; element.&lt;br /&gt;
According to the specification &lt;br /&gt;
(http://www.cs.rpi.edu/~puninj/XGMML/draft-&lt;br /&gt;
xgmml.html#GraphA) this attribute is optional &lt;br /&gt;
(#IMPLIED).&lt;/p&gt;
&lt;p&gt;I have verified this with the two first sample &lt;br /&gt;
documents given in the XGMML Draft Specification &lt;br /&gt;
(http://www.cs.rpi.edu/~puninj/XGMML/draft-&lt;br /&gt;
xgmml.html#StructXGMMLDoc).&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Graph.createEdge  - deprecation warning</title><link href="https://sourceforge.net/p/openjgraph/bugs/9/" rel="alternate"/><published>2002-01-24T23:30:17Z</published><updated>2002-01-24T23:30:17Z</updated><author><name>Jürgen Failenschmid</name><uri>https://sourceforge.net/u/jfai/</uri></author><id>https://sourceforge.netbe26bf72809b8fe0dc49caeb24ff6b639309f204</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The deprecation warning for Graph.createEdge() is &lt;br /&gt;
caused in the graph package even if one of the &lt;br /&gt;
implementation classes in the graph package is &lt;br /&gt;
subclassed. The Graph.createEdge() method was &lt;br /&gt;
deprecated, but GraphImpl still implements the method. &lt;br /&gt;
Unless createEdge() is removed from the interface and &lt;br /&gt;
any implementation class in the graph package(s), the &lt;br /&gt;
warning will occur for any subclass that includes the &lt;br /&gt;
Graph interface or any of its derived interfaces in &lt;br /&gt;
its &amp;amp;quot;implements&amp;amp;quot; list.&lt;/p&gt;
&lt;p&gt;Our code audit requires that no deprecation warnings &lt;br /&gt;
exist at release time. The only workaround we have at &lt;br /&gt;
this point is to create a copy of the graph package(s) &lt;br /&gt;
and modify.&lt;/p&gt;
&lt;p&gt;Please either remove the method or remove the &lt;br /&gt;
deprecated keyword.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Remove dependency for Log4J</title><link href="https://sourceforge.net/p/openjgraph/bugs/8/" rel="alternate"/><published>2002-01-01T15:45:26Z</published><updated>2002-01-01T15:45:26Z</updated><author><name>Egon Willighagen</name><uri>https://sourceforge.net/u/egonw/</uri></author><id>https://sourceforge.net8c1565c11c317b1d028ace268e8e479d95bdc62d</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;It would be nice if the dependency for Log4J was removed.&lt;br /&gt;
The Apache license interferes with the GPL license, i.e.&lt;br /&gt;
this is argued by some legal people at Debian. The&lt;br /&gt;
CDK project (cdk.sf.net) uses it too... I made a&lt;br /&gt;
wrapper (LoggerTool) that uses Log4J is available, and&lt;br /&gt;
redirects output to STDOUT if not available. In this&lt;br /&gt;
way we can distribute GPLed software without any trouble.&lt;br /&gt;
Bug #498302 includes a patch that includes this class&lt;br /&gt;
and changes in the OpenJGraph CVS code (of 29 dec)&lt;br /&gt;
to use the wrapper instead of Cathegory directly.&lt;/p&gt;
&lt;p&gt;I was not able to get XGMML reading completely working&lt;br /&gt;
yet, but want to send my patch anyway... Contact me&lt;br /&gt;
if you are interested in the patch and want to get&lt;br /&gt;
it working in CVS...&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Possibility to use other SAX2 parsers</title><link href="https://sourceforge.net/p/openjgraph/bugs/7/" rel="alternate"/><published>2002-01-01T15:40:33Z</published><updated>2002-01-01T15:40:33Z</updated><author><name>Egon Willighagen</name><uri>https://sourceforge.net/u/egonw/</uri></author><id>https://sourceforge.neteb3c05fac0a095e7257ef05d4777043229748c16</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Possibility to use other SAX2 parsers would be a nice&lt;br /&gt;
feature, like the GPL Aelfred2. &lt;br /&gt;
Since the interface to SAX2 parsers is always identical&lt;br /&gt;
this is not hard. Attached is a patch that partly&lt;br /&gt;
changes the CVS version for Aelfred2 support. &lt;br /&gt;
It also adds a new class LoggerTool that detects if the&lt;br /&gt;
Log4J tool is installed, and if not redirects log info&lt;br /&gt;
to STDOUT. (Aelfred2 is part of Gnu JAXP: &lt;br /&gt;
&lt;a href="http://www.gnu.org/software/classpathx/jaxp/\" rel="nofollow"&gt;http://www.gnu.org/software/classpathx/jaxp/\&lt;/a&gt;)&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Avoid package private visibility</title><link href="https://sourceforge.net/p/openjgraph/bugs/6/" rel="alternate"/><published>2001-10-25T21:33:42Z</published><updated>2001-10-25T21:33:42Z</updated><author><name>Jürgen Failenschmid</name><uri>https://sourceforge.net/u/jfai/</uri></author><id>https://sourceforge.net24a1ffc789914c60266692dd5872ec8ff09c3d59</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;There are several members that have package private &lt;br /&gt;
visibility. This makes it cumbersome to use the &lt;br /&gt;
member's class as parent class in other packages. &lt;/p&gt;
&lt;p&gt;For example, GraphTraversal has a package private &lt;br /&gt;
graph attribute but no access methods. A subclass of &lt;br /&gt;
GraphTraversal in another package cannot access the &lt;br /&gt;
graph attribute, and has to resort to define its own &lt;br /&gt;
graph attribute.&lt;/p&gt;
&lt;p&gt;It is generally better to use protected visibility &lt;br /&gt;
instead of package private, especially if the owning &lt;br /&gt;
class is abstract.&lt;/p&gt;&lt;/div&gt;</summary></entry></feed>