Search Mailing List Archives
[protege-discussion] Bug in OntoViz
tudorache at stanford.edu
Tue Feb 5 16:20:28 PST 2008
I did the change that you suggested, but it still does not seem to work
well if you have non-ASCII characters in the label of a node. It seems
to work fine for other things than labels.
Even though the dot-input is generated as UTF-8, the dot file is
generated by the "dot" command. I don't know how to specify to the dot
command that the character set should be UTF-8.
The fix that I have done will be available in the next beta release. If
you find out more about how to fix the UTF-8 issue, I am happy to
integrate it into the ontoviz code.
> I found a problem about OntoViz installed with protege 3.3.1.
> A Similar problem is preveously mentioned in:
> but it seems unrepaired yet.
> >> OntoViz writes its intermediate .dot-input files using the standard
> >> FileWriter class, which uses the 'default encoding' (whatever that may
> >> be).
> Not only .dot-input files but also .dot file depend on 'default
> encoding' and it is not utf8 in my environment.
> So, charactor in JIS X 0212 (e.g. E-acute of Emile Durkheim) can't be
> represented in the .dot file.
> Moreover, we can't use Japanese charactor because dot.exe can
> recognize those charactor in UTF8 encoded file ONLY.
> I think I should repair this problem like below:
> >> You could make a change in the OntoViz ExportDot.java file and
> >>change the PrintWriter line (in the export method) into something like
> >>pw = new OutputStreamWriter(new FileOutputStream(fileName), "UTF-8");
> >>(make sure to declare the global variable pw as an OutputStreamWriter
> >>instead of a PrintWriter)
> >>This will make the OntoViz plugin _always_ write UTF-8 encoded files,
> >>independent of the default encoding.
> but I couldn't. I'm not good at programming and a lot of compile error
> hit me. (I googled error _expression and tried many times.)
> I hope someone repair this problem.
> protege-discussion mailing list
> protege-discussion at lists.stanford.edu
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
More information about the protege-discussion