svn commit: r1233753 - /incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStream.java

[jk...@apache.org]

Author: jkosin
Date: Fri Jan 20 04:20:25 2012
New Revision: 1233753

URL: http://svn.apache.org/viewvc?rev=1233753&view=rev
Log:
OPENNLP-367: CONLL 03 data for German & English must be in UTF-8.  I tried many combinations of ISO and the German doesn't come out correctly without UTF-8.  I'm hoping this fixes the last of the encoding issues for this round.

Modified:
    incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStream.java

Modified: incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStream.java
URL: http://svn.apache.org/viewvc/incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStream.java?rev=1233753&r1=1233752&r2=1233753&view=diff
==============================================================================
--- incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStream.java (original)
+++ incubator/opennlp/trunk/opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStream.java Fri Jan 20 04:20:25 2012
@@ -67,8 +67,8 @@ public class Conll03NameSampleStream imp
 
     this.lang = lang;
     try {
-      this.lineStream = new PlainTextByLineStream(in, "ISO-8859-1");
-      System.setOut(new PrintStream(System.out, true, "ISO-8859-1"));
+      this.lineStream = new PlainTextByLineStream(in, "UTF-8");
+      System.setOut(new PrintStream(System.out, true, "UTF-8"));
     } catch (UnsupportedEncodingException e) {
       // UTF-8 is available on all JVMs, will never happen
       throw new IllegalStateException(e);