svn commit: r1603856 - in /lucene/dev/branches/lucene_solr_4_9: ./ lucene/ lucene/analysis/ lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/ lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/

[rm...@apache.org]

Author: rmuir
Date: Thu Jun 19 12:16:23 2014
New Revision: 1603856

URL: http://svn.apache.org/r1603856
Log:
LUCENE-5777: fix double escaping of dash in hunspell conditions

Added:
    lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestDoubleEscape.java
      - copied unchanged from r1603853, lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestDoubleEscape.java
    lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/double-escaped.aff
      - copied unchanged from r1603853, lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/double-escaped.aff
    lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/double-escaped.dic
      - copied unchanged from r1603853, lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/double-escaped.dic
Modified:
    lucene/dev/branches/lucene_solr_4_9/   (props changed)
    lucene/dev/branches/lucene_solr_4_9/lucene/   (props changed)
    lucene/dev/branches/lucene_solr_4_9/lucene/CHANGES.txt   (contents, props changed)
    lucene/dev/branches/lucene_solr_4_9/lucene/analysis/   (props changed)
    lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java

Modified: lucene/dev/branches/lucene_solr_4_9/lucene/CHANGES.txt
URL: http://svn.apache.org/viewvc/lucene/dev/branches/lucene_solr_4_9/lucene/CHANGES.txt?rev=1603856&r1=1603855&r2=1603856&view=diff
==============================================================================
--- lucene/dev/branches/lucene_solr_4_9/lucene/CHANGES.txt (original)
+++ lucene/dev/branches/lucene_solr_4_9/lucene/CHANGES.txt Thu Jun 19 12:16:23 2014
@@ -244,7 +244,7 @@ Bug fixes
 * LUCENE-5747: Project-specific settings for the eclipse development
   environment will prevent automatic code reformatting. (Shawn Heisey)
 
-* LUCENE-5768: Hunspell condition checks containing character classes
+* LUCENE-5768, LUCENE-5777: Hunspell condition checks containing character classes
   were buggy. (Clinton Gormley, Robert Muir)
 
 Test Framework

Modified: lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java?rev=1603856&r1=1603855&r2=1603856&view=diff
==============================================================================
--- lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java (original)
+++ lucene/dev/branches/lucene_solr_4_9/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java Thu Jun 19 12:16:23 2014
@@ -356,6 +356,25 @@ public class Dictionary {
     }
     return builder.finish();
   }
+  
+  static String escapeDash(String re) {
+    // we have to be careful, even though dash doesn't have a special meaning,
+    // some dictionaries already escape it (e.g. pt_PT), so we don't want to nullify it
+    StringBuilder escaped = new StringBuilder();
+    for (int i = 0; i < re.length(); i++) {
+      char c = re.charAt(i);
+      if (c == '-') {
+        escaped.append("\\-");
+      } else {
+        escaped.append(c);
+        if (c == '\\' && i + 1 < re.length()) {
+          escaped.append(re.charAt(i+1));
+          i++;
+        }
+      }
+    }
+    return escaped.toString();
+  }
 
   /**
    * Parses a specific affix rule putting the result into the provided affix map
@@ -425,7 +444,7 @@ public class Dictionary {
       }
       // "dash hasn't got special meaning" (we must escape it)
       if (condition.indexOf('-') >= 0) {
-        condition = condition.replace("-", "\\-");
+        condition = escapeDash(condition);
       }
 
       final String regex;