[de] properly consider spelling.txt for Swiss German, i.e. don't sugg…

…est words with "ß" even when listed in spelling.txt
languagetool-org · Feb 10, 2015 · 32b7e4c · 32b7e4c
1 parent 68c7bce
commit 32b7e4c
Show file tree

Hide file tree

Showing 3 changed files with 27 additions and 7 deletions.
diff --git a/...ore/src/main/java/org/languagetool/rules/spelling/hunspell/CompoundAwareHunspellRule.java b/...ore/src/main/java/org/languagetool/rules/spelling/hunspell/CompoundAwareHunspellRule.java
@@ -37,7 +37,9 @@ public abstract class CompoundAwareHunspellRule extends HunspellRule {
 
   private final CompoundWordTokenizer wordSplitter;
   private final MorfologikMultiSpeller morfoSpeller;
-
+
+  protected abstract void filterForLanguage(List<String> suggestions);
+
   public CompoundAwareHunspellRule(ResourceBundle messages, Language language, CompoundWordTokenizer wordSplitter, MorfologikMultiSpeller morfoSpeller) {
     super(messages, language);
     this.wordSplitter = wordSplitter;
@@ -47,7 +49,7 @@ public CompoundAwareHunspellRule(ResourceBundle messages, Language language, Com
   /**
    * As a hunspell-based approach is too slow, we use Morfologik to create suggestions. As this
    * won't work for compounds not in the dictionary, we split the word and also get suggestions
-   * on the compound parts. In the end, all candidates are filtered against Hunspell again (which 
+   * on the compound parts. In the end, all candidates are filtered against Hunspell again (which
    * supports compounds).
    */
   @Override
@@ -95,6 +97,7 @@ public List<String> getSuggestions(String word) throws IOException {
     suggestions.addAll(0, noSplitSuggestions);
 
     filterDupes(suggestions);
+    filterForLanguage(suggestions);
     final List<String> sortedSuggestions = sortSuggestionByQuality(word, suggestions);
     return sortedSuggestions.subList(0, Math.min(MAX_SUGGESTIONS, sortedSuggestions.size()));
   }

diff --git a/...getool-language-modules/de/src/main/java/org/languagetool/rules/de/GermanSpellerRule.java b/...getool-language-modules/de/src/main/java/org/languagetool/rules/de/GermanSpellerRule.java
@@ -101,6 +101,16 @@ private static MorfologikMultiSpeller getSpeller(Language language) {
     }
   }
 
+  @Override
+  protected void filterForLanguage(List<String> suggestions) {
+    if (language.getShortNameWithCountryAndVariant().equals("de-CH")) {
+      for (int i = 0; i < suggestions.size(); i++) {
+        String s = suggestions.get(i);
+        suggestions.set(i, s.replace("ß", "ss"));
+      }
+    }
+  }
+
   // Use hunspell-style replacements to get good suggestions for "heisse", namely "heiße" etc
   // TODO: remove this when the Morfologik speller can do this directly during tree iteration:
   @Override

diff --git a/...ol-language-modules/de/src/test/java/org/languagetool/rules/de/GermanSpellerRuleTest.java b/...ol-language-modules/de/src/test/java/org/languagetool/rules/de/GermanSpellerRuleTest.java
@@ -32,6 +32,7 @@
 import java.util.List;
 import java.util.ResourceBundle;
 
+import static junit.framework.TestCase.assertFalse;
 import static org.hamcrest.CoreMatchers.is;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertThat;
@@ -40,6 +41,7 @@
 public class GermanSpellerRuleTest {
 
   private static final GermanyGerman GERMAN_DE = new GermanyGerman();
+  private static final SwissGerman GERMAN_CH = new SwissGerman();
 
   @Test
   public void testSortSuggestion() throws Exception {
@@ -71,15 +73,20 @@ public void testDashAndHyphen() throws Exception {
 
   @Test
   public void testGetSuggestionsFromSpellingTxt() throws Exception {
-    MyGermanSpellerRule rule = new MyGermanSpellerRule(TestTools.getMessages("de"), GERMAN_DE);
-    assertThat(rule.getSuggestions("Ligafußboll").toString(), is("[Ligafußball, Ligafußballs]"));  // from spelling.txt
+    MyGermanSpellerRule ruleGermany = new MyGermanSpellerRule(TestTools.getMessages("de"), GERMAN_DE);
+    assertThat(ruleGermany.getSuggestions("Ligafußboll").toString(), is("[Ligafußball, Ligafußballs]"));  // from spelling.txt
+    MyGermanSpellerRule ruleSwiss = new MyGermanSpellerRule(TestTools.getMessages("de"), GERMAN_CH);
+    assertThat(ruleSwiss.getSuggestions("Ligafußboll").toString(), is("[Ligafussball, Ligafussballs]"));
   }
 
   @Test
   public void testIgnoreWord() throws Exception {
-    MyGermanSpellerRule rule = new MyGermanSpellerRule(TestTools.getMessages("de"), GERMAN_DE);
-    assertTrue(rule.doIgnoreWord("einPseudoWortFürLanguageToolTests"));  // from ignore.txt
-    assertTrue(rule.doIgnoreWord("Ligafußball"));  // from spelling.txt
+    MyGermanSpellerRule ruleGermany = new MyGermanSpellerRule(TestTools.getMessages("de"), GERMAN_DE);
+    assertTrue(ruleGermany.doIgnoreWord("einPseudoWortFürLanguageToolTests"));  // from ignore.txt
+    assertTrue(ruleGermany.doIgnoreWord("Ligafußball"));  // from spelling.txt
+    MyGermanSpellerRule ruleSwiss = new MyGermanSpellerRule(TestTools.getMessages("de"), GERMAN_CH);
+    assertTrue(ruleSwiss.doIgnoreWord("einPseudoWortFürLanguageToolTests"));
+    assertFalse(ruleSwiss.doIgnoreWord("Ligafußball"));  // 'ß' never accepted for Swiss
   }
 
   private class MyGermanSpellerRule extends GermanSpellerRule {