Fix false positives for some Slovak rules

Fixes valentjn/vscode-ltex#221.
valentjn · Jan 24, 2021 · 42b5124 · 42b5124
1 parent 3a7f027
commit 42b5124
Show file tree

Hide file tree

Showing 3 changed files with 27 additions and 1 deletion.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -8,6 +8,10 @@
 
 # Changelog
 
+## 9.1.1 (upcoming)
+
+- Fix false positives for words added by `Add to dictionary` for Slovak rule IDs `MUZSKY_ROD_NEZIV_A`, `ZENSKY_ROD_A`, and `STREDNY_ROD_A` (fixes [vscode-ltex#221](https://github.com/valentjn/vscode-ltex/issues/221))
+
 ## 9.1.0 (January 24, 2021)
 
 - Add support for BibT<sub>E</sub>X files (language code `bibtex`, fixes [vscode-ltex#211](https://github.com/valentjn/vscode-ltex/issues/211))

diff --git a/ltexls-core/src/main/java/org/bsplines/ltexls/languagetool/LanguageToolJavaInterface.java b/ltexls-core/src/main/java/org/bsplines/ltexls/languagetool/LanguageToolJavaInterface.java
@@ -39,6 +39,7 @@
 import org.xml.sax.SAXException;
 
 public class LanguageToolJavaInterface extends LanguageToolInterface {
+  private Set<String> dictionary;
   private @MonotonicNonNull ResultCache resultCache;
   private @MonotonicNonNull JLanguageTool languageTool;
 
@@ -54,6 +55,8 @@ public class LanguageToolJavaInterface extends LanguageToolInterface {
    */
   public LanguageToolJavaInterface(String languageShortCode, String motherTongueShortCode,
         int sentenceCacheSize, Set<String> dictionary) {
+    this.dictionary = dictionary;
+
     if (!Languages.isLanguageSupported(languageShortCode)) {
       Tools.logger.severe(Tools.i18n("notARecognizedLanguage", languageShortCode));
       return;
@@ -157,10 +160,20 @@ public void write(int b) {
       return Collections.emptyList();
     }
 
+    String code = annotatedTextFragment.getCodeFragment().getCode();
     List<LanguageToolRuleMatch> result = new ArrayList<>();
 
     for (RuleMatch match : matches) {
-      result.add(new LanguageToolRuleMatch(match, annotatedTextFragment));
+      LanguageToolRuleMatch languageToolRuleMatch =
+          new LanguageToolRuleMatch(match, annotatedTextFragment);
+
+      if (languageToolRuleMatch.isUnknownWordRule()
+            && this.dictionary.contains(code.substring(languageToolRuleMatch.getFromPos(),
+              languageToolRuleMatch.getToPos()))) {
+        continue;
+      }
+
+      result.add(languageToolRuleMatch);
     }
 
     return result;

diff --git a/ltexls-core/src/test/java/org/bsplines/ltexls/server/DocumentCheckerTest.java b/ltexls-core/src/test/java/org/bsplines/ltexls/server/DocumentCheckerTest.java
@@ -235,6 +235,15 @@ public void testDictionary() {
     Pair<List<LanguageToolRuleMatch>, List<AnnotatedTextFragment>> checkingResult =
         checkDocument(document, settings);
     Assertions.assertEquals(1, checkingResult.getKey().size());
+
+    document = createDocument("latex", "S pekn\u00e9 inteligentn\u00fdmi dubmi.\n");
+    settings = settings.withLanguageShortCode("sk-SK");
+    checkingResult = checkDocument(document, settings);
+    Assertions.assertEquals(1, checkingResult.getKey().size());
+
+    settings = settings.withDictionary(Collections.singleton("pekn\u00e9"));
+    checkingResult = checkDocument(document, settings);
+    Assertions.assertEquals(0, checkingResult.getKey().size());
   }
 
   @Test