Improve regexes and test names

opentripplanner · Aug 9, 2024 · 70f99d4 · 70f99d4
1 parent edea05e
commit 70f99d4
Show file tree

Hide file tree

Showing 3 changed files with 8 additions and 2 deletions.
diff --git a/src/ext-test/java/org/opentripplanner/ext/geocoder/EnglishNgramAnalyzerTest.java b/src/ext-test/java/org/opentripplanner/ext/geocoder/EnglishNgramAnalyzerTest.java
@@ -135,6 +135,12 @@ void numberSuffixes(String input, String expected) {
     assertEquals(List.of(expected), result);
   }
 
+  @Test
+  void wordBoundary() {
+    var result = tokenize("1stst");
+    assertEquals(List.of("1sts", "1stst", "stst"), result);
+  }
+
   public List<String> tokenize(String text) {
     try (var analyzer = new EnglishNGramAnalyzer()) {
       List<String> result;

diff --git a/src/ext-test/java/org/opentripplanner/ext/geocoder/LuceneIndexTest.java b/src/ext-test/java/org/opentripplanner/ext/geocoder/LuceneIndexTest.java
@@ -316,7 +316,7 @@ void agenciesAndFeedPublisher() {
         "meridian av 148",
       }
     )
-    void shortTokens(String query) {
+    void numericAdjectives(String query) {
       var names = index.queryStopClusters(query).map(c -> c.primary().name()).toList();
       assertEquals(
         Stream.of(MERIDIAN_AVE, MERIDIAN_N2, MERIDIAN_N1).map(s -> s.getName().toString()).toList(),

diff --git a/src/ext/java/org/opentripplanner/ext/geocoder/EnglishNGramAnalyzer.java b/src/ext/java/org/opentripplanner/ext/geocoder/EnglishNGramAnalyzer.java
@@ -26,7 +26,7 @@
 class EnglishNGramAnalyzer extends Analyzer {
 
   // matches one or more numbers followed by the English suffixes "st", "nd", "rd", "th"
-  private static final Pattern NUMBER_SUFFIX_PATTERN = Pattern.compile("(\\d+)[st|nd|rd|th]+");
+  private static final Pattern NUMBER_SUFFIX_PATTERN = Pattern.compile("(\\d+)(st|nd|rd|th)\\b");
 
   @Override
   protected TokenStreamComponents createComponents(String fieldName) {