Benchmark string search approaches

2 files changed, 30 insertions, 0 deletions

diff --git a/src/main/java/org/javacs/Parser.java b/src/main/java/org/javacs/Parser.java
index 27aa40e..9a04f0a 100644
--- a/src/main/java/org/javacs/Parser.java
+++ b/src/main/java/org/javacs/Parser.java
@@ -95,6 +95,7 @@ class Parser {
 
     private static final Pattern WORD = Pattern.compile("\\b\\w+\\b");
 
+    // TODO cache the progress made by searching shorter queries
     static boolean containsWordMatching(Path java, String query) {
         try {
             var reader = Files.newBufferedReader(java);
@@ -111,6 +112,7 @@ class Parser {
         }
     }
 
+    // TODO try working on inputstream rather than lines
     static boolean containsText(Path java, String query) {
         try {
             var search = new StringSearch(query);
diff --git a/src/test/java/org/javacs/StringSearchBenchmark.java b/src/test/java/org/javacs/StringSearchBenchmark.java
new file mode 100644
index 0000000..015dfdd
--- /dev/null
+++ b/src/test/java/org/javacs/StringSearchBenchmark.java
@@ -0,0 +1,28 @@
+package org.javacs;
+
+import java.nio.file.Path;
+import java.nio.file.Paths;
+import java.util.concurrent.TimeUnit;
+import org.openjdk.jmh.annotations.*;
+
+// TODO this is coloring badly
+@Warmup(iterations = 3, time = 1, timeUnit = TimeUnit.SECONDS)
+@Measurement(iterations = 3, time = 1, timeUnit = TimeUnit.SECONDS)
+@Fork(1)
+public class StringSearchBenchmark {
+    private static final Path largeFile = Paths.get(FindResource.uri("/org/javacs/example/LargeFile.java"));
+    // "removeMethodBodies" appears late in the file, so stopping early will not be very effective
+    private static final String query = "removeMethodBodies";
+
+    @Benchmark
+    public void regex() {
+        var found = Parser.containsWordMatching(largeFile, query);
+        assert found;
+    }
+
+    @Benchmark
+    public void boyerMoore() {
+        var found = Parser.containsText(largeFile, query);
+        assert found;
+    }
+}