From 9d6fd98810aff7311e124be98f8c8f8cc66520be Mon Sep 17 00:00:00 2001
From: Peter Gromov <peter@jetbrains.com>
Date: Wed, 24 Feb 2021 17:41:50 +0100
Subject: [PATCH] LUCENE-9811: Hunspell suggestions: speed up ngram calculation
 by not searching for substrings in impossible places (#2428)

---
 .../hunspell/GeneratingSuggester.java         | 27 ++++++++++++++++---
 .../analysis/hunspell/TestPerformance.java    |  2 +-
 2 files changed, 25 insertions(+), 4 deletions(-)

diff --git a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/GeneratingSuggester.java b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/GeneratingSuggester.java
index c1bbdf9e555..cb5286132fc 100644
--- a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/GeneratingSuggester.java
+++ b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/GeneratingSuggester.java
@@ -355,12 +355,20 @@ class GeneratingSuggester {
     if (l2 == 0) {
       return 0;
     }
+
+    int[] lastStarts = new int[l1];
     for (int j = 1; j <= n; j++) {
       int ns = 0;
       for (int i = 0; i <= (l1 - j); i++) {
-        if (s2.contains(s1.substring(i, i + j))) {
-          ns++;
-        } else if (opt.contains(NGramOptions.WEIGHTED)) {
+        if (lastStarts[i] >= 0) {
+          int pos = indexOfSubstring(s2, lastStarts[i], s1, i, j);
+          lastStarts[i] = pos;
+          if (pos >= 0) {
+            ns++;
+            continue;
+          }
+        }
+        if (opt.contains(NGramOptions.WEIGHTED)) {
           ns--;
           if (i == 0 || i == l1 - j) {
             ns--; // side weight
@@ -383,6 +391,19 @@ class GeneratingSuggester {
     return score - Math.max(ns, 0);
   }
 
+  private static int indexOfSubstring(
+      String haystack, int haystackPos, String needle, int needlePos, int len) {
+    char c = needle.charAt(needlePos);
+    int limit = haystack.length() - len;
+    for (int i = haystackPos; i <= limit; i++) {
+      if (haystack.charAt(i) == c
+          && haystack.regionMatches(i + 1, needle, needlePos + 1, len - 1)) {
+        return i;
+      }
+    }
+    return -1;
+  }
+
   private static int lcs(String s1, String s2) {
     int[] lengths = new int[s2.length() + 1];
 
diff --git a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestPerformance.java b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestPerformance.java
index 2dfa2a8cdbc..eda0f7659fe 100644
--- a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestPerformance.java
+++ b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestPerformance.java
@@ -62,7 +62,7 @@ public class TestPerformance extends LuceneTestCase {
 
   @Test
   public void en_suggest() throws Exception {
-    checkSuggestionPerformance("en", 1_000);
+    checkSuggestionPerformance("en", 1_200);
   }
 
   @Test