From 13e747f95f7b8277cc961e83b4990e6d152dd351 Mon Sep 17 00:00:00 2001
From: Peter Gromov <peter@jetbrains.com>
Date: Tue, 8 Aug 2023 22:40:42 +0200
Subject: [PATCH] hunspell: simplify TrigramAutomaton to speed up the
 suggestion enumeration (#12491)

* hunspell: simplify TrigramAutomaton to speed up the suggestion enumeration

avoid the automaton access on definitely absent characters;
count the scores for all substring lengths together
---
 lucene/CHANGES.txt                            |  2 +-
 .../analysis/hunspell/TrigramAutomaton.java   | 20 +++++++------------
 .../analysis/hunspell/TestPerformance.java    |  2 +-
 3 files changed, 9 insertions(+), 15 deletions(-)

diff --git a/lucene/CHANGES.txt b/lucene/CHANGES.txt
index b3a7df1b5be..13fd71a2b88 100644
--- a/lucene/CHANGES.txt
+++ b/lucene/CHANGES.txt
@@ -75,7 +75,7 @@ Improvements
 
 * LUCENE-10652: Add a top-n range faceting example to RangeFacetsExample. (Yuting Gan)
 
-* GITHUB#12447: Hunspell: speed up the dictionary enumeration (Peter Gromov)
+* GITHUB#12447, GITHUB#12491: Hunspell: speed up the dictionary enumeration on suggestion (Peter Gromov)
 
 * GITHUB#12464: Hunspell: allow customizing the hash table load factor (Peter Gromov)
 
diff --git a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/TrigramAutomaton.java b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/TrigramAutomaton.java
index 2e6091541d2..06d528fc001 100644
--- a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/TrigramAutomaton.java
+++ b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/TrigramAutomaton.java
@@ -32,7 +32,7 @@ class TrigramAutomaton {
   private final CharacterRunAutomaton automaton;
   private final int[] state2Score;
   private final FixedBitSet countedSubstrings;
-  private final char minChar;
+  private final char minChar, maxChar;
 
   TrigramAutomaton(String s1) {
     Map<String, Integer> substringCounts = new HashMap<>();
@@ -41,6 +41,7 @@ class TrigramAutomaton {
     int initialState = builder.createState();
 
     minChar = (char) s1.chars().min().orElseThrow();
+    maxChar = (char) s1.chars().max().orElseThrow();
 
     for (int start = 0; start < s1.length(); start++) {
       int limit = Math.min(s1.length(), start + N);
@@ -80,7 +81,7 @@ class TrigramAutomaton {
   int ngramScore(CharSequence s2) {
     countedSubstrings.clear();
 
-    int score1 = 0, score2 = 0, score3 = 0; // scores for substrings of length 1, 2 and 3
+    int score = 0;
 
     // states of running the automaton on substrings [i-1, i) and [i-2, i)
     int state1 = -1, state2 = -1;
@@ -88,7 +89,7 @@ class TrigramAutomaton {
     int limit = s2.length();
     for (int i = 0; i < limit; i++) {
       char c = s2.charAt(i);
-      if (c < minChar) {
+      if (c < minChar || c > maxChar) {
         state1 = state2 = -1;
         continue;
       }
@@ -96,27 +97,20 @@ class TrigramAutomaton {
 
       int state3 = state2 <= 0 ? 0 : automaton.step(state2, c);
       if (state3 > 0) {
-        score3 += substringScore(state3, countedSubstrings);
+        score += substringScore(state3, countedSubstrings);
       }
 
       state2 = state1 <= 0 ? 0 : automaton.step(state1, c);
       if (state2 > 0) {
-        score2 += substringScore(state2, countedSubstrings);
+        score += substringScore(state2, countedSubstrings);
       }
 
       state1 = automaton.step(0, c);
       if (state1 > 0) {
-        score1 += substringScore(state1, countedSubstrings);
+        score += substringScore(state1, countedSubstrings);
       }
     }
 
-    int score = score1;
-    if (score1 >= 2) {
-      score += score2;
-      if (score2 >= 2) {
-        score += score3;
-      }
-    }
     return score;
   }
 
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 232f12e13c1..04989954591 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
@@ -106,7 +106,7 @@ public class TestPerformance extends LuceneTestCase {
 
   @Test
   public void uk_suggest() throws Exception {
-    checkSuggestionPerformance("uk", 700);
+    checkSuggestionPerformance("uk", 800);
   }
 
   private Dictionary loadDictionary(String code) throws IOException, ParseException {