небольшой тюнинг поиска

теперь точные совпадения (без учета
морфологии) будут выше в выдаче

src/main/java/ru/org/linux/search/SearchController.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 1998-2022 Linux.org.ru
+ * Copyright 1998-2024 Linux.org.ru
  *    Licensed under the Apache License, Version 2.0 (the "License");
  *    you may not use this file except in compliance with the License.
  *    You may obtain a copy of the License at
@@ -25,7 +25,6 @@
 import com.sksamuel.elastic4s.requests.searches.aggs.responses.bucket.TermBucket;
 import com.sksamuel.elastic4s.requests.searches.aggs.responses.bucket.Terms;
 import org.joda.time.DateTimeZone;
-import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.stereotype.Controller;
 import org.springframework.ui.Model;
 import org.springframework.validation.BindingResult;
@@ -57,20 +56,20 @@
 
 @Controller
 public class SearchController {
-  @Autowired
-  private SectionService sectionService;
-
-  @Autowired
-  private UserService userService;
-
-  @Autowired
-  private GroupDao groupDao;
-
-  @Autowired
-  private ElasticClient client;
-
-  @Autowired
-  private SearchResultsService resultsService;
+  private final SectionService sectionService;
+  private final UserService userService;
+  private final GroupDao groupDao;
+  private final ElasticClient client;
+  private final SearchResultsService resultsService;
+
+  public SearchController(SectionService sectionService, UserService userService, GroupDao groupDao,
+                          ElasticClient client, SearchResultsService resultsService) {
+    this.sectionService = sectionService;
+    this.userService = userService;
+    this.groupDao = groupDao;
+    this.client = client;
+    this.resultsService = resultsService;
+  }
 
   @ModelAttribute("sorts")
   public static Map<String, String> getSorts() {

src/main/scala/ru/org/linux/search/ElasticsearchIndexService.scala

@@ -16,14 +16,13 @@
 package ru.org.linux.search
 
 import com.sksamuel.elastic4s.ElasticDsl.*
-import com.sksamuel.elastic4s.requests.analyzers.*
+import com.sksamuel.elastic4s.analysis.{Analysis, CustomAnalyzer, LengthTokenFilter, MappingCharFilter, SnowballTokenFilter, StandardTokenizer}
 import com.sksamuel.elastic4s.requests.bulk.{BulkCompatibleRequest, BulkRequest}
 import com.sksamuel.elastic4s.requests.indexes.IndexRequest
 import com.sksamuel.elastic4s.requests.mappings.{MappingDefinition, TermVector}
 import com.sksamuel.elastic4s.{ElasticClient, Index}
 import com.typesafe.scalalogging.StrictLogging
 import org.apache.commons.text.StringEscapeUtils
-import org.joda.time.DateTime
 import org.springframework.stereotype.Service
 import ru.org.linux.comment.{Comment, CommentReadService}
 import ru.org.linux.group.{Group, GroupDao}
@@ -54,19 +53,34 @@ object ElasticsearchIndexService {
     longField("topic_id"),
     textField("topic_title").index(false),
     textField("title").analyzer("text_analyzer"),
-    textField("message").analyzer("text_analyzer").termVector(TermVector.WithPositionsOffsets),
+    textField("message").analyzer("text_analyzer").termVector(TermVector.WithPositionsOffsets).fields {
+      textField("raw").termVector(TermVector.WithPositionsOffsets).analyzer("exact_analyzer")
+    },
     booleanField("topic_awaits_commit"))
 
-  private val Analyzers: Seq[CustomAnalyzerDefinition] = Seq(
-    CustomAnalyzerDefinition(
-      "text_analyzer",
-      tokenizer = StandardTokenizer("text_tokenizer"),
-      filters = Seq(
-        LengthTokenFilter("m_long_word").max(100),
-        LowercaseTokenFilter,
-        MappingCharFilter("m_ee", "ё" -> "е", "Ё" -> "Е"),
-        SnowballTokenFilter("m_my_snow_ru", "Russian"),
-        SnowballTokenFilter("m_my_snow_en", "English")))
+  private val Analyzers = Analysis(
+    analyzers = List(
+      CustomAnalyzer(
+        name = "text_analyzer",
+        tokenizer = "text_tokenizer",
+        tokenFilters = List("m_long_word", "lowercase", "m_my_snow_ru", "m_my_snow_en"),
+        charFilters = List("m_ee")),
+      CustomAnalyzer(
+        name = "exact_analyzer",
+        tokenizer = "text_tokenizer",
+        tokenFilters = List("m_long_word", "lowercase"),
+        charFilters = List("m_ee"))),
+    tokenizers = List(
+      StandardTokenizer("text_tokenizer")
+    ),
+    tokenFilters = List(
+      LengthTokenFilter("m_long_word").max(100),
+      SnowballTokenFilter("m_my_snow_ru", "Russian"),
+      SnowballTokenFilter("m_my_snow_en", "English")
+    ),
+    charFilters = List(
+      MappingCharFilter("m_ee", Map("ё" -> "е", "Ё" -> "Е"))
+    )
   )
 }
 
@@ -197,7 +211,7 @@ class ElasticsearchIndexService(sectionService: SectionService, groupDao: GroupD
         "topic_title" -> topicTitle,
         COLUMN_TOPIC_AWAITS_COMMIT -> topicAwaitsCommit(topic),
         "message" -> message,
-        "postdate" -> new DateTime(comment.postdate),
+        "postdate" -> comment.postdate.toInstant,
         "tag" -> topicTagService.getTags(topic),
         "is_comment" -> true) ++ title.map("title" -> _)
       )
@@ -213,7 +227,7 @@ class ElasticsearchIndexService(sectionService: SectionService, groupDao: GroupD
     val section = sectionService.getSection(topic.sectionId)
     val author = userService.getUserCached(topic.authorUserId)
 
-    indexInto(MessageIndexType) id topic.id.toString fields(
+    indexInto(MessageIndexType).id(topic.id.toString).fields(
       "section" -> section.getUrlName,
       "topic_author" -> author.getNick,
       "topic_id" -> topic.id,
@@ -222,7 +236,7 @@ class ElasticsearchIndexService(sectionService: SectionService, groupDao: GroupD
       "title" -> topic.getTitleUnescaped,
       "topic_title" -> topic.getTitleUnescaped,
       "message" -> messageTextService.extractPlainText(msgbaseDao.getMessageText(topic.id)),
-      "postdate" -> new DateTime(topic.postdate),
+      "postdate" -> topic.postdate.toInstant,
       "tag" -> topicTagService.getTags(topic),
       COLUMN_TOPIC_AWAITS_COMMIT -> topicAwaitsCommit(topic),
       "is_comment" -> false)

src/main/scala/ru/org/linux/search/SearchViewer.scala

@@ -33,16 +33,20 @@ class SearchViewer(query: SearchRequest, elastic: ElasticClient) {
     if (queryText.isEmpty) {
       matchAllQuery()
     } else {
-      boolQuery().
-        should(
-          MatchQuery("title", queryText).minimumShouldMatch("2"),
-          MatchQuery("message", queryText).minimumShouldMatch("2"),
-          matchPhraseQuery("message", queryText)).minimumShouldMatch(1)
+      boolQuery()
+        .must(
+          should(
+            MatchQuery("title", queryText).minimumShouldMatch("2"),
+            MatchQuery("message", queryText).minimumShouldMatch("2")))
+        .should(
+          matchPhraseQuery("message", queryText),
+          MatchQuery("message.raw", queryText).minimumShouldMatch("2")
+        ).minimumShouldMatch(0)
     }
   }
 
   private def boost(query: Query) = {
-    functionScoreQuery(query) functions(
+    functionScoreQuery(query).functions(
       WeightScore(TopicBoost).filter(termQuery("is_comment", "false")),
       WeightScore(RecentBoost).filter(rangeQuery("postdate").gte("now/d-3y"))
     )
@@ -56,7 +60,7 @@ class SearchViewer(query: SearchRequest, elastic: ElasticClient) {
     }
   }
 
-  def performSearch(tz:DateTimeZone): SearchResponse = {
+  def performSearch(tz: DateTimeZone): SearchResponse = {
     val typeFilter = Option(query.getRange.getValue) map { value =>
       termQuery(query.getRange.getColumn, value)
     }