fix: 修复知识库索引并发与向量化边界

- 复用 Lucene 和 Elasticsearch 客户端并支持有界批量写入与删除

- 记录脱敏后的 Embedding 失败请求与完整响应

- 为 BGE-M3 分块统一增加上下文硬上限

easy-agents-embedding/easy-agents-embedding-openai/src/main/java/com/easyagents/embedding/openai/OpenAIEmbeddingModel.java

@@ -11,13 +11,19 @@ import com.easyagents.core.store.VectorData;
 import com.easyagents.core.util.JSONUtil;
 import com.easyagents.core.util.Maps;
 import com.easyagents.core.util.StringUtil;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 import java.util.HashMap;
+import java.util.LinkedHashMap;
 import java.util.Locale;
 import java.util.Map;
 
 public class OpenAIEmbeddingModel extends BaseEmbeddingModel<OpenAIEmbeddingConfig> {
 
+    private static final Logger LOG = LoggerFactory.getLogger(OpenAIEmbeddingModel.class);
+    private static final String REDACTED_HEADER_VALUE = "[REDACTED]";
+
     private HttpClient httpClient = new HttpClient();
 
     public OpenAIEmbeddingModel(OpenAIEmbeddingConfig config) {
@@ -40,27 +46,45 @@ public class OpenAIEmbeddingModel extends BaseEmbeddingModel<OpenAIEmbeddingConf
 
         String payload = promptToEmbeddingsPayload(document, options, config);
         String endpoint = config.getEndpoint();
+        String requestUrl = endpoint + config.getRequestPath();
         // https://platform.openai.com/docs/api-reference/embeddings/create
-        String response = httpClient.post(endpoint + config.getRequestPath(), headers, payload);
+        String response = httpClient.post(requestUrl, headers, payload);
 
         if (StringUtil.noText(response)) {
+            logResponseParsingFailure(
+                "response is null or empty",
+                requestUrl,
+                headers,
+                payload,
+                response
+            );
             throw new ModelException("response is null or empty.");
         }
 
-        JSONObject jsonObject = JSON.parseObject(response);
+        try {
-        String errorMessage = JSONUtil.detectErrorMessage(jsonObject);
+            JSONObject jsonObject = JSON.parseObject(response);
-        if (errorMessage != null) {
+            String errorMessage = JSONUtil.detectErrorMessage(jsonObject);
-            throw new ModelException(errorMessage);
+            if (errorMessage != null) {
-        }
+                logResponseParsingFailure(errorMessage, requestUrl, headers, payload, response);
+                throw new ModelException(errorMessage);
+            }
 
-        VectorData vectorData = new VectorData();
+            VectorData vectorData = new VectorData();
-        double[] embedding = JSONUtil.readDoubleArray(jsonObject, "$.data[0].embedding");
+            double[] embedding = JSONUtil.readDoubleArray(jsonObject, "$.data[0].embedding");
-        if (embedding == null || embedding.length == 0) {
+            if (embedding == null || embedding.length == 0) {
-            throw new ModelException(buildMissingEmbeddingMessage());
+                String missingEmbeddingMessage = buildMissingEmbeddingMessage();
-        }
+                logResponseParsingFailure(missingEmbeddingMessage, requestUrl, headers, payload, response);
-        vectorData.setVector(embedding);
+                throw new ModelException(missingEmbeddingMessage);
+            }
+            vectorData.setVector(embedding);
 
-        return vectorData;
+            return vectorData;
+        } catch (ModelException e) {
+            throw e;
+        } catch (RuntimeException e) {
+            logResponseParsingFailure(e.getMessage(), requestUrl, headers, payload, response);
+            throw new ModelException("Failed to parse embedding response.", e);
+        }
     }
 
 
@@ -114,4 +138,35 @@ public class OpenAIEmbeddingModel extends BaseEmbeddingModel<OpenAIEmbeddingConf
             + ", endpoint=" + config.getEndpoint()
             + ", requestPath=" + config.getRequestPath();
     }
+
+    private void logResponseParsingFailure(String reason,
+                                           String requestUrl,
+                                           Map<String, String> requestHeaders,
+                                           String requestBody,
+                                           String responseBody) {
+        LOG.error(
+            "Embedding response parsing failed: reason={}\n"
+                + "requestUrl={}\n"
+                + "requestHeaders={}\n"
+                + "requestBody={}\n"
+                + "responseBody={}",
+            reason,
+            requestUrl,
+            sanitizeHeadersForLogging(requestHeaders),
+            requestBody,
+            responseBody
+        );
+    }
+
+    static Map<String, String> sanitizeHeadersForLogging(Map<String, String> headers) {
+        Map<String, String> sanitizedHeaders = new LinkedHashMap<>();
+        if (headers == null) {
+            return sanitizedHeaders;
+        }
+        headers.forEach((name, value) -> sanitizedHeaders.put(
+            name,
+            "Authorization".equalsIgnoreCase(name) ? REDACTED_HEADER_VALUE : value
+        ));
+        return sanitizedHeaders;
+    }
 }

easy-agents-embedding/easy-agents-embedding-openai/src/test/java/com/easyagents/embedding/openai/OpenAIEmbeddingModelTest.java

@@ -7,6 +7,7 @@ import com.easyagents.core.model.exception.ModelException;
 import org.junit.Assert;
 import org.junit.Test;
 
+import java.util.LinkedHashMap;
 import java.util.Map;
 
 /**
@@ -68,4 +69,46 @@ public class OpenAIEmbeddingModelTest {
 
         Assert.assertTrue(exception.getMessage().contains("data[0].embedding"));
     }
+
+    /**
+     * Verifies that invalid JSON responses are reported as embedding response parsing failures.
+     */
+    @Test
+    public void shouldThrowModelExceptionWhenResponseIsInvalidJson() {
+        OpenAIEmbeddingConfig config = new OpenAIEmbeddingConfig();
+        config.setProvider("test-provider");
+        config.setModel("BAAI/bge-m3");
+        config.setApiKey("test-key");
+        OpenAIEmbeddingModel model = new OpenAIEmbeddingModel(config);
+        model.setHttpClient(new HttpClient() {
+            @Override
+            public String post(String url, Map<String, String> headers, String payload) {
+                return "not-json";
+            }
+        });
+
+        ModelException exception = Assert.assertThrows(
+            ModelException.class,
+            () -> model.embed(Document.of("hello"))
+        );
+
+        Assert.assertEquals("Failed to parse embedding response.", exception.getMessage());
+        Assert.assertNotNull(exception.getCause());
+    }
+
+    /**
+     * Verifies that diagnostic headers retain ordinary values without exposing credentials.
+     */
+    @Test
+    public void shouldRedactAuthorizationHeaderForFailureLogging() {
+        Map<String, String> headers = new LinkedHashMap<>();
+        headers.put("Content-Type", "application/json");
+        headers.put("Authorization", "Bearer test-key");
+
+        Map<String, String> sanitizedHeaders = OpenAIEmbeddingModel.sanitizeHeadersForLogging(headers);
+
+        Assert.assertEquals("application/json", sanitizedHeaders.get("Content-Type"));
+        Assert.assertEquals("[REDACTED]", sanitizedHeaders.get("Authorization"));
+        Assert.assertEquals("Bearer test-key", headers.get("Authorization"));
+    }
 }

easy-agents-rag/easy-agents-rag-core/src/main/java/com/easyagents/rag/core/BgeM3ChunkSafety.java

@@ -0,0 +1,230 @@
+package com.easyagents.rag.core;
+
+import java.text.Normalizer;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.List;
+
+/**
+ * Provides conservative BGE-M3 token estimation and hard-limit text splitting.
+ *
+ * <p>BGE-M3 uses an XLM-R SentencePiece tokenizer. Loading the model vocabulary
+ * during ingestion would introduce runtime model downloads, so this guard counts
+ * NFKC-normalized Unicode code points. The estimate is intentionally conservative
+ * for ordinary Chinese and mixed-language knowledge documents.</p>
+ */
+public final class BgeM3ChunkSafety {
+
+    private static final double PREFERRED_BOUNDARY_MIN_RATIO = 0.8D;
+
+    private BgeM3ChunkSafety() {
+    }
+
+    /**
+     * Estimates the content token count after tokenizer-compatible normalization.
+     *
+     * @param content chunk content
+     * @return conservative token estimate, or zero for empty content
+     */
+    public static int estimateContentTokens(String content) {
+        if (content == null || content.isEmpty()) {
+            return 0;
+        }
+        String normalized = Normalizer.normalize(content, Normalizer.Form.NFKC);
+        return normalized.codePointCount(0, normalized.length());
+    }
+
+    /**
+     * Checks whether content stays within the shared BGE-M3 embedding budget.
+     *
+     * @param content chunk content
+     * @return true when the content can be sent to BGE-M3 safely
+     */
+    public static boolean isWithinHardLimit(String content) {
+        return estimateContentTokens(content)
+            <= RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT;
+    }
+
+    /**
+     * Splits content using the shared BGE-M3 hard token limit.
+     *
+     * @param content source content
+     * @return ordered, non-overlapping source ranges
+     */
+    public static List<ChunkRange> splitToHardLimit(String content) {
+        return splitToTokenLimit(content, RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT);
+    }
+
+    /**
+     * Splits content to an explicit conservative token limit.
+     *
+     * @param content source content
+     * @param maxTokens maximum estimated tokens per part
+     * @return ordered, non-overlapping source ranges
+     * @throws IllegalArgumentException when maxTokens is not positive
+     */
+    public static List<ChunkRange> splitToTokenLimit(String content, int maxTokens) {
+        if (maxTokens <= 0) {
+            throw new IllegalArgumentException("maxTokens must be positive");
+        }
+        if (content == null || content.isEmpty()) {
+            return Collections.emptyList();
+        }
+        if (estimateContentTokens(content) <= maxTokens) {
+            return Collections.singletonList(new ChunkRange(0, content.length()));
+        }
+
+        List<ChunkRange> parts = new ArrayList<ChunkRange>();
+        int start = 0;
+        while (start < content.length()) {
+            int end = findSplitEnd(content, start, maxTokens);
+            if (end <= start) {
+                throw new IllegalStateException("BGE-M3 hard split did not advance the text cursor");
+            }
+            parts.add(new ChunkRange(start, end));
+            start = end;
+        }
+        return parts;
+    }
+
+    /**
+     * Finds a safe source boundary while preferring nearby sentence endings.
+     *
+     * @param content complete source content
+     * @param start current part start
+     * @param maxTokens maximum estimated tokens
+     * @return exclusive end offset
+     */
+    private static int findSplitEnd(String content, int start, int maxTokens) {
+        int cursor = start;
+        int currentTokens = 0;
+        int preferredEnd = -1;
+        int preferredBoundaryThreshold = Math.max(
+            1, (int) Math.floor(maxTokens * PREFERRED_BOUNDARY_MIN_RATIO));
+        while (cursor < content.length()) {
+            int codePoint = content.codePointAt(cursor);
+            int next = cursor + Character.charCount(codePoint);
+            int nextTokens = currentTokens + estimateCodePointTokens(codePoint);
+            if (nextTokens > maxTokens) {
+                break;
+            }
+            cursor = next;
+            currentTokens = nextTokens;
+            if (currentTokens >= preferredBoundaryThreshold
+                && isPreferredBoundary(codePoint)) {
+                preferredEnd = cursor;
+            }
+        }
+        int candidateEnd = preferredEnd > start ? preferredEnd : cursor;
+        if (candidateEnd <= start) {
+            candidateEnd = start + Character.charCount(content.codePointAt(start));
+        }
+        return fitWithinTokenLimit(content, start, candidateEnd, maxTokens);
+    }
+
+    /**
+     * Shrinks a candidate when Unicode normalization expands its token estimate.
+     *
+     * @param content complete source content
+     * @param start part start
+     * @param candidateEnd proposed exclusive end
+     * @param maxTokens maximum estimated tokens
+     * @return exclusive end within the requested token limit
+     */
+    private static int fitWithinTokenLimit(
+        String content,
+        int start,
+        int candidateEnd,
+        int maxTokens) {
+        if (estimateContentTokens(content.substring(start, candidateEnd))
+            <= maxTokens) {
+            return candidateEnd;
+        }
+        int low = 1;
+        int high = content.codePointCount(start, candidateEnd);
+        while (low < high) {
+            int middle = low + (high - low + 1) / 2;
+            int end = content.offsetByCodePoints(start, middle);
+            if (estimateContentTokens(content.substring(start, end)) <= maxTokens) {
+                low = middle;
+            } else {
+                high = middle - 1;
+            }
+        }
+        return content.offsetByCodePoints(start, low);
+    }
+
+    /**
+     * Estimates the normalized token cost of one source code point.
+     *
+     * @param codePoint source Unicode code point
+     * @return conservative token estimate for the code point
+     */
+    private static int estimateCodePointTokens(int codePoint) {
+        if (codePoint >= 0 && codePoint <= 0x7F) {
+            return 1;
+        }
+        String source = new String(Character.toChars(codePoint));
+        String normalized = Normalizer.normalize(source, Normalizer.Form.NFKC);
+        return Math.max(1, normalized.codePointCount(0, normalized.length()));
+    }
+
+    /**
+     * Determines whether a code point is a suitable semantic boundary.
+     *
+     * @param codePoint current Unicode code point
+     * @return true for sentence, paragraph, or whitespace boundaries
+     */
+    private static boolean isPreferredBoundary(int codePoint) {
+        return Character.isWhitespace(codePoint)
+            || codePoint == '。'
+            || codePoint == '！'
+            || codePoint == '？'
+            || codePoint == '；'
+            || codePoint == '.'
+            || codePoint == '!'
+            || codePoint == '?'
+            || codePoint == ';';
+    }
+
+    /**
+     * Immutable UTF-16 source range returned by the hard splitter.
+     */
+    public static final class ChunkRange {
+
+        private final int start;
+        private final int end;
+
+        /**
+         * Creates a source range.
+         *
+         * @param start inclusive UTF-16 offset
+         * @param end exclusive UTF-16 offset
+         */
+        public ChunkRange(int start, int end) {
+            if (start < 0 || end < start) {
+                throw new IllegalArgumentException("Invalid chunk range");
+            }
+            this.start = start;
+            this.end = end;
+        }
+
+        /**
+         * Returns the inclusive start offset.
+         *
+         * @return inclusive UTF-16 offset
+         */
+        public int getStart() {
+            return start;
+        }
+
+        /**
+         * Returns the exclusive end offset.
+         *
+         * @return exclusive UTF-16 offset
+         */
+        public int getEnd() {
+            return end;
+        }
+    }
+}

easy-agents-rag/easy-agents-rag-core/src/main/java/com/easyagents/rag/core/RagDefaults.java

@@ -1,12 +1,26 @@
 package com.easyagents.rag.core;
 
+/**
+ * RAG ingestion default values and safety limits.
+ */
 public final class RagDefaults {
 
     private RagDefaults() {
     }
 
+    /** Default strategy chunk size. */
     public static final int CHUNK_SIZE = 512;
+    /** Default overlap size. */
     public static final int OVERLAP_SIZE = 128;
+    /** Default Markdown heading level. */
     public static final int MD_SPLITTER_LEVEL = 2;
+    /** Default spreadsheet rows per chunk. */
     public static final int ROWS_PER_CHUNK = 1;
+    /** BGE-M3 maximum sequence length. */
+    public static final int BGE_M3_MAX_SEQUENCE_TOKENS = 8192;
+    /** Reserved budget for model special tokens and tokenizer estimation differences. */
+    public static final int BGE_M3_RESERVED_TOKENS = 128;
+    /** Hard upper token estimate for chunk content sent to BGE-M3. */
+    public static final int BGE_M3_HARD_CHUNK_TOKEN_LIMIT =
+        BGE_M3_MAX_SEQUENCE_TOKENS - BGE_M3_RESERVED_TOKENS;
 }

easy-agents-rag/easy-agents-rag-ingestion/src/main/java/com/easyagents/rag/ingestion/chunk/RagSplitStrategyRegistry.java

@@ -21,19 +21,19 @@ public class RagSplitStrategyRegistry {
             strategyCode = analysisResult.getRecommendedStrategyCode();
         }
         String normalizedContent = analysisResult.getNormalizedContent();
+        List<RagChunk> chunks;
         if (RagStrategyCodes.MARKDOWN_SECTION.equals(strategyCode)) {
-            return buildMarkdownChunks(normalizedContent, strategyConfig);
+            chunks = buildMarkdownChunks(normalizedContent, strategyConfig);
+        } else if (RagStrategyCodes.OUTLINE_SECTION.equals(strategyCode)) {
+            chunks = buildOutlineChunks(normalizedContent, strategyConfig);
+        } else if (RagStrategyCodes.QA_PAIR.equals(strategyCode)) {
+            chunks = buildQaChunks(normalizedContent, strategyConfig);
+        } else if (RagStrategyCodes.CUSTOM_REGEX.equals(strategyCode)) {
+            chunks = buildRegexChunks(normalizedContent, strategyConfig);
+        } else {
+            chunks = buildParagraphChunks(normalizedContent, strategyConfig);
         }
-        if (RagStrategyCodes.OUTLINE_SECTION.equals(strategyCode)) {
+        return postProcess(enforceBgeM3HardLimit(chunks));
-            return buildOutlineChunks(normalizedContent, strategyConfig);
-        }
-        if (RagStrategyCodes.QA_PAIR.equals(strategyCode)) {
-            return buildQaChunks(normalizedContent, strategyConfig);
-        }
-        if (RagStrategyCodes.CUSTOM_REGEX.equals(strategyCode)) {
-            return buildRegexChunks(normalizedContent, strategyConfig);
-        }
-        return buildParagraphChunks(normalizedContent, strategyConfig);
     }
 
     private List<RagChunk> buildMarkdownChunks(String content, StrategyConfig strategyConfig) {
@@ -143,7 +143,7 @@ public class RagSplitStrategyRegistry {
                 ));
             }
         }
-        return postProcess(result);
+        return result;
     }
 
     private List<RagChunk> buildQaChunks(String content, StrategyConfig strategyConfig) {
@@ -162,7 +162,14 @@ public class RagSplitStrategyRegistry {
             Matcher questionMatcher = QUESTION_PREFIX.matcher(line);
             Matcher answerMatcher = ANSWER_PREFIX.matcher(line);
             if (questionMatcher.matches()) {
-                qaIndex = flushQaChunk(result, currentQuestion, questionSlices, answerSlices, qaIndex, strategyConfig);
+                qaIndex = flushQaChunk(
+                    result,
+                    content,
+                    currentQuestion,
+                    questionSlices,
+                    answerSlices,
+                    qaIndex,
+                    strategyConfig);
                 currentQuestion = questionMatcher.group(2).trim();
                 questionSlices = new ArrayList<LineSlice>();
                 answerSlices = new ArrayList<LineSlice>();
@@ -187,11 +194,19 @@ public class RagSplitStrategyRegistry {
                 questionSlices.add(lineSlice);
             }
         }
-        flushQaChunk(result, currentQuestion, questionSlices, answerSlices, qaIndex, strategyConfig);
+        flushQaChunk(
-        return postProcess(result);
+            result,
+            content,
+            currentQuestion,
+            questionSlices,
+            answerSlices,
+            qaIndex,
+            strategyConfig);
+        return result;
     }
 
     private int flushQaChunk(List<RagChunk> result,
+                             String content,
                              String currentQuestion,
                              List<LineSlice> questionSlices,
                              List<LineSlice> answerSlices,
@@ -203,27 +218,59 @@ public class RagSplitStrategyRegistry {
         if (answerSlices == null || answerSlices.isEmpty()) {
             return qaIndex;
         }
-        String question = joinLineSlices(questionSlices);
+        TextRange questionRange = mergeLineSlices(questionSlices);
-        String answer = joinLineSlices(answerSlices);
+        TextRange rawAnswerRange = mergeLineSlices(answerSlices);
-        String baseContent = "问题：" + question + "\n答案：" + answer;
+        TextRange answerRange = rawAnswerRange == null
-        List<String> subContents = baseContent.length() > safeChunkSize(strategyConfig)
+            ? null
-            ? splitLongContent(baseContent, strategyConfig.getChunkSize())
+            : trimRange(content, rawAnswerRange.start, rawAnswerRange.end);
-            : Collections.singletonList(baseContent);
+        if (questionRange == null || answerRange == null) {
-        int total = subContents.size();
+            return qaIndex;
-        List<TextRange> sourceRanges = buildQaSourceRanges(questionSlices, answerSlices);
+        }
-        for (int i = 0; i < subContents.size(); i++) {
+
+        String question = currentQuestion.trim();
+        String answer = content.substring(answerRange.start, answerRange.end);
+        String qaPrefix = "问题：" + question + "\n答案：";
+        int requestedTokens = Math.min(
+            safeChunkSize(strategyConfig),
+            RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT);
+        int prefixTokens = BgeM3ChunkSafety.estimateContentTokens(qaPrefix);
+        boolean includePrefix = prefixTokens
+            < RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT;
+        int answerTokenLimit = includePrefix
+            ? Math.max(1, Math.max(requestedTokens, prefixTokens + 1) - prefixTokens)
+            : RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT;
+        List<TextRange> answerPartRanges = new ArrayList<TextRange>();
+        for (BgeM3ChunkSafety.ChunkRange answerPart
+            : BgeM3ChunkSafety.splitToTokenLimit(answer, answerTokenLimit)) {
+            TextRange absoluteAnswerRange = trimRange(
+                content,
+                answerRange.start + answerPart.getStart(),
+                answerRange.start + answerPart.getEnd());
+            if (absoluteAnswerRange != null) {
+                answerPartRanges.add(absoluteAnswerRange);
+            }
+        }
+        int total = answerPartRanges.size();
+        for (int i = 0; i < answerPartRanges.size(); i++) {
+            TextRange absoluteAnswerRange = answerPartRanges.get(i);
+            String answerFragment = content.substring(
+                absoluteAnswerRange.start,
+                absoluteAnswerRange.end);
+            List<TextRange> sourceRanges = new ArrayList<TextRange>();
+            sourceRanges.add(questionRange);
+            sourceRanges.add(absoluteAnswerRange);
             RagChunk chunk = createChunk(
                 RagChunkTypes.QA_PAIR,
                 "Q" + qaIndex + " " + question,
                 Collections.<String>emptyList(),
-                subContents.get(i),
+                includePrefix ? qaPrefix + answerFragment : answerFragment,
                 result.size() + 1,
                 i + 1,
                 total,
                 sourceRanges
             );
             chunk.setQuestion(question);
-            chunk.setAnswer(answer);
+            chunk.setAnswer(answerFragment);
             chunk.getOptions().put(RagMetadataKeys.QA_GROUP_ID, "qa-" + qaIndex);
             result.add(chunk);
         }
@@ -254,7 +301,7 @@ public class RagSplitStrategyRegistry {
             ));
             index++;
         }
-        return postProcess(result);
+        return result;
     }
 
     private List<RagChunk> buildRegexChunks(String content, StrategyConfig strategyConfig) {
@@ -264,42 +311,129 @@ public class RagSplitStrategyRegistry {
         Pattern pattern = Pattern.compile(regex);
         Matcher matcher = pattern.matcher(content);
         int segmentStart = 0;
+        boolean retainRegexMatch = Boolean.TRUE.equals(
+            strategyConfig.getRetainRegexMatch());
         while (matcher.find()) {
             index = addRegexChunk(result, content, segmentStart, matcher.start(), index);
-            segmentStart = matcher.end();
+            // 保留时将匹配内容归入下一分块；默认继续从匹配结束位置开始。
+            segmentStart = retainRegexMatch ? matcher.start() : matcher.end();
         }
         addRegexChunk(result, content, segmentStart, content.length(), index);
-        return postProcess(result);
+        return result;
     }
 
     private List<String> splitLongContent(String content, Integer chunkSize) {
-        int size = chunkSize == null || chunkSize.intValue() <= 0 ? RagDefaults.CHUNK_SIZE : chunkSize.intValue();
+        int size = chunkSize == null || chunkSize.intValue() <= 0
-        String[] paragraphs = content.split("\\n\\s*\\n");
+            ? RagDefaults.CHUNK_SIZE
+            : Math.min(chunkSize.intValue(), RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT);
         List<String> parts = new ArrayList<String>();
-        StringBuilder current = new StringBuilder();
+        for (BgeM3ChunkSafety.ChunkRange range
-        for (String paragraph : paragraphs) {
+            : BgeM3ChunkSafety.splitToTokenLimit(content, size)) {
-            String text = paragraph.trim();
+            String part = content.substring(range.getStart(), range.getEnd()).trim();
-            if (!StringUtil.hasText(text)) {
+            if (StringUtil.hasText(part)) {
-                continue;
+                parts.add(part);
             }
-            if (current.length() > 0 && current.length() + text.length() + 2 > size) {
-                parts.add(current.toString().trim());
-                current = new StringBuilder();
-            }
-            if (current.length() > 0) {
-                current.append("\n\n");
-            }
-            current.append(text);
-        }
-        if (current.length() > 0) {
-            parts.add(current.toString().trim());
-        }
-        if (parts.isEmpty()) {
-            parts.add(content);
         }
         return parts;
     }
 
+    /**
+     * Applies the shared BGE-M3 context safety guard after strategy-specific splitting.
+     *
+     * @param chunks chunks produced by a semantic or length strategy
+     * @return chunks whose conservative token estimates stay within the hard limit
+     */
+    private List<RagChunk> enforceBgeM3HardLimit(List<RagChunk> chunks) {
+        List<RagChunk> result = new ArrayList<RagChunk>();
+        for (RagChunk chunk : chunks) {
+            String content = chunk.getContent();
+            if (BgeM3ChunkSafety.isWithinHardLimit(content)) {
+                result.add(chunk);
+                continue;
+            }
+            List<BgeM3ChunkSafety.ChunkRange> parts =
+                BgeM3ChunkSafety.splitToHardLimit(content);
+            for (int i = 0; i < parts.size(); i++) {
+                BgeM3ChunkSafety.ChunkRange range = parts.get(i);
+                result.add(copyHardSplitChunk(
+                    chunk,
+                    content,
+                    new TextRange(range.getStart(), range.getEnd()),
+                    i + 1,
+                    parts.size()));
+            }
+        }
+        return result;
+    }
+
+    /**
+     * Copies a strategy chunk while replacing only content and hard-split metadata.
+     *
+     * @param source original strategy chunk
+     * @param sourceContent original chunk content
+     * @param range selected content range
+     * @param partNo forced part number
+     * @param partTotal forced part count
+     * @return copied hard-split chunk
+     */
+    private RagChunk copyHardSplitChunk(RagChunk source,
+                                        String sourceContent,
+                                        TextRange range,
+                                        int partNo,
+                                        int partTotal) {
+        RagChunk copy = new RagChunk();
+        copy.setChunkType(source.getChunkType());
+        copy.setSourceLabel(source.getSourceLabel());
+        copy.setHeadingPath(new ArrayList<String>(source.getHeadingPath()));
+        copy.setContent(sourceContent.substring(range.start, range.end));
+        copy.setQuestion(source.getQuestion());
+        copy.setAnswer(RagChunkTypes.QA_PAIR.equals(source.getChunkType())
+            ? copy.getContent()
+            : source.getAnswer());
+        copy.setPartNo(Integer.valueOf(partNo));
+        copy.setPartTotal(Integer.valueOf(partTotal));
+        copy.setWarnings(new ArrayList<String>(source.getWarnings()));
+        copy.setOptions(new LinkedHashMap<String, Object>(source.getOptions()));
+        copy.getOptions().put("hardSplit", Boolean.TRUE);
+        copy.getOptions().put("hardSplitTokenLimit",
+            Integer.valueOf(RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT));
+        adjustSingleSourceRange(copy, sourceContent.length(), range);
+        return copy;
+    }
+
+    /**
+     * Narrows an exact single source range to the forced subrange.
+     *
+     * @param chunk copied hard-split chunk
+     * @param sourceLength original chunk content length
+     * @param relativeRange forced range relative to the original chunk
+     */
+    @SuppressWarnings("unchecked")
+    private void adjustSingleSourceRange(RagChunk chunk, int sourceLength, TextRange relativeRange) {
+        Object rawRanges = chunk.getOptions().get(RagMetadataKeys.SOURCE_RANGES);
+        if (!(rawRanges instanceof List) || ((List<?>) rawRanges).size() != 1) {
+            return;
+        }
+        Object rawRange = ((List<?>) rawRanges).get(0);
+        if (!(rawRange instanceof Map)) {
+            return;
+        }
+        Map<String, Object> sourceRange = (Map<String, Object>) rawRange;
+        Object rawStart = sourceRange.get("start");
+        Object rawEnd = sourceRange.get("end");
+        if (!(rawStart instanceof Number) || !(rawEnd instanceof Number)) {
+            return;
+        }
+        int sourceStart = ((Number) rawStart).intValue();
+        int sourceEnd = ((Number) rawEnd).intValue();
+        if (sourceEnd - sourceStart != sourceLength) {
+            return;
+        }
+        chunk.getOptions().put(RagMetadataKeys.SOURCE_RANGES,
+            toSourceRangeMaps(Collections.singletonList(
+                new TextRange(sourceStart + relativeRange.start, sourceStart + relativeRange.end))));
+    }
+
     private List<RagChunk> postProcess(List<RagChunk> chunks) {
         List<RagChunk> result = new ArrayList<RagChunk>();
         Set<String> dedup = new HashSet<String>();
@@ -318,7 +452,8 @@ public class RagSplitStrategyRegistry {
             }
             chunk.setChunkId("chunk-" + index);
             chunk.setCharCount(Integer.valueOf(content.length()));
-            chunk.setTokenEstimate(Integer.valueOf(Math.max(1, content.length() / 4)));
+            chunk.setTokenEstimate(Integer.valueOf(
+                Math.max(1, BgeM3ChunkSafety.estimateContentTokens(content))));
             result.add(chunk);
             index++;
         }
@@ -460,7 +595,14 @@ public class RagSplitStrategyRegistry {
 
     private int safeOverlap(StrategyConfig strategyConfig) {
         Integer overlapSize = strategyConfig.getOverlapSize();
-        return overlapSize == null || overlapSize.intValue() < 0 ? RagDefaults.OVERLAP_SIZE : overlapSize.intValue();
+        int overlap = overlapSize == null || overlapSize.intValue() < 0
+            ? RagDefaults.OVERLAP_SIZE
+            : overlapSize.intValue();
+        int chunkSize = safeChunkSize(strategyConfig);
+        if (overlap >= chunkSize) {
+            throw new IllegalArgumentException("overlapSize must be smaller than chunkSize");
+        }
+        return overlap;
     }
 
     private String joinAndTrim(List<String> lines) {

easy-agents-rag/easy-agents-rag-ingestion/src/main/java/com/easyagents/rag/ingestion/model/StrategyConfig.java

@@ -11,6 +11,10 @@ public class StrategyConfig implements Serializable {
     private Integer chunkSize = RagDefaults.CHUNK_SIZE;
     private Integer overlapSize = RagDefaults.OVERLAP_SIZE;
     private String regex;
+    /**
+     * Whether a custom-regex match is retained at the beginning of the next chunk.
+     */
+    private Boolean retainRegexMatch;
     private Integer rowsPerChunk = RagDefaults.ROWS_PER_CHUNK;
     private Integer mdSplitterLevel = RagDefaults.MD_SPLITTER_LEVEL;
 
@@ -24,6 +28,7 @@ public class StrategyConfig implements Serializable {
         copy.setChunkSize(this.chunkSize);
         copy.setOverlapSize(this.overlapSize);
         copy.setRegex(this.regex);
+        copy.setRetainRegexMatch(this.retainRegexMatch);
         copy.setRowsPerChunk(this.rowsPerChunk);
         copy.setMdSplitterLevel(this.mdSplitterLevel);
         return copy;
@@ -61,6 +66,25 @@ public class StrategyConfig implements Serializable {
         this.regex = regex;
     }
 
+    /**
+     * Returns whether custom-regex matches are retained in the next chunk.
+     *
+     * @return {@code true} to retain each match at the beginning of the next chunk;
+     * otherwise {@code false} or {@code null}
+     */
+    public Boolean getRetainRegexMatch() {
+        return retainRegexMatch;
+    }
+
+    /**
+     * Sets whether custom-regex matches are retained in the next chunk.
+     *
+     * @param retainRegexMatch {@code true} to retain each match; {@code false} to discard it
+     */
+    public void setRetainRegexMatch(Boolean retainRegexMatch) {
+        this.retainRegexMatch = retainRegexMatch;
+    }
+
     public Integer getRowsPerChunk() {
         return rowsPerChunk;
     }

easy-agents-rag/easy-agents-rag-ingestion/src/test/java/com/easyagents/rag/ingestion/RagIngestionPipelineTest.java

@@ -2,6 +2,7 @@ package com.easyagents.rag.ingestion;
 
 import com.easyagents.rag.core.RagChunk;
 import com.easyagents.rag.core.RagChunkTypes;
+import com.easyagents.rag.core.RagDefaults;
 import com.easyagents.rag.core.RagStrategyCodes;
 import com.easyagents.rag.ingestion.analysis.DocumentStructureAnalyzer;
 import com.easyagents.rag.ingestion.chunk.RagSplitStrategyRegistry;
@@ -97,6 +98,157 @@ public class RagIngestionPipelineTest {
         assertHasValidSourceRanges(analysis, chunks.get(0));
     }
 
+    @Test
+    public void shouldEnforceBgeM3HardLimitForAutoQaTxt() {
+        StringBuilder qa = new StringBuilder("问：自动导入为什么失败？\n答：");
+        int i = 0;
+        while (qa.length() < 77946) {
+            qa.append((char) ('一' + i % 20));
+            if (i > 0 && i % 700 == 0) {
+                qa.append('。');
+            }
+            i++;
+        }
+        qa.setLength(77946);
+        AnalysisResult analysis = recommender.recommend(analyzer.analyze(qa.toString(), "txt"));
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.AUTO);
+
+        List<RagChunk> chunks = registry.split(analysis, config);
+
+        Assert.assertEquals(RagStrategyCodes.QA_PAIR, analysis.getRecommendedStrategyCode());
+        Assert.assertEquals(77946, analysis.getNormalizedContent().length());
+        Assert.assertTrue(chunks.size() > 1);
+        assertWithinBgeM3HardLimit(chunks);
+        Assert.assertTrue(chunks.stream().allMatch(
+            chunk -> chunk.getTokenEstimate().intValue() <= RagDefaults.CHUNK_SIZE));
+        Assert.assertTrue(chunks.stream().allMatch(
+            chunk -> chunk.getAnswer() != null
+                && chunk.getAnswer().length() < analysis.getNormalizedContent().length()));
+        assertHasValidSourceRanges(analysis, chunks.get(0));
+    }
+
+    @Test
+    public void shouldEnforceBgeM3HardLimitForCustomRegexChunk() {
+        StringBuilder content = new StringBuilder();
+        for (int i = 0; i < 20000; i++) {
+            content.append((char) ('甲' + i % 16));
+            if (i > 0 && i % 997 == 0) {
+                content.append('；');
+            }
+        }
+        AnalysisResult analysis = recommender.recommend(analyzer.analyze(content.toString(), "txt"));
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.CUSTOM_REGEX);
+        config.setRegex("\\|NEVER_MATCH\\|");
+
+        List<RagChunk> chunks = registry.split(analysis, config);
+
+        Assert.assertTrue(chunks.size() > 1);
+        assertWithinBgeM3HardLimit(chunks);
+        assertHasValidSourceRanges(analysis, chunks.get(0));
+    }
+
+    /**
+     * Verifies that custom-regex matches remain discarded when the new option is omitted.
+     */
+    @Test
+    public void shouldDiscardCustomRegexMatchByDefault() {
+        AnalysisResult analysis = new AnalysisResult();
+        analysis.setNormalizedContent(
+            "<QUESTION>第一个问题及其答案内容。<QUESTION>第二个问题及其答案内容。");
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.CUSTOM_REGEX);
+        config.setRegex("<QUESTION>");
+
+        List<RagChunk> chunks = registry.split(analysis, config);
+
+        Assert.assertEquals(2, chunks.size());
+        Assert.assertEquals("第一个问题及其答案内容。", chunks.get(0).getContent());
+        Assert.assertEquals("第二个问题及其答案内容。", chunks.get(1).getContent());
+    }
+
+    /**
+     * Verifies that an enabled option retains each match at the beginning of the next chunk.
+     */
+    @Test
+    public void shouldRetainCustomRegexMatchInNextChunk() {
+        AnalysisResult analysis = new AnalysisResult();
+        analysis.setNormalizedContent(
+            "<QUESTION>第一个问题及其答案内容。<QUESTION>第二个问题及其答案内容。");
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.CUSTOM_REGEX);
+        config.setRegex("<QUESTION>");
+        config.setRetainRegexMatch(Boolean.TRUE);
+
+        List<RagChunk> chunks = registry.split(analysis, config);
+
+        Assert.assertEquals(2, chunks.size());
+        Assert.assertEquals(
+            "<QUESTION>第一个问题及其答案内容。",
+            chunks.get(0).getContent());
+        Assert.assertEquals(
+            "<QUESTION>第二个问题及其答案内容。",
+            chunks.get(1).getContent());
+        assertHasValidSourceRanges(analysis, chunks.get(1));
+    }
+
+    @Test
+    public void shouldUseChineseAwareTokenEstimate() {
+        String content = "知识库自动导入向量化失败，需要检查分块大小和模型上下文窗口。";
+        AnalysisResult analysis = recommender.recommend(analyzer.analyze(content, "txt"));
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.PARAGRAPH_LENGTH);
+
+        List<RagChunk> chunks = registry.split(analysis, config);
+
+        Assert.assertEquals(Integer.valueOf(content.codePointCount(0, content.length())),
+            chunks.get(0).getTokenEstimate());
+    }
+
+    @Test
+    public void shouldRejectOverlapEqualToChunkSize() {
+        AnalysisResult analysis = recommender.recommend(
+            analyzer.analyze("知识库分块参数需要保证游标持续向前推进。", "txt"));
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.PARAGRAPH_LENGTH);
+        config.setChunkSize(128);
+        config.setOverlapSize(128);
+
+        try {
+            registry.split(analysis, config);
+            Assert.fail("overlapSize 等于 chunkSize 时应拒绝分块");
+        } catch (IllegalArgumentException expected) {
+            Assert.assertTrue(expected.getMessage().contains("overlapSize"));
+        }
+    }
+
+    @Test
+    public void shouldRejectOverlapGreaterThanChunkSize() {
+        AnalysisResult analysis = recommender.recommend(
+            analyzer.analyze("知识库分块参数需要保证游标持续向前推进。", "txt"));
+        StrategyConfig config = StrategyConfig.defaults();
+        config.setStrategyCode(RagStrategyCodes.PARAGRAPH_LENGTH);
+        config.setChunkSize(128);
+        config.setOverlapSize(512);
+
+        try {
+            registry.split(analysis, config);
+            Assert.fail("overlapSize 大于 chunkSize 时应拒绝分块");
+        } catch (IllegalArgumentException expected) {
+            Assert.assertTrue(expected.getMessage().contains("overlapSize"));
+        }
+    }
+
+    private void assertWithinBgeM3HardLimit(List<RagChunk> chunks) {
+        for (RagChunk chunk : chunks) {
+            Assert.assertNotNull(chunk.getTokenEstimate());
+            Assert.assertTrue(
+                "chunk token estimate exceeds BGE-M3 hard limit: " + chunk.getTokenEstimate(),
+                chunk.getTokenEstimate().intValue() <= RagDefaults.BGE_M3_HARD_CHUNK_TOKEN_LIMIT);
+        }
+    }
+
     @SuppressWarnings("unchecked")
     private void assertHasValidSourceRanges(AnalysisResult analysis, RagChunk chunk) {
         Object rawRanges = chunk.getOptions().get("sourceRanges");

easy-agents-search-engine/easy-agents-search-engine-es/src/main/java/com/easyagents/engine/es/ElasticSearcher.java

@@ -2,8 +2,6 @@ package com.easyagents.engine.es;
 
 import co.elastic.clients.elasticsearch.ElasticsearchClient;
 import co.elastic.clients.elasticsearch.core.*;
-import co.elastic.clients.elasticsearch.core.bulk.BulkOperation;
-import co.elastic.clients.elasticsearch.core.bulk.IndexOperation;
 import co.elastic.clients.elasticsearch.core.search.SourceConfig;
 import co.elastic.clients.json.JsonData;
 import co.elastic.clients.json.jackson.JacksonJsonpMapper;
@@ -25,19 +23,46 @@ import org.slf4j.LoggerFactory;
 import javax.net.ssl.SSLContext;
 import javax.net.ssl.TrustManager;
 import javax.net.ssl.X509TrustManager;
+import java.io.IOException;
 import java.security.KeyManagementException;
 import java.security.NoSuchAlgorithmException;
 import java.security.cert.X509Certificate;
 import java.util.*;
 
-public class ElasticSearcher implements DocumentSearcher {
+/**
+ * 基于 Elasticsearch 的关键词搜索器。
+ *
+ * <p>底层客户端线程安全并与搜索器实例共享生命周期，避免逐文档重复创建网络连接。</p>
+ */
+public class ElasticSearcher implements DocumentSearcher, AutoCloseable {
 
     private static final Logger LOG = LoggerFactory.getLogger(ElasticSearcher.class);
+    static final int MAX_BULK_OPERATIONS = 200;
+    static final long MAX_BULK_ESTIMATED_BYTES = 5L * 1024L * 1024L;
+    private static final long BULK_OPERATION_METADATA_BYTES = 128L;
 
     private final ESConfig esConfig;
+    private final JacksonJsonpMapper jsonpMapper = new JacksonJsonpMapper();
+    private final ElasticsearchTransport transport;
+    private final ElasticsearchClient client;
 
+    /**
+     * 创建 Elasticsearch 搜索器及其共享客户端。
+     *
+     * @param esConfig Elasticsearch 配置
+     * @throws IllegalStateException 客户端初始化失败时抛出
+     */
     public ElasticSearcher(ESConfig esConfig) {
-        this.esConfig = esConfig;
+        this.esConfig = Objects.requireNonNull(esConfig, "ESConfig 不能为 null");
+        RestClient openedRestClient = null;
+        try {
+            openedRestClient = buildRestClient();
+            this.transport = new RestClientTransport(openedRestClient, jsonpMapper);
+            this.client = new ElasticsearchClient(transport);
+        } catch (Exception e) {
+            closeQuietly(openedRestClient);
+            throw new IllegalStateException("初始化 Elasticsearch 客户端失败", e);
+        }
     }
 
     // 忽略 SSL 的 client 构建逻辑
@@ -76,52 +101,128 @@ public class ElasticSearcher implements DocumentSearcher {
 
 
     /**
-     * 添加文档到Elasticsearch
+     * 添加文档到 Elasticsearch。
+     *
+     * @param document 待写入文档
+     * @return 写入成功时返回 {@code true}
      */
     @Override
     public boolean addDocument(Document document) {
-        if (document == null || document.getContent() == null) {
+        return addDocuments(Collections.singletonList(document));
+    }
+
+    /**
+     * 使用有界 Bulk 请求批量写入文档。
+     *
+     * @param documents 待写入文档
+     * @return 全部文档写入成功时返回 {@code true}
+     */
+    @Override
+    public boolean addDocuments(List<Document> documents) {
+        if (!areValidDocuments(documents)) {
             return false;
         }
-
-        RestClient restClient = null;
-        ElasticsearchTransport transport = null;
         try {
-            restClient = buildRestClient();
+            List<List<Document>> batches = partitionDocuments(documents);
-            transport = new RestClientTransport(restClient, new JacksonJsonpMapper());
+            for (int batchIndex = 0; batchIndex < batches.size(); batchIndex++) {
-            ElasticsearchClient client = new ElasticsearchClient(transport);
+                List<Document> batch = batches.get(batchIndex);
-
+                BulkResponse response = client.bulk(buildBulkRequest(batch));
-            Map<String, Object> source = buildSource(document);
+                if (!isBulkSuccessful(response, "写入", batchIndex, batch.size())) {
-            String documentId = document.getId().toString();
+                    return false;
-            IndexOperation<?> indexOp = IndexOperation.of(i -> i
+                }
-                .index(esConfig.getIndexName())
+            }
-                .id(documentId)
+            return true;
-                .document(JsonData.of(source))
-            );
-
-            BulkOperation bulkOp = BulkOperation.of(b -> b.index(indexOp));
-            BulkRequest request = BulkRequest.of(b -> b.operations(Collections.singletonList(bulkOp)));
-            BulkResponse response = client.bulk(request);
-            return !response.errors();
-
         } catch (Exception e) {
-            LOG.error(e.getMessage(), e);
+            LOG.error("Elasticsearch 批量写入异常: count={}", documents.size(), e);
             return false;
-        } finally {
-            closeResources(transport, restClient);
         }
     }
 
+    /**
+     * 按操作数量和序列化后的估算字节数拆分 Bulk 批次。
+     *
+     * @param documents 已完成校验的文档
+     * @return 有界文档批次
+     * @throws IOException 文档序列化失败时抛出
+     * @throws IllegalArgumentException 单个文档超过批次字节限制时抛出
+     */
+    List<List<Document>> partitionDocuments(List<Document> documents) throws IOException {
+        List<List<Document>> batches = new ArrayList<>();
+        List<Document> currentBatch = new ArrayList<>(Math.min(documents.size(), MAX_BULK_OPERATIONS));
+        long currentBatchBytes = 0L;
+        for (Document document : documents) {
+            long documentBytes = estimateBulkOperationBytes(document);
+            if (documentBytes > MAX_BULK_ESTIMATED_BYTES) {
+                throw new IllegalArgumentException(
+                    "单个 Elasticsearch 索引文档超过 Bulk 字节限制: id=" + document.getId());
+            }
+            if (!currentBatch.isEmpty()
+                && (currentBatch.size() >= MAX_BULK_OPERATIONS
+                || currentBatchBytes + documentBytes > MAX_BULK_ESTIMATED_BYTES)) {
+                batches.add(currentBatch);
+                currentBatch = new ArrayList<>(Math.min(documents.size(), MAX_BULK_OPERATIONS));
+                currentBatchBytes = 0L;
+            }
+            currentBatch.add(document);
+            currentBatchBytes += documentBytes;
+        }
+        if (!currentBatch.isEmpty()) {
+            batches.add(currentBatch);
+        }
+        return batches;
+    }
+
+    private long estimateBulkOperationBytes(Document document) throws IOException {
+        long sourceBytes = jsonpMapper.objectMapper().writeValueAsBytes(buildSource(document)).length;
+        return sourceBytes
+            + utf8Length(esConfig.getIndexName())
+            + utf8Length(document.getId().toString())
+            + BULK_OPERATION_METADATA_BYTES;
+    }
+
+    private int utf8Length(String value) {
+        return value == null ? 0 : value.getBytes(java.nio.charset.StandardCharsets.UTF_8).length;
+    }
+
+    /**
+     * 构建包含全部文档的 Bulk 请求。
+     *
+     * @param documents 已完成校验的文档
+     * @return Bulk 请求
+     */
+    BulkRequest buildBulkRequest(List<Document> documents) {
+        BulkRequest.Builder builder = new BulkRequest.Builder();
+        for (Document document : documents) {
+            builder.operations(operation -> operation.index(index -> index
+                .index(esConfig.getIndexName())
+                .id(document.getId().toString())
+                .document(JsonData.of(buildSource(document)))
+            ));
+        }
+        return builder.build();
+    }
+
+    private boolean areValidDocuments(List<Document> documents) {
+        if (documents == null || documents.isEmpty()) {
+            return false;
+        }
+        for (Document document : documents) {
+            if (document == null || document.getId() == null || document.getContent() == null) {
+                return false;
+            }
+        }
+        return true;
+    }
+
+    /**
+     * 按请求条件搜索文档。
+     *
+     * @param request 搜索请求
+     * @return 命中文档
+     */
     @Override
     public List<Document> searchDocuments(KeywordSearchRequest request) {
-        RestClient restClient = null;
-        ElasticsearchTransport transport = null;
-
         try {
-            restClient = buildRestClient();
-            transport = new RestClientTransport(restClient, new JacksonJsonpMapper());
-            ElasticsearchClient client = new ElasticsearchClient(transport);
-
             SearchResponse<Map> response = client.search(buildSearchRequest(request), Map.class);
             List<Document> results = new ArrayList<>();
             response.hits().hits().forEach(hit -> {
@@ -137,54 +238,114 @@ public class ElasticSearcher implements DocumentSearcher {
         } catch (Exception e) {
             LOG.error(e.getMessage(), e);
             return Collections.emptyList();
-        } finally {
-            closeResources(transport, restClient);
         }
     }
 
+    /**
+     * 删除指定文档。
+     *
+     * @param id 文档 ID
+     * @return 删除成功时返回 {@code true}
+     */
     @Override
     public boolean deleteDocument(Object id) {
-        if (id == null) {
+        return deleteDocuments(Collections.singletonList(id));
+    }
+
+    /**
+     * 使用有界 Bulk 请求批量删除文档。
+     *
+     * <p>删除不存在的文档按幂等成功处理；仅 Bulk 条目包含实际错误时返回失败。</p>
+     *
+     * @param ids 文档 ID 集合
+     * @return 全部删除操作成功时返回 {@code true}
+     */
+    @Override
+    public boolean deleteDocuments(Collection<?> ids) {
+        if (ids == null || ids.isEmpty()) {
             return false;
         }
-
+        List<String> documentIds = new ArrayList<>(ids.size());
-        RestClient restClient = null;
+        for (Object id : ids) {
-        ElasticsearchTransport transport = null;
+            if (id == null) {
+                return false;
+            }
+            documentIds.add(id.toString());
+        }
         try {
-            restClient = buildRestClient();
+            for (int start = 0, batchIndex = 0;
-            transport = new RestClientTransport(restClient, new JacksonJsonpMapper());
+                 start < documentIds.size();
-            ElasticsearchClient client = new ElasticsearchClient(transport);
+                 start += MAX_BULK_OPERATIONS, batchIndex++) {
-
+                int end = Math.min(start + MAX_BULK_OPERATIONS, documentIds.size());
-            DeleteRequest request = DeleteRequest.of(d -> d
+                List<String> batch = documentIds.subList(start, end);
-                .index(esConfig.getIndexName())
+                BulkResponse response = client.bulk(buildDeleteBulkRequest(batch));
-                .id(id.toString())
+                if (!isBulkSuccessful(response, "删除", batchIndex, batch.size())) {
-            );
+                    return false;
-
+                }
-            DeleteResponse response = client.delete(request);
+            }
-            return response.result() == co.elastic.clients.elasticsearch._types.Result.Deleted;
+            return true;
-
         } catch (Exception e) {
-            LOG.error("Error deleting document with id: " + id, e);
+            LOG.error("Elasticsearch 批量删除异常: count={}", documentIds.size(), e);
             return false;
-        } finally {
-            closeResources(transport, restClient);
         }
     }
 
+    /**
+     * 构建批量删除请求。
+     *
+     * @param ids 文档 ID
+     * @return Bulk 删除请求
+     */
+    BulkRequest buildDeleteBulkRequest(List<String> ids) {
+        BulkRequest.Builder builder = new BulkRequest.Builder();
+        for (String id : ids) {
+            builder.operations(operation -> operation.delete(delete -> delete
+                .index(esConfig.getIndexName())
+                .id(id)
+            ));
+        }
+        return builder.build();
+    }
+
+    private boolean isBulkSuccessful(BulkResponse response,
+                                     String operation,
+                                     int batchIndex,
+                                     int batchSize) {
+        if (response != null && !response.errors()) {
+            return true;
+        }
+        if (response == null) {
+            LOG.error("Elasticsearch 批量{}未返回结果: batchIndex={}, batchSize={}",
+                operation, batchIndex, batchSize);
+            return false;
+        }
+        response.items().stream()
+            .filter(item -> item.error() != null)
+            .forEach(item -> LOG.error(
+                "Elasticsearch 批量{}失败: batchIndex={}, index={}, id={}, status={}, reason={}",
+                operation,
+                batchIndex,
+                item.index(),
+                item.id(),
+                item.status(),
+                item.error().reason()
+            ));
+        return false;
+    }
+
+    /**
+     * 更新指定文档。
+     *
+     * @param document 待更新文档
+     * @return 更新成功时返回 {@code true}
+     */
     @Override
     public boolean updateDocument(Document document) {
         if (document == null || document.getId() == null) {
             return false;
         }
 
-        RestClient restClient = null;
-        ElasticsearchTransport transport = null;
-
         try {
-            restClient = buildRestClient();
-            transport = new RestClientTransport(restClient, new JacksonJsonpMapper());
-            ElasticsearchClient client = new ElasticsearchClient(transport);
-
             UpdateRequest<Map<String, Object>, Map<String, Object>> request = UpdateRequest.of(u -> u
                 .index(esConfig.getIndexName())
                 .id(document.getId().toString())
@@ -199,20 +360,17 @@ public class ElasticSearcher implements DocumentSearcher {
         } catch (Exception e) {
             LOG.error("Error updating document with id: " + document.getId(), e);
             return false;
-        } finally {
-            closeResources(transport, restClient);
         }
     }
 
-
+    private static void closeQuietly(AutoCloseable closeable) {
-    private void closeResources(AutoCloseable... closeables) {
+        if (closeable == null) {
-        for (AutoCloseable closeable : closeables) {
+            return;
-            try {
+        }
-                if (closeable != null)
+        try {
-                    closeable.close();
+            closeable.close();
-            } catch (Exception e) {
+        } catch (Exception ignored) {
-                LOG.error("Error closing resource", e);
+            // 初始化失败时仅执行尽力清理，原始异常由构造方法继续抛出。
-            }
         }
     }
 
@@ -284,19 +442,27 @@ public class ElasticSearcher implements DocumentSearcher {
         );
     }
 
+    /**
+     * 检查 Elasticsearch 服务是否可用。
+     *
+     * @return 服务可访问时返回 {@code true}
+     */
     public boolean checkAvailable() {
-        RestClient restClient = null;
-        ElasticsearchTransport transport = null;
         try {
-            restClient = buildRestClient();
-            transport = new RestClientTransport(restClient, new JacksonJsonpMapper());
-            ElasticsearchClient client = new ElasticsearchClient(transport);
             return client.info() != null;
         } catch (Exception e) {
             LOG.error("Elasticsearch availability check failed", e);
             return false;
-        } finally {
-            closeResources(transport, restClient);
         }
     }
+
+    /**
+     * 关闭共享传输层及其底层 RestClient。
+     *
+     * @throws IOException 关闭客户端资源失败时抛出
+     */
+    @Override
+    public void close() throws IOException {
+        transport.close();
+    }
 }

easy-agents-search-engine/easy-agents-search-engine-es/src/test/java/com/easyagents/engine/es/ElasticSearcherQueryBuilderTest.java

@@ -1,48 +1,185 @@
 package com.easyagents.engine.es;
 
 import co.elastic.clients.elasticsearch.core.SearchRequest;
+import co.elastic.clients.elasticsearch.core.BulkRequest;
 import com.easyagents.core.document.Document;
 import com.easyagents.search.engine.service.KeywordSearchMetadataKeys;
 import com.easyagents.search.engine.service.KeywordSearchRequest;
 import org.junit.Assert;
 import org.junit.Test;
 
+import java.util.ArrayList;
+import java.util.List;
 import java.util.Map;
 
+/**
+ * {@link ElasticSearcher} 请求构建回归测试。
+ */
 public class ElasticSearcherQueryBuilderTest {
 
+    /**
+     * 验证搜索请求同时包含多字段检索和知识库过滤。
+     *
+     * @throws Exception Elasticsearch 客户端关闭失败时抛出
+     */
     @Test
-    public void shouldBuildSearchRequestWithMultiMatchAndKnowledgeFilter() {
+    public void shouldBuildSearchRequestWithMultiMatchAndKnowledgeFilter() throws Exception {
-        ElasticSearcher searcher = new ElasticSearcher(config());
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
-        KeywordSearchRequest request = KeywordSearchRequest.of("客服", 5);
+            KeywordSearchRequest request = KeywordSearchRequest.of("客服", 5);
-        request.setKnowledgeId("100");
+            request.setKnowledgeId("100");
 
-        SearchRequest searchRequest = searcher.buildSearchRequest(request);
+            SearchRequest searchRequest = searcher.buildSearchRequest(request);
 
-        Assert.assertEquals(5, searchRequest.size().intValue());
+            Assert.assertEquals(5, searchRequest.size().intValue());
-        Assert.assertNotNull(searchRequest.query().bool());
+            Assert.assertNotNull(searchRequest.query().bool());
-        Assert.assertEquals(1, searchRequest.query().bool().must().size());
+            Assert.assertEquals(1, searchRequest.query().bool().must().size());
-        Assert.assertNotNull(searchRequest.query().bool().must().get(0).multiMatch());
+            Assert.assertNotNull(searchRequest.query().bool().must().get(0).multiMatch());
-        Assert.assertEquals(2, searchRequest.query().bool().must().get(0).multiMatch().fields().size());
+            Assert.assertEquals(2, searchRequest.query().bool().must().get(0).multiMatch().fields().size());
-        Assert.assertEquals(1, searchRequest.query().bool().filter().size());
+            Assert.assertEquals(1, searchRequest.query().bool().filter().size());
-        Assert.assertEquals("knowledgeId", searchRequest.query().bool().filter().get(0).term().field());
+            Assert.assertEquals("knowledgeId", searchRequest.query().bool().filter().get(0).term().field());
+        }
     }
 
+    /**
+     * 验证知识库 ID 会写入顶层字段。
+     *
+     * @throws Exception Elasticsearch 客户端关闭失败时抛出
+     */
     @Test
-    public void shouldExtractKnowledgeIdToTopLevelSource() {
+    public void shouldExtractKnowledgeIdToTopLevelSource() throws Exception {
-        ElasticSearcher searcher = new ElasticSearcher(config());
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
+            Document document = new Document();
+            document.setId("1");
+            document.setTitle("title");
+            document.setContent("content");
+            document.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "100");
+
+            Map<String, Object> source = searcher.buildSource(document);
+
+            Assert.assertEquals("100", source.get(KeywordSearchMetadataKeys.KNOWLEDGE_ID));
+            Assert.assertTrue(source.get("metadataMap") instanceof Map);
+        }
+    }
+
+    /**
+     * 验证多个文档会进入同一个 Bulk 请求并保留稳定文档 ID。
+     *
+     * @throws Exception Elasticsearch 客户端关闭失败时抛出
+     */
+    @Test
+    public void shouldBuildSingleBulkRequestForMultipleDocuments() throws Exception {
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
+            Document first = new Document();
+            first.setId("first");
+            first.setContent("first content");
+            Document second = new Document();
+            second.setId("second");
+            second.setContent("second content");
+
+            BulkRequest request = searcher.buildBulkRequest(List.of(first, second));
+
+            Assert.assertEquals(2, request.operations().size());
+            Assert.assertEquals("first", request.operations().get(0).index().id());
+            Assert.assertEquals("second", request.operations().get(1).index().id());
+        }
+    }
+
+    /**
+     * 验证 Bulk 写入会按最大操作数拆分，避免单次请求无界增长。
+     *
+     * @throws Exception Elasticsearch 客户端关闭或文档序列化失败时抛出
+     */
+    @Test
+    public void shouldPartitionBulkRequestsByOperationCount() throws Exception {
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
+            List<Document> documents = new ArrayList<>();
+            for (int index = 0; index <= ElasticSearcher.MAX_BULK_OPERATIONS; index++) {
+                documents.add(document("count-" + index, "content"));
+            }
+
+            List<List<Document>> batches = searcher.partitionDocuments(documents);
+
+            Assert.assertEquals(2, batches.size());
+            Assert.assertEquals(ElasticSearcher.MAX_BULK_OPERATIONS, batches.get(0).size());
+            Assert.assertEquals(1, batches.get(1).size());
+        }
+    }
+
+    /**
+     * 验证 Bulk 写入会按序列化字节数拆分。
+     *
+     * @throws Exception Elasticsearch 客户端关闭或文档序列化失败时抛出
+     */
+    @Test
+    public void shouldPartitionBulkRequestsByEstimatedBytes() throws Exception {
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
+            int contentLength = (int) (ElasticSearcher.MAX_BULK_ESTIMATED_BYTES / 2L);
+            Document first = document("bytes-first", "a".repeat(contentLength));
+            Document second = document("bytes-second", "b".repeat(contentLength));
+
+            List<List<Document>> batches = searcher.partitionDocuments(List.of(first, second));
+
+            Assert.assertEquals(2, batches.size());
+            Assert.assertEquals(1, batches.get(0).size());
+            Assert.assertEquals(1, batches.get(1).size());
+        }
+    }
+
+    /**
+     * 验证单个超限文档会在发送请求前失败。
+     *
+     * @throws Exception Elasticsearch 客户端关闭或文档序列化失败时抛出
+     */
+    @Test
+    public void shouldRejectSingleDocumentOverByteLimit() throws Exception {
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
+            int contentLength = (int) ElasticSearcher.MAX_BULK_ESTIMATED_BYTES;
+            Document oversized = document("oversized", "a".repeat(contentLength));
+
+            try {
+                searcher.partitionDocuments(List.of(oversized));
+                Assert.fail("单个超限文档应拒绝构建 Bulk 请求");
+            } catch (IllegalArgumentException expected) {
+                Assert.assertTrue(expected.getMessage().contains("oversized"));
+            }
+        }
+    }
+
+    /**
+     * 验证批量删除请求包含全部稳定文档 ID。
+     *
+     * @throws Exception Elasticsearch 客户端关闭失败时抛出
+     */
+    @Test
+    public void shouldBuildBulkDeleteRequest() throws Exception {
+        try (ElasticSearcher searcher = new ElasticSearcher(config())) {
+            BulkRequest request = searcher.buildDeleteBulkRequest(List.of("first", "second"));
+
+            Assert.assertEquals(2, request.operations().size());
+            Assert.assertEquals("first", request.operations().get(0).delete().id());
+            Assert.assertEquals("second", request.operations().get(1).delete().id());
+        }
+    }
+
+    /**
+     * 创建测试文档。
+     *
+     * @param id 文档 ID
+     * @param content 文档内容
+     * @return 文档
+     */
+    private Document document(String id, String content) {
         Document document = new Document();
-        document.setId("1");
+        document.setId(id);
-        document.setTitle("title");
+        document.setContent(content);
-        document.setContent("content");
+        return document;
-        document.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "100");
-
-        Map<String, Object> source = searcher.buildSource(document);
-
-        Assert.assertEquals("100", source.get(KeywordSearchMetadataKeys.KNOWLEDGE_ID));
-        Assert.assertTrue(source.get("metadataMap") instanceof Map);
     }
 
+    /**
+     * 创建测试用 Elasticsearch 配置。
+     *
+     * @return Elasticsearch 配置
+     */
     private ESConfig config() {
         ESConfig config = new ESConfig();
         config.setHost("http://127.0.0.1:9200");

easy-agents-search-engine/easy-agents-search-engine-lucene/src/main/java/com/easyagents/search/engine/lucene/LuceneSearcher.java

@@ -29,7 +29,7 @@ import org.apache.lucene.queryparser.classic.QueryParser;
 import org.apache.lucene.search.*;
 import org.apache.lucene.store.Directory;
 import org.apache.lucene.store.FSDirectory;
-import org.jetbrains.annotations.NotNull;
+import org.apache.lucene.util.IOUtils;
 import org.lionsoul.jcseg.ISegment;
 import org.lionsoul.jcseg.analyzer.JcsegAnalyzer;
 import org.lionsoul.jcseg.dic.DictionaryFactory;
@@ -40,17 +40,38 @@ import org.slf4j.LoggerFactory;
 import java.io.File;
 import java.io.IOException;
 import java.util.ArrayList;
+import java.util.Collection;
+import java.util.Collections;
+import java.util.LinkedHashSet;
 import java.util.List;
 import java.util.Objects;
+import java.util.Set;
 
-public class LuceneSearcher implements DocumentSearcher {
+/**
+ * 基于 Lucene 的本地关键词搜索器。
+ *
+ * <p>每个实例长期复用一个线程安全的 {@link IndexWriter}，避免并发任务重复抢占
+ * 同一索引目录的 {@code write.lock}。</p>
+ */
+public class LuceneSearcher implements DocumentSearcher, AutoCloseable {
 
     private static final Logger LOG = LoggerFactory.getLogger(LuceneSearcher.class);
 
-    private Directory directory;
+    private final Directory directory;
+    private final Analyzer analyzer;
+    private final IndexWriter indexWriter;
 
+    /**
+     * 创建 Lucene 搜索器并打开索引写入器。
+     *
+     * @param config Lucene 配置
+     * @throws IllegalStateException 索引目录或写入器初始化失败时抛出
+     */
     public LuceneSearcher(LuceneConfig config) {
         Objects.requireNonNull(config, "LuceneConfig 不能为 null");
+        Directory openedDirectory = null;
+        Analyzer openedAnalyzer = null;
+        IndexWriter openedWriter = null;
         try {
             String indexDirPath = config.getIndexDirPath(); // 索引目录路径
             File indexDir = new File(indexDirPath);
@@ -58,89 +79,123 @@ public class LuceneSearcher implements DocumentSearcher {
                 throw new IllegalStateException("can not mkdirs for path: " + indexDirPath);
             }
 
-            this.directory = FSDirectory.open(indexDir.toPath());
+            openedDirectory = FSDirectory.open(indexDir.toPath());
-        } catch (IOException e) {
+            openedAnalyzer = createAnalyzer();
+            openedWriter = new IndexWriter(openedDirectory, new IndexWriterConfig(openedAnalyzer));
+        } catch (IOException | RuntimeException e) {
+            IOUtils.closeWhileHandlingException(openedWriter, openedAnalyzer, openedDirectory);
             LOG.error("初始化 Lucene 索引失败", e);
-            throw new RuntimeException(e);
+            throw new IllegalStateException("初始化 Lucene 索引失败", e);
         }
+        this.directory = openedDirectory;
+        this.analyzer = openedAnalyzer;
+        this.indexWriter = openedWriter;
     }
 
+    /**
+     * 以文档 ID 为键写入或覆盖单个文档。
+     *
+     * @param document 待写入文档
+     * @return 写入成功时返回 {@code true}
+     */
     @Override
     public boolean addDocument(Document document) {
-        if (document == null || document.getContent() == null) return false;
+        return addDocuments(Collections.singletonList(document));
+    }
 
-        IndexWriter indexWriter = null;
+    /**
-        try {
+     * 批量写入或覆盖文档，并在批次结束后统一提交。
-            indexWriter = createIndexWriter();
+     *
-
+     * @param documents 待写入文档
-            org.apache.lucene.document.Document luceneDoc = new org.apache.lucene.document.Document();
+     * @return 全部文档写入成功时返回 {@code true}
-            luceneDoc.add(new StringField("id", document.getId().toString(), Field.Store.YES));
+     */
-            luceneDoc.add(new TextField("content", document.getContent(), Field.Store.YES));
+    @Override
-
+    public boolean addDocuments(List<Document> documents) {
-            if (document.getTitle() != null) {
+        if (documents == null || documents.isEmpty()) {
-                luceneDoc.add(new TextField("title", document.getTitle(), Field.Store.YES));
+            return false;
+        }
+        List<org.apache.lucene.document.Document> luceneDocuments = new ArrayList<>(documents.size());
+        for (Document document : documents) {
+            if (document == null || document.getId() == null || document.getContent() == null) {
+                return false;
+            }
+            luceneDocuments.add(toLuceneDocument(document));
+        }
+        try {
+            for (org.apache.lucene.document.Document luceneDocument : luceneDocuments) {
+                String documentId = luceneDocument.get("id");
+                // updateDocument 以 ID 执行 upsert，确保失败重试不会生成重复索引。
+                indexWriter.updateDocument(new Term("id", documentId), luceneDocument);
             }
-            appendKnowledgeId(document, luceneDoc);
-
-            indexWriter.addDocument(luceneDoc);
             indexWriter.commit();
             return true;
         } catch (Exception e) {
-            LOG.error("添加文档失败", e);
+            LOG.error("批量添加文档失败: count={}", documents.size(), e);
             return false;
-        } finally {
-            close(indexWriter);
         }
     }
 
-
+    /**
+     * 删除指定文档。
+     *
+     * @param id 文档 ID
+     * @return 删除成功时返回 {@code true}
+     */
     @Override
     public boolean deleteDocument(Object id) {
-        if (id == null) return false;
+        return deleteDocuments(Collections.singletonList(id));
+    }
 
-        IndexWriter indexWriter = null;
+    /**
+     * 批量删除指定文档，并在批次结束后统一提交。
+     *
+     * @param ids 文档 ID 集合
+     * @return 全部文档删除成功时返回 {@code true}
+     */
+    @Override
+    public boolean deleteDocuments(Collection<?> ids) {
+        if (ids == null || ids.isEmpty()) {
+            return false;
+        }
+        Set<String> uniqueIds = new LinkedHashSet<>();
+        for (Object id : ids) {
+            if (id == null) {
+                return false;
+            }
+            uniqueIds.add(id.toString());
+        }
         try {
-            indexWriter = createIndexWriter();
+            Term[] terms = new Term[uniqueIds.size()];
-            Term term = new Term("id", id.toString());
+            int index = 0;
-            indexWriter.deleteDocuments(term);
+            for (String id : uniqueIds) {
+                terms[index++] = new Term("id", id);
+            }
+            indexWriter.deleteDocuments(terms);
             indexWriter.commit();
             return true;
         } catch (IOException e) {
-            LOG.error("删除文档失败", e);
+            LOG.error("批量删除文档失败: count={}", uniqueIds.size(), e);
             return false;
-        } finally {
-            close(indexWriter);
         }
     }
 
+    /**
+     * 以文档 ID 为键更新文档。
+     *
+     * @param document 待更新文档
+     * @return 更新成功时返回 {@code true}
+     */
     @Override
     public boolean updateDocument(Document document) {
-        if (document == null || document.getId() == null) return false;
+        return addDocument(document);
-
-        IndexWriter indexWriter = null;
-        try {
-            indexWriter = createIndexWriter();
-            Term term = new Term("id", document.getId().toString());
-
-            org.apache.lucene.document.Document luceneDoc = new org.apache.lucene.document.Document();
-            luceneDoc.add(new StringField("id", document.getId().toString(), Field.Store.YES));
-            luceneDoc.add(new TextField("content", document.getContent(), Field.Store.YES));
-
-            if (document.getTitle() != null) {
-                luceneDoc.add(new TextField("title", document.getTitle(), Field.Store.YES));
-            }
-            appendKnowledgeId(document, luceneDoc);
-            indexWriter.updateDocument(term, luceneDoc);
-            indexWriter.commit();
-            return true;
-        } catch (IOException e) {
-            LOG.error("更新文档失败", e);
-            return false;
-        } finally {
-            close(indexWriter);
-        }
     }
 
+    /**
+     * 按请求条件搜索文档。
+     *
+     * @param request 搜索请求
+     * @return 命中文档
+     */
     @Override
     public List<Document> searchDocuments(KeywordSearchRequest request) {
         List<Document> results = new ArrayList<>();
@@ -171,7 +226,6 @@ public class LuceneSearcher implements DocumentSearcher {
 
     Query buildQuery(KeywordSearchRequest request) {
         try {
-            Analyzer analyzer = createAnalyzer();
             String keyword = request == null ? null : request.getKeyword();
 
             QueryParser titleQueryParser = new QueryParser("title", analyzer);
@@ -195,20 +249,22 @@ public class LuceneSearcher implements DocumentSearcher {
         return null;
     }
 
-
-    @NotNull
-    private IndexWriter createIndexWriter() throws IOException {
-        Analyzer analyzer = createAnalyzer();
-        IndexWriterConfig indexWriterConfig = new IndexWriterConfig(analyzer);
-        return new IndexWriter(directory, indexWriterConfig);
-    }
-
-
     private static Analyzer createAnalyzer() {
         SegmenterConfig config = new SegmenterConfig(true);
         return new JcsegAnalyzer(ISegment.Type.NLP, config, DictionaryFactory.createSingletonDictionary(config));
     }
 
+    private org.apache.lucene.document.Document toLuceneDocument(Document document) {
+        org.apache.lucene.document.Document luceneDoc = new org.apache.lucene.document.Document();
+        luceneDoc.add(new StringField("id", document.getId().toString(), Field.Store.YES));
+        luceneDoc.add(new TextField("content", document.getContent(), Field.Store.YES));
+        if (document.getTitle() != null) {
+            luceneDoc.add(new TextField("title", document.getTitle(), Field.Store.YES));
+        }
+        appendKnowledgeId(document, luceneDoc);
+        return luceneDoc;
+    }
+
     private void appendKnowledgeId(Document document, org.apache.lucene.document.Document luceneDoc) {
         if (document == null || document.getMetadataMap() == null) {
             return;
@@ -219,13 +275,13 @@ public class LuceneSearcher implements DocumentSearcher {
         }
     }
 
-    public void close(IndexWriter indexWriter) {
+    /**
-        try {
+     * 关闭写入器、分词器和索引目录。
-            if (indexWriter != null) {
+     *
-                indexWriter.close();
+     * @throws IOException 任一 Lucene 资源关闭失败时抛出
-            }
+     */
-        } catch (IOException e) {
+    @Override
-            LOG.error("关闭 Lucene 失败", e);
+    public void close() throws IOException {
-        }
+        IOUtils.close(indexWriter, analyzer, directory);
     }
 }

easy-agents-search-engine/easy-agents-search-engine-lucene/src/test/java/com/easyagents/search/engine/lucene/LuceneSearcherTest.java

@@ -8,38 +8,145 @@ import org.junit.Test;
 
 import java.nio.file.Files;
 import java.nio.file.Path;
+import java.util.ArrayList;
 import java.util.List;
+import java.util.concurrent.CountDownLatch;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
+import java.util.concurrent.Future;
 
+/**
+ * {@link LuceneSearcher} 回归测试。
+ */
 public class LuceneSearcherTest {
 
+    /**
+     * 验证知识库过滤同时覆盖标题和正文。
+     *
+     * @throws Exception 临时目录或 Lucene 资源操作失败时抛出
+     */
     @Test
     public void shouldFilterByKnowledgeIdAndSearchTitleAndContent() throws Exception {
         Path tempDir = Files.createTempDirectory("lucene-searcher-test");
         LuceneConfig config = new LuceneConfig();
         config.setIndexDirPath(tempDir.toString());
-        LuceneSearcher searcher = new LuceneSearcher(config);
+        try (LuceneSearcher searcher = new LuceneSearcher(config)) {
+            Document first = new Document();
+            first.setId("1");
+            first.setTitle("客服标题");
+            first.setContent("这里没有关键字");
+            first.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "100");
 
-        Document first = new Document();
+            Document second = new Document();
-        first.setId("1");
+            second.setId("2");
-        first.setTitle("客服标题");
+            second.setTitle("别的知识库");
-        first.setContent("这里没有关键字");
+            second.setContent("客服内容");
-        first.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "100");
+            second.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "200");
 
-        Document second = new Document();
+            Assert.assertTrue(searcher.addDocument(first));
-        second.setId("2");
+            Assert.assertTrue(searcher.addDocument(second));
-        second.setTitle("别的知识库");
-        second.setContent("客服内容");
-        second.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "200");
 
-        Assert.assertTrue(searcher.addDocument(first));
+            KeywordSearchRequest request = KeywordSearchRequest.of("客服", 10);
-        Assert.assertTrue(searcher.addDocument(second));
+            request.setKnowledgeId("100");
+            List<Document> results = searcher.searchDocuments(request);
 
-        KeywordSearchRequest request = KeywordSearchRequest.of("客服", 10);
+            Assert.assertEquals(1, results.size());
-        request.setKnowledgeId("100");
+            Assert.assertEquals("1", String.valueOf(results.get(0).getId()));
-        List<Document> results = searcher.searchDocuments(request);
+            Assert.assertEquals("100", String.valueOf(
+                results.get(0).getMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID)));
+        }
+    }
 
-        Assert.assertEquals(1, results.size());
+    /**
-        Assert.assertEquals("1", String.valueOf(results.get(0).getId()));
+     * 验证多个导入线程可共享同一个 IndexWriter 完成批量写入。
-        Assert.assertEquals("100", String.valueOf(results.get(0).getMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID)));
+     *
+     * @throws Exception 并发执行或 Lucene 资源操作失败时抛出
+     */
+    @Test
+    public void shouldWriteConcurrentBatchesWithoutLockConflict() throws Exception {
+        Path tempDir = Files.createTempDirectory("lucene-searcher-concurrent-test");
+        LuceneConfig config = new LuceneConfig();
+        config.setIndexDirPath(tempDir.toString());
+        int workerCount = 4;
+        int documentsPerWorker = 20;
+        ExecutorService executor = Executors.newFixedThreadPool(workerCount);
+        try (LuceneSearcher searcher = new LuceneSearcher(config)) {
+            CountDownLatch startSignal = new CountDownLatch(1);
+            List<Future<Boolean>> futures = new ArrayList<>();
+            for (int worker = 0; worker < workerCount; worker++) {
+                int currentWorker = worker;
+                futures.add(executor.submit(() -> {
+                    startSignal.await();
+                    List<Document> documents = new ArrayList<>();
+                    for (int offset = 0; offset < documentsPerWorker; offset++) {
+                        Document document = new Document();
+                        document.setId(currentWorker + "-" + offset);
+                        document.setContent("concurrent indexing content");
+                        document.addMetadata(KeywordSearchMetadataKeys.KNOWLEDGE_ID, "concurrent-kb");
+                        documents.add(document);
+                    }
+                    return searcher.addDocuments(documents);
+                }));
+            }
+            startSignal.countDown();
+            for (Future<Boolean> future : futures) {
+                Assert.assertTrue(future.get());
+            }
+
+            int expectedCount = workerCount * documentsPerWorker;
+            KeywordSearchRequest request = KeywordSearchRequest.of("concurrent", expectedCount);
+            request.setKnowledgeId("concurrent-kb");
+            Assert.assertEquals(expectedCount, searcher.searchDocuments(request).size());
+        } finally {
+            executor.shutdownNow();
+        }
+    }
+
+    /**
+     * 验证相同文档 ID 重试写入时执行覆盖而不会保留重复旧索引。
+     *
+     * @throws Exception 临时目录或 Lucene 资源操作失败时抛出
+     */
+    @Test
+    public void shouldUpsertDocumentByIdOnRetry() throws Exception {
+        Path tempDir = Files.createTempDirectory("lucene-searcher-upsert-test");
+        LuceneConfig config = new LuceneConfig();
+        config.setIndexDirPath(tempDir.toString());
+        try (LuceneSearcher searcher = new LuceneSearcher(config)) {
+            Document document = new Document();
+            document.setId("retry-id");
+            document.setContent("oldkeyword");
+            Assert.assertTrue(searcher.addDocument(document));
+
+            document.setContent("newkeyword");
+            Assert.assertTrue(searcher.addDocument(document));
+
+            Assert.assertTrue(searcher.searchDocuments("oldkeyword", 10).isEmpty());
+            Assert.assertEquals(1, searcher.searchDocuments("newkeyword", 10).size());
+        }
+    }
+
+    /**
+     * 验证多个文档可在一次提交中批量删除。
+     *
+     * @throws Exception 临时目录或 Lucene 资源操作失败时抛出
+     */
+    @Test
+    public void shouldDeleteDocumentsInSingleBatch() throws Exception {
+        Path tempDir = Files.createTempDirectory("lucene-searcher-delete-test");
+        LuceneConfig config = new LuceneConfig();
+        config.setIndexDirPath(tempDir.toString());
+        try (LuceneSearcher searcher = new LuceneSearcher(config)) {
+            Document first = new Document();
+            first.setId("delete-first");
+            first.setContent("batchdelete");
+            Document second = new Document();
+            second.setId("delete-second");
+            second.setContent("batchdelete");
+
+            Assert.assertTrue(searcher.addDocuments(List.of(first, second)));
+            Assert.assertTrue(searcher.deleteDocuments(List.of(first.getId(), second.getId())));
+            Assert.assertTrue(searcher.searchDocuments("batchdelete", 10).isEmpty());
+        }
     }
 }

easy-agents-search-engine/easy-agents-search-engine-service/src/main/java/com/easyagents/search/engine/service/DocumentSearcher.java

@@ -17,23 +17,106 @@ package com.easyagents.search.engine.service;
 
 import com.easyagents.core.document.Document;
 
+import java.util.Collection;
 import java.util.List;
 
+/**
+ * 关键词搜索引擎统一接口。
+ */
 public interface DocumentSearcher {
 
+    /**
+     * 写入单个文档。
+     *
+     * @param document 待写入文档
+     * @return 写入成功时返回 {@code true}
+     */
     boolean addDocument(Document document);
 
+    /**
+     * 批量写入文档。
+     *
+     * <p>默认实现保持现有搜索引擎兼容性；支持原生批量写入的实现应覆盖该方法。</p>
+     *
+     * @param documents 待写入文档
+     * @return 全部文档写入成功时返回 {@code true}
+     */
+    default boolean addDocuments(List<Document> documents) {
+        if (documents == null || documents.isEmpty()) {
+            return false;
+        }
+        for (Document document : documents) {
+            if (!addDocument(document)) {
+                return false;
+            }
+        }
+        return true;
+    }
+
+    /**
+     * 删除指定文档。
+     *
+     * @param id 文档 ID
+     * @return 删除成功时返回 {@code true}
+     */
     boolean deleteDocument(Object id);
 
+    /**
+     * 批量删除指定文档。
+     *
+     * <p>默认实现保持现有搜索引擎兼容性，并确保所有文档都会尝试删除；
+     * 支持原生批量删除的实现应覆盖该方法。</p>
+     *
+     * @param ids 文档 ID 集合
+     * @return 全部文档删除成功时返回 {@code true}
+     */
+    default boolean deleteDocuments(Collection<?> ids) {
+        if (ids == null || ids.isEmpty()) {
+            return false;
+        }
+        boolean success = true;
+        for (Object id : ids) {
+            if (!deleteDocument(id)) {
+                success = false;
+            }
+        }
+        return success;
+    }
+
+    /**
+     * 更新指定文档。
+     *
+     * @param document 待更新文档
+     * @return 更新成功时返回 {@code true}
+     */
     boolean updateDocument(Document document);
 
+    /**
+     * 使用默认返回数量搜索文档。
+     *
+     * @param keyword 搜索关键词
+     * @return 命中文档
+     */
     default List<Document> searchDocuments(String keyword) {
         return searchDocuments(KeywordSearchRequest.of(keyword, 10));
     }
 
+    /**
+     * 搜索指定数量的文档。
+     *
+     * @param keyword 搜索关键词
+     * @param count 最大返回数量
+     * @return 命中文档
+     */
     default List<Document> searchDocuments(String keyword, int count) {
         return searchDocuments(KeywordSearchRequest.of(keyword, count));
     }
 
+    /**
+     * 按请求条件搜索文档。
+     *
+     * @param request 搜索请求
+     * @return 命中文档
+     */
     List<Document> searchDocuments(KeywordSearchRequest request);
 }