lukeye/doc-parser-server
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

兼容doc/xls格式的文件

Browse Source
This commit is contained in:
luke 2025-06-21 10:00:35 +08:00
parent 9ec83eec49
commit 9670477f51
3 changed files with 42 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
pom.xml
View File

@ -39,6 +39,7 @@
<pdfbox.version>2.0.29</pdfbox.version>
<poi.version>5.2.3</poi.version>
<poi-ooxml.version>5.2.3</poi-ooxml.version>
<poi-scratchpad.version>5.2.3</poi-scratchpad.version>
<mysql-connector-java.version>8.0.33</mysql-connector-java.version>
<mybatis-plus-boot-starter.version>3.5.5</mybatis-plus-boot-starter.version>
@ -190,6 +191,11 @@
<artifactId>poi-ooxml</artifactId>
<version>${poi-ooxml.version}</version>
</dependency>
<dependency>
<groupId>org.apache.poi</groupId>
<artifactId>poi-scratchpad</artifactId>
<version>${poi-scratchpad.version}</version>
</dependency>
<!-- MySQL JDBC 驱动 -->
<dependency>
@ -346,6 +352,10 @@
<groupId>org.apache.poi</groupId>
<artifactId>poi-ooxml</artifactId>
</dependency>
<dependency>
<groupId>org.apache.poi</groupId>
<artifactId>poi-scratchpad</artifactId>
</dependency>
<dependency>
<groupId>mysql</groupId>

15
src/main/java/com/knowledge/base/domain/doc/service/impl/ExcelImporter.java
View File

@ -1,9 +1,7 @@
package com.knowledge.base.domain.doc.service.impl;
/**
* @author Luke.ye
* @date 2025/5/25 14:54
*/
import org.apache.poi.hssf.usermodel.HSSFWorkbook;
import org.apache.poi.ss.usermodel.*;
import org.apache.poi.xssf.usermodel.XSSFWorkbook;
import org.springframework.beans.factory.annotation.Value;
@ -15,6 +13,10 @@ import java.nio.file.Path;
import java.util.Iterator;
import java.util.Set;
/**
* @author Luke.ye
* @date 2025/5/25 14:54
*/
@Component
public class ExcelImporter extends AbstractBaseFileImporter {
@ -43,8 +45,11 @@ public class ExcelImporter extends AbstractBaseFileImporter {
@Override
protected String extractContent(Path path) throws IOException {
String fileName = path.getFileName().toString().toLowerCase();
boolean isXlsx = fileName.endsWith(".xlsx");
try (FileInputStream fis = new FileInputStream(path.toFile());
Workbook workbook = new XSSFWorkbook(fis)) {
Workbook workbook = isXlsx ? new XSSFWorkbook(fis) : new HSSFWorkbook(fis)) {
Sheet sheet = workbook.getSheetAt(0);
StringBuilder sb = new StringBuilder();

27
src/main/java/com/knowledge/base/domain/doc/service/impl/WordImporter.java
View File

@ -1,6 +1,7 @@
package com.knowledge.base.domain.doc.service.impl;
import com.knowledge.base.domain.common.enums.DocTypeEnum;
import org.apache.poi.hwpf.HWPFDocument;
import org.springframework.stereotype.Component;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFParagraph;
@ -44,12 +45,28 @@ public class WordImporter extends AbstractBaseFileImporter {
@Override
protected String extractContent(Path path) throws IOException {
try (XWPFDocument document = new XWPFDocument(new FileInputStream(path.toFile()))) {
StringBuilder content = new StringBuilder();
for (XWPFParagraph para : document.getParagraphs()) {
content.append(para.getText()).append("\n");
String fileName = path.getFileName().toString().toLowerCase();
if (fileName.endsWith(".docx")) {
// 处理 .docx
try (FileInputStream fis = new FileInputStream(path.toFile());
XWPFDocument document = new XWPFDocument(fis)) {
StringBuilder content = new StringBuilder();
for (XWPFParagraph para : document.getParagraphs()) {
content.append(para.getText()).append("\n");
}
return content.toString().trim();
}
return content.toString().trim();
} else if (fileName.endsWith(".doc")) {
// 处理 .doc
try (FileInputStream fis = new FileInputStream(path.toFile());
HWPFDocument doc = new HWPFDocument(fis)) {
return doc.getDocumentText().trim();
}
} else {
throw new IllegalArgumentException("不支持的 Word 文件类型: " + fileName);
}
}
}