* 修复部分xlsx无法读取超链接的bug

README.md

@@ -42,7 +42,7 @@ Java解析、生成Excel比较有名的框架有Apache poi、jxl。但他们都
  <dependency>
             <groupId>com.alibaba</groupId>
             <artifactId>easyexcel</artifactId>
-            <version>3.1.0</version>
+            <version>3.1.1</version>
             <exclusions>
                 <exclusion>
                     <artifactId>poi-ooxml-schemas</artifactId>

easyexcel-core/src/main/java/com/alibaba/excel/analysis/v07/XlsxSaxAnalyser.java

@@ -8,6 +8,7 @@ import java.util.HashMap;
 import java.util.Iterator;
 import java.util.List;
 import java.util.Map;
+import java.util.Optional;
 import java.util.UUID;
 
 import javax.xml.parsers.ParserConfigurationException;
@@ -25,17 +26,23 @@ import com.alibaba.excel.metadata.CellExtra;
 import com.alibaba.excel.read.metadata.ReadSheet;
 import com.alibaba.excel.read.metadata.holder.xlsx.XlsxReadWorkbookHolder;
 import com.alibaba.excel.util.FileUtils;
+import com.alibaba.excel.util.MapUtils;
 import com.alibaba.excel.util.SheetUtils;
 import com.alibaba.excel.util.StringUtils;
 
 import lombok.extern.slf4j.Slf4j;
 import org.apache.commons.collections4.CollectionUtils;
+import org.apache.poi.openxml4j.exceptions.InvalidFormatException;
 import org.apache.poi.openxml4j.opc.OPCPackage;
 import org.apache.poi.openxml4j.opc.PackageAccess;
 import org.apache.poi.openxml4j.opc.PackagePart;
+import org.apache.poi.openxml4j.opc.PackagePartName;
+import org.apache.poi.openxml4j.opc.PackageRelationshipCollection;
+import org.apache.poi.openxml4j.opc.PackagingURIHelper;
 import org.apache.poi.ss.util.CellAddress;
 import org.apache.poi.xssf.eventusermodel.XSSFReader;
 import org.apache.poi.xssf.model.CommentsTable;
+import org.apache.poi.xssf.model.SharedStringsTable;
 import org.apache.poi.xssf.usermodel.XSSFComment;
 import org.apache.poi.xssf.usermodel.XSSFRelation;
 import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTWorkbook;
@@ -52,6 +59,20 @@ import org.xml.sax.XMLReader;
 @Slf4j
 public class XlsxSaxAnalyser implements ExcelReadExecutor {
 
+    /**
+     * Storage sheet SharedStrings
+     */
+    public static final PackagePartName SHARED_STRINGS_PART_NAME;
+
+    static {
+        try {
+            SHARED_STRINGS_PART_NAME = PackagingURIHelper.createPartName("/xl/sharedStrings.xml");
+        } catch (InvalidFormatException e) {
+            log.error("Initialize the XlsxSaxAnalyser failure", e);
+            throw new ExcelAnalysisException("Initialize the XlsxSaxAnalyser failure", e);
+        }
+    }
+
     private final XlsxReadContext xlsxReadContext;
     private final List<ReadSheet> sheetList;
     private final Map<Integer, InputStream> sheetMap;
@@ -68,11 +89,9 @@ public class XlsxSaxAnalyser implements ExcelReadExecutor {
         OPCPackage pkg = readOpcPackage(xlsxReadWorkbookHolder, decryptedStream);
         xlsxReadWorkbookHolder.setOpcPackage(pkg);
 
-        ArrayList<PackagePart> packageParts = pkg.getPartsByContentType(XSSFRelation.SHARED_STRINGS.getContentType());
-
-        if (!CollectionUtils.isEmpty(packageParts)) {
-            PackagePart sharedStringsTablePackagePart = packageParts.get(0);
-
+        // Read the Shared information Strings
+        PackagePart sharedStringsTablePackagePart = pkg.getPart(SHARED_STRINGS_PART_NAME);
+        if (sharedStringsTablePackagePart != null) {
             // Specify default cache
             defaultReadCache(xlsxReadWorkbookHolder, sharedStringsTablePackagePart);
 
@@ -89,6 +108,9 @@ public class XlsxSaxAnalyser implements ExcelReadExecutor {
         sheetList = new ArrayList<>();
         sheetMap = new HashMap<>();
         commentsTableMap = new HashMap<>();
+        Map<Integer, PackageRelationshipCollection> packageRelationshipCollectionMap = MapUtils.newHashMap();
+        xlsxReadWorkbookHolder.setPackageRelationshipCollectionMap(packageRelationshipCollectionMap);
+
         XSSFReader.SheetIterator ite = (XSSFReader.SheetIterator)xssfReader.getSheetsData();
         int index = 0;
         if (!ite.hasNext()) {
@@ -104,6 +126,20 @@ public class XlsxSaxAnalyser implements ExcelReadExecutor {
                     commentsTableMap.put(index, commentsTable);
                 }
             }
+            if (xlsxReadContext.readWorkbookHolder().getExtraReadSet().contains(CellExtraTypeEnum.HYPERLINK)) {
+                PackageRelationshipCollection packageRelationshipCollection = Optional.ofNullable(ite.getSheetPart())
+                    .map(packagePart -> {
+                        try {
+                            return packagePart.getRelationships();
+                        } catch (InvalidFormatException e) {
+                            log.warn("Reading the Relationship failed", e);
+                            return null;
+                        }
+                    }).orElse(null);
+                if (packageRelationshipCollection != null) {
+                    packageRelationshipCollectionMap.put(index, packageRelationshipCollection);
+                }
+            }
             index++;
         }
     }

easyexcel-core/src/main/java/com/alibaba/excel/analysis/v07/handlers/HyperlinkTagHandler.java

@@ -1,5 +1,9 @@
 package com.alibaba.excel.analysis.v07.handlers;
 
+import java.util.Optional;
+
+import org.apache.poi.openxml4j.opc.PackageRelationship;
+import org.apache.poi.openxml4j.opc.PackageRelationshipCollection;
 import org.xml.sax.Attributes;
 
 import com.alibaba.excel.constant.ExcelXmlConstants;
@@ -23,13 +27,32 @@ public class HyperlinkTagHandler extends AbstractXlsxTagHandler {
     @Override
     public void startElement(XlsxReadContext xlsxReadContext, String name, Attributes attributes) {
         String ref = attributes.getValue(ExcelXmlConstants.ATTRIBUTE_REF);
-        String location = attributes.getValue(ExcelXmlConstants.ATTRIBUTE_LOCATION);
         if (StringUtils.isEmpty(ref)) {
             return;
         }
+        // Hyperlink has 2 case:
+        // case 1，In the 'location' tag
+        String location = attributes.getValue(ExcelXmlConstants.ATTRIBUTE_LOCATION);
+        if (location != null) {
             CellExtra cellExtra = new CellExtra(CellExtraTypeEnum.HYPERLINK, location, ref);
             xlsxReadContext.readSheetHolder().setCellExtra(cellExtra);
             xlsxReadContext.analysisEventProcessor().extra(xlsxReadContext);
+            return;
+        }
+        // case 2, In the 'r:id' tag, Then go to 'PackageRelationshipCollection' to get inside
+        String rId = attributes.getValue(ExcelXmlConstants.ATTRIBUTE_RID);
+        PackageRelationshipCollection packageRelationshipCollection = xlsxReadContext.xlsxReadSheetHolder()
+            .getPackageRelationshipCollection();
+        if (rId == null || packageRelationshipCollection == null) {
+            return;
+        }
+        Optional.ofNullable(packageRelationshipCollection.getRelationshipByID(rId))
+            .map(PackageRelationship::getTargetURI)
+            .ifPresent(uri -> {
+                CellExtra cellExtra = new CellExtra(CellExtraTypeEnum.HYPERLINK, uri.toString(), ref);
+                xlsxReadContext.readSheetHolder().setCellExtra(cellExtra);
+                xlsxReadContext.analysisEventProcessor().extra(xlsxReadContext);
+            });
     }
 
 }

easyexcel-core/src/main/java/com/alibaba/excel/constant/ExcelXmlConstants.java

@@ -49,6 +49,11 @@ public class ExcelXmlConstants {
      */
     public static final String ATTRIBUTE_LOCATION = "location";
 
+    /**
+     * rId attribute
+     */
+    public static final String ATTRIBUTE_RID = "r:id";
+
     /**
      * Cell range split
      */

easyexcel-core/src/main/java/com/alibaba/excel/read/metadata/holder/xlsx/XlsxReadSheetHolder.java

@@ -10,6 +10,7 @@ import com.alibaba.excel.read.metadata.holder.ReadWorkbookHolder;
 import lombok.EqualsAndHashCode;
 import lombok.Getter;
 import lombok.Setter;
+import org.apache.poi.openxml4j.opc.PackageRelationshipCollection;
 
 /**
  * sheet holder
@@ -36,9 +37,16 @@ public class XlsxReadSheetHolder extends ReadSheetHolder {
      * Formula for current label.
      */
     private StringBuilder tempFormula;
+    /**
+     * excel Relationship
+     */
+    private PackageRelationshipCollection packageRelationshipCollection;
 
     public XlsxReadSheetHolder(ReadSheet readSheet, ReadWorkbookHolder readWorkbookHolder) {
         super(readSheet, readWorkbookHolder);
         this.tagDeque = new LinkedList<String>();
+        packageRelationshipCollection
+            = ((XlsxReadWorkbookHolder)readWorkbookHolder).getPackageRelationshipCollectionMap().get(
+            readSheet.getSheetNo());
     }
 }

easyexcel-core/src/main/java/com/alibaba/excel/read/metadata/holder/xlsx/XlsxReadWorkbookHolder.java

@@ -15,6 +15,7 @@ import lombok.EqualsAndHashCode;
 import lombok.Getter;
 import lombok.Setter;
 import org.apache.poi.openxml4j.opc.OPCPackage;
+import org.apache.poi.openxml4j.opc.PackageRelationshipCollection;
 import org.apache.poi.xssf.model.StylesTable;
 import org.apache.poi.xssf.usermodel.XSSFCellStyle;
 
@@ -51,6 +52,11 @@ public class XlsxReadWorkbookHolder extends ReadWorkbookHolder {
      */
     private Map<Integer, DataFormatData> dataFormatDataCache;
 
+    /**
+     * excel Relationship, key: sheetNo value: PackageRelationshipCollection
+     */
+    private Map<Integer, PackageRelationshipCollection> packageRelationshipCollectionMap;
+
     public XlsxReadWorkbookHolder(ReadWorkbook readWorkbook) {
         super(readWorkbook);
         this.saxParserFactoryName = readWorkbook.getXlsxSAXParserFactoryName();

easyexcel-test/src/test/java/com/alibaba/easyexcel/test/core/extra/ExtraDataTest.java

@@ -2,6 +2,7 @@ package com.alibaba.easyexcel.test.core.extra;
 
 import java.io.File;
 
+import org.junit.Assert;
 import org.junit.BeforeClass;
 import org.junit.Test;
 import org.slf4j.Logger;
@@ -9,10 +10,13 @@ import org.slf4j.LoggerFactory;
 
 import com.alibaba.easyexcel.test.util.TestFileUtil;
 import com.alibaba.excel.EasyExcel;
+import com.alibaba.excel.context.AnalysisContext;
 import com.alibaba.excel.enums.CellExtraTypeEnum;
+import com.alibaba.excel.metadata.CellExtra;
+import com.alibaba.excel.read.listener.ReadListener;
+import com.alibaba.fastjson.JSON;
 
 /**
- *
  * @author Jiaju Zhuang
  */
 public class ExtraDataTest {
@@ -20,10 +24,13 @@ public class ExtraDataTest {
     private static File file03;
     private static File file07;
 
+    private static File extraRelationships;
+
     @BeforeClass
     public static void init() {
         file03 = TestFileUtil.readFile("extra" + File.separator + "extra.xls");
         file07 = TestFileUtil.readFile("extra" + File.separator + "extra.xlsx");
+        extraRelationships = TestFileUtil.readFile("extra" + File.separator + "extraRelationships.xlsx");
     }
 
     @Test
@@ -36,6 +43,41 @@ public class ExtraDataTest {
         read(file03);
     }
 
+    @Test
+    public void t03Read() {
+        EasyExcel.read(extraRelationships, ExtraData.class, new ReadListener() {
+                @Override
+                public void invoke(Object data, AnalysisContext context) {
+                }
+
+                @Override
+                public void doAfterAllAnalysed(AnalysisContext context) {
+                }
+
+                @Override
+                public void extra(CellExtra extra, AnalysisContext context) {
+                    LOGGER.info("extra data:{}", JSON.toJSONString(extra));
+                    switch (extra.getType()) {
+                        case HYPERLINK:
+                            if ("222222222".equals(extra.getText())) {
+                                Assert.assertEquals(1, (int)extra.getRowIndex());
+                                Assert.assertEquals(0, (int)extra.getColumnIndex());
+                            } else if ("333333333333".equals(extra.getText())) {
+                                Assert.assertEquals(1, (int)extra.getRowIndex());
+                                Assert.assertEquals(1, (int)extra.getColumnIndex());
+                            } else {
+                                Assert.fail("Unknown hyperlink!");
+                            }
+                            break;
+                        default:
+                    }
+                }
+            })
+            .extraRead(CellExtraTypeEnum.HYPERLINK)
+            .sheet()
+            .doRead();
+    }
+
     private void read(File file) {
         EasyExcel.read(file, ExtraData.class, new ExtraDataListener()).extraRead(CellExtraTypeEnum.COMMENT)
             .extraRead(CellExtraTypeEnum.HYPERLINK).extraRead(CellExtraTypeEnum.MERGE).sheet().doRead();

pom.xml

@@ -20,7 +20,7 @@
 
 
     <properties>
-        <revision>3.1.0</revision>
+        <revision>3.1.1</revision>
         <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
         <jdk.version>1.8</jdk.version>
         <gpg.skip>true</gpg.skip>
@@ -145,7 +145,7 @@
             <dependency>
                 <groupId>com.alibaba</groupId>
                 <artifactId>fastjson</artifactId>
-                <version>1.2.78</version>
+                <version>2.0.6</version>
             </dependency>
             <dependency>
                 <groupId>org.springframework.boot</groupId>

update.md

@@ -1,3 +1,6 @@
+# 3.1.1
+* 修复部分xlsx无法读取超链接的bug
+
 # 3.1.0
 * 支持jdk17,去除cglib&asm依赖，改成重新拷贝一份 [Issue #2240](https://github.com/alibaba/easyexcel/issues/2240)
 * 升级ehcache 到 3.9.9 ,为了兼容jdk17