* 读csv会忽略BOM数据 [Issue #3137](#3137)

* 解决csv用office打开乱码的问题，写csv默认带上BOM数据

easyexcel-core/pom.xml

@@ -40,5 +40,9 @@
  <groupId>org.ehcache</groupId>
  <artifactId>ehcache</artifactId>
  </dependency>
+ <dependency>
+ <groupId>commons-io</groupId>
+ <artifactId>commons-io</artifactId>
+ </dependency>
  </dependencies>
 </project>

easyexcel-core/src/main/java/com/alibaba/excel/analysis/csv/CsvExcelReadExecutor.java

@@ -1,6 +1,7 @@
 package com.alibaba.excel.analysis.csv;
 
 import java.io.IOException;
+import java.io.InputStream;
 import java.io.InputStreamReader;
 import java.nio.file.Files;
 import java.util.ArrayList;
@@ -11,6 +12,7 @@
 
 import com.alibaba.excel.analysis.ExcelReadExecutor;
 import com.alibaba.excel.context.csv.CsvReadContext;
+import com.alibaba.excel.enums.ByteOrderMarkEnum;
 import com.alibaba.excel.enums.CellDataTypeEnum;
 import com.alibaba.excel.enums.RowTypeEnum;
 import com.alibaba.excel.exception.ExcelAnalysisException;
@@ -27,6 +29,7 @@
 import org.apache.commons.csv.CSVFormat;
 import org.apache.commons.csv.CSVParser;
 import org.apache.commons.csv.CSVRecord;
+import org.apache.commons.io.input.BOMInputStream;
 
 /**
  * read executor
@@ -82,17 +85,26 @@ public void execute() {
  private CSVParser csvParser() throws IOException {
  CsvReadWorkbookHolder csvReadWorkbookHolder = csvReadContext.csvReadWorkbookHolder();
  CSVFormat csvFormat = csvReadWorkbookHolder.getCsvFormat();
-
+ ByteOrderMarkEnum byteOrderMark = ByteOrderMarkEnum.valueOfByCharsetName(
+ csvReadContext.csvReadWorkbookHolder().getCharset().name());
  if (csvReadWorkbookHolder.getMandatoryUseInputStream()) {
- return csvFormat.parse(
- new InputStreamReader(csvReadWorkbookHolder.getInputStream(), csvReadWorkbookHolder.getCharset()));
+ return buildCsvParser(csvFormat, csvReadWorkbookHolder.getInputStream(), byteOrderMark);
  }
  if (csvReadWorkbookHolder.getFile() != null) {
- return csvFormat.parse(new InputStreamReader(Files.newInputStream(csvReadWorkbookHolder.getFile().toPath()),
- csvReadWorkbookHolder.getCharset()));
+ return buildCsvParser(csvFormat, Files.newInputStream(csvReadWorkbookHolder.getFile().toPath()),
+ byteOrderMark);
+ }
+ return buildCsvParser(csvFormat, csvReadWorkbookHolder.getInputStream(), byteOrderMark);
+ }
+
+ private CSVParser buildCsvParser(CSVFormat csvFormat, InputStream inputStream, ByteOrderMarkEnum byteOrderMark)
+ throws IOException {
+ if (byteOrderMark == null) {
+ return csvFormat.parse(
+ new InputStreamReader(inputStream, csvReadContext.csvReadWorkbookHolder().getCharset()));
  }
- return csvFormat.parse(
- new InputStreamReader(csvReadWorkbookHolder.getInputStream(), csvReadWorkbookHolder.getCharset()));
+ return csvFormat.parse(new InputStreamReader(new BOMInputStream(inputStream, byteOrderMark.getByteOrderMark()),
+ csvReadContext.csvReadWorkbookHolder().getCharset()));
  }
 
  private void dealRecord(CSVRecord record, int rowIndex) {

easyexcel-core/src/main/java/com/alibaba/excel/enums/ByteOrderMarkEnum.java

@@ -0,0 +1,51 @@
+package com.alibaba.excel.enums;
+
+import java.nio.charset.Charset;
+import java.util.Map;
+
+import com.alibaba.excel.util.MapUtils;
+
+import lombok.Getter;
+import org.apache.commons.io.ByteOrderMark;
+
+/**
+ * byte order mark
+ *
+ * @author Jiaju Zhuang
+ */
+@Getter
+public enum ByteOrderMarkEnum {
+
+ UTF_8(ByteOrderMark.UTF_8),
+ UTF_16BE(ByteOrderMark.UTF_16BE),
+ UTF_16LE(ByteOrderMark.UTF_16LE),
+ UTF_32BE(ByteOrderMark.UTF_32BE),
+ UTF_32LE(ByteOrderMark.UTF_32LE),
+
+ ;
+
+ final ByteOrderMark byteOrderMark;
+ final String stringPrefix;
+
+ ByteOrderMarkEnum(ByteOrderMark byteOrderMark) {
+ this.byteOrderMark = byteOrderMark;
+ Charset charset = Charset.forName(byteOrderMark.getCharsetName());
+ this.stringPrefix = new String(byteOrderMark.getBytes(), charset);
+ }
+
+ /**
+ * store character aliases corresponding to `ByteOrderMark` prefix
+ */
+ private static final Map<String, ByteOrderMarkEnum> CHARSET_BYTE_ORDER_MARK_MAP = MapUtils.newHashMap();
+
+ static {
+ for (ByteOrderMarkEnum value : ByteOrderMarkEnum.values()) {
+ CHARSET_BYTE_ORDER_MARK_MAP.put(value.getByteOrderMark().getCharsetName(), value);
+ }
+ }
+
+ public static ByteOrderMarkEnum valueOfByCharsetName(String charsetName) {
+ return CHARSET_BYTE_ORDER_MARK_MAP.get(charsetName);
+ }
+
+}

easyexcel-core/src/main/java/com/alibaba/excel/metadata/csv/CsvSheet.java

@@ -3,24 +3,30 @@
 import java.io.Closeable;
 import java.io.IOException;
 import java.math.BigDecimal;
+import java.nio.charset.Charset;
+import java.nio.charset.StandardCharsets;
 import java.util.Collection;
 import java.util.Iterator;
 import java.util.List;
 import java.util.Map;
 
 import com.alibaba.excel.constant.BuiltinFormats;
+import com.alibaba.excel.enums.ByteOrderMarkEnum;
 import com.alibaba.excel.enums.NumericCellTypeEnum;
 import com.alibaba.excel.exception.ExcelGenerateException;
 import com.alibaba.excel.util.DateUtils;
 import com.alibaba.excel.util.ListUtils;
+import com.alibaba.excel.util.MapUtils;
 import com.alibaba.excel.util.NumberDataFormatterUtils;
 import com.alibaba.excel.util.StringUtils;
 
 import lombok.EqualsAndHashCode;
 import lombok.Getter;
 import lombok.Setter;
+import org.apache.commons.collections4.CollectionUtils;
 import org.apache.commons.csv.CSVFormat;
 import org.apache.commons.csv.CSVPrinter;
+import org.apache.commons.io.ByteOrderMark;
 import org.apache.poi.ss.usermodel.AutoFilter;
 import org.apache.poi.ss.usermodel.Cell;
 import org.apache.poi.ss.usermodel.CellRange;
@@ -51,6 +57,7 @@
 @Setter
 @EqualsAndHashCode
 public class CsvSheet implements Sheet, Closeable {
+
  /**
  * workbook
  */
@@ -109,6 +116,13 @@ private void initSheet() {
  }
  rowCache = ListUtils.newArrayListWithExpectedSize(rowCacheCount);
  try {
+ if (csvWorkbook.getWithBom()) {
+ ByteOrderMarkEnum byteOrderMark = ByteOrderMarkEnum.valueOfByCharsetName(
+ csvWorkbook.getCharset().name());
+ if (byteOrderMark != null) {
+ out.append(byteOrderMark.getStringPrefix());
+ }
+ }
  csvPrinter = csvFormat.print(out);
  } catch (IOException e) {
  throw new ExcelGenerateException(e);