基于HttpClient上传文件中文名乱码的解决

现象

使用HttpClient工具上传文件时，如果文件名是中文，文件名会乱码

文件名乱码的代码：

private HttpEntity buildEntity(Long scenarioId, List<String> groupIds, String extension,File fileToUpload) {
         MultipartEntityBuilder builder = MultipartEntityBuilder.create();
         builder.addTextBody("scenarioId", scenarioId.toString());
         for (String groupId : groupIds) {
             builder.addTextBody("groupIds", groupId);
         }
         builder.addTextBody("extension", extension);
         builder.addPart("fileToUpload", new FileBody(fileToUpload));
         builder.addTextBody("type", AssetFileTypeEnum.CSV.getName());
         builder.addTextBody("isSplit", "false");
         builder.addTextBody("isRefresh", "false");
         return builder.build();

乱码原因：

HttpClient上传文件时，会调用doWriteTo方法，写一个输出流，但是在调用formatMultipartHeader方法时，底层主要有3种不同的实现，3种方式的采用的字符集不一样

HttpClient中的doWriteTo方法：

void doWriteTo(
      final OutputStream out,
      final boolean writeContent) throws IOException {
      final ByteArrayBuffer boundaryEncoded = encode(this.charset, this.boundary);
      for (final FormBodyPart part: getBodyParts()) {
          writeBytes(TWO_DASHES, out);
          writeBytes(boundaryEncoded, out);
          writeBytes(CR_LF, out);
          //此处代码主要有3种不同的实现,不同的mode,实现方式不一样，采用的字符集也不同
          formatMultipartHeader(part, out);
          writeBytes(CR_LF, out);
          if (writeContent) {
              part.getBody().writeTo(out);
          }
          writeBytes(CR_LF, out);
      }
      writeBytes(TWO_DASHES, out);
      writeBytes(boundaryEncoded, out);
      writeBytes(TWO_DASHES, out);
      writeBytes(CR_LF, out);
  }

其中的formatMultipartHeader方法，不同的模式有不同的实现方式

MultipartEntityBuilder

    MultipartFormEntity buildEntity() {
        String boundaryCopy = boundary;
        if (boundaryCopy == null && contentType != null) {
            boundaryCopy = contentType.getParameter("boundary");
        }
        if (boundaryCopy == null) {
            boundaryCopy = generateBoundary();
        }
        Charset charsetCopy = charset;
        if (charsetCopy == null && contentType != null) {
            charsetCopy = contentType.getCharset();
        }
        final List<NameValuePair> paramsList = new ArrayList<NameValuePair>(2);
        paramsList.add(new BasicNameValuePair("boundary", boundaryCopy));
        if (charsetCopy != null) {
            paramsList.add(new BasicNameValuePair("charset", charsetCopy.name()));
        }
        final NameValuePair[] params = paramsList.toArray(new NameValuePair[paramsList.size()]);
        final ContentType contentTypeCopy = contentType != null ?
                contentType.withParameters(params) :
                ContentType.create("multipart/" + DEFAULT_SUBTYPE, params);
        final List<FormBodyPart> bodyPartsCopy = bodyParts != null ? new ArrayList<FormBodyPart>(bodyParts) :
                Collections.<FormBodyPart>emptyList();
        //此处将mode赋值给modeCopy
        final HttpMultipartMode modeCopy = mode != null ? mode : HttpMultipartMode.STRICT;
        final AbstractMultipartForm form;
        //此处根据modeCopy的值不同，构造3种form，每种的字符集都不一样，也是产生乱码的根源
        switch (modeCopy) {
            case BROWSER_COMPATIBLE:
                form = new HttpBrowserCompatibleMultipart(charsetCopy, boundaryCopy, bodyPartsCopy);
                break;
            case RFC6532:
                form = new HttpRFC6532Multipart(charsetCopy, boundaryCopy, bodyPartsCopy);
                break;
            default:
                form = new HttpStrictMultipart(charsetCopy, boundaryCopy, bodyPartsCopy);
        }
        return new MultipartFormEntity(form, contentTypeCopy, form.getTotalLength());
    }
    public HttpEntity build() {
        return buildEntity();
    }

BROWSER_COMPATIBLE模式中的formatMultipartHeader方法

class HttpBrowserCompatibleMultipart extends AbstractMultipartForm {
    private final List<FormBodyPart> parts;
    public HttpBrowserCompatibleMultipart(
            final Charset charset,
            final String boundary,
            final List<FormBodyPart> parts) {
        super(charset, boundary);
        this.parts = parts;
    }
    @Override
    public List<FormBodyPart> getBodyParts() {
        return this.parts;
    }
    /**
      * Write the multipart header fields; depends on the style.
      */
    @Override
    protected void formatMultipartHeader(
            final FormBodyPart part,
            final OutputStream out) throws IOException {
        // For browser-compatible, only write Content-Disposition
        // Use content charset
        final Header header = part.getHeader();
        final MinimalField cd = header.getField(MIME.CONTENT_DISPOSITION);
        //可以看到此处的字符集采用的是设置的字符集
        writeField(cd, this.charset, out);
        final String filename = part.getBody().getFilename();
        if (filename != null) {
            final MinimalField ct = header.getField(MIME.CONTENT_TYPE);
            //可以看到此处的字符集采用的也是设置的字符集
            writeField(ct, this.charset, out);
        }
    }
}

RFC6532模式中的formatMultipartHeader方法

class HttpRFC6532Multipart extends AbstractMultipartForm {
    private final List<FormBodyPart> parts;
    public HttpRFC6532Multipart(
            final Charset charset,
            final String boundary,
            final List<FormBodyPart> parts) {
        super(charset, boundary);
        this.parts = parts;
    }
    @Override
    public List<FormBodyPart> getBodyParts() {
        return this.parts;
    }
    @Override
    protected void formatMultipartHeader(
        final FormBodyPart part,
        final OutputStream out) throws IOException {
        // For RFC6532, we output all fields with UTF-8 encoding.
        final Header header = part.getHeader();
        for (final MinimalField field: header) {
            //可以看到此处的字符集默认采用UTF8
            writeField(field, MIME.UTF8_CHARSET, out);
        }
    }
}

默认模式中的formatMultipartHeader方法

class HttpStrictMultipart extends AbstractMultipartForm {
    private final List<FormBodyPart> parts;
    public HttpStrictMultipart(
            final Charset charset,
            final String boundary,
            final List<FormBodyPart> parts) {
        super(charset, boundary);
        this.parts = parts;
    }
    @Override
    public List<FormBodyPart> getBodyParts() {
        return this.parts;
    }
    @Override
    protected void formatMultipartHeader(
        final FormBodyPart part,
        final OutputStream out) throws IOException {
        // For strict, we output all fields with MIME-standard encoding.
        //从上面注释中可以看到，此处的字符集采用的是默认字符集即ASCII（下面MIME类中可以看到）
        final Header header = part.getHeader();
        for (final MinimalField field: header) {
            writeField(field, out);
        }
    }
}

MIME类

public final class MIME {
    public static final String CONTENT_TYPE          = "Content-Type";
    public static final String CONTENT_TRANSFER_ENC  = "Content-Transfer-Encoding";
    public static final String CONTENT_DISPOSITION   = "Content-Disposition";
    public static final String ENC_8BIT              = "8bit";
    public static final String ENC_BINARY            = "binary";
    /** The default character set to be used, i.e. "US-ASCII" */
    public static final Charset DEFAULT_CHARSET      = Consts.ASCII;
    /** UTF-8 is used for RFC6532 */
    public static final Charset UTF8_CHARSET         = Consts.UTF_8;
}

解决方法

知道乱码产生的根源，乱码问题也就好解决了，解决方式有两种

设置mode为：BROWSER_COMPATIBLE，并设置字符集为UTF8

private HttpEntity buildEntity(Long scenarioId, List<String> groupIds, String extension,
                                   File fileToUpload) {
        MultipartEntityBuilder builder = MultipartEntityBuilder.create();
        //设置模式为BROWSER_COMPATIBLE，并设置字符集为UTF8
        builder.setMode(HttpMultipartMode.BROWSER_COMPATIBLE);
        builder.setCharset(Charset.forName("UTF-8"));
        builder.addTextBody("scenarioId", scenarioId.toString());
        for (String groupId : groupIds) {
            builder.addTextBody("groupIds", groupId);
        }
        builder.addTextBody("extension", extension);
        builder.addPart("fileToUpload", new FileBody(fileToUpload));
        builder.addTextBody("type", AssetFileTypeEnum.CSV.getName());
        builder.addTextBody("isSplit", "false");
        builder.addTextBody("isRefresh", "false");
        return builder.build();
    }

设置模式为：RFC6532

    private HttpEntity buildEntity(Long scenarioId, List<String> groupIds, String extension,
                                   File fileToUpload) {
        MultipartEntityBuilder builder = MultipartEntityBuilder.create();
        //设置模式为RFC6532
        builder.setMode(HttpMultipartMode.RFC6532);
        builder.addTextBody("scenarioId", scenarioId.toString());
        for (String groupId : groupIds) {
            builder.addTextBody("groupIds", groupId);
        }
        builder.addTextBody("extension", extension);
        builder.addPart("fileToUpload", new FileBody(fileToUpload));
        builder.addTextBody("type", AssetFileTypeEnum.CSV.getName());
        builder.addTextBody("isSplit", "false");
        builder.addTextBody("isRefresh", "false");
        return builder.build();
    }

以上为个人经验，希望能给大家一个参考，也希望大家多多支持我们。