前言
有时候，我们需要用拦截器对Request或者Response流里面的数据进行拦截，读取里面的一些信息，也许是作为日志检索，也许是做一些校验，但是当我们读取里请求或者回调的流数据后，会发现这些流数据在下游就无法再次被消费了，这里面是其实存在着两个潜在的坑。
坑一
Request的 getInputStream()、getReader()、getParameter()方法互斥，也就是使用了其中一个，再使用另外的两，是获取不到数据的。除了互斥外，getInputStream()和getReader()都只能使用一次，getParameter单线程上可重复使用。
三个方法互斥原因
org.apache.catalina.connector.Request方法实现了javax.servlet.http.HttpServletRequest接口，我们来看看这三个方法的实现：
getInputStream
@Override
public ServletInputStream getInputStream() throws IOException {
if (usingReader) {
throw new IllegalStateException
(sm.getString("coyoteRequest.getInputStream.ise"));
}
usingInputStream = true;
if (inputStream == null) {
inputStream = new CoyoteInputStream(inputBuffer);
}
return inputStream;
}
getReader
@Override
public BufferedReader getReader() throws IOException {
if (usingInputStream) {
throw new IllegalStateException
(sm.getString("coyoteRequest.getReader.ise"));
}
usingReader = true;
inputBuffer.checkConverter();
if (reader == null) {
reader = new CoyoteReader(inputBuffer);
}
return reader;
}
首先来看getInputStream()和getReader()这两个方法，可以看到，在读流时分别用usingReader和usingInputStream标志做了限制，这两个方法的互斥很好理解。下面看一看getParameter()方法是怎么跟他们互斥的。
getParameter
@Override
public String getParameter(String name) {
// 只会解析一遍Parameter
if (!parametersParsed) {
parseParameters();
}
// 从coyoteRequest中获取参数
return coyoteRequest.getParameters().getParameter(name);
}
粗略一看好像没有互斥，别着急，继续往下看，我们进到parseParameters()方法中来看一看（可以直接看源码中间部分）：
protected void parseParameters() {
//标识位，标志已经被解析过。
parametersParsed = true;
Parameters parameters = coyoteRequest.getParameters();
boolean success = false;
try {
// Set this every time in case limit has been changed via JMX
parameters.setLimit(getConnector().getMaxParameterCount());
// getCharacterEncoding() may have been overridden to search for
// hidden form field containing request encoding
String enc = getCharacterEncoding();
boolean useBodyEncodingForURI = connector.getUseBodyEncodingForURI();
if (enc != null) {
parameters.setEncoding(enc);
if (useBodyEncodingForURI) {
parameters.setQueryStringEncoding(enc);
}
} else {
parameters.setEncoding
(org.apache.coyote.Constants.DEFAULT_CHARACTER_ENCODING);
if (useBodyEncodingForURI) {
parameters.setQueryStringEncoding
(org.apache.coyote.Constants.DEFAULT_CHARACTER_ENCODING);
}
}
parameters.handleQueryParameters();
// 重点看这里：这里会判断是否有读取过流。如果有，则直接return。
if (usingInputStream || usingReader) {
success = true;
return;
}
if( !getConnector().isParseBodyMethod(getMethod()) ) {
success = true;
return;
}
String contentType = getContentType();
if (contentType == null) {
contentType = "";
}
int semicolon = contentType.indexOf(';');
if (semicolon >= 0) {
contentType = contentType.substring(0, semicolon).trim();
} else {
contentType = contentType.trim();
}
if ("multipart/form-data".equals(contentType)) {
parseParts(false);
success = true;
return;
}
if (!("application/x-www-form-urlencoded".equals(contentType))) {
success = true;
return;
}
int len = getContentLength();
if (len > 0) {
int maxPostSize = connector.getMaxPostSize();
if ((maxPostSize > 0) && (len > maxPostSize)) {
Context context = getContext();
if (context != null && context.getLogger().isDebugEnabled()) {
context.getLogger().debug(
sm.getString("coyoteRequest.postTooLarge"));
}
checkSwallowInput();
return;
}
byte[] formData = null;
if (len < CACHED_POST_LEN) {
if (postData == null) {
postData = new byte[CACHED_POST_LEN];
}
formData = postData;
} else {
formData = new byte[len];
}
try {
if (readPostBody(formData, len) != len) {
return;
}
} catch (IOException e) {
// Client disconnect
Context context = getContext();
if (context != null && context.getLogger().isDebugEnabled()) {
context.getLogger().debug(
sm.getString("coyoteRequest.parseParameters"),
e);
}
return;
}
parameters.processParameters(formData, 0, len);
} else if ("chunked".equalsIgnoreCase(
coyoteRequest.getHeader("transfer-encoding"))) {
byte[] formData = null;
try {
formData = readChunkedPostBody();
} catch (IOException e) {
// Client disconnect or chunkedPostTooLarge error
Context context = getContext();
if (context != null && context.getLogger().isDebugEnabled()) {
context.getLogger().debug(
sm.getString("coyoteRequest.parseParameters"),
e);
}
return;
}
if (formData != null) {
parameters.processParameters(formData, 0, formData.length);
}
}
success = true;
} finally {
if (!success) {
parameters.setParseFailed(true);
}
}
}
这样一来，就说明了getParameter()方法也不能随意读取的。那么为什么它们都只能读取一次呢？
只能读取一次的原因
getInputStream()和getReader()方法都只能读取一次，而getParameter()是在单线程上可重复使用，主要是因为getParameter()中会解析流中的数据后存放在了一个LinkedHashMap中，相关的内容可以看Parameters类中的封装，在上面parseParameters()方法的源码中也可以看到一开始就生成了一个Parameters对象。后续读取的数据都存在了这个对象中。但是getInputStream()和getReader()方法就没有这样做，getInputStream()方法返回CoyoteInputStream，getReader()返回CoyoteReader，CoyoteInputStream继承了InputStream，CoyoteReader继承了BufferedReader，从源码看InputStream和BufferedReader在读取数据后，记录数据读取的坐标不会被重置，因为CoyoteInputStream和CoyoteReader都没有实现reset方法，这导致数据只能被读取一次。
坑二
Response与Request一样，getOutputStream()和getWriter()方法也是互斥的，并且Response中的body数据也只能消费一次。
互斥原因
getOutputStream
@Override
public ServletOutputStream getOutputStream()
throws IOException {
if (usingWriter) {
throw new IllegalStateException
(sm.getString("coyoteResponse.getOutputStream.ise"));
}
usingOutputStream = true;
if (outputStream == null) {
outputStream = new CoyoteOutputStream(outputBuffer);
}
return outputStream;
}
getWriter
@Override
public PrintWriter getWriter()
throws IOException {
if (usingOutputStream) {
throw new IllegalStateException
(sm.getString("coyoteResponse.getWriter.ise"));
}
if (ENFORCE_ENCODING_IN_GET_WRITER) {
setCharacterEncoding(getCharacterEncoding());
}
usingWriter = true;
outputBuffer.checkConverter();
if (writer == null) {
writer = new CoyoteWriter(outputBuffer);
}
return writer;
}
只能读取一次的原因
在Response中，读取是指从OutputStream中重新把body数据读出来，而OutputStream也和InputStream存在同样的问题，流只能读取一次，这里就不展开讲了。
解决方案
在Spring库中，提供了ContentCachingResponseWrapper和ContentCachingRequestWrapper两个类，分别解决了Response和Request不能重复读以及方法互斥问题。我们可以直接用ContentCachingRequestWrapper来包装Request，ContentCachingResponseWrapper来包装Response，包装后，在读取流数据的时候会将这个数据缓存一份，等读完以后，再将流数据重新写入Request或者Response就可以了。下面是一个简单的使用示例：
ContentCachingResponseWrapper responseToCache = new ContentCachingResponseWrapper(response);
String responseBody = new String(responseToCache.getContentAsByteArray());
responseToCache.copyBodyToResponse();
缓存一份流数据，这就是基本的解决思路，下面我们从源码层面来看一看，主要关注getContentAsByteArray()、copyBodyToResponse()方法就行：
public class ContentCachingResponseWrapper extends HttpServletResponseWrapper {
private final FastByteArrayOutputStream content = new FastByteArrayOutputStream(1024);
private final ServletOutputStream outputStream = new ResponseServletOutputStream();
private PrintWriter writer;
private int statusCode = HttpServletResponse.SC_OK;
private Integer contentLength;
/**
* Create a new ContentCachingResponseWrapper for the given servlet response.
* @param response the original servlet response
*/
public ContentCachingResponseWrapper(HttpServletResponse response) {
super(response);
}
@Override
public void setStatus(int sc) {
super.setStatus(sc);
this.statusCode = sc;
}
@SuppressWarnings("deprecation")
@Override
public void setStatus(int sc, String sm) {
super.setStatus(sc, sm);
this.statusCode = sc;
}
@Override
public void sendError(int sc) throws IOException {
copyBodyToResponse(false);
try {
super.sendError(sc);
}
catch (IllegalStateException ex) {
// Possibly on Tomcat when called too late: fall back to silent setStatus
super.setStatus(sc);
}
this.statusCode = sc;
}
@Override
@SuppressWarnings("deprecation")
public void sendError(int sc, String msg) throws IOException {
copyBodyToResponse(false);
try {
super.sendError(sc, msg);
}
catch (IllegalStateException ex) {
// Possibly on Tomcat when called too late: fall back to silent setStatus
super.setStatus(sc, msg);
}
this.statusCode = sc;
}
@Override
public void sendRedirect(String location) throws IOException {
copyBodyToResponse(false);
super.sendRedirect(location);
}
@Override
public ServletOutputStream getOutputStream() throws IOException {
return this.outputStream;
}
@Override
public PrintWriter getWriter() throws IOException {
if (this.writer == null) {
String characterEncoding = getCharacterEncoding();
this.writer = (characterEncoding != null ? new ResponsePrintWriter(characterEncoding) :
new ResponsePrintWriter(WebUtils.DEFAULT_CHARACTER_ENCODING));
}
return this.writer;
}
@Override
public void flushBuffer() throws IOException {
// do not flush the underlying response as the content as not been copied to it yet
}
@Override
public void setContentLength(int len) {
if (len > this.content.size()) {
this.content.resize(len);
}
this.contentLength = len;
}
// Overrides Servlet 3.1 setContentLengthLong(long) at runtime
public void setContentLengthLong(long len) {
if (len > Integer.MAX_VALUE) {
throw new IllegalArgumentException("Content-Length exceeds ContentCachingResponseWrapper's maximum (" +
Integer.MAX_VALUE + "): " + len);
}
int lenInt = (int) len;
if (lenInt > this.content.size()) {
this.content.resize(lenInt);
}
this.contentLength = lenInt;
}
@Override
public void setBufferSize(int size) {
if (size > this.content.size()) {
this.content.resize(size);
}
}
@Override
public void resetBuffer() {
this.content.reset();
}
@Override
public void reset() {
super.reset();
this.content.reset();
}
/**
* Return the status code as specified on the response.
*/
public int getStatusCode() {
return this.statusCode;
}
/**
* Return the cached response content as a byte array.
*/
public byte[] getContentAsByteArray() {
return this.content.toByteArray();
}
/**
* Return an {@link InputStream} to the cached content.
* @since 4.2
*/
public InputStream getContentInputStream() {
return this.content.getInputStream();
}
/**
* Return the current size of the cached content.
* @since 4.2
*/
public int getContentSize() {
return this.content.size();
}
/**
* Copy the complete cached body content to the response.
* @since 4.2
*/
public void copyBodyToResponse() throws IOException {
copyBodyToResponse(true);
}
/**
* Copy the cached body content to the response.
* @param complete whether to set a corresponding content length
* for the complete cached body content
* @since 4.2
*/
protected void copyBodyToResponse(boolean complete) throws IOException {
if (this.content.size() > 0) {
HttpServletResponse rawResponse = (HttpServletResponse) getResponse();
if ((complete || this.contentLength != null) && !rawResponse.isCommitted()) {
rawResponse.setContentLength(complete ? this.content.size() : this.contentLength);
this.contentLength = null;
}
this.content.writeTo(rawResponse.getOutputStream());
this.content.reset();
if (complete) {
super.flushBuffer();
}
}
}
private class ResponseServletOutputStream extends ServletOutputStream {
@Override
public void write(int b) throws IOException {
content.write(b);
}
@Override
public void write(byte[] b, int off, int len) throws IOException {
content.write(b, off, len);
}
}
private class ResponsePrintWriter extends PrintWriter {
public ResponsePrintWriter(String characterEncoding) throws UnsupportedEncodingException {
super(new OutputStreamWriter(content, characterEncoding));
}
@Override
public void write(char buf[], int off, int len) {
super.write(buf, off, len);
super.flush();
}
@Override
public void write(String s, int off, int len) {
super.write(s, off, len);
super.flush();
}
@Override
public void write(int c) {
super.write(c);
super.flush();
}
}
}
而ContentCachingRequestWrapper的解决思路也是差不多，我这里就不展开了，有兴趣的可以直接查看源码。
