我正在尝试在
Spring MVC
应用程序中实现用于记录请求和响应的过滤器。
我使用以下代码:
@Component
public class LoggingFilter extends OncePerRequestFilter {
private static final Logger LOGGER = LoggerFactory.getLogger(LoggingFilter.class);
@Override
protected void doFilterInternal(HttpServletRequest request, HttpServletResponse response, FilterChain filterChain) throws ServletException, IOException {
ContentCachingRequestWrapper requestWrapper = new ContentCachingRequestWrapper(request);
ContentCachingResponseWrapper responseWrapper = new ContentCachingResponseWrapper(response);
LOGGER.debug(REQUEST_MESSAGE_FORMAT, requestWrapper.getRequestURI(), requestWrapper.getMethod(), requestWrapper.getContentType(),
new ServletServerHttpRequest(requestWrapper).getHeaders(), IOUtils.toString(requestWrapper.getInputStream(), UTF_8));
filterChain.doFilter(requestWrapper, responseWrapper);
LOGGER.debug(RESPONSE_MESSAGE_FORMAT, responseWrapper.getStatus(), responseWrapper.getContentType(),
new ServletServerHttpResponse(responseWrapper).getHeaders(), IOUtils.toString(responseWrapper.getContentInputStream(), UTF_8));
}
}
因此,我收到了我的请求并按预期记录了响应。这是日志:
2016-10-08 19:10:11.212 DEBUG 11072 --- [qtp108982313-19] by.kolodyuk.logging.LoggingFilter
----------------------------
ID: 1
URI: /resources/1
Http-Method: GET
Content-Type: null
Headers: {User-Agent=[curl/7.41.0], Accept=[*/*], Host=[localhost:9015]}
Body:
--------------------------------------
2016-10-08 19:10:11.277 DEBUG 11072 --- [qtp108982313-19] by.kolodyuk.logging.LoggingFilter
----------------------------
ID: 1
Response-Code: 200
Content-Type: application/json;charset=UTF-8
Headers: {}
Body: {"id":"1"}
--------------------------------------
但是,返回的是空响应。这是
curl
的输出:
$ curl http://localhost:9015/resources/1 --verbose
* Trying ::1...
* Connected to localhost (::1) port 9015 (#0)
> GET /resources/1 HTTP/1.1
> User-Agent: curl/7.41.0
> Host: localhost:9015
> Accept: */*
>
< HTTP/1.1 200 OK
< Date: Sat, 08 Oct 2016 17:10:11 GMT
< Content-Type: application/json;charset=UTF-8
< Content-Length: 0
<
* Connection #0 to host localhost left intact
有什么想法吗?
谢谢
经过几个小时的努力,我终于找到了解决方案。
简而言之,
ContentCachingResponseWrapper.copyBodyToResponse()
应该在filter方法的最后调用。
ContentCachingResponseWrapper
通过从响应输出流读取响应正文来缓存响应正文。因此,流变空了。要将响应写回输出流,应使用 ContentCachingResponseWrapper.copyBodyToResponse()
。
终于解决了问题。这是完美的解决方案:
import com.fasterxml.jackson.databind.JsonNode;
import com.fasterxml.jackson.databind.ObjectMapper;
import org.apache.commons.io.IOUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.http.*;
import org.springframework.stereotype.Component;
import org.springframework.web.filter.OncePerRequestFilter;
import org.springframework.web.util.ContentCachingRequestWrapper;
import org.springframework.web.util.ContentCachingResponseWrapper;
import javax.servlet.FilterChain;
import javax.servlet.ServletException;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.net.URI;
import java.util.Enumeration;
import java.util.Map;
import static java.nio.charset.StandardCharsets.UTF_8;
import static net.logstash.logback.marker.Markers.appendFields;
@Component
public class LoggingFilter extends OncePerRequestFilter {
private static final Logger LOGGER = LoggerFactory.getLogger(LoggingFilter.class);
@Autowired
private ObjectMapper objectMapper;
@Override
protected void doFilterInternal(HttpServletRequest httpServletRequest, HttpServletResponse httpServletResponse, FilterChain filterChain) throws ServletException, IOException {
ContentCachingRequestWrapper requestWrapper = new ContentCachingRequestWrapper(httpServletRequest);
ContentCachingResponseWrapper responseWrapper = new ContentCachingResponseWrapper(httpServletResponse);
filterChain.doFilter(requestWrapper, responseWrapper);
String requestUrl = requestWrapper.getRequestURL().toString();
HttpHeaders requestHeaders = new HttpHeaders();
Enumeration headerNames = requestWrapper.getHeaderNames();
while (headerNames.hasMoreElements()) {
String headerName = (String) headerNames.nextElement();
requestHeaders.add(headerName, requestWrapper.getHeader(headerName));
}
HttpMethod httpMethod = HttpMethod.valueOf(requestWrapper.getMethod());
Map<String, String[]> requestParams = requestWrapper.getParameterMap();
String requestBody = IOUtils.toString(requestWrapper.getInputStream(),UTF_8);
JsonNode requestJson = objectMapper.readTree(requestBody);
RequestEntity<JsonNode> requestEntity = new RequestEntity<>(requestJson,requestHeaders, httpMethod, URI.create(requestUrl));
LOGGER.info(appendFields(requestEntity),"Logging Http Request");
HttpStatus responseStatus = HttpStatus.valueOf(responseWrapper.getStatusCode());
HttpHeaders responseHeaders = new HttpHeaders();
for (String headerName : responseWrapper.getHeaderNames()) {
responseHeaders.add(headerName, responseWrapper.getHeader(headerName));
}
String responseBody = IOUtils.toString(responseWrapper.getContentInputStream(), UTF_8);
JsonNode responseJson = objectMapper.readTree(responseBody);
ResponseEntity<JsonNode> responseEntity = new ResponseEntity<>(responseJson,responseHeaders,responseStatus);
LOGGER.info(appendFields(responseEntity),"Logging Http Response");
responseWrapper.copyBodyToResponse();
}
}
我喜欢使用的模式是将其分为 2 个过滤器:一个用于提取原始主体,另一个用于进行日志记录。感觉更SRP。
@Slf4j // Lombok logging
@Component // Spring loads filter into its filter chain
@Order(1) // Best if this goes first (or early in filter chain)
public class CachingBodyFilter implements Filter {
@Override
public void doFilter(ServletRequest req, ServletResponse res, FilterChain chain) {
ContentCachingRequestWrapper reqWrapper = new ContentCachingRequestWrapper((HttpServletRequest) req);
ContentCachingResponseWrapper resWrapper = new ContentCachingResponseWrapper((HttpServletResponse) res);
try {
chain.doFilter(reqWrapper, resWrapper);
resWrapper.copyBodyToResponse(); // Necessary (see answer by StasKolodyuk above)
} catch (IOException | ServletException e) {
log.error("Error extracting body", e);
}
}
}
然后我们创建另一个过滤器来执行日志记录部分:
@Slf4j
@Component
@Order(2) // This needs to come after `CachingBodyFilter`
public class PayloadLogFilter implements Filter {
public void doFilter(ServletRequest req, ServletResponse res, FilterChain chain) throws IOException, ServletException {
chain.doFilter(req, res);
if (req instanceof ContentCachingRequestWrapper) {
ContentCachingRequestWrapper reqWrapper = (ContentCachingRequestWrapper) req;
String payload = new String (reqWrapper.getContentAsByteArray(), "UTF-8");
log.debug("Request [ {} ] has payload [ {} ]", reqWrapper.getRequestURI(), payload);
}
}
}
将它们分开的一个好处是其他类(例如 Spring AOP 拦截器或 Spring 控制器)也可以访问/使用 HTTP 主体。