Request body extraction

pcap2har/har.py

@@ -34,6 +34,20 @@ def HTTPRequestJsonRepr(self):
     '''
     self = http.Request
     '''
+    content = {
+        'size': self.body_length,
+        'mimeType': self.mimeType
+    }
+    if self.compression_amount is not None:
+        content['compression'] = self.compression_amount
+    if self.text:
+        if self.encoding:
+            content['text'] = self.text
+            content['encoding'] = self.encoding
+        else:
+            content['text'] = self.text.encode('utf8')  # must transcode to utf-8
+    # FIXME: should process postData further and replace text
+    # with params if URL encoded parameters
     return {
         'method': self.msg.method,
         'url': self.url,
@@ -43,6 +57,7 @@ def HTTPRequestJsonRepr(self):
         'headersSize': -1,
         'headers': header_json_repr(self.msg.headers),
         'bodySize': len(self.msg.body),
+        'postData': content,
     }
 http.Request.json_repr = HTTPRequestJsonRepr
 

pcap2har/http/request.py

@@ -2,10 +2,13 @@
 
 # dpkt.http is buggy, so we use our modified replacement
 from .. import dpkt_http_replacement as dpkt_http
+from .. import settings
+from ..mediatype import MediaType
 import message as http
+from response import Response
 
 
-class Request(http.Message):
+class Request(Response):
     '''
     HTTP request. Parses higher-level info out of dpkt.http.Request
     Members:
@@ -24,3 +27,20 @@ def __init__(self, tcpdir, pointer):
         self.fullurl = fullurl.geturl()
         self.url, frag = urlparse.urldefrag(self.fullurl)
         self.query = urlparse.parse_qs(uri.query, keep_blank_values=True)
+
+        if 'content-type' in self.msg.headers:
+            self.mediaType = MediaType(self.msg.headers['content-type'])
+        else:
+            self.mediaType = MediaType('application/x-unknown-content-type')
+        self.mimeType = self.mediaType.mimeType()
+        self.body_length = len(self.msg.body)
+        self.compression_amount = None
+        self.text = None
+        # handle body stuff
+        if settings.drop_bodies:
+            self.clear_body()
+        else:
+            # uncompress body if necessary
+            self.handle_compression()
+            # try to get out unicode
+            self.handle_text()