Adding sniffing of MHTML files.

net/base/mime_sniffer.cc

 // Copyright (c) 2011 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 // Detecting mime types is a tricky business because we need to balance
 // compatibility concerns with security issues.  Here is a survey of how other
 // browsers behave and then a description of how we intend to behave.
 //
 // HTML payload, no Content-Type header:
 // * IE 7: Render as HTML
@@ skipping 516 matching lines @@
   // sniffing gives us less room for error. If the version number ever changes,
   // we can just add an entry to this list.
   //
   // TODO(aa): If we ever have another magic number, we'll want to pass a
   // histogram into CheckForMagicNumbers(), below, to see which one matched.
   static const struct MagicNumber kCRXMagicNumbers[] = {
     MAGIC_NUMBER("application/x-chrome-extension", "Cr24\x02\x00\x00\x00")
   };
 
   // Only consider files that have the extension ".crx".
-  static const char kCRXExtension[] = ".crx";
-  // Ignore null by subtracting 1.
-  static const int kExtensionLength = arraysize(kCRXExtension) - 1;
-  if (url.path().rfind(kCRXExtension, std::string::npos, kExtensionLength) ==
-      url.path().size() - kExtensionLength) {
-    counter->Add(1);
-  } else {
+  if (!EndsWith(url.path(), ".crx", true))
     return false;
-  }
+
+  counter->Add(1);
 
   *have_enough_content &= TruncateSize(kBytesRequiredForMagic, &size);
   if (CheckForMagicNumbers(content, size,
                            kCRXMagicNumbers, arraysize(kCRXMagicNumbers),
                            NULL, result)) {
     counter->Add(2);
   } else {
     return false;
   }
 
@@ skipping 100 matching lines @@
       return true;
     return have_enough_content;
   }
 
   // CRX files (chrome extensions) have a special sniffing algorithm. It is
   // tighter than the others because we don't have to match legacy behavior.
   if (SniffCRX(content, content_size, url, type_hint,
                &have_enough_content, result))
     return true;
 
+  // MHTML cannot be easily sniffed as its header might be relatively long
+  // before the "Content-Type: multipart/related".
+  std::string path = url.path();
+  if (EndsWith(path, ".mht", false) || EndsWith(path, ".mhtml", false)) {
+    result->assign("multipart/related");
+    return true;
+  }
+
   // We're not interested in sniffing for magic numbers when the type_hint
   // is application/octet-stream.  Time to bail out.
   if (type_hint == "application/octet-stream")
     return have_enough_content;
 
   // Now we look in our large table of magic numbers to see if we can find
   // anything that matches the content.
   if (SniffForMagicNumbers(content, content_size,
                            &have_enough_content, result))
     return true;  // We've matched a magic number.  No more content needed.
 
   return have_enough_content;
 }
 
 }  // namespace net