Add UMAs for checking header values against RFC 7230 in //net

net/http/http_util.cc

 // Copyright (c) 2012 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.
 
 // The rules for parsing content-types were borrowed from Firefox:
 // http://lxr.mozilla.org/mozilla/source/netwerk/base/src/nsURLHelper.cpp#834
 
 #include "net/http/http_util.h"
 
 #include <algorithm>
@@ skipping 328 matching lines @@
 }
 
 // static
 bool HttpUtil::IsValidHeaderValue(const std::string& value) {
   // Just a sanity check: disallow NUL and CRLF.
   return value.find('\0') == std::string::npos &&
       value.find("\r\n") == std::string::npos;
 }
 
 // static
+bool HttpUtil::IsValidHeaderValueRFC7230(base::StringPiece value) {

[davidben, 2015/09/30 16:38:18]

Optional: Now that you have a base::StringPiece, I think this would be more
readable if you wrote this using those APIs.

if (value.empty())
  return true;

if (IsLWS(value[0]) || IsLWS(value[value.size() - 1]))
  return false;

// Can you just write "for (unsigned char c : value) {"? I'm not
// sure. Did I mention I really hate how C messed up char
// signedness? :-)
for (char c_signed : value) {
  unsigned char c = static_cast<unsigned char>(c_signed);
  if (c == .......)
    return false;
}
return true;

[hiroshige, 2015/12/15 12:00:26]

Done.

+  base::StringPiece::const_iterator value_begin = value.begin();
+  base::StringPiece::const_iterator value_end = value.end();
+
+  // Empty string is a valid header-value.

[davidben, 2015/09/30 16:38:18]

Nit: "This empty string is a valid header-value.

[hiroshige, 2015/12/15 12:00:26]

Done.

+  if (value_begin == value_end)
+    return true;
+
+  // Check leading/trailing whitespaces.
+  if (IsLWS(*value_begin) || IsLWS(*(value_end - 1)))
+    return false;
+
+  // Check each octet is |field-vchar|, |SP| or |HTAB|.
+  for (; value_begin != value_end; ++value_begin) {
+    unsigned char c = *value_begin;
+    if (c == 0x7F || c > 0xFF || (c < 0x20 && c != '\t'))

[davidben, 2015/09/30 16:38:18]

c > 0xFF isn't possible. char's are one byte.

[hiroshige, 2015/12/15 12:00:26]

Done.

+      return false;
+  }
+  return true;
+}
+
+// static
 std::string HttpUtil::StripHeaders(const std::string& headers,
                                    const char* const headers_to_remove[],
                                    size_t headers_to_remove_len) {
   std::string stripped_headers;
   HttpUtil::HeadersIterator it(headers.begin(), headers.end(), "\r\n");
 
   while (it.GetNext()) {
     bool should_remove = false;
     for (size_t i = 0; i < headers_to_remove_len; ++i) {
       if (base::LowerCaseEqualsASCII(
@@ skipping 582 matching lines @@
       value_is_quoted_ = true;
       // Do not store iterators into this. See declaration of unquoted_value_.
       unquoted_value_ = HttpUtil::Unquote(value_begin_, value_end_);
     }
   }
 
   return true;
 }
 
 }  // namespace net