Improve parsing of the Content-Disposition header

Previous, we were using GetHeaderParamValue to parse the Content-Disposition
header, which describes itself as a "quick and dirty implementation."  After
this patch, we use more of our normal HTTP parsing machinery, making our
parsing much less quirky and better aligned with RFC 6266.

Some notes:

1) Many of the test cases for parsing the Content-Disposition header included
   the string "Content-Disposition: " in the input.  I've looked through all
   of the callers of these functions, and that seems to be completely bogus.
   The old parser wasn't careful enough to see that as a problem, but the new
   one follows the spec more closely.  I've updated the test cases to remove
   this string.

2) After this patch, there's a bunch of code in net_util.cc that really should
   be moved to http_content_disposition.cc.  I didn't move that code in this
   patch because I didn't want this path to be too large.  I'll move it in a
   future patch.

3) In a future patch, I'll audit the codebase for callers of
   GetHeaderParamValue.  With any luck, we'll be able to remove them all and
   delete this less-than-amazing function.

BUG=65423

Review URL: http://codereview.chromium.org/9234055

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@119378 0039d316-1c4b-4281-b951-d872f2087c98

1 files changed, 93 insertions, 0 deletions

diff --git a/net/http/http_content_disposition.cc b/net/http/http_content_disposition.cc
new file mode 100644
index 0000000..4e5d94e
--- /dev/null
+++ b/net/http/http_content_disposition.cc
@@ -0,0 +1,93 @@
+// 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.
+
+#include "net/http/http_content_disposition.h"
+
+#include "base/logging.h"
+#include "base/string_util.h"
+#include "net/base/net_util.h"
+#include "net/http/http_util.h"
+
+namespace net {
+
+HttpContentDisposition::HttpContentDisposition(
+    const std::string& header, const std::string& referrer_charset)
+  : type_(INLINE) {
+  Parse(header, referrer_charset);
+}
+
+HttpContentDisposition::~HttpContentDisposition() {
+}
+
+std::string::const_iterator HttpContentDisposition::ConsumeDispositionType(
+    std::string::const_iterator begin, std::string::const_iterator end) {
+  DCHECK(type_ == INLINE);
+
+  std::string::const_iterator delimiter = std::find(begin, end, ';');
+
+  // If there's an '=' in before the first ';', then the Content-Disposition
+  // header is malformed, and we treat the first bytes as a parameter rather
+  // than a disposition-type.
+  if (std::find(begin, delimiter, '=') != delimiter)
+    return begin;
+
+  std::string::const_iterator type_begin = begin;
+  std::string::const_iterator type_end = delimiter;
+  HttpUtil::TrimLWS(&type_begin, &type_end);
+  if (!LowerCaseEqualsASCII(type_begin, type_end, "inline"))
+    type_ = ATTACHMENT;
+  return delimiter;
+}
+
+// http://tools.ietf.org/html/rfc6266
+//
+//  content-disposition = "Content-Disposition" ":"
+//                         disposition-type *( ";" disposition-parm )
+//
+//  disposition-type    = "inline" | "attachment" | disp-ext-type
+//                      ; case-insensitive
+//  disp-ext-type       = token
+//
+//  disposition-parm    = filename-parm | disp-ext-parm
+//
+//  filename-parm       = "filename" "=" value
+//                      | "filename*" "=" ext-value
+//
+//  disp-ext-parm       = token "=" value
+//                      | ext-token "=" ext-value
+//  ext-token           = <the characters in token, followed by "*">
+//
+void HttpContentDisposition::Parse(const std::string& header,
+                                   const std::string& referrer_charset) {
+  DCHECK(type_ == INLINE);
+  DCHECK(filename_.empty());
+
+  std::string::const_iterator pos = header.begin();
+  std::string::const_iterator end = header.end();
+  pos = ConsumeDispositionType(pos, end);
+
+  std::string filename;
+  std::string ext_filename;
+
+  HttpUtil::NameValuePairsIterator iter(pos, end, ';');
+  while (iter.GetNext()) {
+    if (LowerCaseEqualsASCII(iter.name_begin(),
+                             iter.name_end(),
+                             "filename")) {
+      DecodeFilenameValue(iter.value(), referrer_charset, &filename);
+    } else if (LowerCaseEqualsASCII(iter.name_begin(),
+                             iter.name_end(),
+                             "name")) {
+      DecodeFilenameValue(iter.value(), referrer_charset, &filename);
+    } else if (LowerCaseEqualsASCII(iter.name_begin(),
+                                    iter.name_end(),
+                                    "filename*")) {
+      DecodeExtValue(iter.raw_value(), &ext_filename);
+    }
+  }
+
+  filename_ = ext_filename.empty() ? filename : ext_filename;
+}
+
+}  // namespace net