author Jonathan Watt <>
Tue, 07 Jun 2016 21:10:18 +0100
changeset 301499 232a786cf87fb0f68f9bda6e9316515524d2ac8a
parent 271385 9aab415d2bd9b8ee29f71e3a7b3bd7616bb006be
child 363728 e155fa765af299f0e8cfb42e0a1709e5b04928b9
permissions -rw-r--r--
Bug 1279451 - Remove a lot of unnecessary includes of nsAutoPtr.h. rs=sparky

/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim:expandtab:shiftwidth=2:tabstop=4: 
/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at */
#include "nsString.h"
#include "nsUTF8ConverterService.h"
#include "nsEscape.h"
#include "nsIUnicodeDecoder.h"
#include "mozilla/dom/EncodingUtils.h"
#include "mozilla/UniquePtr.h"

using mozilla::dom::EncodingUtils;

NS_IMPL_ISUPPORTS(nsUTF8ConverterService, nsIUTF8ConverterService)

static nsresult 
ToUTF8(const nsACString &aString, const char *aCharset,
       bool aAllowSubstitution, nsACString &aResult)
  nsresult rv;
  if (!aCharset || !*aCharset)

  nsDependentCString label(aCharset);
  nsAutoCString encoding;
  if (!EncodingUtils::FindEncodingForLabelNoReplacement(label, encoding)) {
  nsCOMPtr<nsIUnicodeDecoder> unicodeDecoder =

  if (!aAllowSubstitution)

  int32_t srcLen = aString.Length();
  int32_t dstLen;
  const nsAFlatCString& inStr = PromiseFlatCString(aString);
  rv = unicodeDecoder->GetMaxLength(inStr.get(), srcLen, &dstLen);

  auto ustr = mozilla::MakeUnique<char16_t[]>(dstLen);

  rv = unicodeDecoder->Convert(inStr.get(), &srcLen, ustr.get(), &dstLen);
  if (NS_SUCCEEDED(rv)){
    CopyUTF16toUTF8(Substring(ustr.get(), ustr.get() + dstLen), aResult);
  return rv;

nsUTF8ConverterService::ConvertStringToUTF8(const nsACString &aString, 
                                            const char *aCharset, 
                                            bool aSkipCheck, 
                                            bool aAllowSubstitution,
                                            uint8_t aOptionalArgc,
                                            nsACString &aUTF8String)
  bool allowSubstitution = (aOptionalArgc == 1) ? aAllowSubstitution : true;

  // return if ASCII only or valid UTF-8 providing that the ASCII/UTF-8
  // check is requested. It may not be asked for if a caller suspects
  // that the input is in non-ASCII 7bit charset (ISO-2022-xx, HZ) or 
  // it's in a charset other than UTF-8 that can be mistaken for UTF-8.
  if (!aSkipCheck && (IsASCII(aString) || IsUTF8(aString))) {
    aUTF8String = aString;
    return NS_OK;


  nsresult rv = ToUTF8(aString, aCharset, allowSubstitution, aUTF8String);

  // additional protection for cases where check is skipped and  the input
  // is actually in UTF-8 as opposed to aCharset. (i.e. caller's hunch
  // was wrong.) We don't check ASCIIness assuming there's no charset
  // incompatible with ASCII (we don't support EBCDIC).
  if (aSkipCheck && NS_FAILED(rv) && IsUTF8(aString)) {
    aUTF8String = aString;
    return NS_OK;

  return rv;

nsUTF8ConverterService::ConvertURISpecToUTF8(const nsACString &aSpec, 
                                             const char *aCharset, 
                                             nsACString &aUTF8Spec)
  // assume UTF-8 if the spec contains unescaped non-ASCII characters.
  // No valid spec in Mozilla would break this assumption.
  if (!IsASCII(aSpec)) {
    aUTF8Spec = aSpec;
    return NS_OK;


  nsAutoCString unescapedSpec; 
  // NS_UnescapeURL does not fill up unescapedSpec unless there's at least 
  // one character to unescape.
  bool written = NS_UnescapeURL(PromiseFlatCString(aSpec).get(), aSpec.Length(), 
                                  esc_OnlyNonASCII, unescapedSpec);

  if (!written) {
    aUTF8Spec = aSpec;
    return NS_OK;
  // return if ASCII only or escaped UTF-8
  if (IsASCII(unescapedSpec) || IsUTF8(unescapedSpec)) {
    aUTF8Spec = unescapedSpec;
    return NS_OK;

  return ToUTF8(unescapedSpec, aCharset, true, aUTF8Spec);