1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/uconv/src/nsUTF8ConverterService.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,121 @@
1.4 +/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* vim:expandtab:shiftwidth=2:tabstop=4:
1.6 + */
1.7 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.8 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.9 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.10 +#include "nsString.h"
1.11 +#include "nsICharsetConverterManager.h"
1.12 +#include "nsUTF8ConverterService.h"
1.13 +#include "nsEscape.h"
1.14 +#include "nsAutoPtr.h"
1.15 +#include "nsServiceManagerUtils.h"
1.16 +
1.17 +NS_IMPL_ISUPPORTS(nsUTF8ConverterService, nsIUTF8ConverterService)
1.18 +
1.19 +static nsresult
1.20 +ToUTF8(const nsACString &aString, const char *aCharset,
1.21 + bool aAllowSubstitution, nsACString &aResult)
1.22 +{
1.23 + nsresult rv;
1.24 + if (!aCharset || !*aCharset)
1.25 + return NS_ERROR_INVALID_ARG;
1.26 +
1.27 + nsCOMPtr<nsICharsetConverterManager> ccm;
1.28 +
1.29 + ccm = do_GetService(NS_CHARSETCONVERTERMANAGER_CONTRACTID, &rv);
1.30 + NS_ENSURE_SUCCESS(rv, rv);
1.31 +
1.32 + nsCOMPtr<nsIUnicodeDecoder> unicodeDecoder;
1.33 + rv = ccm->GetUnicodeDecoder(aCharset,
1.34 + getter_AddRefs(unicodeDecoder));
1.35 + NS_ENSURE_SUCCESS(rv, rv);
1.36 +
1.37 + if (!aAllowSubstitution)
1.38 + unicodeDecoder->SetInputErrorBehavior(nsIUnicodeDecoder::kOnError_Signal);
1.39 +
1.40 + int32_t srcLen = aString.Length();
1.41 + int32_t dstLen;
1.42 + const nsAFlatCString& inStr = PromiseFlatCString(aString);
1.43 + rv = unicodeDecoder->GetMaxLength(inStr.get(), srcLen, &dstLen);
1.44 + NS_ENSURE_SUCCESS(rv, rv);
1.45 +
1.46 + nsAutoArrayPtr<char16_t> ustr(new char16_t[dstLen]);
1.47 + NS_ENSURE_TRUE(ustr, NS_ERROR_OUT_OF_MEMORY);
1.48 +
1.49 + rv = unicodeDecoder->Convert(inStr.get(), &srcLen, ustr, &dstLen);
1.50 + if (NS_SUCCEEDED(rv)){
1.51 + // Tru64 Cxx needs an explicit get()
1.52 + CopyUTF16toUTF8(Substring(ustr.get(), ustr + dstLen), aResult);
1.53 + }
1.54 + return rv;
1.55 +}
1.56 +
1.57 +NS_IMETHODIMP
1.58 +nsUTF8ConverterService::ConvertStringToUTF8(const nsACString &aString,
1.59 + const char *aCharset,
1.60 + bool aSkipCheck,
1.61 + bool aAllowSubstitution,
1.62 + uint8_t aOptionalArgc,
1.63 + nsACString &aUTF8String)
1.64 +{
1.65 + bool allowSubstitution = (aOptionalArgc == 1) ? aAllowSubstitution : true;
1.66 +
1.67 + // return if ASCII only or valid UTF-8 providing that the ASCII/UTF-8
1.68 + // check is requested. It may not be asked for if a caller suspects
1.69 + // that the input is in non-ASCII 7bit charset (ISO-2022-xx, HZ) or
1.70 + // it's in a charset other than UTF-8 that can be mistaken for UTF-8.
1.71 + if (!aSkipCheck && (IsASCII(aString) || IsUTF8(aString))) {
1.72 + aUTF8String = aString;
1.73 + return NS_OK;
1.74 + }
1.75 +
1.76 + aUTF8String.Truncate();
1.77 +
1.78 + nsresult rv = ToUTF8(aString, aCharset, allowSubstitution, aUTF8String);
1.79 +
1.80 + // additional protection for cases where check is skipped and the input
1.81 + // is actually in UTF-8 as opposed to aCharset. (i.e. caller's hunch
1.82 + // was wrong.) We don't check ASCIIness assuming there's no charset
1.83 + // incompatible with ASCII (we don't support EBCDIC).
1.84 + if (aSkipCheck && NS_FAILED(rv) && IsUTF8(aString)) {
1.85 + aUTF8String = aString;
1.86 + return NS_OK;
1.87 + }
1.88 +
1.89 + return rv;
1.90 +}
1.91 +
1.92 +NS_IMETHODIMP
1.93 +nsUTF8ConverterService::ConvertURISpecToUTF8(const nsACString &aSpec,
1.94 + const char *aCharset,
1.95 + nsACString &aUTF8Spec)
1.96 +{
1.97 + // assume UTF-8 if the spec contains unescaped non-ASCII characters.
1.98 + // No valid spec in Mozilla would break this assumption.
1.99 + if (!IsASCII(aSpec)) {
1.100 + aUTF8Spec = aSpec;
1.101 + return NS_OK;
1.102 + }
1.103 +
1.104 + aUTF8Spec.Truncate();
1.105 +
1.106 + nsAutoCString unescapedSpec;
1.107 + // NS_UnescapeURL does not fill up unescapedSpec unless there's at least
1.108 + // one character to unescape.
1.109 + bool written = NS_UnescapeURL(PromiseFlatCString(aSpec).get(), aSpec.Length(),
1.110 + esc_OnlyNonASCII, unescapedSpec);
1.111 +
1.112 + if (!written) {
1.113 + aUTF8Spec = aSpec;
1.114 + return NS_OK;
1.115 + }
1.116 + // return if ASCII only or escaped UTF-8
1.117 + if (IsASCII(unescapedSpec) || IsUTF8(unescapedSpec)) {
1.118 + aUTF8Spec = unescapedSpec;
1.119 + return NS_OK;
1.120 + }
1.121 +
1.122 + return ToUTF8(unescapedSpec, aCharset, true, aUTF8Spec);
1.123 +}
1.124 +
