1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/uconv/src/nsTextToSubURI.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,239 @@
1.4 +/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +#include "nsString.h"
1.9 +#include "nsIUnicodeEncoder.h"
1.10 +#include "nsICharsetConverterManager.h"
1.11 +#include "nsITextToSubURI.h"
1.12 +#include "nsEscape.h"
1.13 +#include "nsTextToSubURI.h"
1.14 +#include "nsCRT.h"
1.15 +#include "nsServiceManagerUtils.h"
1.16 +
1.17 +static NS_DEFINE_CID(kCharsetConverterManagerCID, NS_ICHARSETCONVERTERMANAGER_CID);
1.18 +
1.19 +nsTextToSubURI::nsTextToSubURI()
1.20 +{
1.21 +}
1.22 +nsTextToSubURI::~nsTextToSubURI()
1.23 +{
1.24 +}
1.25 +
1.26 +NS_IMPL_ISUPPORTS(nsTextToSubURI, nsITextToSubURI)
1.27 +
1.28 +NS_IMETHODIMP nsTextToSubURI::ConvertAndEscape(
1.29 + const char *charset, const char16_t *text, char **_retval)
1.30 +{
1.31 + if(nullptr == _retval)
1.32 + return NS_ERROR_NULL_POINTER;
1.33 + *_retval = nullptr;
1.34 + nsresult rv = NS_OK;
1.35 +
1.36 + // Get Charset, get the encoder.
1.37 + nsICharsetConverterManager *ccm;
1.38 + rv = CallGetService(kCharsetConverterManagerCID, &ccm);
1.39 + if(NS_SUCCEEDED(rv)) {
1.40 + nsIUnicodeEncoder *encoder;
1.41 + rv = ccm->GetUnicodeEncoder(charset, &encoder);
1.42 + NS_RELEASE(ccm);
1.43 + if (NS_SUCCEEDED(rv)) {
1.44 + rv = encoder->SetOutputErrorBehavior(nsIUnicodeEncoder::kOnError_Replace, nullptr, (char16_t)'?');
1.45 + if(NS_SUCCEEDED(rv))
1.46 + {
1.47 + char buf[256];
1.48 + char *pBuf = buf;
1.49 + int32_t ulen = text ? NS_strlen(text) : 0;
1.50 + int32_t outlen = 0;
1.51 + if(NS_SUCCEEDED(rv = encoder->GetMaxLength(text, ulen, &outlen)))
1.52 + {
1.53 + if(outlen >= 256) {
1.54 + pBuf = (char*)NS_Alloc(outlen+1);
1.55 + }
1.56 + if(nullptr == pBuf) {
1.57 + outlen = 255;
1.58 + pBuf = buf;
1.59 + }
1.60 + int32_t bufLen = outlen;
1.61 + if(NS_SUCCEEDED(rv = encoder->Convert(text,&ulen, pBuf, &outlen))) {
1.62 + // put termination characters (e.g. ESC(B of ISO-2022-JP) if necessary
1.63 + int32_t finLen = bufLen - outlen;
1.64 + if (finLen > 0) {
1.65 + if (NS_SUCCEEDED(encoder->Finish((char *)(pBuf+outlen), &finLen)))
1.66 + outlen += finLen;
1.67 + }
1.68 + pBuf[outlen] = '\0';
1.69 + *_retval = nsEscape(pBuf, url_XPAlphas);
1.70 + if(nullptr == *_retval)
1.71 + rv = NS_ERROR_OUT_OF_MEMORY;
1.72 + }
1.73 + }
1.74 + if(pBuf != buf)
1.75 + NS_Free(pBuf);
1.76 + }
1.77 + NS_RELEASE(encoder);
1.78 + }
1.79 + }
1.80 +
1.81 + return rv;
1.82 +}
1.83 +
1.84 +NS_IMETHODIMP nsTextToSubURI::UnEscapeAndConvert(
1.85 + const char *charset, const char *text, char16_t **_retval)
1.86 +{
1.87 + if(nullptr == _retval)
1.88 + return NS_ERROR_NULL_POINTER;
1.89 + if(nullptr == text) {
1.90 + // set empty string instead of returning error
1.91 + // due to compatibility for old version
1.92 + text = "";
1.93 + }
1.94 + *_retval = nullptr;
1.95 + nsresult rv = NS_OK;
1.96 +
1.97 + // unescape the string, unescape changes the input
1.98 + char *unescaped = NS_strdup(text);
1.99 + if (nullptr == unescaped)
1.100 + return NS_ERROR_OUT_OF_MEMORY;
1.101 + unescaped = nsUnescape(unescaped);
1.102 + NS_ASSERTION(unescaped, "nsUnescape returned null");
1.103 +
1.104 + // Convert from the charset to unicode
1.105 + nsCOMPtr<nsICharsetConverterManager> ccm =
1.106 + do_GetService(kCharsetConverterManagerCID, &rv);
1.107 + if (NS_SUCCEEDED(rv)) {
1.108 + nsIUnicodeDecoder *decoder;
1.109 + rv = ccm->GetUnicodeDecoder(charset, &decoder);
1.110 + if (NS_SUCCEEDED(rv)) {
1.111 + char16_t *pBuf = nullptr;
1.112 + int32_t len = strlen(unescaped);
1.113 + int32_t outlen = 0;
1.114 + if (NS_SUCCEEDED(rv = decoder->GetMaxLength(unescaped, len, &outlen))) {
1.115 + pBuf = (char16_t *) NS_Alloc((outlen+1)*sizeof(char16_t));
1.116 + if (nullptr == pBuf)
1.117 + rv = NS_ERROR_OUT_OF_MEMORY;
1.118 + else {
1.119 + if (NS_SUCCEEDED(rv = decoder->Convert(unescaped, &len, pBuf, &outlen))) {
1.120 + pBuf[outlen] = 0;
1.121 + *_retval = pBuf;
1.122 + }
1.123 + else
1.124 + NS_Free(pBuf);
1.125 + }
1.126 + }
1.127 + NS_RELEASE(decoder);
1.128 + }
1.129 + }
1.130 + NS_Free(unescaped);
1.131 +
1.132 + return rv;
1.133 +}
1.134 +
1.135 +static bool statefulCharset(const char *charset)
1.136 +{
1.137 + if (!nsCRT::strncasecmp(charset, "ISO-2022-", sizeof("ISO-2022-")-1) ||
1.138 + !nsCRT::strcasecmp(charset, "UTF-7") ||
1.139 + !nsCRT::strcasecmp(charset, "HZ-GB-2312"))
1.140 + return true;
1.141 +
1.142 + return false;
1.143 +}
1.144 +
1.145 +nsresult nsTextToSubURI::convertURItoUnicode(const nsAFlatCString &aCharset,
1.146 + const nsAFlatCString &aURI,
1.147 + bool aIRI,
1.148 + nsAString &_retval)
1.149 +{
1.150 + nsresult rv = NS_OK;
1.151 +
1.152 + // check for 7bit encoding the data may not be ASCII after we decode
1.153 + bool isStatefulCharset = statefulCharset(aCharset.get());
1.154 +
1.155 + if (!isStatefulCharset && IsASCII(aURI)) {
1.156 + CopyASCIItoUTF16(aURI, _retval);
1.157 + return rv;
1.158 + }
1.159 +
1.160 + if (!isStatefulCharset && aIRI) {
1.161 + if (IsUTF8(aURI)) {
1.162 + CopyUTF8toUTF16(aURI, _retval);
1.163 + return rv;
1.164 + }
1.165 + }
1.166 +
1.167 + // empty charset could indicate UTF-8, but aURI turns out not to be UTF-8.
1.168 + NS_ENSURE_FALSE(aCharset.IsEmpty(), NS_ERROR_INVALID_ARG);
1.169 +
1.170 + nsCOMPtr<nsICharsetConverterManager> charsetConverterManager;
1.171 +
1.172 + charsetConverterManager = do_GetService(NS_CHARSETCONVERTERMANAGER_CONTRACTID, &rv);
1.173 + NS_ENSURE_SUCCESS(rv, rv);
1.174 +
1.175 + nsCOMPtr<nsIUnicodeDecoder> unicodeDecoder;
1.176 + rv = charsetConverterManager->GetUnicodeDecoder(aCharset.get(),
1.177 + getter_AddRefs(unicodeDecoder));
1.178 + NS_ENSURE_SUCCESS(rv, rv);
1.179 + unicodeDecoder->SetInputErrorBehavior(nsIUnicodeDecoder::kOnError_Signal);
1.180 +
1.181 + int32_t srcLen = aURI.Length();
1.182 + int32_t dstLen;
1.183 + rv = unicodeDecoder->GetMaxLength(aURI.get(), srcLen, &dstLen);
1.184 + NS_ENSURE_SUCCESS(rv, rv);
1.185 +
1.186 + char16_t *ustr = (char16_t *) NS_Alloc(dstLen * sizeof(char16_t));
1.187 + NS_ENSURE_TRUE(ustr, NS_ERROR_OUT_OF_MEMORY);
1.188 +
1.189 + rv = unicodeDecoder->Convert(aURI.get(), &srcLen, ustr, &dstLen);
1.190 +
1.191 + if (NS_SUCCEEDED(rv))
1.192 + _retval.Assign(ustr, dstLen);
1.193 +
1.194 + NS_Free(ustr);
1.195 +
1.196 + return rv;
1.197 +}
1.198 +
1.199 +NS_IMETHODIMP nsTextToSubURI::UnEscapeURIForUI(const nsACString & aCharset,
1.200 + const nsACString &aURIFragment,
1.201 + nsAString &_retval)
1.202 +{
1.203 + nsAutoCString unescapedSpec;
1.204 + // skip control octets (0x00 - 0x1f and 0x7f) when unescaping
1.205 + NS_UnescapeURL(PromiseFlatCString(aURIFragment),
1.206 + esc_SkipControl | esc_AlwaysCopy, unescapedSpec);
1.207 +
1.208 + // in case of failure, return escaped URI
1.209 + // Test for != NS_OK rather than NS_FAILED, because incomplete multi-byte
1.210 + // sequences are also considered failure in this context
1.211 + if (convertURItoUnicode(
1.212 + PromiseFlatCString(aCharset), unescapedSpec, true, _retval)
1.213 + != NS_OK)
1.214 + // assume UTF-8 instead of ASCII because hostname (IDN) may be in UTF-8
1.215 + CopyUTF8toUTF16(aURIFragment, _retval);
1.216 + return NS_OK;
1.217 +}
1.218 +
1.219 +NS_IMETHODIMP nsTextToSubURI::UnEscapeNonAsciiURI(const nsACString & aCharset,
1.220 + const nsACString & aURIFragment,
1.221 + nsAString &_retval)
1.222 +{
1.223 + nsAutoCString unescapedSpec;
1.224 + NS_UnescapeURL(PromiseFlatCString(aURIFragment),
1.225 + esc_AlwaysCopy | esc_OnlyNonASCII, unescapedSpec);
1.226 + // leave the URI as it is if it's not UTF-8 and aCharset is not a ASCII
1.227 + // superset since converting "http:" with such an encoding is always a bad
1.228 + // idea.
1.229 + if (!IsUTF8(unescapedSpec) &&
1.230 + (aCharset.LowerCaseEqualsLiteral("utf-16") ||
1.231 + aCharset.LowerCaseEqualsLiteral("utf-16be") ||
1.232 + aCharset.LowerCaseEqualsLiteral("utf-16le") ||
1.233 + aCharset.LowerCaseEqualsLiteral("utf-7") ||
1.234 + aCharset.LowerCaseEqualsLiteral("x-imap4-modified-utf7"))){
1.235 + CopyASCIItoUTF16(aURIFragment, _retval);
1.236 + return NS_OK;
1.237 + }
1.238 +
1.239 + return convertURItoUnicode(PromiseFlatCString(aCharset), unescapedSpec, true, _retval);
1.240 +}
1.241 +
1.242 +//----------------------------------------------------------------------
