intl/uconv/src/nsCharsetConverterManager.cpp

   1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
   2 /* ***** BEGIN LICENSE BLOCK *****
   3  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   4  *
   5  * The contents of this file are subject to the Mozilla Public License Version
   6  * 1.1 (the "License"); you may not use this file except in compliance with
   7  * the License. You may obtain a copy of the License at
   8  * http://www.mozilla.org/MPL/
   9  *
  10  * Software distributed under the License is distributed on an "AS IS" basis,
  11  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  12  * for the specific language governing rights and limitations under the
  13  * License.
  14  *
  15  * The Original Code is Mozilla Communicator client code.
  16  *
  17  * The Initial Developer of the Original Code is
  18  * Netscape Communications Corporation.
  19  * Portions created by the Initial Developer are Copyright (C) 1998
  20  * the Initial Developer. All Rights Reserved.
  21  *
  22  * Contributor(s):
  23  *   Pierre Phaneuf <pp@ludusdesign.com>
  24  *
  25  * Alternatively, the contents of this file may be used under the terms of
  26  * either of the GNU General Public License Version 2 or later (the "GPL"),
  27  * or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  28  * in which case the provisions of the GPL or the LGPL are applicable instead
  29  * of those above. If you wish to allow use of your version of this file only
  30  * under the terms of either the GPL or the LGPL, and not to allow others to
  31  * use your version of this file under the terms of the MPL, indicate your
  32  * decision by deleting the provisions above and replace them with the notice
  33  * and other provisions required by the GPL or the LGPL. If you do not delete
  34  * the provisions above, a recipient may use your version of this file under
  35  * the terms of any one of the MPL, the GPL or the LGPL.
  36  *
  37  * ***** END LICENSE BLOCK ***** */
  38
  39 #include "nsCOMPtr.h"
  40 #include "nsString.h"
  41 #include "nsReadableUtils.h"
  42 #include "nsUnicharUtils.h"
  43 #include "nsICharsetAlias.h"
  44 #include "nsIServiceManager.h"
  45 #include "nsICategoryManager.h"
  46 #include "nsICharsetConverterManager.h"
  47 #include "nsEncoderDecoderUtils.h"
  48 #include "nsIStringBundle.h"
  49 #include "nsILocaleService.h"
  50 #include "nsUConvDll.h"
  51 #include "prmem.h"
  52 #include "nsCRT.h"
  53 #include "nsVoidArray.h"
  54 #include "nsStringEnumerator.h"
  55
  56 #include "nsXPCOM.h"
  57 #include "nsISupportsPrimitives.h"
  58
  59 // just for CONTRACTIDs
  60 #include "nsCharsetConverterManager.h"
  61
  62 #ifdef MOZ_USE_NATIVE_UCONV
  63 #include "nsNativeUConvService.h"
  64 #endif
  65
  66 // Pattern of cached, commonly used, single byte decoder
  67 #define NS_1BYTE_CODER_PATTERN "ISO-8859"
  68 #define NS_1BYTE_CODER_PATTERN_LEN 8
  69
  70 // Class nsCharsetConverterManager [implementation]
  71
  72 NS_IMPL_THREADSAFE_ISUPPORTS1(nsCharsetConverterManager,
  73                               nsICharsetConverterManager)
  74
  75 nsCharsetConverterManager::nsCharsetConverterManager()
  76   :mDataBundle(NULL), mTitleBundle(NULL)
  77 {
  78 #ifdef MOZ_USE_NATIVE_UCONV
  79   mNativeUC = do_GetService(NS_NATIVE_UCONV_SERVICE_CONTRACT_ID);
  80 #endif
  81 }
  82
  83 nsCharsetConverterManager::~nsCharsetConverterManager()
  84 {
  85   NS_IF_RELEASE(mDataBundle);
  86   NS_IF_RELEASE(mTitleBundle);
  87 }
  88
  89 nsresult nsCharsetConverterManager::Init()
  90 {
  91   if (!mDecoderHash.Init())
  92     return NS_ERROR_OUT_OF_MEMORY;
  93   return NS_OK;
  94 }
  95
  96 nsresult nsCharsetConverterManager::RegisterConverterManagerData()
  97 {
  98   nsresult rv;
  99   nsCOMPtr<nsICategoryManager> catman = do_GetService(NS_CATEGORYMANAGER_CONTRACTID, &rv);
 100   if (NS_FAILED(rv))
 101     return rv;
 102
 103   RegisterConverterCategory(catman, NS_TITLE_BUNDLE_CATEGORY,
 104                             "chrome://global/locale/charsetTitles.properties");
 105   RegisterConverterCategory(catman, NS_DATA_BUNDLE_CATEGORY,
 106                             "resource://gre/res/charsetData.properties");
 107
 108   return NS_OK;
 109 }
 110
 111 nsresult
 112 nsCharsetConverterManager::RegisterConverterCategory(nsICategoryManager* catman,
 113                                                      const char* aCategory,
 114                                                      const char* aURL)
 115 {
 116   return catman->AddCategoryEntry(aCategory, aURL, "",
 117                                   PR_TRUE, PR_TRUE, nsnull);
 118 }
 119
 120 nsresult nsCharsetConverterManager::LoadExtensibleBundle(
 121                                     const char* aCategory,
 122                                     nsIStringBundle ** aResult)
 123 {
 124   nsresult rv = NS_OK;
 125
 126   nsCOMPtr<nsIStringBundleService> sbServ =
 127            do_GetService(NS_STRINGBUNDLE_CONTRACTID, &rv);
 128   if (NS_FAILED(rv))
 129     return rv;
 130
 131   return sbServ->CreateExtensibleBundle(aCategory, aResult);
 132 }
 133
 134 nsresult nsCharsetConverterManager::GetBundleValue(nsIStringBundle * aBundle,
 135                                                    const char * aName,
 136                                                    const nsAFlatString& aProp,
 137                                                    PRUnichar ** aResult)
 138 {
 139   nsAutoString key;
 140
 141   key.AssignWithConversion(aName);
 142   ToLowerCase(key); // we lowercase the main comparison key
 143   key.Append(aProp);
 144
 145   return aBundle->GetStringFromName(key.get(), aResult);
 146 }
 147
 148 nsresult nsCharsetConverterManager::GetBundleValue(nsIStringBundle * aBundle,
 149                                                    const char * aName,
 150                                                    const nsAFlatString& aProp,
 151                                                    nsAString& aResult)
 152 {
 153   nsresult rv = NS_OK;
 154
 155   nsXPIDLString value;
 156   rv = GetBundleValue(aBundle, aName, aProp, getter_Copies(value));
 157   if (NS_FAILED(rv))
 158     return rv;
 159
 160   aResult = value;
 161
 162   return NS_OK;
 163 }
 164
 165
 166 //----------------------------------------------------------------------------//----------------------------------------------------------------------------
 167 // Interface nsICharsetConverterManager [implementation]
 168
 169 NS_IMETHODIMP
 170 nsCharsetConverterManager::GetUnicodeEncoder(const char * aDest,
 171                                              nsIUnicodeEncoder ** aResult)
 172 {
 173   // resolve the charset first
 174   nsCAutoString charset;
 175
 176   // fully qualify to possibly avoid vtable call
 177   nsCharsetConverterManager::GetCharsetAlias(aDest, charset);
 178
 179   return nsCharsetConverterManager::GetUnicodeEncoderRaw(charset.get(),
 180                                                          aResult);
 181 }
 182
 183
 184 NS_IMETHODIMP
 185 nsCharsetConverterManager::GetUnicodeEncoderRaw(const char * aDest,
 186                                                 nsIUnicodeEncoder ** aResult)
 187 {
 188   *aResult= nsnull;
 189   nsCOMPtr<nsIUnicodeEncoder> encoder;
 190
 191 #ifdef MOZ_USE_NATIVE_UCONV
 192   if (mNativeUC) {
 193     nsCOMPtr<nsISupports> supports;
 194     mNativeUC->GetNativeConverter("UCS-2",
 195                                   aDest,
 196                                   getter_AddRefs(supports));
 197
 198     encoder = do_QueryInterface(supports);
 199
 200     if (encoder) {
 201       NS_ADDREF(*aResult = encoder);
 202       return NS_OK;
 203     }
 204   }
 205 #endif
 206   nsresult rv = NS_OK;
 207
 208   nsCAutoString
 209     contractid(NS_LITERAL_CSTRING(NS_UNICODEENCODER_CONTRACTID_BASE) +
 210                nsDependentCString(aDest));
 211
 212   // Always create an instance since encoders hold state.
 213   encoder = do_CreateInstance(contractid.get(), &rv);
 214
 215   if (NS_FAILED(rv))
 216     rv = NS_ERROR_UCONV_NOCONV;
 217   else
 218   {
 219     *aResult = encoder.get();
 220     NS_ADDREF(*aResult);
 221   }
 222   return rv;
 223 }
 224
 225 NS_IMETHODIMP
 226 nsCharsetConverterManager::GetUnicodeDecoder(const char * aSrc,
 227                                              nsIUnicodeDecoder ** aResult)
 228 {
 229   // resolve the charset first
 230   nsCAutoString charset;
 231
 232   // fully qualify to possibly avoid vtable call
 233   nsCharsetConverterManager::GetCharsetAlias(aSrc, charset);
 234
 235   return nsCharsetConverterManager::GetUnicodeDecoderRaw(charset.get(),
 236                                                          aResult);
 237 }
 238
 239 NS_IMETHODIMP
 240 nsCharsetConverterManager::GetUnicodeDecoderRaw(const char * aSrc,
 241                                                 nsIUnicodeDecoder ** aResult)
 242 {
 243   *aResult= nsnull;
 244   nsCOMPtr<nsIUnicodeDecoder> decoder;
 245
 246 #ifdef MOZ_USE_NATIVE_UCONV
 247   if (mNativeUC) {
 248     nsCOMPtr<nsISupports> supports;
 249     mNativeUC->GetNativeConverter(aSrc,
 250                                   "UCS-2",
 251                                   getter_AddRefs(supports));
 252
 253     decoder = do_QueryInterface(supports);
 254
 255     if (decoder) {
 256       NS_ADDREF(*aResult = decoder);
 257       return NS_OK;
 258     }
 259   }
 260 #endif
 261   nsresult rv = NS_OK;
 262
 263   NS_NAMED_LITERAL_CSTRING(contractbase, NS_UNICODEDECODER_CONTRACTID_BASE);
 264   nsDependentCString src(aSrc);
 265
 266   if (!strncmp(aSrc, NS_1BYTE_CODER_PATTERN, NS_1BYTE_CODER_PATTERN_LEN))
 267   {
 268     // Single byte decoders don't hold state. Optimize by using a service, and
 269     // cache it in our hash to avoid repeated trips through the service manager.
 270     if (!mDecoderHash.Get(aSrc, getter_AddRefs(decoder))) {
 271       decoder = do_GetService(PromiseFlatCString(contractbase + src).get(),
 272                               &rv);
 273       if (NS_SUCCEEDED(rv))
 274         mDecoderHash.Put(aSrc, decoder);
 275     }
 276   }
 277   else
 278   {
 279     decoder = do_CreateInstance(PromiseFlatCString(contractbase + src).get(),
 280                                 &rv);
 281   }
 282   NS_ENSURE_SUCCESS(rv, NS_ERROR_UCONV_NOCONV);
 283
 284   decoder.forget(aResult);
 285   return rv;
 286 }
 287
 288 nsresult
 289 nsCharsetConverterManager::GetList(const nsACString& aCategory,
 290                                    const nsACString& aPrefix,
 291                                    nsIUTF8StringEnumerator** aResult)
 292 {
 293   if (aResult == NULL)
 294     return NS_ERROR_NULL_POINTER;
 295   *aResult = NULL;
 296
 297   nsresult rv;
 298   nsCAutoString alias;
 299
 300   nsCOMPtr<nsICategoryManager> catman = do_GetService(NS_CATEGORYMANAGER_CONTRACTID, &rv);
 301   if (NS_FAILED(rv))
 302     return rv;
 303
 304   nsCStringArray* array = new nsCStringArray;
 305   if (!array)
 306     return NS_ERROR_OUT_OF_MEMORY;
 307
 308   nsCOMPtr<nsISimpleEnumerator> enumerator;
 309   catman->EnumerateCategory(PromiseFlatCString(aCategory).get(),
 310                             getter_AddRefs(enumerator));
 311
 312   PRBool hasMore;
 313   while (NS_SUCCEEDED(enumerator->HasMoreElements(&hasMore)) && hasMore) {
 314     nsCOMPtr<nsISupports> supports;
 315     if (NS_FAILED(enumerator->GetNext(getter_AddRefs(supports))))
 316       continue;
 317
 318     nsCOMPtr<nsISupportsCString> supStr = do_QueryInterface(supports);
 319     if (!supStr)
 320       continue;
 321
 322     nsCAutoString fullName(aPrefix);
 323
 324     nsCAutoString name;
 325     if (NS_FAILED(supStr->GetData(name)))
 326       continue;
 327
 328     fullName += name;
 329     rv = GetCharsetAlias(fullName.get(), alias);
 330     if (NS_FAILED(rv))
 331       continue;
 332
 333     rv = array->AppendCString(alias);
 334   }
 335
 336   return NS_NewAdoptingUTF8StringEnumerator(aResult, array);
 337 }
 338
 339 // we should change the interface so that we can just pass back a enumerator!
 340 NS_IMETHODIMP
 341 nsCharsetConverterManager::GetDecoderList(nsIUTF8StringEnumerator ** aResult)
 342 {
 343   return GetList(NS_LITERAL_CSTRING(NS_UNICODEDECODER_NAME),
 344                  EmptyCString(), aResult);
 345 }
 346
 347 NS_IMETHODIMP
 348 nsCharsetConverterManager::GetEncoderList(nsIUTF8StringEnumerator ** aResult)
 349 {
 350   return GetList(NS_LITERAL_CSTRING(NS_UNICODEENCODER_NAME),
 351                  EmptyCString(), aResult);
 352 }
 353
 354 NS_IMETHODIMP
 355 nsCharsetConverterManager::GetCharsetDetectorList(nsIUTF8StringEnumerator** aResult)
 356 {
 357   return GetList(NS_LITERAL_CSTRING("charset-detectors"),
 358                  NS_LITERAL_CSTRING("chardet."), aResult);
 359 }
 360
 361 // XXX Improve the implementation of this method. Right now, it is build on
 362 // top of the nsCharsetAlias service. We can make the nsCharsetAlias
 363 // better, with its own hash table (not the StringBundle anymore) and
 364 // a nicer file format.
 365 NS_IMETHODIMP
 366 nsCharsetConverterManager::GetCharsetAlias(const char * aCharset,
 367                                            nsACString& aResult)
 368 {
 369   NS_PRECONDITION(aCharset, "null param");
 370   if (!aCharset)
 371     return NS_ERROR_NULL_POINTER;
 372
 373   // We try to obtain the preferred name for this charset from the charset
 374   // aliases. If we don't get it from there, we just use the original string
 375   nsDependentCString charset(aCharset);
 376   nsCOMPtr<nsICharsetAlias> csAlias(do_GetService(NS_CHARSETALIAS_CONTRACTID));
 377   NS_ASSERTION(csAlias, "failed to get the CharsetAlias service");
 378   if (csAlias) {
 379     nsAutoString pref;
 380     nsresult rv = csAlias->GetPreferred(charset, aResult);
 381     if (NS_SUCCEEDED(rv)) {
 382       return (!aResult.IsEmpty()) ? NS_OK : NS_ERROR_OUT_OF_MEMORY;
 383     }
 384   }
 385
 386   aResult = charset;
 387   return NS_OK;
 388 }
 389
 390
 391 NS_IMETHODIMP
 392 nsCharsetConverterManager::GetCharsetTitle(const char * aCharset,
 393                                            nsAString& aResult)
 394 {
 395   if (aCharset == NULL) return NS_ERROR_NULL_POINTER;
 396
 397   if (mTitleBundle == NULL) {
 398     nsresult rv = LoadExtensibleBundle(NS_TITLE_BUNDLE_CATEGORY, &mTitleBundle);
 399     if (NS_FAILED(rv))
 400       return rv;
 401   }
 402
 403   return GetBundleValue(mTitleBundle, aCharset, NS_LITERAL_STRING(".title"), aResult);
 404 }
 405
 406 NS_IMETHODIMP
 407 nsCharsetConverterManager::GetCharsetData(const char * aCharset,
 408                                           const PRUnichar * aProp,
 409                                           nsAString& aResult)
 410 {
 411   if (aCharset == NULL)
 412     return NS_ERROR_NULL_POINTER;
 413   // aProp can be NULL
 414
 415   if (mDataBundle == NULL) {
 416     nsresult rv = LoadExtensibleBundle(NS_DATA_BUNDLE_CATEGORY, &mDataBundle);
 417     if (NS_FAILED(rv))
 418       return rv;
 419   }
 420
 421   return GetBundleValue(mDataBundle, aCharset, nsDependentString(aProp), aResult);
 422 }
 423
 424 NS_IMETHODIMP
 425 nsCharsetConverterManager::GetCharsetLangGroup(const char * aCharset,
 426                                                nsIAtom** aResult)
 427 {
 428   // resolve the charset first
 429   nsCAutoString charset;
 430
 431   nsresult rv = GetCharsetAlias(aCharset, charset);
 432   if (NS_FAILED(rv))
 433     return rv;
 434
 435   // fully qualify to possibly avoid vtable call
 436   return nsCharsetConverterManager::GetCharsetLangGroupRaw(charset.get(),
 437                                                            aResult);
 438 }
 439
 440 NS_IMETHODIMP
 441 nsCharsetConverterManager::GetCharsetLangGroupRaw(const char * aCharset,
 442                                                   nsIAtom** aResult)
 443 {
 444
 445   *aResult = nsnull;
 446   if (aCharset == NULL)
 447     return NS_ERROR_NULL_POINTER;
 448
 449   nsresult rv = NS_OK;
 450
 451   if (mDataBundle == NULL) {
 452     rv = LoadExtensibleBundle(NS_DATA_BUNDLE_CATEGORY, &mDataBundle);
 453     if (NS_FAILED(rv))
 454       return rv;
 455   }
 456
 457   nsAutoString langGroup;
 458   rv = GetBundleValue(mDataBundle, aCharset, NS_LITERAL_STRING(".LangGroup"), langGroup);
 459
 460   if (NS_SUCCEEDED(rv))
 461     *aResult = NS_NewAtom(langGroup);
 462
 463   return rv;
 464 }