more work on UTF-7 converters

intl/uconv/ucvlatin/Makefile.in

@@ -64,6 +64,7 @@ CPPSRCS		= \
                   nsVISCIIToUnicode.cpp           \
                   nsVPSToUnicode.cpp              \
                   nsUTF8ToUnicode.cpp \
+                  nsUTF7ToUnicode.cpp \
                   nsMUTF7ToUnicode.cpp \
                   nsUnicodeToISO88591.cpp \
                   nsUnicodeToISO88592.cpp \
@@ -100,6 +101,7 @@ CPPSRCS		= \
                   nsUnicodeToVISCII.cpp           \
                   nsUnicodeToVPS.cpp              \
                   nsUnicodeToUTF8.cpp \
+                  nsUnicodeToUTF7.cpp \
                   nsUnicodeToMUTF7.cpp \
                   ns1ByteToUnicodeBase.cpp \
 		  nsUCvLatinSupport.cpp \

intl/uconv/ucvlatin/makefile.win

@@ -62,6 +62,7 @@ CPPSRCS =				\
 	nsVISCIIToUnicode.cpp		\
 	nsVPSToUnicode.cpp		\
 	nsUTF8ToUnicode.cpp		\
+	nsUTF7ToUnicode.cpp		\
 	nsMUTF7ToUnicode.cpp		\
 	nsUnicodeToISO88591.cpp		\
 	nsUnicodeToISO88592.cpp		\
@@ -94,6 +95,7 @@ CPPSRCS =				\
 	nsUnicodeToMacUkrainian.cpp	\
 	nsUnicodeToMacIcelandic.cpp	\
 	nsUnicodeToUTF8.cpp		\
+	nsUnicodeToUTF7.cpp		\
 	nsUnicodeToMUTF7.cpp		\
 	nsUnicodeToARMSCII8.cpp		\
 	nsUnicodeToTCVN5712.cpp		\
@@ -131,6 +133,7 @@ CPP_OBJS=						\
 	.\$(OBJDIR)\nsMacGreekToUnicode.obj 		\
 	.\$(OBJDIR)\nsMacTurkishToUnicode.obj 		\
 	.\$(OBJDIR)\nsUTF8ToUnicode.obj 		\
+	.\$(OBJDIR)\nsUTF7ToUnicode.obj 		\
 	.\$(OBJDIR)\nsMUTF7ToUnicode.obj 		\
 	.\$(OBJDIR)\nsMacCroatianToUnicode.obj		\
 	.\$(OBJDIR)\nsMacRomanianToUnicode.obj		\
@@ -167,6 +170,7 @@ CPP_OBJS=						\
 	.\$(OBJDIR)\nsUnicodeToMacGreek.obj		\
 	.\$(OBJDIR)\nsUnicodeToMacTurkish.obj		\
 	.\$(OBJDIR)\nsUnicodeToUTF8.obj			\
+	.\$(OBJDIR)\nsUnicodeToUTF7.obj			\
 	.\$(OBJDIR)\nsUnicodeToMUTF7.obj		\
 	.\$(OBJDIR)\nsUnicodeToMacCroatian.obj		\
 	.\$(OBJDIR)\nsUnicodeToMacRomanian.obj		\

intl/uconv/ucvlatin/nsMUTF7ToUnicode.cpp

@@ -19,24 +19,11 @@
 
 #include "nsMUTF7ToUnicode.h"
 
-//----------------------------------------------------------------------
-// Global functions and data [declaration]
-
-static PRUint16 g_utMappingTable[] = {
-#include "cp1252.ut"
-};
-
-static PRInt16 g_utShiftTable[] =  {
-  0, u1ByteCharset ,
-  ShiftCell(0,0,0,0,0,0,0,0)
-};
-
 //----------------------------------------------------------------------
 // Class nsMUTF7ToUnicode [implementation]
 
 nsMUTF7ToUnicode::nsMUTF7ToUnicode() 
-: nsOneByteDecoderSupport((uShiftTable*) &g_utShiftTable, 
+: nsBasicUTF7Decoder(',', '&')
-                          (uMappingTable*) &g_utMappingTable)
 {
 }
 

intl/uconv/ucvlatin/nsMUTF7ToUnicode.h

@@ -20,18 +20,18 @@
 #ifndef nsMUTF7ToUnicode_h___
 #define nsMUTF7ToUnicode_h___
 
-#include "nsUCvLatinSupport.h"
+#include "nsUTF7ToUnicode.h"
 
 //----------------------------------------------------------------------
 // Class nsMUTF7ToUnicode [declaration]
 
 /**
- * A character set converter from MUTF7 to Unicode.
+ * A character set converter from Modified UTF7 to Unicode.
  *
  * @created         18/May/1999
  * @author  Catalin Rotaru [CATA]
  */
-class nsMUTF7ToUnicode : public nsOneByteDecoderSupport
+class nsMUTF7ToUnicode : public nsBasicUTF7Decoder 
 {
 public:
 

intl/uconv/ucvlatin/nsUCvLatinCID.h

@@ -198,6 +198,11 @@ NS_DECLARE_ID(kVPSToUnicodeCID,
 NS_DECLARE_ID(kUTF8ToUnicodeCID, 
   0x5534ddc0, 0xdd96, 0x11d2, 0x8a, 0xac, 0x0, 0x60, 0x8, 0x11, 0xa8, 0x36);
 
+// Class ID for our UTF7ToUnicode charset converter
+// {77CFAAF1-1CF4-11d3-8AAF-00600811A836}
+NS_DECLARE_ID(kUTF7ToUnicodeCID, 
+  0x77cfaaf1, 0x1cf4, 0x11d3, 0x8a, 0xaf, 0x0, 0x60, 0x8, 0x11, 0xa8, 0x36);
+
 // Class ID for our MUTF7ToUnicode charset converter
 // {B57F97C1-0D70-11d3-8AAE-00600811A836}
 NS_DECLARE_ID(kMUTF7ToUnicodeCID, 
@@ -377,6 +382,11 @@ NS_DECLARE_ID(kUnicodeToVPSCID,
 NS_DECLARE_ID(kUnicodeToUTF8CID, 
   0x7c657d18, 0xec5e, 0x11d2, 0x8a, 0xac, 0x0, 0x60, 0x8, 0x11, 0xa8, 0x36);
 
+// Class ID for our UnicodeToUTF7 charset converter
+// {77CFAAF2-1CF4-11d3-8AAF-00600811A836}
+NS_DECLARE_ID(kUnicodeToUTF7CID, 
+  0x77cfaaf2, 0x1cf4, 0x11d3, 0x8a, 0xaf, 0x0, 0x60, 0x8, 0x11, 0xa8, 0x36);
+
 // Class ID for our UnicodeToMUTF7 charset converter
 // {B57F97C2-0D70-11d3-8AAE-00600811A836}
 NS_DECLARE_ID(kUnicodeToMUTF7CID, 

intl/uconv/ucvlatin/nsUCvLatinDll.cpp

@@ -62,6 +62,7 @@
 #include "nsVISCIIToUnicode.h"
 #include "nsVPSToUnicode.h"
 #include "nsUTF8ToUnicode.h"
+#include "nsUTF7ToUnicode.h"
 #include "nsMUTF7ToUnicode.h"
 #include "nsUnicodeToISO88591.h"
 #include "nsUnicodeToISO88592.h"
@@ -98,6 +99,7 @@
 #include "nsUnicodeToVISCII.h"
 #include "nsUnicodeToVPS.h"
 #include "nsUnicodeToUTF8.h"
+#include "nsUnicodeToUTF7.h"
 #include "nsUnicodeToMUTF7.h"
 
 // just for NS_IMPL_IDS; this is a good, central place to implement GUIDs
@@ -340,10 +342,16 @@ FactoryData g_FactoryData[] =
     "UTF-8",
     "Unicode"
   },
+  {
+    &kUTF7ToUnicodeCID,
+    nsUTF7ToUnicode::CreateInstance,
+    "UTF-7",
+    "Unicode"
+  },
   {
     &kMUTF7ToUnicodeCID,
     nsMUTF7ToUnicode::CreateInstance,
-    "MUTF-7",
+    "x-imap4-modified-utf7",
     "Unicode"
   },
   {
@@ -556,11 +564,17 @@ FactoryData g_FactoryData[] =
     "Unicode",
     "UTF-8"
   },
+  {
+    &kUnicodeToUTF7CID,
+    nsUnicodeToUTF7::CreateInstance,
+    "Unicode",
+    "UTF-7"
+  },
   {
     &kUnicodeToMUTF7CID,
     nsUnicodeToMUTF7::CreateInstance,
     "Unicode",
-    "MUTF-7"
+    "x-imap4-modified-utf7"
   }
 };
 

intl/uconv/ucvlatin/nsUTF7ToUnicode.cpp

@@ -0,0 +1,259 @@
+/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*-
+ *
+ * The contents of this file are subject to the Netscape Public License
+ * Version 1.0 (the "License"); you may not use this file except in
+ * compliance with the License.  You may obtain a copy of the License at
+ * http://www.mozilla.org/NPL/
+ *
+ * Software distributed under the License is distributed on an "AS IS"
+ * basis, WITHOUT WARRANTY OF ANY KIND, either express or implied.  See
+ * the License for the specific language governing rights and limitations
+ * under the License.
+ *
+ * The Original Code is Mozilla Communicator client code.
+ *
+ * The Initial Developer of the Original Code is Netscape Communications
+ * Corporation.  Portions created by Netscape are Copyright (C) 1998
+ * Netscape Communications Corporation.  All Rights Reserved.
+ */
+
+#include "nsUTF7ToUnicode.h"
+
+#define ENC_DIRECT      0
+#define ENC_BASE64      1
+
+//----------------------------------------------------------------------
+// Class nsBasicUTF7Decoder [implementation]
+
+nsBasicUTF7Decoder::nsBasicUTF7Decoder(char aLastChar, char aEscChar) 
+: nsBufferDecoderSupport()
+{
+  mLastChar = aLastChar;
+  mEscChar = aEscChar;
+  Reset();
+}
+
+nsresult nsBasicUTF7Decoder::DecodeDirect(
+                             const char * aSrc, 
+                             PRInt32 * aSrcLength, 
+                             PRUnichar * aDest, 
+                             PRInt32 * aDestLength)
+{
+  const char * srcEnd = aSrc + *aSrcLength;
+  const char * src = aSrc;
+  PRUnichar * destEnd = aDest + *aDestLength;
+  PRUnichar * dest = aDest;
+  nsresult res = NS_OK;
+  char ch;
+
+  while (src < srcEnd) {
+    ch = *src;
+
+    // stop when we meet other chars or end of direct encoded seq.
+    if ((ch < 0x20) || (ch > 0x7e) || (ch == mEscChar)) {
+      res = NS_ERROR_UDEC_ILLEGALINPUT;
+      break;
+    }
+
+    if (dest >= destEnd) {
+      res = NS_OK_UDEC_MOREOUTPUT;
+      break;
+    } else {
+      *dest++ = ch;
+      src++;
+    }
+  }
+
+  *aSrcLength = src - aSrc;
+  *aDestLength = dest - aDest;
+  return res;
+}
+
+nsresult nsBasicUTF7Decoder::DecodeBase64(
+                             const char * aSrc, 
+                             PRInt32 * aSrcLength, 
+                             PRUnichar * aDest, 
+                             PRInt32 * aDestLength)
+{
+  const char * srcEnd = aSrc + *aSrcLength;
+  const char * src = aSrc;
+  PRUnichar * destEnd = aDest + *aDestLength;
+  PRUnichar * dest = aDest;
+  nsresult res = NS_OK;
+  char ch;
+  PRUint32 value;
+
+  while (src < srcEnd) {
+    ch = *src;
+
+    // stop when we meet other chars or end of direct encoded seq.
+    value = CharToValue(ch);
+    if (value == -1) {
+      res = NS_ERROR_UDEC_ILLEGALINPUT;
+      break;
+    }
+
+    switch (mEncStep) {
+      case 0:
+        mEncBits = value << 10;
+        break;
+      case 1:
+        mEncBits += value << 4;
+        break;
+      case 2:
+        if (dest >= destEnd) {
+          res = NS_OK_UDEC_MOREOUTPUT;
+          break;
+        }
+        mEncBits += value >> 2;
+        *(dest++) = (PRUnichar) mEncBits;
+        mEncBits = (value & 0x03) << 14;
+        break;
+      case 3:
+        mEncBits += value << 8;
+        break;
+      case 4:
+        mEncBits += value << 2;
+        break;
+      case 5:
+        if (dest >= destEnd) {
+          res = NS_OK_UDEC_MOREOUTPUT;
+          break;
+        }
+        mEncBits += value >> 4;
+        *(dest++) = (PRUnichar) mEncBits;
+        mEncBits = (value & 0x03) << 12;
+        break;
+      case 6:
+        mEncBits += value << 6;
+        break;
+      case 7:
+        if (dest >= destEnd) {
+          res = NS_OK_UDEC_MOREOUTPUT;
+          break;
+        }
+        mEncBits += value;
+        *(dest++) = (PRUnichar) mEncBits;
+        mEncBits = 0;
+        break;
+    }
+
+    if (res != NS_OK) break;
+
+    src++;
+    (++mEncStep)%=8;
+  }
+
+  *aSrcLength = src - aSrc;
+  *aDestLength = dest - aDest;
+  return res;
+}
+
+PRUint32 nsBasicUTF7Decoder::CharToValue(char aChar) {
+  if ((aChar>='A')&&(aChar<='Z'))
+    return (PRUint8)(aChar-'A');
+  else if ((aChar>='a')&&(aChar<='z'))
+    return (PRUint8)(26+aChar-'a');
+  else if ((aChar>='0')&&(aChar<='9'))
+    return (PRUint8)(26+26+aChar-'0');
+  else if (aChar=='+')
+    return (PRUint8)(26+26+10);
+  else if (aChar=='/')
+    return (PRUint8)(26+26+10+1);
+  else
+    return -1;
+}
+
+//----------------------------------------------------------------------
+// Subclassing of nsBufferDecoderSupport class [implementation]
+
+NS_IMETHODIMP nsBasicUTF7Decoder::ConvertNoBuff(const char * aSrc, 
+                                                PRInt32 * aSrcLength, 
+                                                PRUnichar * aDest, 
+                                                PRInt32 * aDestLength)
+{
+  const char * srcEnd = aSrc + *aSrcLength;
+  const char * src = aSrc;
+  PRUnichar * destEnd = aDest + *aDestLength;
+  PRUnichar * dest = aDest;
+  PRInt32 bcr,bcw;
+  nsresult res = NS_OK;
+  char ch;
+
+  while (src < srcEnd) {
+    ch = *src;
+
+    // fist, attept to decode in the current mode
+    bcr = srcEnd - src;
+    bcw = destEnd - dest;
+    if (mEncoding == ENC_DIRECT) 
+      res = DecodeDirect(src, &bcr, dest, &bcw);
+    else if ((mFreshBase64) && (*src == '-')) {
+      *dest = mEscChar;
+      bcr = 0;
+      bcw = 1;
+      res = NS_ERROR_UDEC_ILLEGALINPUT;
+    } else {
+      mFreshBase64 = PR_FALSE;
+      res = DecodeBase64(src, &bcr, dest, &bcw);
+    }
+    src += bcr;
+    dest += bcw;
+
+    // if an illegal char was encountered, test if it is an escape seq.
+    if (res == NS_ERROR_UDEC_ILLEGALINPUT) {
+      if (mEncoding == ENC_DIRECT) {
+        if (*src == mEscChar) {
+          mEncoding = ENC_BASE64;
+          mFreshBase64 = PR_TRUE;
+          src++;
+          res = NS_OK;
+        } else break;
+      } else {
+        if (*src == '-') {
+          mEncoding = ENC_DIRECT;
+          src++;
+          res = NS_OK;
+        } else break;
+      }
+    } else if (res != NS_OK) break;
+  }
+
+  *aSrcLength = src - aSrc;
+  *aDestLength = dest - aDest;
+  return res;
+}
+
+NS_IMETHODIMP nsBasicUTF7Decoder::GetMaxLength(const char * aSrc, 
+                                               PRInt32 aSrcLength, 
+                                               PRInt32 * aDestLength)
+{
+  // worst case
+  *aDestLength = aSrcLength;
+  return NS_OK;
+}
+
+NS_IMETHODIMP nsBasicUTF7Decoder::Reset()
+{
+  mEncoding = ENC_DIRECT;
+  mEncBits = 0;
+  mEncStep = 0;
+  return nsBufferDecoderSupport::Reset();
+}
+
+//----------------------------------------------------------------------
+// Class nsUTF7ToUnicode [implementation]
+
+// XXX Ok, you know that this is too close to the M-UTF-8. 
+// You need to change it according to the spec.
+
+nsUTF7ToUnicode::nsUTF7ToUnicode() 
+: nsBasicUTF7Decoder('/', '+')
+{
+}
+
+nsresult nsUTF7ToUnicode::CreateInstance(nsISupports ** aResult) 
+{
+  *aResult = new nsUTF7ToUnicode();
+  return (*aResult == NULL)? NS_ERROR_OUT_OF_MEMORY : NS_OK;
+}

intl/uconv/ucvlatin/nsUTF7ToUnicode.h

@@ -0,0 +1,92 @@
+/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*-
+ *
+ * The contents of this file are subject to the Netscape Public License
+ * Version 1.0 (the "License"); you may not use this file except in
+ * compliance with the License.  You may obtain a copy of the License at
+ * http://www.mozilla.org/NPL/
+ *
+ * Software distributed under the License is distributed on an "AS IS"
+ * basis, WITHOUT WARRANTY OF ANY KIND, either express or implied.  See
+ * the License for the specific language governing rights and limitations
+ * under the License.
+ *
+ * The Original Code is Mozilla Communicator client code.
+ *
+ * The Initial Developer of the Original Code is Netscape Communications
+ * Corporation.  Portions created by Netscape are Copyright (C) 1998
+ * Netscape Communications Corporation.  All Rights Reserved.
+ */
+
+#ifndef nsUTF7ToUnicode_h___
+#define nsUTF7ToUnicode_h___
+
+#include "nsUCvLatinSupport.h"
+
+//----------------------------------------------------------------------
+// Class nsBasicUTF7Decoder [declaration]
+
+/**
+ * Basic class for a character set converter from UTF-7 to Unicode.
+ *
+ * @created         03/Jun/1999
+ * @author  Catalin Rotaru [CATA]
+ */
+class nsBasicUTF7Decoder : public nsBufferDecoderSupport
+{
+public:
+
+  /**
+   * Class constructor.
+   */
+  nsBasicUTF7Decoder(char aLastChar, char aEscChar);
+
+protected:
+
+  PRInt32                   mEncoding;      // current encoding
+  PRUint32                  mEncBits;
+  PRInt32                   mEncStep;
+  char                      mLastChar;
+  char                      mEscChar;
+  PRBool                    mFreshBase64;
+
+  nsresult DecodeDirect(const char * aSrc, PRInt32 * aSrcLength, 
+      PRUnichar * aDest, PRInt32 * aDestLength);
+  nsresult DecodeBase64(const char * aSrc, PRInt32 * aSrcLength, 
+      PRUnichar * aDest, PRInt32 * aDestLength);
+  PRUint32 CharToValue(char aChar);
+
+  //--------------------------------------------------------------------
+  // Subclassing of nsBufferDecoderSupport class [declaration]
+
+  NS_IMETHOD ConvertNoBuff(const char * aSrc, PRInt32 * aSrcLength, 
+      PRUnichar * aDest, PRInt32 * aDestLength);
+  NS_IMETHOD GetMaxLength(const char * aSrc, PRInt32 aSrcLength, 
+      PRInt32 * aDestLength);
+  NS_IMETHOD Reset();
+};
+
+//----------------------------------------------------------------------
+// Class nsUTF7ToUnicode [declaration]
+
+/**
+ * A character set converter from Modified UTF7 to Unicode.
+ *
+ * @created         18/May/1999
+ * @author  Catalin Rotaru [CATA]
+ */
+class nsUTF7ToUnicode : public nsBasicUTF7Decoder 
+{
+public:
+
+  /**
+   * Class constructor.
+   */
+  nsUTF7ToUnicode();
+
+  /**
+   * Static class constructor.
+   */
+  static nsresult CreateInstance(nsISupports **aResult);
+};
+
+#endif /* nsUTF7ToUnicode_h___ */

intl/uconv/ucvlatin/nsUnicodeToMUTF7.cpp

@@ -19,24 +19,11 @@
 
 #include "nsUnicodeToMUTF7.h"
 
-//----------------------------------------------------------------------
-// Global functions and data [declaration]
-
-static PRUint16 g_ufMappingTable[] = {
-#include "8859-1.uf"
-};
-
-static PRInt16 g_ufShiftTable[] =  {
-  0, u1ByteCharset ,
-  ShiftCell(0,0,0,0,0,0,0,0)
-};
-
 //----------------------------------------------------------------------
 // Class nsUnicodeToMUTF7 [implementation]
 
 nsUnicodeToMUTF7::nsUnicodeToMUTF7() 
-: nsTableEncoderSupport((uShiftTable*) &g_ufShiftTable, 
+: nsBasicUTF7Encoder(',', '&')
-                        (uMappingTable*) &g_ufMappingTable)
 {
 }
 
@@ -45,14 +32,3 @@ nsresult nsUnicodeToMUTF7::CreateInstance(nsISupports ** aResult)
   *aResult = new nsUnicodeToMUTF7();
   return (*aResult == NULL)? NS_ERROR_OUT_OF_MEMORY : NS_OK;
 }
-
-//----------------------------------------------------------------------
-// Subclassing of nsTableEncoderSupport class [implementation]
-
-NS_IMETHODIMP nsUnicodeToMUTF7::GetMaxLength(const PRUnichar * aSrc, 
-                                             PRInt32 aSrcLength,
-                                             PRInt32 * aDestLength)
-{
-  *aDestLength = aSrcLength;
-  return NS_OK_UENC_EXACTLENGTH;
-}

intl/uconv/ucvlatin/nsUnicodeToMUTF7.h

@@ -20,18 +20,18 @@
 #ifndef nsUnicodeToMUTF7_h___
 #define nsUnicodeToMUTF7_h___
 
-#include "nsUCvLatinSupport.h"
+#include "nsUnicodeToUTF7.h"
 
 //----------------------------------------------------------------------
 // Class nsUnicodeToMUTF7 [declaration]
 
 /**
- * A character set converter from Unicode to MUTF7.
+ * A character set converter from Unicode to Modified UTF-7.
  *
  * @created         18/May/1999
  * @author  Catalin Rotaru [CATA]
  */
-class nsUnicodeToMUTF7 : public nsTableEncoderSupport
+class nsUnicodeToMUTF7 : public nsBasicUTF7Encoder
 {
 public:
 
@@ -44,14 +44,6 @@ public:
    * Static class constructor.
    */
   static nsresult CreateInstance(nsISupports **aResult);
-
-protected:
-
-  //--------------------------------------------------------------------
-  // Subclassing of nsEncoderSupport class [declaration]
-
-  NS_IMETHOD GetMaxLength(const PRUnichar * aSrc, PRInt32 aSrcLength, 
-      PRInt32 * aDestLength);
 };
 
 #endif /* nsUnicodeToMUTF7_h___ */

intl/uconv/ucvlatin/nsUnicodeToUTF7.cpp

@@ -0,0 +1,295 @@
+/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*-
+ *
+ * The contents of this file are subject to the Netscape Public License
+ * Version 1.0 (the "License"); you may not use this file except in
+ * compliance with the License.  You may obtain a copy of the License at
+ * http://www.mozilla.org/NPL/
+ *
+ * Software distributed under the License is distributed on an "AS IS"
+ * basis, WITHOUT WARRANTY OF ANY KIND, either express or implied.  See
+ * the License for the specific language governing rights and limitations
+ * under the License.
+ *
+ * The Original Code is Mozilla Communicator client code.
+ *
+ * The Initial Developer of the Original Code is Netscape Communications
+ * Corporation.  Portions created by Netscape are Copyright (C) 1998
+ * Netscape Communications Corporation.  All Rights Reserved.
+ */
+
+#include "nsUnicodeToUTF7.h"
+
+//----------------------------------------------------------------------
+// Global functions and data [declaration]
+
+#define ENC_DIRECT      0
+#define ENC_BASE64      1
+
+//----------------------------------------------------------------------
+// Class nsBasicUTF7Encoder [implementation]
+
+nsBasicUTF7Encoder::nsBasicUTF7Encoder(char aLastChar, char aEscChar) 
+: nsEncoderSupport()
+{
+  mLastChar = aLastChar;
+  mEscChar = aEscChar;
+  Reset();
+}
+
+nsresult nsBasicUTF7Encoder::ShiftEncoding(PRInt32 aEncoding,
+                                          char * aDest, 
+                                          PRInt32 * aDestLength)
+{
+  if (aEncoding == mEncoding) {
+    *aDestLength = 0;
+    return NS_OK;
+  } 
+
+  nsresult res = NS_OK;
+  char * dest = aDest;
+  char * destEnd = aDest + *aDestLength;
+
+  if (mEncStep != 0) {
+    if (dest >= destEnd) return NS_OK_UENC_MOREOUTPUT;
+    *(dest++)=ValueToChar(mEncBits);
+  }
+
+  if (dest >= destEnd) {
+    res = NS_OK_UENC_MOREOUTPUT;
+  } else {
+    switch (aEncoding) {
+      case 0:
+        *(dest++) = '-';
+        break;
+      case 1:
+        *(dest++) = mEscChar;
+        break;
+    }
+    mEncoding = aEncoding;
+  }
+
+  *aDestLength  = dest - aDest;
+  return res;
+}
+
+nsresult nsBasicUTF7Encoder::EncodeDirect(
+                            const PRUnichar * aSrc, 
+                            PRInt32 * aSrcLength, 
+                            char * aDest, 
+                            PRInt32 * aDestLength)
+{
+  nsresult res = NS_OK;
+  const PRUnichar * src = aSrc;
+  const PRUnichar * srcEnd = aSrc + *aSrcLength;
+  char * dest = aDest;
+  char * destEnd = aDest + *aDestLength;
+  PRUnichar ch;
+
+  while (src < srcEnd) {
+    ch = *src;
+
+    // stop when we reach Unicode chars
+    if ((ch < 0x20) || (ch > 0x7e)) break;
+
+    if (ch == mEscChar) {
+      // special case for the escape char
+      if (destEnd - dest < 1) {
+        res = NS_OK_UENC_MOREOUTPUT;
+        break;
+      } else {
+        *dest++ = (char)ch;
+        *dest++ = (char)'-';
+        src++;
+      }
+    } else {
+      //classic direct encoding
+      if (dest >= destEnd) {
+        res = NS_OK_UENC_MOREOUTPUT;
+        break;
+      } else {
+        *dest++ = (char)ch;
+        src++;
+      }
+    }
+  }
+
+  *aSrcLength = src - aSrc;
+  *aDestLength  = dest - aDest;
+  return res;
+}
+
+nsresult nsBasicUTF7Encoder::EncodeBase64(
+                             const PRUnichar * aSrc, 
+                             PRInt32 * aSrcLength, 
+                             char * aDest, 
+                             PRInt32 * aDestLength)
+{
+  nsresult res = NS_OK;
+  const PRUnichar * src = aSrc;
+  const PRUnichar * srcEnd = aSrc + *aSrcLength;
+  char * dest = aDest;
+  char * destEnd = aDest + *aDestLength;
+  PRUnichar ch;
+  PRUint32 value;
+
+  while (src < srcEnd) {
+    ch = *src;
+
+    // stop when we reach printable US-ASCII chars
+    if ((ch >= 0x20) && (ch <= 0x7e)) break;
+
+    switch (mEncStep) {
+      case 0:
+        if (destEnd - dest < 2) {
+          res = NS_OK_UENC_MOREOUTPUT;
+          break;
+        }
+        value=ch>>10;
+        *(dest++)=ValueToChar(value);
+        value=(ch>>4)&0x3f;
+        *(dest++)=ValueToChar(value);
+        mEncBits=(ch&0x0f)<<2;
+        break;
+      case 1:
+        if (destEnd - dest < 3) {
+          res = NS_OK_UENC_MOREOUTPUT;
+          break;
+        }
+        value=mEncBits+(ch>>14);
+        *(dest++)=ValueToChar(value);
+        value=(ch>>8)&0x3f;
+        *(dest++)=ValueToChar(value);
+        value=(ch>>2)&0x3f;
+        *(dest++)=ValueToChar(value);
+        mEncBits=(ch&0x03)<<4;
+        break;
+      case 2:
+        if (destEnd - dest < 3) {
+          res = NS_OK_UENC_MOREOUTPUT;
+          break;
+        }
+        value=mEncBits+(ch>>12);
+        *(dest++)=ValueToChar(value);
+        value=(ch>>6)&0x3f;
+        *(dest++)=ValueToChar(value);
+        value=ch&0x3f;
+        *(dest++)=ValueToChar(value);
+        mEncBits=0;
+        break;
+    }
+
+    if (res != NS_OK) break;
+
+    src++;
+    (++mEncStep)%=3;
+  }
+
+  *aSrcLength = src - aSrc;
+  *aDestLength  = dest - aDest;
+  return res;
+}
+
+char nsBasicUTF7Encoder::ValueToChar(PRUint32 aValue) { 
+  if (aValue < 26) 
+    return (char)('A'+aValue);
+  else if (aValue < 26 + 26) 
+    return (char)('a' + aValue - 26);
+  else if (aValue < 26 + 26 + 10)
+    return (char)('0' + aValue - 26 - 26);
+  else if (aValue == 26 + 26 + 10)
+    return '+';
+  else if (aValue == 26 + 26 + 10 + 1)
+    return mLastChar;
+  else
+    return -1;
+}
+
+//----------------------------------------------------------------------
+// Subclassing of nsEncoderSupport class [implementation]
+
+NS_IMETHODIMP nsBasicUTF7Encoder::ConvertNoBuffNoErr(
+                                  const PRUnichar * aSrc, 
+                                  PRInt32 * aSrcLength, 
+                                  char * aDest, 
+                                  PRInt32 * aDestLength)
+{
+  nsresult res = NS_OK;
+  const PRUnichar * src = aSrc;
+  const PRUnichar * srcEnd = aSrc + *aSrcLength;
+  char * dest = aDest;
+  char * destEnd = aDest + *aDestLength;
+  PRInt32 bcr,bcw;
+  PRUnichar ch;
+  PRInt32 enc;
+
+  while (src < srcEnd) {
+    // find the encoding for the next char
+    ch = *src;
+    if ((ch >= 0x20) && (ch <= 0x7e)) 
+      enc = ENC_DIRECT;
+    else
+      enc = ENC_BASE64;
+
+    // if necessary, shift into the required encoding
+    bcw = destEnd - dest;
+    res = ShiftEncoding(enc, dest, &bcw);
+    dest += bcw;
+    if (res != NS_OK) break;
+
+    // now encode (as much as you can)
+    bcr = srcEnd - src;
+    bcw = destEnd - dest;
+    if (enc == ENC_DIRECT) 
+      res = EncodeDirect(src, &bcr, dest, &bcw);
+    else 
+      res = EncodeBase64(src, &bcr, dest, &bcw);
+    src += bcr;
+    dest += bcw;
+
+    if (res != NS_OK) break;
+  }
+
+  *aSrcLength = src - aSrc;
+  *aDestLength  = dest - aDest;
+  return res;
+}
+
+NS_IMETHODIMP nsBasicUTF7Encoder::FinishNoBuff(char * aDest, 
+                                               PRInt32 * aDestLength)
+{
+  return ShiftEncoding(ENC_DIRECT, aDest, aDestLength);
+}
+
+NS_IMETHODIMP nsBasicUTF7Encoder::GetMaxLength(const PRUnichar * aSrc, 
+                                               PRInt32 aSrcLength,
+                                               PRInt32 * aDestLength)
+{
+  // worst case
+  *aDestLength = 5*aSrcLength;
+  return NS_OK;
+}
+
+NS_IMETHODIMP nsBasicUTF7Encoder::Reset()
+{
+  mEncoding = ENC_DIRECT;
+  mEncBits = 0;
+  mEncStep = 0;
+  return nsEncoderSupport::Reset();
+}
+
+//----------------------------------------------------------------------
+// Class nsUnicodeToUTF7 [implementation]
+
+// XXX Ok, you know that this is too close to the M-UTF-8. 
+// You need to change it according to the spec.
+
+nsUnicodeToUTF7::nsUnicodeToUTF7() 
+: nsBasicUTF7Encoder('/', '+')
+{
+}
+
+nsresult nsUnicodeToUTF7::CreateInstance(nsISupports ** aResult) 
+{
+  *aResult = new nsUnicodeToUTF7();
+  return (*aResult == NULL)? NS_ERROR_OUT_OF_MEMORY : NS_OK;
+}

intl/uconv/ucvlatin/nsUnicodeToUTF7.h

@@ -0,0 +1,94 @@
+/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*-
+ *
+ * The contents of this file are subject to the Netscape Public License
+ * Version 1.0 (the "License"); you may not use this file except in
+ * compliance with the License.  You may obtain a copy of the License at
+ * http://www.mozilla.org/NPL/
+ *
+ * Software distributed under the License is distributed on an "AS IS"
+ * basis, WITHOUT WARRANTY OF ANY KIND, either express or implied.  See
+ * the License for the specific language governing rights and limitations
+ * under the License.
+ *
+ * The Original Code is Mozilla Communicator client code.
+ *
+ * The Initial Developer of the Original Code is Netscape Communications
+ * Corporation.  Portions created by Netscape are Copyright (C) 1998
+ * Netscape Communications Corporation.  All Rights Reserved.
+ */
+
+#ifndef nsUnicodeToUTF7_h___
+#define nsUnicodeToUTF7_h___
+
+#include "nsUCvLatinSupport.h"
+
+//----------------------------------------------------------------------
+// Class nsBasicUTF7Encoder [declaration]
+
+/**
+ * Basic class for a character set converter from Unicode to UTF-7.
+ *
+ * @created         03/Jun/1999
+ * @author  Catalin Rotaru [CATA]
+ */
+class nsBasicUTF7Encoder : public nsEncoderSupport
+{
+public:
+
+  /**
+   * Class constructor.
+   */
+  nsBasicUTF7Encoder(char aLastChar, char aEscChar);
+
+protected:
+
+  PRInt32                   mEncoding;      // current encoding
+  PRUint32                  mEncBits;
+  PRInt32                   mEncStep;
+  char                      mLastChar;
+  char                      mEscChar;
+
+  nsresult ShiftEncoding(PRInt32 aEncoding, char * aDest, 
+      PRInt32 * aDestLength);
+  nsresult EncodeDirect(const PRUnichar * aSrc, PRInt32 * aSrcLength, 
+      char * aDest, PRInt32 * aDestLength);
+  nsresult EncodeBase64(const PRUnichar * aSrc, PRInt32 * aSrcLength, 
+      char * aDest, PRInt32 * aDestLength);
+  char ValueToChar(PRUint32 aValue);
+
+  //--------------------------------------------------------------------
+  // Subclassing of nsEncoderSupport class [declaration]
+
+  NS_IMETHOD ConvertNoBuffNoErr(const PRUnichar * aSrc, PRInt32 * aSrcLength, 
+      char * aDest, PRInt32 * aDestLength);
+  NS_IMETHOD FinishNoBuff(char * aDest, PRInt32 * aDestLength);
+  NS_IMETHOD GetMaxLength(const PRUnichar * aSrc, PRInt32 aSrcLength, 
+      PRInt32 * aDestLength);
+  NS_IMETHOD Reset();
+};
+
+//----------------------------------------------------------------------
+// Class nsUnicodeToUTF7 [declaration]
+
+/**
+ * A character set converter from Unicode to UTF-7.
+ *
+ * @created         03/Jun/1999
+ * @author  Catalin Rotaru [CATA]
+ */
+class nsUnicodeToUTF7 : public nsBasicUTF7Encoder
+{
+public:
+
+  /**
+   * Class constructor.
+   */
+  nsUnicodeToUTF7();
+
+  /**
+   * Static class constructor.
+   */
+  static nsresult CreateInstance(nsISupports **aResult);
+};
+
+#endif /* nsUnicodeToUTF7_h___ */