[188] | 1 | /*
|
---|
| 2 | * The Apache Software License, Version 1.1
|
---|
| 3 | *
|
---|
| 4 | * Copyright (c) 1999-2000 The Apache Software Foundation. All rights
|
---|
| 5 | * reserved.
|
---|
| 6 | *
|
---|
| 7 | * Redistribution and use in source and binary forms, with or without
|
---|
| 8 | * modification, are permitted provided that the following conditions
|
---|
| 9 | * are met:
|
---|
| 10 | *
|
---|
| 11 | * 1. Redistributions of source code must retain the above copyright
|
---|
| 12 | * notice, this list of conditions and the following disclaimer.
|
---|
| 13 | *
|
---|
| 14 | * 2. Redistributions in binary form must reproduce the above copyright
|
---|
| 15 | * notice, this list of conditions and the following disclaimer in
|
---|
| 16 | * the documentation and/or other materials provided with the
|
---|
| 17 | * distribution.
|
---|
| 18 | *
|
---|
| 19 | * 3. The end-user documentation included with the redistribution,
|
---|
| 20 | * if any, must include the following acknowledgment:
|
---|
| 21 | * "This product includes software developed by the
|
---|
| 22 | * Apache Software Foundation (http://www.apache.org/)."
|
---|
| 23 | * Alternately, this acknowledgment may appear in the software itself,
|
---|
| 24 | * if and wherever such third-party acknowledgments normally appear.
|
---|
| 25 | *
|
---|
| 26 | * 4. The names "Xerces" and "Apache Software Foundation" must
|
---|
| 27 | * not be used to endorse or promote products derived from this
|
---|
| 28 | * software without prior written permission. For written
|
---|
| 29 | * permission, please contact apache\@apache.org.
|
---|
| 30 | *
|
---|
| 31 | * 5. Products derived from this software may not be called "Apache",
|
---|
| 32 | * nor may "Apache" appear in their name, without prior written
|
---|
| 33 | * permission of the Apache Software Foundation.
|
---|
| 34 | *
|
---|
| 35 | * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
|
---|
| 36 | * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
---|
| 37 | * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
---|
| 38 | * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
|
---|
| 39 | * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
---|
| 40 | * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
---|
| 41 | * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
|
---|
| 42 | * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
|
---|
| 43 | * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
---|
| 44 | * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
|
---|
| 45 | * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
---|
| 46 | * SUCH DAMAGE.
|
---|
| 47 | * ====================================================================
|
---|
| 48 | *
|
---|
| 49 | * This software consists of voluntary contributions made by many
|
---|
| 50 | * individuals on behalf of the Apache Software Foundation, and was
|
---|
| 51 | * originally based on software copyright (c) 1999, International
|
---|
| 52 | * Business Machines, Inc., http://www.ibm.com . For more information
|
---|
| 53 | * on the Apache Software Foundation, please see
|
---|
| 54 | * <http://www.apache.org/>.
|
---|
| 55 | */
|
---|
| 56 |
|
---|
| 57 | /*
|
---|
| 58 | * $Id: XMLRecognizer.hpp,v 1.6 2004/01/29 11:46:29 cargilld Exp $
|
---|
| 59 | */
|
---|
| 60 |
|
---|
| 61 | #if !defined(XMLRECOGNIZER_HPP)
|
---|
| 62 | #define XMLRECOGNIZER_HPP
|
---|
| 63 |
|
---|
| 64 | #include <xercesc/util/XercesDefs.hpp>
|
---|
| 65 | #include <xercesc/util/PlatformUtils.hpp>
|
---|
| 66 |
|
---|
| 67 | XERCES_CPP_NAMESPACE_BEGIN
|
---|
| 68 |
|
---|
| 69 | /**
|
---|
| 70 | * This class provides some simple code to recognize the encodings of
|
---|
| 71 | * XML files. This recognition only does very basic sensing of the encoding
|
---|
| 72 | * in a broad sense. Basically its just enough to let us get started and
|
---|
| 73 | * read the XMLDecl line. The scanner, once it reads the XMLDecl, will
|
---|
| 74 | * tell the reader any actual encoding string it found and the reader can
|
---|
| 75 | * update itself to be more specific at that point.
|
---|
| 76 | */
|
---|
| 77 | class XMLPARSER_EXPORT XMLRecognizer
|
---|
| 78 | {
|
---|
| 79 | public :
|
---|
| 80 | // -----------------------------------------------------------------------
|
---|
| 81 | // Class types
|
---|
| 82 | //
|
---|
| 83 | // This enum represents the various encoding families that we have to
|
---|
| 84 | // deal with individually at the scanner level. This does not indicate
|
---|
| 85 | // the exact encoding, just the rough family that would let us scan
|
---|
| 86 | // the XML/TextDecl to find the encoding string.
|
---|
| 87 | //
|
---|
| 88 | // The 'L's and 'B's stand for little or big endian. We conditionally
|
---|
| 89 | // create versions that will automatically map to the local UTF-16 and
|
---|
| 90 | // UCS-4 endian modes.
|
---|
| 91 | //
|
---|
| 92 | // OtherEncoding means that its some transcoder based encoding, i.e. not
|
---|
| 93 | // one of the ones that we do internally. Its a special case and should
|
---|
| 94 | // never be used directly outside of the reader.
|
---|
| 95 | //
|
---|
| 96 | // NOTE: Keep this in sync with the name map array in the Cpp file!!
|
---|
| 97 | // -----------------------------------------------------------------------
|
---|
| 98 | enum Encodings
|
---|
| 99 | {
|
---|
| 100 | EBCDIC = 0
|
---|
| 101 | , UCS_4B = 1
|
---|
| 102 | , UCS_4L = 2
|
---|
| 103 | , US_ASCII = 3
|
---|
| 104 | , UTF_8 = 4
|
---|
| 105 | , UTF_16B = 5
|
---|
| 106 | , UTF_16L = 6
|
---|
| 107 | , XERCES_XMLCH = 7
|
---|
| 108 |
|
---|
| 109 | , Encodings_Count
|
---|
| 110 | , Encodings_Min = EBCDIC
|
---|
| 111 | , Encodings_Max = XERCES_XMLCH
|
---|
| 112 |
|
---|
| 113 | , OtherEncoding = 999
|
---|
| 114 |
|
---|
| 115 | #if defined(ENDIANMODE_BIG)
|
---|
| 116 | , Def_UTF16 = UTF_16B
|
---|
| 117 | , Def_UCS4 = UCS_4B
|
---|
| 118 | #else
|
---|
| 119 | , Def_UTF16 = UTF_16L
|
---|
| 120 | , Def_UCS4 = UCS_4L
|
---|
| 121 | #endif
|
---|
| 122 | };
|
---|
| 123 |
|
---|
| 124 |
|
---|
| 125 | // -----------------------------------------------------------------------
|
---|
| 126 | // Public, const static data
|
---|
| 127 | //
|
---|
| 128 | // These are the byte sequences for each of the encodings that we can
|
---|
| 129 | // auto sense, and their lengths.
|
---|
| 130 | // -----------------------------------------------------------------------
|
---|
| 131 | static const char fgASCIIPre[];
|
---|
| 132 | static const unsigned int fgASCIIPreLen;
|
---|
| 133 | static const XMLByte fgEBCDICPre[];
|
---|
| 134 | static const unsigned int fgEBCDICPreLen;
|
---|
| 135 | static const XMLByte fgUTF16BPre[];
|
---|
| 136 | static const XMLByte fgUTF16LPre[];
|
---|
| 137 | static const unsigned int fgUTF16PreLen;
|
---|
| 138 | static const XMLByte fgUCS4BPre[];
|
---|
| 139 | static const XMLByte fgUCS4LPre[];
|
---|
| 140 | static const unsigned int fgUCS4PreLen;
|
---|
| 141 | static const char fgUTF8BOM[];
|
---|
| 142 | static const unsigned int fgUTF8BOMLen;
|
---|
| 143 |
|
---|
| 144 |
|
---|
| 145 | // -----------------------------------------------------------------------
|
---|
| 146 | // Encoding recognition methods
|
---|
| 147 | // -----------------------------------------------------------------------
|
---|
| 148 | static Encodings basicEncodingProbe
|
---|
| 149 | (
|
---|
| 150 | const XMLByte* const rawBuffer
|
---|
| 151 | , const unsigned int rawByteCount
|
---|
| 152 | );
|
---|
| 153 |
|
---|
| 154 | static Encodings encodingForName
|
---|
| 155 | (
|
---|
| 156 | const XMLCh* const theEncName
|
---|
| 157 | );
|
---|
| 158 |
|
---|
| 159 | static const XMLCh* nameForEncoding(const Encodings theEncoding
|
---|
| 160 | , MemoryManager* const manager = XMLPlatformUtils::fgMemoryManager);
|
---|
| 161 |
|
---|
| 162 |
|
---|
| 163 | protected :
|
---|
| 164 | // -----------------------------------------------------------------------
|
---|
| 165 | // Unimplemented constructors, operators, and destructor
|
---|
| 166 | //
|
---|
| 167 | // This class is effectively being used as a namespace for some static
|
---|
| 168 | // methods.
|
---|
| 169 | //
|
---|
| 170 | // (these functions are protected rather than private only to get rid of
|
---|
| 171 | // some annoying compiler warnings.)
|
---|
| 172 | //
|
---|
| 173 | // -----------------------------------------------------------------------
|
---|
| 174 | XMLRecognizer();
|
---|
| 175 | ~XMLRecognizer();
|
---|
| 176 |
|
---|
| 177 | private:
|
---|
| 178 | // -----------------------------------------------------------------------
|
---|
| 179 | // Unimplemented constructors and operators
|
---|
| 180 | // -----------------------------------------------------------------------
|
---|
| 181 | XMLRecognizer(const XMLRecognizer&);
|
---|
| 182 | XMLRecognizer& operator=(const XMLRecognizer&);
|
---|
| 183 | };
|
---|
| 184 |
|
---|
| 185 | XERCES_CPP_NAMESPACE_END
|
---|
| 186 |
|
---|
| 187 | #endif
|
---|