encoding.c

ReactOS是一些高手根据Windows XP的内核编写出的类XP。内核实现机理和API函数调用几乎相同。甚至可以兼容XP的程序。喜欢研究系统内核的人可以看一看。

	    return(XML_CHAR_ENCODING_UCS4_3412);
	if ((in[0] == 0x4C) && (in[1] == 0x6F) &&
	    (in[2] == 0xA7) && (in[3] == 0x94))
	    return(XML_CHAR_ENCODING_EBCDIC);
	if ((in[0] == 0x3C) && (in[1] == 0x3F) &&
	    (in[2] == 0x78) && (in[3] == 0x6D))
	    return(XML_CHAR_ENCODING_UTF8);
	/*
	 * Although not part of the recommendation, we also
	 * attempt an "auto-recognition" of UTF-16LE and
	 * UTF-16BE encodings.
	 */
	if ((in[0] == 0x3C) && (in[1] == 0x00) &&
	    (in[2] == 0x3F) && (in[3] == 0x00))
	    return(XML_CHAR_ENCODING_UTF16LE);
	if ((in[0] == 0x00) && (in[1] == 0x3C) &&
	    (in[2] == 0x00) && (in[3] == 0x3F))
	    return(XML_CHAR_ENCODING_UTF16BE);
    }
    if (len >= 3) {
	/*
	 * Errata on XML-1.0 June 20 2001
	 * We now allow an UTF8 encoded BOM
	 */
	if ((in[0] == 0xEF) && (in[1] == 0xBB) &&
	    (in[2] == 0xBF))
	    return(XML_CHAR_ENCODING_UTF8);
    }
    /* For UTF-16 we can recognize by the BOM */
    if (len >= 2) {
	if ((in[0] == 0xFE) && (in[1] == 0xFF))
	    return(XML_CHAR_ENCODING_UTF16BE);
	if ((in[0] == 0xFF) && (in[1] == 0xFE))
	    return(XML_CHAR_ENCODING_UTF16LE);
    }
    return(XML_CHAR_ENCODING_NONE);
}

/**
 * xmlCleanupEncodingAliases:
 *
 * Unregisters all aliases
 */
void
xmlCleanupEncodingAliases(void) {
    int i;

    if (xmlCharEncodingAliases == NULL)
	return;

    for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
	if (xmlCharEncodingAliases[i].name != NULL)
	    xmlFree((char *) xmlCharEncodingAliases[i].name);
	if (xmlCharEncodingAliases[i].alias != NULL)
	    xmlFree((char *) xmlCharEncodingAliases[i].alias);
    }
    xmlCharEncodingAliasesNb = 0;
    xmlCharEncodingAliasesMax = 0;
    xmlFree(xmlCharEncodingAliases);
    xmlCharEncodingAliases = NULL;
}

/**
 * xmlGetEncodingAlias:
 * @alias:  the alias name as parsed, in UTF-8 format (ASCII actually)
 *
 * Lookup an encoding name for the given alias.
 * 
 * Returns NULL if not found, otherwise the original name
 */
const char *
xmlGetEncodingAlias(const char *alias) {
    int i;
    char upper[100];

    if (alias == NULL)
	return(NULL);

    if (xmlCharEncodingAliases == NULL)
	return(NULL);

    for (i = 0;i < 99;i++) {
        upper[i] = toupper(alias[i]);
	if (upper[i] == 0) break;
    }
    upper[i] = 0;

    /*
     * Walk down the list looking for a definition of the alias
     */
    for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
	if (!strcmp(xmlCharEncodingAliases[i].alias, upper)) {
	    return(xmlCharEncodingAliases[i].name);
	}
    }
    return(NULL);
}

/**
 * xmlAddEncodingAlias:
 * @name:  the encoding name as parsed, in UTF-8 format (ASCII actually)
 * @alias:  the alias name as parsed, in UTF-8 format (ASCII actually)
 *
 * Registers an alias @alias for an encoding named @name. Existing alias
 * will be overwritten.
 * 
 * Returns 0 in case of success, -1 in case of error
 */
int
xmlAddEncodingAlias(const char *name, const char *alias) {
    int i;
    char upper[100];

    if ((name == NULL) || (alias == NULL))
	return(-1);

    for (i = 0;i < 99;i++) {
        upper[i] = toupper(alias[i]);
	if (upper[i] == 0) break;
    }
    upper[i] = 0;

    if (xmlCharEncodingAliases == NULL) {
	xmlCharEncodingAliasesNb = 0;
	xmlCharEncodingAliasesMax = 20;
	xmlCharEncodingAliases = (xmlCharEncodingAliasPtr) 
	      xmlMalloc(xmlCharEncodingAliasesMax * sizeof(xmlCharEncodingAlias));
	if (xmlCharEncodingAliases == NULL)
	    return(-1);
    } else if (xmlCharEncodingAliasesNb >= xmlCharEncodingAliasesMax) {
	xmlCharEncodingAliasesMax *= 2;
	xmlCharEncodingAliases = (xmlCharEncodingAliasPtr) 
	      xmlRealloc(xmlCharEncodingAliases,
		         xmlCharEncodingAliasesMax * sizeof(xmlCharEncodingAlias));
    }
    /*
     * Walk down the list looking for a definition of the alias
     */
    for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
	if (!strcmp(xmlCharEncodingAliases[i].alias, upper)) {
	    /*
	     * Replace the definition.
	     */
	    xmlFree((char *) xmlCharEncodingAliases[i].name);
	    xmlCharEncodingAliases[i].name = xmlMemStrdup(name);
	    return(0);
	}
    }
    /*
     * Add the definition
     */
    xmlCharEncodingAliases[xmlCharEncodingAliasesNb].name = xmlMemStrdup(name);
    xmlCharEncodingAliases[xmlCharEncodingAliasesNb].alias = xmlMemStrdup(upper);
    xmlCharEncodingAliasesNb++;
    return(0);
}

/**
 * xmlDelEncodingAlias:
 * @alias:  the alias name as parsed, in UTF-8 format (ASCII actually)
 *
 * Unregisters an encoding alias @alias
 * 
 * Returns 0 in case of success, -1 in case of error
 */
int
xmlDelEncodingAlias(const char *alias) {
    int i;

    if (alias == NULL)
	return(-1);

    if (xmlCharEncodingAliases == NULL)
	return(-1);
    /*
     * Walk down the list looking for a definition of the alias
     */
    for (i = 0;i < xmlCharEncodingAliasesNb;i++) {
	if (!strcmp(xmlCharEncodingAliases[i].alias, alias)) {
	    xmlFree((char *) xmlCharEncodingAliases[i].name);
	    xmlFree((char *) xmlCharEncodingAliases[i].alias);
	    xmlCharEncodingAliasesNb--;
	    memmove(&xmlCharEncodingAliases[i], &xmlCharEncodingAliases[i + 1],
		    sizeof(xmlCharEncodingAlias) * (xmlCharEncodingAliasesNb - i));
	    return(0);
	}
    }
    return(-1);
}

/**
 * xmlParseCharEncoding:
 * @name:  the encoding name as parsed, in UTF-8 format (ASCII actually)
 *
 * Compare the string to the encoding schemes already known. Note
 * that the comparison is case insensitive accordingly to the section
 * [XML] 4.3.3 Character Encoding in Entities.
 * 
 * Returns one of the XML_CHAR_ENCODING_... values or XML_CHAR_ENCODING_NONE
 * if not recognized.
 */
xmlCharEncoding
xmlParseCharEncoding(const char* name)
{
    const char *alias;
    char upper[500];
    int i;

    if (name == NULL)
	return(XML_CHAR_ENCODING_NONE);

    /*
     * Do the alias resolution
     */
    alias = xmlGetEncodingAlias(name);
    if (alias != NULL)
	name = alias;

    for (i = 0;i < 499;i++) {
        upper[i] = toupper(name[i]);
	if (upper[i] == 0) break;
    }
    upper[i] = 0;

    if (!strcmp(upper, "")) return(XML_CHAR_ENCODING_NONE);
    if (!strcmp(upper, "UTF-8")) return(XML_CHAR_ENCODING_UTF8);
    if (!strcmp(upper, "UTF8")) return(XML_CHAR_ENCODING_UTF8);

    /*
     * NOTE: if we were able to parse this, the endianness of UTF16 is
     *       already found and in use
     */
    if (!strcmp(upper, "UTF-16")) return(XML_CHAR_ENCODING_UTF16LE);
    if (!strcmp(upper, "UTF16")) return(XML_CHAR_ENCODING_UTF16LE);
    
    if (!strcmp(upper, "ISO-10646-UCS-2")) return(XML_CHAR_ENCODING_UCS2);
    if (!strcmp(upper, "UCS-2")) return(XML_CHAR_ENCODING_UCS2);
    if (!strcmp(upper, "UCS2")) return(XML_CHAR_ENCODING_UCS2);

    /*
     * NOTE: if we were able to parse this, the endianness of UCS4 is
     *       already found and in use
     */
    if (!strcmp(upper, "ISO-10646-UCS-4")) return(XML_CHAR_ENCODING_UCS4LE);
    if (!strcmp(upper, "UCS-4")) return(XML_CHAR_ENCODING_UCS4LE);
    if (!strcmp(upper, "UCS4")) return(XML_CHAR_ENCODING_UCS4LE);

    
    if (!strcmp(upper,  "ISO-8859-1")) return(XML_CHAR_ENCODING_8859_1);
    if (!strcmp(upper,  "ISO-LATIN-1")) return(XML_CHAR_ENCODING_8859_1);
    if (!strcmp(upper,  "ISO LATIN 1")) return(XML_CHAR_ENCODING_8859_1);

    if (!strcmp(upper,  "ISO-8859-2")) return(XML_CHAR_ENCODING_8859_2);
    if (!strcmp(upper,  "ISO-LATIN-2")) return(XML_CHAR_ENCODING_8859_2);
    if (!strcmp(upper,  "ISO LATIN 2")) return(XML_CHAR_ENCODING_8859_2);

    if (!strcmp(upper,  "ISO-8859-3")) return(XML_CHAR_ENCODING_8859_3);
    if (!strcmp(upper,  "ISO-8859-4")) return(XML_CHAR_ENCODING_8859_4);
    if (!strcmp(upper,  "ISO-8859-5")) return(XML_CHAR_ENCODING_8859_5);
    if (!strcmp(upper,  "ISO-8859-6")) return(XML_CHAR_ENCODING_8859_6);
    if (!strcmp(upper,  "ISO-8859-7")) return(XML_CHAR_ENCODING_8859_7);
    if (!strcmp(upper,  "ISO-8859-8")) return(XML_CHAR_ENCODING_8859_8);
    if (!strcmp(upper,  "ISO-8859-9")) return(XML_CHAR_ENCODING_8859_9);

    if (!strcmp(upper, "ISO-2022-JP")) return(XML_CHAR_ENCODING_2022_JP);
    if (!strcmp(upper, "SHIFT_JIS")) return(XML_CHAR_ENCODING_SHIFT_JIS);
    if (!strcmp(upper, "EUC-JP")) return(XML_CHAR_ENCODING_EUC_JP);

#ifdef DEBUG_ENCODING
    xmlGenericError(xmlGenericErrorContext, "Unknown encoding %s\n", name);
#endif
    return(XML_CHAR_ENCODING_ERROR);
}

/**
 * xmlGetCharEncodingName:
 * @enc:  the encoding
 *
 * The "canonical" name for XML encoding.
 * C.f. http://www.w3.org/TR/REC-xml#charencoding
 * Section 4.3.3  Character Encoding in Entities
 *
 * Returns the canonical name for the given encoding
 */

const char*
xmlGetCharEncodingName(xmlCharEncoding enc) {
    switch (enc) {
        case XML_CHAR_ENCODING_ERROR:
	    return(NULL);
        case XML_CHAR_ENCODING_NONE:
	    return(NULL);
        case XML_CHAR_ENCODING_UTF8:
	    return("UTF-8");
        case XML_CHAR_ENCODING_UTF16LE:
	    return("UTF-16");
        case XML_CHAR_ENCODING_UTF16BE:
	    return("UTF-16");
        case XML_CHAR_ENCODING_EBCDIC:
            return("EBCDIC");
        case XML_CHAR_ENCODING_UCS4LE:
            return("ISO-10646-UCS-4");
        case XML_CHAR_ENCODING_UCS4BE:
            return("ISO-10646-UCS-4");
        case XML_CHAR_ENCODING_UCS4_2143:
            return("ISO-10646-UCS-4");
        case XML_CHAR_ENCODING_UCS4_3412:
            return("ISO-10646-UCS-4");
        case XML_CHAR_ENCODING_UCS2:
            return("ISO-10646-UCS-2");
        case XML_CHAR_ENCODING_8859_1:
	    return("ISO-8859-1");
        case XML_CHAR_ENCODING_8859_2:
	    return("ISO-8859-2");
        case XML_CHAR_ENCODING_8859_3:
	    return("ISO-8859-3");
        case XML_CHAR_ENCODING_8859_4:
	    return("ISO-8859-4");
        case XML_CHAR_ENCODING_8859_5:
	    return("ISO-8859-5");
        case XML_CHAR_ENCODING_8859_6:
	    return("ISO-8859-6");
        case XML_CHAR_ENCODING_8859_7:
	    return("ISO-8859-7");
        case XML_CHAR_ENCODING_8859_8:
	    return("ISO-8859-8");
        case XML_CHAR_ENCODING_8859_9:
	    return("ISO-8859-9");
        case XML_CHAR_ENCODING_2022_JP:
            return("ISO-2022-JP");
        case XML_CHAR_ENCODING_SHIFT_JIS:
            return("Shift-JIS");
        case XML_CHAR_ENCODING_EUC_JP:
            return("EUC-JP");
	case XML_CHAR_ENCODING_ASCII:
	    return(NULL);
    }
    return(NULL);
}

/************************************************************************
 *									*
 *			Char encoding handlers				*
 *									*
 ************************************************************************/


/* the size should be growable, but it's not a big deal ... */
#define MAX_ENCODING_HANDLERS 50
static xmlCharEncodingHandlerPtr *handlers = NULL;
static int nbCharEncodingHandler = 0;

/*
 * The default is UTF-8 for XML, that's also the default used for the
 * parser internals, so the default encoding handler is NULL
 */

static xmlCharEncodingHandlerPtr xmlDefaultCharEncodingHandler = NULL;

/**
 * xmlNewCharEncodingHandler:
 * @name:  the encoding name, in UTF-8 format (ASCII actually)
 * @input:  the xmlCharEncodingInputFunc to read that encoding
 * @output:  the xmlCharEncodingOutputFunc to write that encoding
 *
 * Create and registers an xmlCharEncodingHandler.
 *
 * Returns the xmlCharEncodingHandlerPtr created (or NULL in case of error).
 */
xmlCharEncodingHandlerPtr
xmlNewCharEncodingHandler(const char *name, 
                          xmlCharEncodingInputFunc input,
                          xmlCharEncodingOutputFunc output) {
    xmlCharEncodingHandlerPtr handler;
    const char *alias;
    char upper[500];
    int i;
    char *up = 0;

    /*
     * Do the alias resolution
     */
    alias = xmlGetEncodingAlias(name);
    if (alias != NULL)
	name = alias;

    /*
     * Keep only the uppercase version of the encoding.
     */
    if (name == NULL) {
        xmlGenericError(xmlGenericErrorContext,
		"xmlNewCharEncodingHandler : no name !\n");
	return(NULL);
    }
    for (i = 0;i < 499;i++) {
        upper[i] = toupper(name[i]);
	if (upper[i] == 0) break;
    }
    upper[i] = 0;
    up = xmlMemStrdup(upper);
    if (up == NULL) {
        xmlGenericError(xmlGenericErrorContext,
		"xmlNewCharEncodingHandler : out of memory !\n");
	return(NULL);
    }

    /*
     * allocate and fill-up an handler block.
     */
    handler = (xmlCharEncodingHandlerPtr)
              xmlMalloc(sizeof(xmlCharEncodingHandler));
    if (handler == NULL) {
        xmlFree(up);
        xmlGenericError(xmlGenericErrorContext,
		"xmlNewCharEncodingHandler : out of memory !\n");
	return(NULL);
    }
    handler->input = input;
    handler->output = output;
    handler->name = up;

#ifdef LIBXML_ICONV_ENABLED
    handler->iconv_in = NULL;
    handler->iconv_out = NULL;
#endif /* LIBXML_ICONV_ENABLED */

    /*
     * registers and returns the handler.
     */
    xmlRegisterCharEncodingHandler(handler);
#ifdef DEBUG_ENCODING
    xmlGenericError(xmlGenericErrorContext,