entities.c

SIP 1.5.0源代码

/*
 * [2] Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD]
 *                  | [#x10000-#x10FFFF]
 * any Unicode character, excluding the surrogate blocks, FFFE, and FFFF.
 */
#define IS_CHAR(c)							\
    (((c) == 0x09) || ((c) == 0x0a) || ((c) == 0x0d) ||			\
     (((c) >= 0x20) && ((c) != 0xFFFE) && ((c) != 0xFFFF)))

/*
 * A buffer used for converting entities to their equivalent and back.
 */
static int buffer_size = 0;
static xmlChar *buffer = NULL;

int growBuffer(void) {
    buffer_size *= 2;
    buffer = (xmlChar *) xmlRealloc(buffer, buffer_size * sizeof(xmlChar));
    if (buffer == NULL) {
        perror("realloc failed");
	return(-1);
    }
    return(0);
}


/**
 * xmlEncodeEntities:
 * @doc:  the document containing the string
 * @input:  A string to convert to XML.
 *
 * Do a global encoding of a string, replacing the predefined entities
 * and non ASCII values with their entities and CharRef counterparts.
 *
 * TODO: remove xmlEncodeEntities, once we are not afraid of breaking binary
 *       compatibility
 *
 * People must migrate their code to xmlEncodeEntitiesReentrant !
 * This routine will issue a warning when encountered.
 * 
 * Returns A newly allocated string with the substitution done.
 */
const xmlChar *
xmlEncodeEntities(xmlDocPtr doc, const xmlChar *input) {
    const xmlChar *cur = input;
    xmlChar *out = buffer;
    static int warning = 1;
    int html = 0;


    if (warning) {
    fprintf(stderr, "Deprecated API xmlEncodeEntities() used\n");
    fprintf(stderr, "   change code to use xmlEncodeEntitiesReentrant()\n");
    warning = 0;
    }

    if (input == NULL) return(NULL);
    if (doc != NULL)
        html = (doc->type == XML_HTML_DOCUMENT_NODE);

    if (buffer == NULL) {
        buffer_size = 1000;
        buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
	if (buffer == NULL) {
	    perror("malloc failed");
            return(NULL);
	}
	out = buffer;
    }
    while (*cur != '\0') {
        if (out - buffer > buffer_size - 100) {
	    int index = out - buffer;

	    growBuffer();
	    out = &buffer[index];
	}

	/*
	 * By default one have to encode at least '<', '>', '"' and '&' !
	 */
	if (*cur == '<') {
	    *out++ = '&';
	    *out++ = 'l';
	    *out++ = 't';
	    *out++ = ';';
	} else if (*cur == '>') {
	    *out++ = '&';
	    *out++ = 'g';
	    *out++ = 't';
	    *out++ = ';';
	} else if (*cur == '&') {
	    *out++ = '&';
	    *out++ = 'a';
	    *out++ = 'm';
	    *out++ = 'p';
	    *out++ = ';';
	} else if (*cur == '"') {
	    *out++ = '&';
	    *out++ = 'q';
	    *out++ = 'u';
	    *out++ = 'o';
	    *out++ = 't';
	    *out++ = ';';
	} else if ((*cur == '\'') && (!html)) {
	    *out++ = '&';
	    *out++ = 'a';
	    *out++ = 'p';
	    *out++ = 'o';
	    *out++ = 's';
	    *out++ = ';';
	} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
	    (*cur == '\n') || (*cur == '\r') || (*cur == '\t')) {
	    /*
	     * default case, just copy !
	     */
	    *out++ = *cur;
#ifndef USE_UTF_8
	} else if ((sizeof(xmlChar) == 1) && (*cur >= 0x80)) {
	    char buf[10], *ptr;
#ifdef HAVE_SNPRINTF
	    snprintf(buf, 9, "&#%d;", *cur);
#else
	    sprintf(buf, "&#%d;", *cur);
#endif
            ptr = buf;
	    while (*ptr != 0) *out++ = *ptr++;
#endif
	} else if (IS_CHAR(*cur)) {
	    char buf[10], *ptr;

#ifdef HAVE_SNPRINTF
	    snprintf(buf, 9, "&#%d;", *cur);
#else
	    sprintf(buf, "&#%d;", *cur);
#endif
            ptr = buf;
	    while (*ptr != 0) *out++ = *ptr++;
	}
#if 0
	else {
	    /*
	     * default case, this is not a valid char !
	     * Skip it...
	     */
	    fprintf(stderr, "xmlEncodeEntities: invalid char %d\n", (int) *cur);
	}
#endif
	cur++;
    }
    *out++ = 0;
    return(buffer);
}

/*
 * Macro used to grow the current buffer.
 */
#define growBufferReentrant() {						\
    buffer_size *= 2;							\
    buffer = (xmlChar *)						\
    		xmlRealloc(buffer, buffer_size * sizeof(xmlChar));	\
    if (buffer == NULL) {						\
	perror("realloc failed");					\
	return(NULL);							\
    }									\
}


/**
 * xmlEncodeEntitiesReentrant:
 * @doc:  the document containing the string
 * @input:  A string to convert to XML.
 *
 * Do a global encoding of a string, replacing the predefined entities
 * and non ASCII values with their entities and CharRef counterparts.
 * Contrary to xmlEncodeEntities, this routine is reentrant, and result
 * must be deallocated.
 *
 * TODO !!!! Once moved to UTF-8 internal encoding, the encoding of non-ascii
 *           get erroneous.
 *
 * Returns A newly allocated string with the substitution done.
 */
xmlChar *
xmlEncodeEntitiesReentrant(xmlDocPtr doc, const xmlChar *input) {
    const xmlChar *cur = input;
    xmlChar *buffer = NULL;
    xmlChar *out = NULL;
    int buffer_size = 0;
    int html = 0;

    if (input == NULL) return(NULL);
    if (doc != NULL)
        html = (doc->type == XML_HTML_DOCUMENT_NODE);

    /*
     * allocate an translation buffer.
     */
    buffer_size = 1000;
    buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
    if (buffer == NULL) {
	perror("malloc failed");
	return(NULL);
    }
    out = buffer;

    while (*cur != '\0') {
        if (out - buffer > buffer_size - 100) {
	    int index = out - buffer;

	    growBufferReentrant();
	    out = &buffer[index];
	}

	/*
	 * By default one have to encode at least '<', '>', '"' and '&' !
	 */
	if (*cur == '<') {
	    *out++ = '&';
	    *out++ = 'l';
	    *out++ = 't';
	    *out++ = ';';
	} else if (*cur == '>') {
	    *out++ = '&';
	    *out++ = 'g';
	    *out++ = 't';
	    *out++ = ';';
	} else if (*cur == '&') {
	    *out++ = '&';
	    *out++ = 'a';
	    *out++ = 'm';
	    *out++ = 'p';
	    *out++ = ';';
	} else if (*cur == '"') {
	    *out++ = '&';
	    *out++ = 'q';
	    *out++ = 'u';
	    *out++ = 'o';
	    *out++ = 't';
	    *out++ = ';';
#if 0
	} else if ((*cur == '\'') && (!html)) {
	    *out++ = '&';
	    *out++ = 'a';
	    *out++ = 'p';
	    *out++ = 'o';
	    *out++ = 's';
	    *out++ = ';';
#endif
	} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
	    (*cur == '\n') || (*cur == '\r') || (*cur == '\t')) {
	    /*
	     * default case, just copy !
	     */
	    *out++ = *cur;
	} else if (*cur >= 0x80) {
	    if (html) {
		char buf[15], *ptr;

		/*
		 * TODO: improve by searching in html40EntitiesTable
		 */
#ifdef HAVE_SNPRINTF
		snprintf(buf, 9, "&#%d;", *cur);
#else
		sprintf(buf, "&#%d;", *cur);
#endif
		ptr = buf;
		while (*ptr != 0) *out++ = *ptr++;
	    } else if (doc->encoding != NULL) {
		/*
		 * TODO !!!
		 */
		*out++ = *cur;
	    } else {
		/*
		 * We assume we have UTF-8 input.
		 */
		char buf[10], *ptr;
		int val = 0, l = 1;

		if (*cur < 0xC0) {
		    fprintf(stderr,
			    "xmlEncodeEntitiesReentrant : input not UTF-8\n");
		    doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
#ifdef HAVE_SNPRINTF
		    snprintf(buf, 9, "&#%d;", *cur);
#else
		    sprintf(buf, "&#%d;", *cur);
#endif
		    ptr = buf;
		    while (*ptr != 0) *out++ = *ptr++;
		    continue;
		} else if (*cur < 0xE0) {
                    val = (cur[0]) & 0x1F;
		    val <<= 6;
		    val |= (cur[1]) & 0x3F;
		    l = 2;
		} else if (*cur < 0xF0) {
                    val = (cur[0]) & 0x0F;
		    val <<= 6;
		    val |= (cur[1]) & 0x3F;
		    val <<= 6;
		    val |= (cur[2]) & 0x3F;
		    l = 3;
		} else if (*cur < 0xF8) {
                    val = (cur[0]) & 0x07;
		    val <<= 6;
		    val |= (cur[1]) & 0x3F;
		    val <<= 6;
		    val |= (cur[2]) & 0x3F;
		    val <<= 6;
		    val |= (cur[3]) & 0x3F;
		    l = 4;
		}
		if ((l == 1) || (!IS_CHAR(val))) {
		    fprintf(stderr,
			"xmlEncodeEntitiesReentrant : char out of range\n");
		    doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
#ifdef HAVE_SNPRINTF
		    snprintf(buf, 9, "&#%d;", *cur);
#else
		    sprintf(buf, "&#%d;", *cur);
#endif
		    ptr = buf;
		    while (*ptr != 0) *out++ = *ptr++;
		    cur++;
		    continue;
		}
		/*
		 * We could do multiple things here. Just save as a char ref
		 */
#ifdef HAVE_SNPRINTF
		snprintf(buf, 14, "&#x%X;", val);
#else
		sprintf(buf, "&#x%X;", val);
#endif
		buf[14] = 0;
		ptr = buf;
		while (*ptr != 0) *out++ = *ptr++;
		cur += l;
		continue;
	    }
	} else if (IS_CHAR(*cur)) {
	    char buf[10], *ptr;

#ifdef HAVE_SNPRINTF
	    snprintf(buf, 9, "&#%d;", *cur);
#else
	    sprintf(buf, "&#%d;", *cur);
#endif
            ptr = buf;
	    while (*ptr != 0) *out++ = *ptr++;
	}
#if 0
	else {
	    /*
	     * default case, this is not a valid char !
	     * Skip it...
	     */
	    fprintf(stderr, "xmlEncodeEntities: invalid char %d\n", (int) *cur);
	}
#endif
	cur++;
    }
    *out++ = 0;
    return(buffer);
}

/**
 * xmlCreateEntitiesTable:
 *
 * create and initialize an empty entities hash table.
 *
 * Returns the xmlEntitiesTablePtr just created or NULL in case of error.
 */
xmlEntitiesTablePtr
xmlCreateEntitiesTable(void) {
    xmlEntitiesTablePtr ret;

    ret = (xmlEntitiesTablePtr) 
         xmlMalloc(sizeof(xmlEntitiesTable));
    if (ret == NULL) {
        fprintf(stderr, "xmlCreateEntitiesTable : xmlMalloc(%ld) failed\n",
	        (long)sizeof(xmlEntitiesTable));
        return(NULL);
    }
    ret->max_entities = XML_MIN_ENTITIES_TABLE;
    ret->nb_entities = 0;
    ret->table = (xmlEntityPtr *) 
         xmlMalloc(ret->max_entities * sizeof(xmlEntityPtr));
    if (ret == NULL) {
        fprintf(stderr, "xmlCreateEntitiesTable : xmlMalloc(%ld) failed\n",
	        ret->max_entities * (long)sizeof(xmlEntityPtr));
	xmlFree(ret);
        return(NULL);
    }
    return(ret);
}

/**
 * xmlFreeEntitiesTable:
 * @table:  An entity table
 *
 * Deallocate the memory used by an entities hash table.
 */
void
xmlFreeEntitiesTable(xmlEntitiesTablePtr table) {
    int i;

    if (table == NULL) return;

    for (i = 0;i < table->nb_entities;i++) {
        xmlFreeEntity(table->table[i]);
    }
    xmlFree(table->table);
    xmlFree(table);
}

/**
 * xmlCopyEntitiesTable:
 * @table:  An entity table
 *
 * Build a copy of an entity table.
 * 
 * Returns the new xmlEntitiesTablePtr or NULL in case of error.
 */
xmlEntitiesTablePtr
xmlCopyEntitiesTable(xmlEntitiesTablePtr table) {
    xmlEntitiesTablePtr ret;
    xmlEntityPtr cur, ent;
    int i;

    ret = (xmlEntitiesTablePtr) xmlMalloc(sizeof(xmlEntitiesTable));
    if (ret == NULL) {
        fprintf(stderr, "xmlCopyEntitiesTable: out of memory !\n");
	return(NULL);
    }
    ret->table = (xmlEntityPtr *) xmlMalloc(table->max_entities *
                                            sizeof(xmlEntityPtr));
    if (ret->table == NULL) {
        fprintf(stderr, "xmlCopyEntitiesTable: out of memory !\n");
	xmlFree(ret);
	return(NULL);
    }
    ret->max_entities = table->max_entities;
    ret->nb_entities = table->nb_entities;
    for (i = 0;i < ret->nb_entities;i++) {
	cur = (xmlEntityPtr) xmlMalloc(sizeof(xmlEntity));
	if (cur == NULL) {
	    fprintf(stderr, "xmlCopyEntityTable: out of memory !\n");
	    xmlFree(ret);
	    xmlFree(ret->table);
	    return(NULL);
	}
	memset(cur, 0, sizeof(xmlEntity));
	cur->type = XML_ELEMENT_DECL;
	ret->table[i] = cur;
	ent = table->table[i];

	cur->etype = ent->etype;
	if (ent->name != NULL)
	    cur->name = xmlStrdup(ent->name);
	if (ent->ExternalID != NULL)
	    cur->ExternalID = xmlStrdup(ent->ExternalID);
	if (ent->SystemID != NULL)
	    cur->SystemID = xmlStrdup(ent->SystemID);
	if (ent->content != NULL)
	    cur->content = xmlStrdup(ent->content);
	if (ent->orig != NULL)
	    cur->orig = xmlStrdup(ent->orig);
    }
    return(ret);
}

/**
 * xmlDumpEntityDecl:
 * @buf:  An XML buffer.
 * @ent:  An entity table
 *
 * This will dump the content of the entity table as an XML DTD definition
 */
void
xmlDumpEntityDecl(xmlBufferPtr buf, xmlEntityPtr ent) {
    switch (ent->etype) {
	case XML_INTERNAL_GENERAL_ENTITY:
	    xmlBufferWriteChar(buf, "<!ENTITY ");
	    xmlBufferWriteCHAR(buf, ent->name);
	    xmlBufferWriteChar(buf, " ");
	    if (ent->orig != NULL)
		xmlBufferWriteQuotedString(buf, ent->orig);
	    else
		xmlBufferWriteQuotedString(buf, ent->content);
	    xmlBufferWriteChar(buf, ">\n");
	    break;
	case XML_EXTERNAL_GENERAL_PARSED_ENTITY:
	    xmlBufferWriteChar(buf, "<!ENTITY ");
	    xmlBufferWriteCHAR(buf, ent->name);
	    if (ent->ExternalID != NULL) {
		 xmlBufferWriteChar(buf, " PUBLIC ");
		 xmlBufferWriteQuotedString(buf, ent->ExternalID);
		 xmlBufferWriteChar(buf, " ");
		 xmlBufferWriteQuotedString(buf, ent->SystemID);
	    } else {
		 xmlBufferWriteChar(buf, " SYSTEM ");
		 xmlBufferWriteQuotedString(buf, ent->SystemID);
	    }
	    xmlBufferWriteChar(buf, ">\n");
	    break;
	case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY:
	    xmlBufferWriteChar(buf, "<!ENTITY ");
	    xmlBufferWriteCHAR(buf, ent->name);
	    if (ent->ExternalID != NULL) {
		 xmlBufferWriteChar(buf, " PUBLIC ");
		 xmlBufferWriteQuotedString(buf, ent->ExternalID);
		 xmlBufferWriteChar(buf, " ");
		 xmlBufferWriteQuotedString(buf, ent->SystemID);
	    } else {
		 xmlBufferWriteChar(buf, " SYSTEM ");
		 xmlBufferWriteQuotedString(buf, ent->SystemID);
	    }
	    if (ent->content != NULL) { /* Should be true ! */
		xmlBufferWriteChar(buf, " NDATA ");
		if (ent->orig != NULL)
		    xmlBufferWriteCHAR(buf, ent->orig);
		else
		    xmlBufferWriteCHAR(buf, ent->content);
	    }
	    xmlBufferWriteChar(buf, ">\n");
	    break;
	case XML_INTERNAL_PARAMETER_ENTITY:
	    xmlBufferWriteChar(buf, "<!ENTITY % ");
	    xmlBufferWriteCHAR(buf, ent->name);
	    xmlBufferWriteChar(buf, " ");
	    if (ent->orig == NULL)
		xmlBufferWriteQuotedString(buf, ent->content);
	    else
		xmlBufferWriteQuotedString(buf, ent->orig);
	    xmlBufferWriteChar(buf, ">\n");
	    break;
	case XML_EXTERNAL_PARAMETER_ENTITY:
	    xmlBufferWriteChar(buf, "<!ENTITY % ");
	    xmlBufferWriteCHAR(buf, ent->name);
	    if (ent->ExternalID != NULL) {
		 xmlBufferWriteChar(buf, " PUBLIC ");
		 xmlBufferWriteQuotedString(buf, ent->ExternalID);
		 xmlBufferWriteChar(buf, " ");
		 xmlBufferWriteQuotedString(buf, ent->SystemID);
	    } else {
		 xmlBufferWriteChar(buf, " SYSTEM ");
		 xmlBufferWriteQuotedString(buf, ent->SystemID);
	    }
	    xmlBufferWriteChar(buf, ">\n");
	    break;
	default:
	    fprintf(stderr,
		"xmlDumpEntitiesTable: internal: unknown type %d\n",
		    ent->etype);
    }
}

/**
 * xmlDumpEntitiesTable:
 * @buf:  An XML buffer.
 * @table:  An entity table
 *
 * This will dump the content of the entity table as an XML DTD definition
 */
void
xmlDumpEntitiesTable(xmlBufferPtr buf, xmlEntitiesTablePtr table) {
    int i;
    xmlEntityPtr cur;

    if (table == NULL) return;

    for (i = 0;i < table->nb_entities;i++) {
        cur = table->table[i];
	xmlDumpEntityDecl(buf, cur);
    }
}