xpointer.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

    oldset = (xmlLocationSetPtr) set->user;

    /*
     * The loop is to compute the covering range for each item and add it
     */
    newset = xmlXPtrLocationSetCreate(NULL);
    for (i = 0;i < oldset->locNr;i++) {
	xmlXPtrLocationSetAdd(newset,
		xmlXPtrCoveringRange(ctxt, oldset->locTab[i]));
    }

    /*
     * Save the new value and cleanup
     */
    valuePush(ctxt, xmlXPtrWrapLocationSet(newset));
    xmlXPathFreeObject(set);
}

/**
 * xmlXPtrInsideRange:
 * @ctxt:  the XPointer Parser context
 * @loc:  the location for which the inside range must be computed
 *
 * A inside range is a range described in the range-inside() description
 *
 * Returns a new location or NULL in case of error
 */
static xmlXPathObjectPtr
xmlXPtrInsideRange(xmlXPathParserContextPtr ctxt, xmlXPathObjectPtr loc) {
    if (loc == NULL)
	return(NULL);
    if ((ctxt == NULL) || (ctxt->context == NULL) ||
	(ctxt->context->doc == NULL))
	return(NULL);
    switch (loc->type) {
        case XPATH_POINT: {
	    xmlNodePtr node = (xmlNodePtr) loc->user;
	    switch (node->type) {
		case XML_PI_NODE:
		case XML_COMMENT_NODE:
		case XML_TEXT_NODE:
		case XML_CDATA_SECTION_NODE: {
		    if (node->content == NULL) {
			return(xmlXPtrNewRange(node, 0, node, 0));
		    } else {
			return(xmlXPtrNewRange(node, 0, node,
					       xmlStrlen(node->content)));
		    }
		}
		case XML_ATTRIBUTE_NODE:
		case XML_ELEMENT_NODE:
		case XML_ENTITY_REF_NODE:
		case XML_DOCUMENT_NODE:
		case XML_NOTATION_NODE:
		case XML_HTML_DOCUMENT_NODE: {
		    return(xmlXPtrNewRange(node, 0, node,
					   xmlXPtrGetArity(node)));
		}
		default:
		    break;
	    }
	    return(NULL);
	}
        case XPATH_RANGE: {
	    xmlNodePtr node = (xmlNodePtr) loc->user;
	    if (loc->user2 != NULL) {
		return(xmlXPtrNewRange(node, loc->index,
			               loc->user2, loc->index2));
	    } else {
		switch (node->type) {
		    case XML_PI_NODE:
		    case XML_COMMENT_NODE:
		    case XML_TEXT_NODE:
		    case XML_CDATA_SECTION_NODE: {
			if (node->content == NULL) {
			    return(xmlXPtrNewRange(node, 0, node, 0));
			} else {
			    return(xmlXPtrNewRange(node, 0, node,
						   xmlStrlen(node->content)));
			}
		    }
		    case XML_ATTRIBUTE_NODE:
		    case XML_ELEMENT_NODE:
		    case XML_ENTITY_REF_NODE:
		    case XML_DOCUMENT_NODE:
		    case XML_NOTATION_NODE:
		    case XML_HTML_DOCUMENT_NODE: {
			return(xmlXPtrNewRange(node, 0, node,
					       xmlXPtrGetArity(node)));
		    }
		    default:
			break;
		}
		return(NULL);
	    }
        }
	default:
	    TODO /* missed one case ??? */
    }
    return(NULL);
}

/**
 * xmlXPtrRangeInsideFunction:
 * @ctxt:  the XPointer Parser context
 * @nargs:  the number of args
 *
 * Function implementing the range-inside() function 5.4.3
 *  location-set range-inside(location-set )
 *
 *  The range-inside function returns ranges covering the contents of
 *  the locations in the argument location-set. For each location x in
 *  the argument location-set, a range location is added to the result
 *  location-set. If x is a range location, then x is added to the
 *  result location-set. If x is not a range location, then x is used
 *  as the container location of the start and end points of the range
 *  location to be added; the index of the start point of the range is
 *  zero; if the end point is a character point then its index is the
 *  length of the string-value of x, and otherwise is the number of
 *  location children of x.
 *
 */
void
xmlXPtrRangeInsideFunction(xmlXPathParserContextPtr ctxt, int nargs) {
    int i;
    xmlXPathObjectPtr set;
    xmlLocationSetPtr oldset;
    xmlLocationSetPtr newset;

    CHECK_ARITY(1);
    if ((ctxt->value == NULL) ||
	((ctxt->value->type != XPATH_LOCATIONSET) &&
	 (ctxt->value->type != XPATH_NODESET)))
        XP_ERROR(XPATH_INVALID_TYPE)

    set = valuePop(ctxt);
    if (set->type == XPATH_NODESET) {
	xmlXPathObjectPtr tmp;

	/*
	 * First convert to a location set
	 */
	tmp = xmlXPtrNewLocationSetNodeSet(set->nodesetval);
	xmlXPathFreeObject(set);
	set = tmp;
    }
    oldset = (xmlLocationSetPtr) set->user;

    /*
     * The loop is to compute the covering range for each item and add it
     */
    newset = xmlXPtrLocationSetCreate(NULL);
    for (i = 0;i < oldset->locNr;i++) {
	xmlXPtrLocationSetAdd(newset,
		xmlXPtrInsideRange(ctxt, oldset->locTab[i]));
    }

    /*
     * Save the new value and cleanup
     */
    valuePush(ctxt, xmlXPtrWrapLocationSet(newset));
    xmlXPathFreeObject(set);
}

/**
 * xmlXPtrRangeToFunction:
 * @ctxt:  the XPointer Parser context
 * @nargs:  the number of args
 *
 * Implement the range-to() XPointer function
 */
void
xmlXPtrRangeToFunction(xmlXPathParserContextPtr ctxt, int nargs) {
    xmlXPathObjectPtr range;
    const xmlChar *cur;
    xmlXPathObjectPtr res, obj;
    xmlXPathObjectPtr tmp;
    xmlLocationSetPtr newset = NULL;
    xmlNodeSetPtr oldset;
    int i;

    CHECK_ARITY(1);
    /*
     * Save the expression pointer since we will have to evaluate
     * it multiple times. Initialize the new set.
     */
    CHECK_TYPE(XPATH_NODESET);
    obj = valuePop(ctxt);
    oldset = obj->nodesetval;
    ctxt->context->node = NULL;

    cur = ctxt->cur;
    newset = xmlXPtrLocationSetCreate(NULL);
    
    for (i = 0; i < oldset->nodeNr; i++) {
	ctxt->cur = cur;

	/*
	 * Run the evaluation with a node list made of a single item
	 * in the nodeset.
	 */
	ctxt->context->node = oldset->nodeTab[i];
	tmp = xmlXPathNewNodeSet(ctxt->context->node);
	valuePush(ctxt, tmp);

	xmlXPathEvalExpr(ctxt);
	CHECK_ERROR;

	/*
	 * The result of the evaluation need to be tested to
	 * decided whether the filter succeeded or not
	 */
	res = valuePop(ctxt);
	range = xmlXPtrNewRangeNodeObject(oldset->nodeTab[i], res);
	if (range != NULL) {
	    xmlXPtrLocationSetAdd(newset, range);
	}

	/*
	 * Cleanup
	 */
	if (res != NULL)
	    xmlXPathFreeObject(res);
	if (ctxt->value == tmp) {
	    res = valuePop(ctxt);
	    xmlXPathFreeObject(res);
	}
	
	ctxt->context->node = NULL;
    }

    /*
     * The result is used as the new evaluation set.
     */
    xmlXPathFreeObject(obj);
    ctxt->context->node = NULL;
    ctxt->context->contextSize = -1;
    ctxt->context->proximityPosition = -1;
    valuePush(ctxt, xmlXPtrWrapLocationSet(newset));
}

/**
 * xmlXPtrAdvanceNode:
 * @cur:  the node
 * @level: incremented/decremented to show level in tree
 *
 * Advance to the next element or text node in document order
 * TODO: add a stack for entering/exiting entities 
 *
 * Returns -1 in case of failure, 0 otherwise
 */
xmlNodePtr
xmlXPtrAdvanceNode(xmlNodePtr cur, int *level) {
next:
    if (cur == NULL)
	return(NULL);
    if (cur->children != NULL) {
        cur = cur->children ;
	if (level != NULL)
	    (*level)++;
	goto found;
    }
skip:		/* This label should only be needed if something is wrong! */
    if (cur->next != NULL) {
	cur = cur->next;
	goto found;
    }
    do {
        cur = cur->parent;
	if (level != NULL)
	    (*level)--;
        if (cur == NULL) return(NULL);
        if (cur->next != NULL) {
	    cur = cur->next;
	    goto found;
	}
    } while (cur != NULL);

found:
    if ((cur->type != XML_ELEMENT_NODE) &&
	(cur->type != XML_TEXT_NODE) &&
	(cur->type != XML_DOCUMENT_NODE) &&
	(cur->type != XML_HTML_DOCUMENT_NODE) &&
	(cur->type != XML_CDATA_SECTION_NODE)) {
	    if (cur->type == XML_ENTITY_REF_NODE) {	/* Shouldn't happen */
		TODO
		goto skip;
	    }
	    goto next;
	}
    return(cur);
}

/**
 * xmlXPtrAdvanceChar:
 * @node:  the node
 * @indx:  the indx
 * @bytes:  the number of bytes
 *
 * Advance a point of the associated number of bytes (not UTF8 chars)
 *
 * Returns -1 in case of failure, 0 otherwise
 */
static int
xmlXPtrAdvanceChar(xmlNodePtr *node, int *indx, int bytes) {
    xmlNodePtr cur;
    int pos;
    int len;

    if ((node == NULL) || (indx == NULL))
	return(-1);
    cur = *node;
    if (cur == NULL)
	return(-1);
    pos = *indx;

    while (bytes >= 0) {
	/*
	 * First position to the beginning of the first text node
	 * corresponding to this point
	 */
	while ((cur != NULL) &&
	       ((cur->type == XML_ELEMENT_NODE) ||
	        (cur->type == XML_DOCUMENT_NODE) ||
	        (cur->type == XML_HTML_DOCUMENT_NODE))) {
	    if (pos > 0) {
		cur = xmlXPtrGetNthChild(cur, pos);
		pos = 0;
	    } else {
		cur = xmlXPtrAdvanceNode(cur, NULL);
		pos = 0;
	    }
	}

	if (cur == NULL) {
	    *node = NULL;
	    *indx = 0;
	    return(-1);
	}

	/*
	 * if there is no move needed return the current value.
	 */
	if (pos == 0) pos = 1;
	if (bytes == 0) {
	    *node = cur;
	    *indx = pos;
	    return(0);
	}
	/*
	 * We should have a text (or cdata) node ... 
	 */
	len = 0;
	if ((cur->type != XML_ELEMENT_NODE) &&
            (cur->content != NULL)) {
	    len = xmlStrlen(cur->content);
	}
	if (pos > len) {
	    /* Strange, the indx in the text node is greater than it's len */
	    STRANGE
	    pos = len;
	}
	if (pos + bytes >= len) {
	    bytes -= (len - pos);
	    cur = xmlXPtrAdvanceNode(cur, NULL);
	    cur = 0;
	} else if (pos + bytes < len) {
	    pos += bytes;
	    *node = cur;
	    *indx = pos;
	    return(0);
	}
    }
    return(-1);
}

/**
 * xmlXPtrMatchString:
 * @string:  the string to search
 * @start:  the start textnode
 * @startindex:  the start index
 * @end:  the end textnode IN/OUT
 * @endindex:  the end index IN/OUT
 *
 * Check whether the document contains @string at the position
 * (@start, @startindex) and limited by the (@end, @endindex) point
 *
 * Returns -1 in case of failure, 0 if not found, 1 if found in which case
 *            (@start, @startindex) will indicate the position of the beginning
 *            of the range and (@end, @endindex) will indicate the end
 *            of the range
 */
static int
xmlXPtrMatchString(const xmlChar *string, xmlNodePtr start, int startindex,
	            xmlNodePtr *end, int *endindex) {
    xmlNodePtr cur;
    int pos; /* 0 based */
    int len; /* in bytes */
    int stringlen; /* in bytes */
    int match;

    if (string == NULL)
	return(-1);
    if (start == NULL)
	return(-1);
    if ((end == NULL) || (endindex == NULL))
	return(-1);
    cur = start;
    if (cur == NULL)
	return(-1);
    pos = startindex - 1;
    stringlen = xmlStrlen(string);

    while (stringlen > 0) {
	if ((cur == *end) && (pos + stringlen > *endindex))
	    return(0);

	if ((cur->type != XML_ELEMENT_NODE) && (cur->content != NULL)) {
	    len = xmlStrlen(cur->content);
	    if (len >= pos + stringlen) {
		match = (!xmlStrncmp(&cur->content[pos], string, stringlen));
		if (match) {
#ifdef DEBUG_RANGES
		    xmlGenericError(xmlGenericErrorContext,
			    "found range %d bytes at index %d of ->",
			    stringlen, pos + 1);
		    xmlDebugDumpString(stdout, cur->content);
		    xmlGenericError(xmlGenericErrorContext, "\n");
#endif
		    *end = cur;
		    *endindex = pos + stringlen;
		    return(1);
		} else {
		    return(0);
		}
	    } else {
                int sub = len - pos;
		match = (!xmlStrncmp(&cur->content[pos], string, sub));
		if (match) {
#ifdef DEBUG_RANGES
		    xmlGenericError(xmlGenericErrorContext,
			    "found subrange %d bytes at index %d of ->",
			    sub, pos + 1);
		    xmlDebugDumpString(stdout, cur->content);
		    xmlGenericError(xmlGenericErrorContext, "\n");
#endif
                    string = &string[sub];
		    stringlen -= sub;
		} else {
		    return(0);
		}
	    }
	}
	cur = xmlXPtrAdvanceNode(cur, NULL);
	if (cur == NULL)
	    return(0);
	pos = 0;
    }
    return(1);
}

/**
 * xmlXPtrSearchString:
 * @string:  the string to search
 * @start:  the start textnode IN/OUT
 * @startindex:  the start index IN/OUT
 * @end:  the end textnode
 * @endindex:  the end index
 *
 * Search the next occurrence of @string within the document content
 * until the (@end, @endindex) point is reached
 *
 * Returns -1 in case of failure, 0 if not found, 1 if found in which case
 *            (@start, @startindex) will indicate the position of the beginning
 *            of the range and (@end, @endindex) will indicate the end
 *            of the range
 */
static int
xmlXPtrSearchString(const xmlChar *string, xmlNodePtr *start, int *startindex,
	            xmlNodePtr *end, int *endindex) {
    xmlNodePtr cur;
    const xmlChar *str;
    int pos; /* 0 based */
    int len; /* in bytes */
    xmlChar first;

    if (string == NULL)
	return(-1);
    if ((start == NULL) || (startindex == NULL))
	return(-1);
    if ((end == NULL) || (endindex == NULL))
	return(-1);
    cur = *start;
    if (cur == NULL)
	return(-1);
    pos = *startindex - 1;
    first = string[0];

    while (cur != NULL) {
	if ((cur->type != XML_ELEMENT_NODE) && (cur->content != NULL)) {
	    len = xmlStrlen(cur->content);
	    while (pos <= len) {
		if (first != 0) {
		    str = xmlStrchr(&cur->content[pos], first);
		    if (str != NULL) {
			pos = (str - (xmlChar *)(cur->content));
#ifdef DEBUG_RANGES
			xmlGenericError(xmlGenericErrorContext,
				"found '%c' at