tree.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

    cur->doc = cur;
    /*
     * The in memory encoding is always UTF8
     * This field will never change and would
     * be obsolete if not for binary compatibility.
     */
    cur->charset = XML_CHAR_ENCODING_UTF8;

    if ((__xmlRegisterCallbacks) && (xmlRegisterNodeDefaultValue))
	xmlRegisterNodeDefaultValue((xmlNodePtr)cur);
    return(cur);
}

/**
 * xmlFreeDoc:
 * @cur:  pointer to the document
 *
 * Free up all the structures used by a document, tree included.
 */
void
xmlFreeDoc(xmlDocPtr cur) {
    xmlDtdPtr extSubset, intSubset;
    xmlDictPtr dict = NULL;

    if (cur == NULL) {
#ifdef DEBUG_TREE
        xmlGenericError(xmlGenericErrorContext,
		"xmlFreeDoc : document == NULL\n");
#endif
	return;
    }
    if (cur != NULL) dict = cur->dict;

    if ((__xmlRegisterCallbacks) && (xmlDeregisterNodeDefaultValue))
	xmlDeregisterNodeDefaultValue((xmlNodePtr)cur);

    /*
     * Do this before freeing the children list to avoid ID lookups
     */
    if (cur->ids != NULL) xmlFreeIDTable((xmlIDTablePtr) cur->ids);
    cur->ids = NULL;
    if (cur->refs != NULL) xmlFreeRefTable((xmlRefTablePtr) cur->refs);
    cur->refs = NULL;
    extSubset = cur->extSubset;
    intSubset = cur->intSubset;
    if (intSubset == extSubset)
	extSubset = NULL;
    if (extSubset != NULL) {
	xmlUnlinkNode((xmlNodePtr) cur->extSubset);
	cur->extSubset = NULL;
	xmlFreeDtd(extSubset);
    }
    if (intSubset != NULL) {
	xmlUnlinkNode((xmlNodePtr) cur->intSubset);
	cur->intSubset = NULL;
	xmlFreeDtd(intSubset);
    }

    if (cur->children != NULL) xmlFreeNodeList(cur->children);
    if (cur->oldNs != NULL) xmlFreeNsList(cur->oldNs);

    DICT_FREE(cur->version)
    DICT_FREE(cur->name)
    DICT_FREE(cur->encoding)
    DICT_FREE(cur->URL)
    xmlFree(cur);
    if (dict) xmlDictFree(dict);
}

/**
 * xmlStringLenGetNodeList:
 * @doc:  the document
 * @value:  the value of the text
 * @len:  the length of the string value
 *
 * Parse the value string and build the node list associated. Should
 * produce a flat tree with only TEXTs and ENTITY_REFs.
 * Returns a pointer to the first child
 */
xmlNodePtr
xmlStringLenGetNodeList(xmlDocPtr doc, const xmlChar *value, int len) {
    xmlNodePtr ret = NULL, last = NULL;
    xmlNodePtr node;
    xmlChar *val;
    const xmlChar *cur = value, *end = cur + len;
    const xmlChar *q;
    xmlEntityPtr ent;

    if (value == NULL) return(NULL);

    q = cur;
    while ((cur < end) && (*cur != 0)) {
	if (cur[0] == '&') {
	    int charval = 0;
	    xmlChar tmp;

	    /*
	     * Save the current text.
	     */
            if (cur != q) {
		if ((last != NULL) && (last->type == XML_TEXT_NODE)) {
		    xmlNodeAddContentLen(last, q, cur - q);
		} else {
		    node = xmlNewDocTextLen(doc, q, cur - q);
		    if (node == NULL) return(ret);
		    if (last == NULL)
			last = ret = node;
		    else {
			last->next = node;
			node->prev = last;
			last = node;
		    }
		}
	    }
	    q = cur;
	    if ((cur + 2 < end) && (cur[1] == '#') && (cur[2] == 'x')) {
		cur += 3;
		if (cur < end)
		    tmp = *cur;
		else
		    tmp = 0;
		while (tmp != ';') { /* Non input consuming loop */
		    if ((tmp >= '0') && (tmp <= '9')) 
			charval = charval * 16 + (tmp - '0');
		    else if ((tmp >= 'a') && (tmp <= 'f'))
			charval = charval * 16 + (tmp - 'a') + 10;
		    else if ((tmp >= 'A') && (tmp <= 'F'))
			charval = charval * 16 + (tmp - 'A') + 10;
		    else {
			xmlTreeErr(XML_TREE_INVALID_HEX, (xmlNodePtr) doc,
			           NULL);
			charval = 0;
			break;
		    }
		    cur++;
		    if (cur < end)
			tmp = *cur;
		    else
			tmp = 0;
		}
		if (tmp == ';')
		    cur++;
		q = cur;
	    } else if ((cur + 1 < end) && (cur[1] == '#')) {
		cur += 2;
		if (cur < end)
		    tmp = *cur;
		else
		    tmp = 0;
		while (tmp != ';') { /* Non input consuming loops */
		    if ((tmp >= '0') && (tmp <= '9')) 
			charval = charval * 10 + (tmp - '0');
		    else {
			xmlTreeErr(XML_TREE_INVALID_DEC, (xmlNodePtr) doc,
			           NULL);
			charval = 0;
			break;
		    }
		    cur++;
		    if (cur < end)
			tmp = *cur;
		    else
			tmp = 0;
		}
		if (tmp == ';')
		    cur++;
		q = cur;
	    } else {
		/*
		 * Read the entity string
		 */
		cur++;
		q = cur;
		while ((cur < end) && (*cur != 0) && (*cur != ';')) cur++;
		if ((cur >= end) || (*cur == 0)) {
		    xmlTreeErr(XML_TREE_UNTERMINATED_ENTITY, (xmlNodePtr) doc,
		               (const char *) q);
		    return(ret);
		}
		if (cur != q) {
		    /*
		     * Predefined entities don't generate nodes
		     */
		    val = xmlStrndup(q, cur - q);
		    ent = xmlGetDocEntity(doc, val);
		    if ((ent != NULL) &&
			(ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) {
			if (last == NULL) {
			    node = xmlNewDocText(doc, ent->content);
			    last = ret = node;
			} else if (last->type != XML_TEXT_NODE) {
			    node = xmlNewDocText(doc, ent->content);
			    last = xmlAddNextSibling(last, node);
			} else
			    xmlNodeAddContent(last, ent->content);
			    
		    } else {
			/*
			 * Create a new REFERENCE_REF node
			 */
			node = xmlNewReference(doc, val);
			if (node == NULL) {
			    if (val != NULL) xmlFree(val);
			    return(ret);
			}
			else if ((ent != NULL) && (ent->children == NULL)) {
			    xmlNodePtr temp;

			    ent->children = xmlStringGetNodeList(doc,
				    (const xmlChar*)node->content);
			    ent->owner = 1;
			    temp = ent->children;
			    while (temp) {
				temp->parent = (xmlNodePtr)ent;
				temp = temp->next;
			    }
			}
			if (last == NULL) {
			    last = ret = node;
			} else {
			    last = xmlAddNextSibling(last, node);
			}
		    }
		    xmlFree(val);
		}
		cur++;
		q = cur;
	    }
	    if (charval != 0) {
		xmlChar buf[10];
		int l;

		l = xmlCopyCharMultiByte(buf, charval);
		buf[l] = 0;
		node = xmlNewDocText(doc, buf);
		if (node != NULL) {
		    if (last == NULL) {
			last = ret = node;
		    } else {
			last = xmlAddNextSibling(last, node);
		    }
		}
		charval = 0;
	    }
	} else
	    cur++;
    }
    if ((cur != q) || (ret == NULL)) {
        /*
	 * Handle the last piece of text.
	 */
	if ((last != NULL) && (last->type == XML_TEXT_NODE)) {
	    xmlNodeAddContentLen(last, q, cur - q);
	} else {
	    node = xmlNewDocTextLen(doc, q, cur - q);
	    if (node == NULL) return(ret);
	    if (last == NULL) {
		last = ret = node;
	    } else {
		last = xmlAddNextSibling(last, node);
	    }
	}
    }
    return(ret);
}

/**
 * xmlStringGetNodeList:
 * @doc:  the document
 * @value:  the value of the attribute
 *
 * Parse the value string and build the node list associated. Should
 * produce a flat tree with only TEXTs and ENTITY_REFs.
 * Returns a pointer to the first child
 */
xmlNodePtr
xmlStringGetNodeList(xmlDocPtr doc, const xmlChar *value) {
    xmlNodePtr ret = NULL, last = NULL;
    xmlNodePtr node;
    xmlChar *val;
    const xmlChar *cur = value;
    const xmlChar *q;
    xmlEntityPtr ent;

    if (value == NULL) return(NULL);

    q = cur;
    while (*cur != 0) {
	if (cur[0] == '&') {
	    int charval = 0;
	    xmlChar tmp;

	    /*
	     * Save the current text.
	     */
            if (cur != q) {
		if ((last != NULL) && (last->type == XML_TEXT_NODE)) {
		    xmlNodeAddContentLen(last, q, cur - q);
		} else {
		    node = xmlNewDocTextLen(doc, q, cur - q);
		    if (node == NULL) return(ret);
		    if (last == NULL)
			last = ret = node;
		    else {
			last->next = node;
			node->prev = last;
			last = node;
		    }
		}
	    }
	    q = cur;
	    if ((cur[1] == '#') && (cur[2] == 'x')) {
		cur += 3;
		tmp = *cur;
		while (tmp != ';') { /* Non input consuming loop */
		    if ((tmp >= '0') && (tmp <= '9')) 
			charval = charval * 16 + (tmp - '0');
		    else if ((tmp >= 'a') && (tmp <= 'f'))
			charval = charval * 16 + (tmp - 'a') + 10;
		    else if ((tmp >= 'A') && (tmp <= 'F'))
			charval = charval * 16 + (tmp - 'A') + 10;
		    else {
			xmlTreeErr(XML_TREE_INVALID_HEX, (xmlNodePtr) doc,
			           NULL);
			charval = 0;
			break;
		    }
		    cur++;
		    tmp = *cur;
		}
		if (tmp == ';')
		    cur++;
		q = cur;
	    } else if  (cur[1] == '#') {
		cur += 2;
		tmp = *cur;
		while (tmp != ';') { /* Non input consuming loops */
		    if ((tmp >= '0') && (tmp <= '9')) 
			charval = charval * 10 + (tmp - '0');
		    else {
			xmlTreeErr(XML_TREE_INVALID_DEC, (xmlNodePtr) doc,
			           NULL);
			charval = 0;
			break;
		    }
		    cur++;
		    tmp = *cur;
		}
		if (tmp == ';')
		    cur++;
		q = cur;
	    } else {
		/*
		 * Read the entity string
		 */
		cur++;
		q = cur;
		while ((*cur != 0) && (*cur != ';')) cur++;
		if (*cur == 0) {
		    xmlTreeErr(XML_TREE_UNTERMINATED_ENTITY,
		               (xmlNodePtr) doc, (const char *) q);
		    return(ret);
		}
		if (cur != q) {
		    /*
		     * Predefined entities don't generate nodes
		     */
		    val = xmlStrndup(q, cur - q);
		    ent = xmlGetDocEntity(doc, val);
		    if ((ent != NULL) &&
			(ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) {
			if (last == NULL) {
			    node = xmlNewDocText(doc, ent->content);
			    last = ret = node;
			} else if (last->type != XML_TEXT_NODE) {
			    node = xmlNewDocText(doc, ent->content);
			    last = xmlAddNextSibling(last, node);
			} else
			    xmlNodeAddContent(last, ent->content);
			    
		    } else {
			/*
			 * Create a new REFERENCE_REF node
			 */
			node = xmlNewReference(doc, val);
			if (node == NULL) {
			    if (val != NULL) xmlFree(val);
			    return(ret);
			}
			else if ((ent != NULL) && (ent->children == NULL)) {
			    xmlNodePtr temp;

			    ent->children = xmlStringGetNodeList(doc,
				    (const xmlChar*)node->content);
			    ent->owner = 1;
			    temp = ent->children;
			    while (temp) {
				temp->parent = (xmlNodePtr)ent;
				temp = temp->next;
			    }
			}
			if (last == NULL) {
			    last = ret = node;
			} else {
			    last = xmlAddNextSibling(last, node);
			}
		    }
		    xmlFree(val);
		}
		cur++;
		q = cur;
	    }
	    if (charval != 0) {
		xmlChar buf[10];
		int len;

		len = xmlCopyCharMultiByte(buf, charval);
		buf[len] = 0;
		node = xmlNewDocText(doc, buf);
		if (node != NULL) {
		    if (last == NULL) {
			last = ret = node;
		    } else {
			last = xmlAddNextSibling(last, node);
		    }
		}

		charval = 0;
	    }
	} else
	    cur++;
    }
    if ((cur != q) || (ret == NULL)) {
        /*
	 * Handle the last piece of text.
	 */
	if ((last != NULL) && (last->type == XML_TEXT_NODE)) {
	    xmlNodeAddContentLen(last, q, cur - q);
	} else {
	    node = xmlNewDocTextLen(doc, q, cur - q);
	    if (node == NULL) return(ret);
	    if (last == NULL) {
		last = ret = node;
	    } else {
		last = xmlAddNextSibling(last, node);
	    }
	}
    }
    return(ret);
}

/**
 * xmlNodeListGetString:
 * @doc:  the document
 * @list:  a Node list
 * @inLine:  should we replace entity contents or show their external form
 *
 * Build the string equivalent to the text contained in the Node list
 * made of TEXTs and ENTITY_REFs
 *
 * Returns a pointer to the string copy, the caller must free it with xmlFree().
 */
xmlChar *
xmlNodeListGetString(xmlDocPtr doc, xmlNodePtr list, int inLine)
{
    xmlNodePtr node = list;
    xmlChar *ret = NULL;
    xmlEntityPtr ent;

    if (list == NULL)
        return (NULL);

    while (node != NULL) {
        if ((node->type == XML_TEXT_NODE) ||
            (node->type == XML_CDATA_SECTION_NODE)) {
            if (inLine) {
                ret = xmlStrcat(ret, node->content);
            } else {
                xmlChar *buffer;

                buffer = xmlEncodeEntitiesReentrant(doc, node->content);
                if (buffer != NULL) {
                    ret = xmlStrcat(ret, buffer);
                    xmlFree(buffer);
                }
            }
        } else if (node->type == XML_ENTITY_REF_NODE) {
            if (inLine) {
                ent = xmlGetDocEntity(doc, node->name);
                if (ent != NULL) {
                    xmlChar *buffer;

                    /* an entity content can be any "well balanced chunk",
                     * i.e. the result of the content [43] production:
                     * http://www.w3.org/TR/REC-xml#NT-content.
                     * So it can contain text, CDATA section or nested
                     * entity reference nodes (among others).
                     * -> we recursive  call xmlNodeListGetString()
                     * which handles these types */
                    buffer = xmlNodeListGetString(doc, ent->children, 1);
                    if (buffer != NULL) {
                        ret = xmlStrcat(ret, buffer);
                        xmlFree(buffer);
                    }
                } else {
                    ret = xmlStrcat(ret, node->content);
                }
            } else {
                xmlChar buf[2];

                buf[0] = '&';
                buf[1] = 0;
                ret = xmlStrncat(ret, buf, 1);
                ret = xmlStrcat(ret, node->name);
                buf[0] = ';';
                buf[1] = 0;
                ret = xmlStrncat(ret, buf, 1);
            }
        }
#if 0
        else {
            xmlGenericError(xmlGenericErrorContext,
                            "xmlGetNodeListString : invalid node type %d\n",
                            node->type);
        }
#endif
        node = node->next;
    }
    return (ret);