-/* Licensed under GPL - see LICENSE file for details */
+#ifndef CCAN_TTXML_H
+#define CCAN_TTXML_H
+
+/**
+ * ttxml - tiny XML library for parsing (trusted!) XML documents.
+ *
+ * This parses an XML file into a convenient data structure.
+ *
+ * Example:
+ * #include <ccan/ttxml/ttxml.h>
+ * #include <stdio.h>
+ *
+ * int main(int argc, char *argv[])
+ * {
+ * XmlNode *xml, *tmp;
+ *
+ * xml = xml_load("./test/test.xml2");
+ * if(!xml)return 1;
+ *
+ * tmp = xml_find(xml, "childnode");
+ *
+ * printf("%s: %s\n", xml->name, xml_attr(tmp, "attribute"));
+ *
+ * xml_free(xml);
+ *
+ * return 0;
+ * }
+ *
+ * Licensed under GPL - see LICENSE file for details.
+ * Author: Daniel Burke <dan.p.burke@gmail.com>
+ */
+
+/* Every node is one of these */
typedef struct XmlNode {
char * name;
char ** attrib;
struct XmlNode * next;
} XmlNode;
+/* It's all pretty straight forward except for the attrib.
+ *
+ * Attrib is an array of char*, that is 2x the size of nattrib.
+ * Each pair of char* points to the attribute name & the attribute value,
+ * if present.
+ *
+ * If it's a text node, then name = "text", and attrib[1] = the body of text.
+ * This is the only case where there will be an attribute with a null name.
+ */
-XmlNode* xml_new(char * name, char * attrib);
XmlNode* xml_load(const char * filename);
void xml_free(XmlNode *target);
char* xml_attr(XmlNode *x, const char *name);
XmlNode * xml_find(XmlNode *xml, const char *name);
+#endif /* CCAN_TTXML_H */