#include <cstdint>
#include <unordered_set>
#include <unordered_map>
-#include <bp/bp.h>
-#include <bp/bp-darray.h>
+#include <libbp/bp.h>
+#include <libbp/bp-darray.h>
#include <libcds/includes/basics.h>
#include <libcds/includes/static_bitsequence.h>
#include <libcds/includes/alphabet_mapper.h>
~xml_tree();
//Counting functions
+ /**
+ * [size()] returns the size of the tree (number of nodes)
+ * Runs in O(1)
+ */
inline uint32_t size() const;
+
+ /**
+ * [num_tags()] returns the number of distinct tags.
+ * Runs in O(1)
+ */
inline uint32_t num_tags() const;
+
+ /**
+ * [subtree_size(n)] returns the size of the subtree (number of nodes)
+ * rooted at n.
+ * Runs in O(1)
+ */
inline uint32_t subtree_size(node_t) const;
+
+ /**
+ * [subtree_tags(n, t)] returns the number of occurences of tag [t] in the
+ * subtree rooted at [n]
+ * Runs in O(1)
+ */
inline uint32_t subtree_tags(node_t, tag_t) const;
- uint32_t subtree_elements(node_t) const;
+
+ /**
+ * [subtree_elements(n)] returns the number of element nodes below [n]
+ * Runs in O(attribute_ids->size()+3)
+ */
+ inline uint32_t subtree_elements(node_t) const;
+
+ /**
+ * [num_children(n)] returns the number of child nodes of [n]
+ * (both text and elements, and including a fake <@> node if
+ * present).
+ * Runs in O(1) (?)
+ */
uint32_t num_children(node_t) const;
+
+ /**
+ * [child_pos(n)] returns the position of [n] amongst its siblings
+ * Runs in O(1) (?)
+ */
uint32_t child_pos(node_t) const;
bool is_child(node_t, node_t) const;
inline bool is_first_child(node_t) const;
inline bool is_nil(node_t) const;
+ inline bool is_open(node_t) const;
uint32_t depth(node_t) const;
uint32_t preorder(node_t) const;
inline node_t prev_sibling(node_t) const;
inline node_t first_element(node_t) const;
inline node_t next_element(node_t) const;
+ inline node_t tagged_next_close(node_t, tag_t) const;
inline node_t tagged_next(node_t, tag_t) const;
inline node_t tagged_descendant(node_t, tag_t) const;
inline node_t tagged_following_before(node_t, tag_t, node_t) const;
inline node_t tagged_child(node_t, tag_t) const;
inline node_t tagged_sibling(node_t, tag_t) const;
- node_t select_child(node_t, std::unordered_set<tag_t>*) const;
- node_t select_descendant(node_t, std::unordered_set<tag_t>*) const;
- node_t select_sibling(node_t, std::unordered_set<tag_t>*) const;
- node_t select_following_before (node_t,
- std::unordered_set<tag_t>*, node_t) const;
+ node_t select_child(node_t, tag_t*) const;
+ inline node_t select_descendant(node_t, tag_t*) const;
+ node_t select_sibling(node_t, tag_t*) const;
+ inline node_t select_following_before (node_t, tag_t*, node_t) const;
inline node_t closing(node_t) const;
//Text functions
SXSI::TextCollection::document_result contains(uchar const *s) const;
SXSI::TextCollection::document_result less_than(uchar const *s) const;
+
+ bool naive_contains(node_t, uchar const *s) const;
+
//I/O functions
void save(int, char*);
static xml_tree* load(int, char*, bool, int);
//Parenthesis sequence
bp *par;
//tag sequence
- static_sequence *tags;
+ std::vector<static_bitsequence_sdarray*> tags;
uint32_t *tag_seq;
uint32_t tag_seq_len;
uint32_t bits_per_tag;
//Mapping from tag_t identifiers to/from tagnames
std::vector<std::string> *tag_names;
std::unordered_map<std::string, tag_t> *tag_ids;
+ //Set of tag ids that map to attribute nodes
+ std::unordered_set<tag_t> *attribute_ids;
//Text index
SXSI::TextCollection *text_collection;
static_bitsequence *text_positions;