+INCLUDE "utils.ml"
+
+
+external init_lib : unit -> unit = "caml_init_lib"
+
+exception CPlusPlusError of string
+
+let () = Callback.register_exception "CPlusPlusError" (CPlusPlusError "")
+
+let () = init_lib ()
+
+
+type tree
+type 'a node = private int
+type node_kind = [`Text | `Tree ]
+
+type t = {
+ doc : tree;
+ ttable : (Tag.t,(Ptset.Int.t*Ptset.Int.t*Ptset.Int.t*Ptset.Int.t)) Hashtbl.t;
+}
+
+external inode : 'a node -> int = "%identity"
+external nodei : int -> 'a node = "%identity"
+let compare_node x y = (inode x) - (inode y)
+let equal_node : 'a node -> 'a node -> bool = (==)
+
+
+external parse_xml_uri : string -> int -> bool -> bool -> tree = "caml_call_shredder_uri"
+external parse_xml_string : string -> int -> bool -> bool -> tree = "caml_call_shredder_string"
+external tree_print_xml_fast3 : tree -> [`Tree ] node -> Unix.file_descr ->unit = "caml_xml_tree_print"
+external tree_save : tree -> Unix.file_descr -> string -> unit = "caml_xml_tree_save"
+external tree_load : Unix.file_descr -> string -> bool -> int -> tree = "caml_xml_tree_load"
+
+external nullt : unit -> 'a node = "caml_xml_tree_nullt"
+
+let nil : [`Tree ] node = nodei ~-1
+let nulldoc : [`Text ] node = nodei ~-1
+let root : [`Tree ] node = nodei 0
+
+external text_get_text : tree -> [`Text] node -> string = "caml_text_collection_get_text"
+external text_is_empty : tree -> [`Text ] node -> bool = "caml_text_collection_empty_text"
+
+let text_is_empty t n = (equal_node nulldoc n) || text_is_empty t n
+
+external text_is_prefix : tree -> string -> bool = "caml_text_collection_is_prefix"
+external text_is_suffix : tree -> string -> bool = "caml_text_collection_is_suffix"
+external text_is_equal : tree -> string -> bool = "caml_text_collection_is_equal"
+external text_is_contains : tree -> string -> bool = "caml_text_collection_is_contains"
+external text_is_lessthan : tree -> string -> bool = "caml_text_collection_is_lessthan"
+
+external text_count : tree -> string -> int = "caml_text_collection_count"
+external text_count_prefix : tree -> string -> int = "caml_text_collection_count_prefix"
+external text_count_suffix : tree -> string -> int = "caml_text_collection_count_suffix"
+external text_count_equal : tree -> string -> int = "caml_text_collection_count_equal"
+external text_count_contains : tree -> string -> int = "caml_text_collection_count_contains"
+external text_count_lessthan : tree -> string -> int = "caml_text_collection_count_lessthan"
+
+external text_prefix : tree -> string -> [`Text ] node array = "caml_text_collection_prefix"
+external text_suffix : tree -> string -> [`Text ] node array = "caml_text_collection_suffix"
+external text_equals : tree -> string -> [`Text ] node array = "caml_text_collection_equals"
+external text_contains : tree -> string -> [`Text ] node array = "caml_text_collection_contains"
+external text_lessthan : tree -> string -> [`Text ] node array = "caml_text_collection_lessthan"
+
+
+external tree_root : tree -> [`Tree] node = "caml_xml_tree_root" "noalloc"
+external tree_size : tree -> int = "caml_xml_tree_size" "noalloc"
+external tree_subtree_size : tree -> [`Tree] node -> int = "caml_xml_tree_subtree_size" "noalloc"
+external tree_subtree_elements : tree -> [`Tree] node -> int = "caml_xml_tree_subtree_elements" "noalloc"
+external tree_subtree_tags : tree -> [`Tree] node -> Tag.t -> int = "caml_xml_tree_subtree_elements" "noalloc"
+
+let tree_is_nil x = equal_node x nil
+external tree_is_leaf : tree -> [`Tree ] node -> bool = "caml_xml_tree_is_leaf" "noalloc"
+external tree_is_ancestor : tree -> [`Tree ] node -> [`Tree ] node -> bool = "caml_xml_tree_is_ancestor" "noalloc"
+external tree_is_child : tree -> [`Tree ] node -> [`Tree ] node -> bool = "caml_xml_tree_is_child" "noalloc"
+external tree_is_first_child : tree -> [`Tree ] node -> bool = "caml_xml_tree_is_first_child" "noalloc"
+external tree_num_children : tree -> [`Tree ] node -> int = "caml_xml_tree_num_children" "noalloc"
+external tree_child_number : tree -> [`Tree ] node -> int = "caml_xml_tree_child_number" "noalloc"
+external tree_depth : tree -> [`Tree ] node -> int = "caml_xml_tree_depth" "noalloc"
+external tree_preorder : tree -> [`Tree ] node -> int = "caml_xml_tree_preorder" "noalloc"
+external tree_postorder : tree -> [`Tree ] node -> int = "caml_xml_tree_postorder" "noalloc"
+external tree_tag : tree -> [`Tree ] node -> Tag.t = "caml_xml_tree_tag" "noalloc"
+external tree_doc_ids : tree -> [`Tree ] node -> [`Text] node*[`Text] node = "caml_xml_tree_doc_ids"
+
+external tree_parent : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_parent" "noalloc"
+external tree_child : tree -> [`Tree] node -> int -> [`Tree] node = "caml_xml_tree_child" "noalloc"
+external tree_first_child : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_first_child" "noalloc"
+external tree_first_element : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_first_element" "noalloc"
+external tree_last_child : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_last_child" "noalloc"
+external tree_next_sibling : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_next_sibling" "noalloc"
+external tree_next_element : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_next_element" "noalloc"
+external tree_prev_sibling : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_prev_sibling" "noalloc"
+external tree_tagged_child : tree -> [`Tree] node -> Tag.t -> [`Tree] node = "caml_xml_tree_tagged_child" "noalloc"
+
+type unordered_set
+external unordered_set_alloc : int -> unordered_set = "caml_unordered_set_alloc"
+external unordered_set_length : unordered_set -> int = "caml_unordered_set_length"
+external unordered_set_insert : unordered_set -> int -> unit = "caml_unordered_set_set" "noalloc"
+
+external tree_select_child : tree -> [`Tree ] node -> unordered_set -> [`Tree] node = "caml_xml_tree_select_child" "noalloc"
+external tree_tagged_following_sibling : tree -> [`Tree] node -> Tag.t -> [`Tree] node = "caml_xml_tree_tagged_following_sibling" "noalloc"
+external tree_select_following_sibling : tree -> [`Tree ] node -> unordered_set -> [`Tree] node = "caml_xml_tree_select_following_sibling" "noalloc"
+external tree_tagged_descendant : tree -> [`Tree ] node -> Tag.t -> [`Tree ] node = "caml_xml_tree_tagged_descendant" "noalloc"
+external tree_select_descendant : tree -> [`Tree ] node -> unordered_set -> [`Tree] node = "caml_xml_tree_select_descendant" "noalloc"
+external tree_tagged_following : tree -> [`Tree ] node -> Tag.t -> [`Tree ] node = "caml_xml_tree_tagged_following" "noalloc"
+external tree_tagged_following_below : tree -> [`Tree ] node -> Tag.t -> [`Tree ] node -> [`Tree ] node = "caml_xml_tree_tagged_following_below" "noalloc"
+external tree_select_following_below : tree -> [`Tree ] node -> unordered_set -> [`Tree] node -> [`Tree] node = "caml_xml_tree_select_following_below" "noalloc"
+
+
+external tree_tagged_following_before : tree -> [`Tree ] node -> Tag.t -> [`Tree ] node -> [`Tree ] node = "caml_xml_tree_tagged_following_before" "noalloc"
+external tree_select_following_below : tree -> [`Tree ] node -> unordered_set -> [`Tree] node -> [`Tree] node = "caml_xml_tree_select_following_before" "noalloc"
+
+external tree_my_text : tree -> [`Tree ] node -> [`Text] node = "caml_xml_tree_my_text" "noalloc"
+external tree_my_text_unsafe : tree -> [`Tree ] node -> [`Text] node = "caml_xml_tree_my_text_unsafe" "noalloc"
+external tree_text_xml_id : tree -> [`Text ] node -> int = "caml_xml_tree_text_xml_id" "noalloc"
+external tree_node_xml_id : tree -> [`Tree ] node -> int = "caml_xml_tree_node_xml_id" "noalloc"
+
+external tree_parent_node : tree -> [`Text ] node -> [`Tree ] node = "caml_xml_tree_parent_node" "noalloc"
+
+(*external tree_prev_doc : tree -> [`Text ] node -> [`Tree ] node = "caml_xml_tree_prev_doc" "noalloc" *)
+
+external tree_closing : tree -> [`Tree] node -> [`Tree] node = "caml_xml_tree_closing" "noalloc"
+external tree_is_open : tree -> [`Tree] node -> bool = "caml_xml_tree_is_open" "noalloc"
+
+
+external benchmark_jump : tree -> Tag.t -> int = "caml_benchmark_jump" "noalloc"
+
+let benchmark_jump t s = benchmark_jump t.doc s
+
+external benchmark_fcns : tree -> int = "caml_benchmark_fcns" "noalloc"
+external benchmark_fene : tree -> int = "caml_benchmark_fene" "noalloc"
+external benchmark_iter : tree -> int = "caml_benchmark_iter" "noalloc"
+
+let benchmark_fcns t = benchmark_fcns t.doc
+
+let benchmark_fene t = benchmark_fene t.doc
+
+let benchmark_iter t = benchmark_iter t.doc
+
+external benchmark_lcps : tree -> unit = "caml_benchmark_lcps" "noalloc"
+
+let benchmark_lcps t = benchmark_lcps t.doc
+
+
+
+
+
+
+
+let text_size tree = inode (snd ( tree_doc_ids tree root ))
+
+let text_get_text t (x:[`Text] node) =
+ if x == nulldoc then ""
+ else text_get_text t x
+
+
+
+
+module HPtset = Hashtbl.Make(Ptset.Int)
+
+let vector_htbl = HPtset.create MED_H_SIZE
+
+let ptset_to_vector s =
+ try
+ HPtset.find vector_htbl s
+ with
+ Not_found ->
+ let v = unordered_set_alloc (Ptset.Int.cardinal s) in
+ let _ = Ptset.Int.iter (fun e -> unordered_set_insert v e) s in
+ HPtset.add vector_htbl s v; v
+
+
+
+let subtree_size t i = tree_subtree_size t.doc i
+let subtree_elements t i = tree_subtree_elements t.doc i
+let text_size t = text_size t.doc
+
+module MemUnion = Hashtbl.Make (struct
+ type t = Ptset.Int.t*Ptset.Int.t
+ let equal (x,y) (z,t) = x == z && y == t
+ let equal a b = equal a b || equal b a
+ let hash (x,y) = (* commutative hash *)
+ let x = Uid.to_int (Ptset.Int.uid x)
+ and y = Uid.to_int (Ptset.Int.uid y)
+ in
+ if x <= y then HASHINT2(x,y) else HASHINT2(y,x)
+ end)
+
+module MemAdd = Hashtbl.Make (
+ struct
+ type t = Tag.t*Ptset.Int.t
+ let equal (x,y) (z,t) = (x == z)&&(y == t)
+ let hash (x,y) = HASHINT2(x,Uid.to_int (Ptset.Int.uid y))
+ end)
+
+module MemUpdate = struct
+include Hashtbl.Make (
+ struct
+ type t = Tag.t*Ptset.Int.t*Ptset.Int.t*Ptset.Int.t*Ptset.Int.t
+ let equal (a1,b1,c1,d1,e1) (a2,b2,c2,d2,e2) = a1==a2 &&
+ b1 == b2 && c1 == c2 && d1 == d2 && e1 == e2
+ let hash (a,b,c,d,e) =
+ HASHINT4(HASHINT2(a,Uid.to_int (Ptset.Int.uid b)),
+ Uid.to_int (Ptset.Int.uid c),
+ Uid.to_int (Ptset.Int.uid d),
+ Uid.to_int (Ptset.Int.uid e))
+ end)
+