X-Git-Url: http://git.nguyen.vg/gitweb/?a=blobdiff_plain;f=tree.ml;h=6c3cc1bf483f0a650010a1524a75269c2263562e;hb=7dea5fd8bedede27d4d601f85630a249bfab420b;hp=9cab2c791f45cf72d13fe8881c6043296495f73f;hpb=6b03c8ef3dac4b1de06ca577e8e0ee07c6c3eae7;p=SXSI%2Fxpathcomp.git diff --git a/tree.ml b/tree.ml index 9cab2c7..6c3cc1b 100644 --- a/tree.ml +++ b/tree.ml @@ -26,11 +26,16 @@ sig val print_xml_fast : out_channel -> t -> unit val compare : t -> t -> int val equal : t -> t -> bool - module DocIdSet : Set.S with type elt = string_content + module DocIdSet : + sig + include Set.S + end + with type elt = string_content val string_below : t -> string_content -> bool val contains : t -> string -> DocIdSet.t val contains_old : t -> string -> bool val dump : t -> unit + val get_string : t -> string_content -> string end module XML = @@ -56,7 +61,8 @@ struct module Text = struct - + let equal : [`Text] node -> [`Text] node -> bool = equal + (* Todo *) external nullt : unit -> [`Text ] node = "caml_xml_tree_nullt" let nil = nullt () @@ -79,7 +85,7 @@ struct module Tree = struct - + let equal : [`Tree ] node -> [`Tree] node -> bool = equal external serialize : t -> string -> unit = "caml_xml_tree_serialize" external unserialize : string -> t = "caml_xml_tree_unserialize" @@ -120,7 +126,7 @@ struct then Printf.eprintf "#\n" else begin - Printf.eprintf "Node %i has tag '%s' DocOrder=%i, DocID of PrevText,MyText,NextText : (%i = %s,%i = %s,%i = %s)\n%!" + Printf.eprintf "Node %i has tag '%s' DocOrder=%i, DocID of PrevText,MyText,NextText : (%i = %s,%i = %s,%i = %s) parent_doc(my_text)=%i\n%!" (int_of_node id) (Tag.to_string (tag_id t id)) (node_xml_id t id) @@ -129,7 +135,9 @@ struct (int_of_node (my_text t id)) (Text.get_text t (my_text t id)) (int_of_node (next_text t id)) - (Text.get_text t (next_text t id)); + (Text.get_text t (next_text t id)) + (int_of_node(parent_doc t (my_text t id))); + aux(first_child t id); aux(next_sibling t id); end @@ -172,10 +180,12 @@ struct node : descr } let dump { doc=t } = Tree.print_skel t - module DocIdSet = Set.Make (struct type t = string_content - let compare = (-) end) - - + module DocIdSet = struct + include Set.Make (struct type t = string_content + let compare = (-) end) + + end + let get_string t (i:string_content) = Text.get_text t.doc i open Tree let node_of_t t = { doc= t; node = Node(NC (root t)) } @@ -296,11 +306,13 @@ struct | _ -> () *) let string_below t id = - let pid = parent_doc t.doc id in + let strid = parent_doc t.doc id in match t.node with - | Node(NC(i)) -> (is_ancestor t.doc i pid) - | Node(SC(i,_)) -> (is_ancestor t.doc (parent_doc t.doc i) pid) + | Node(NC(i)) -> + (Tree.equal i strid) || (is_ancestor t.doc i strid) + | Node(SC(i,_)) -> Text.equal i id | _ -> false + let contains t s = Array.fold_left (fun a i -> DocIdSet.add i a) DocIdSet.empty (Text.contains t.doc s) @@ -324,7 +336,9 @@ struct let rec loop ?(print_right=true) t = match t.node with | Nil -> () | String (s) -> output_string outc (string t) - | Node _ when Tag.equal (tag t) Tag.pcdata -> loop (left t); loop (right t) + | Node _ when Tag.equal (tag t) Tag.pcdata -> + loop (left t); + if print_right then loop (right t) | Node (_) -> let tg = Tag.to_string (tag t) in