Parse HTML document to retrieve title.

ocaml/shell.ml

@@ -15,9 +15,15 @@ let is_youtube_url url =
  * iter on all the items in the list *)
 let evaluate str =
   match str with
-  | str when is_youtube_url str -> SimpleHttp.get_http_title (SimpleHttp.get_body str); str
-  | str when is_url str -> SimpleHttp.get_http_title (SimpleHttp.get_body str); str
-  | _ -> str
+  | str when is_youtube_url str ->
+      (
+        let title = SimpleHttp.get_http_title (SimpleHttp.get_body str) in
+        match title with
+        | Some s -> s
+        | None -> ""
+      )
+  | str when is_url str -> str
+  | _ -> ""
 
 
 let () =

ocaml/simpleHttp.ml

@@ -14,20 +14,54 @@ Convenience.configure_pipeline
 
 let extract_string_value document =
   match document with
-  | Data(s) -> s
-  | _ -> ""
+  | Data(s) -> Some s
+  | _ -> None
 
 
-let rec get_title_element document =
+let find_string_value = function
+  |  Some s -> true
+  | None -> false
+
+
+let rec get_title_element_from_list doc_list =
+
+  try
+    List.find find_string_value (List.map get_title_element doc_list)
+  with
+    Not_found -> None
+
+and get_title_element document =
+
   match document with
-  | Element(e, args, sub) -> printf "%s: %s" "Element\n" e
-  | Data(s) -> printf "%s: %s" "Data\n" s
+  | Element("title", args, sub) ->
+      (
+        let title_candidates = List.map extract_string_value sub in
+        try
+          List.find find_string_value title_candidates
+        with
+          Not_found -> None
+      )
+  | Element(e, args, sub) -> get_title_element_from_list sub
+  | Data(s) -> None
+
+
+let rec print_document document =
+  match document with
+  | Element(e, args, sub) ->
+      printf "Element: %s\n" e;
+      List.iter print_document sub
+  | Data(s) -> printf "Data: %s\n" s
+
+
+let get_http_document body_str =
+  let ch = new Netchannels.input_string body_str in
+  Nethtml.parse ch
 
 
 let get_http_title body =
   let ch = new Netchannels.input_string body in
   let doc = Nethtml.parse ch in
-  get_title_element (List.hd doc)
+  get_title_element_from_list doc
 
 
 (* TODO: Log errors *)