The links table now records links to non-existant pages.

diff --git a/scripts/cocanwiki_links.ml b/scripts/cocanwiki_links.ml
index 9d856d2..101ca64 100644 (file)
--- a/scripts/cocanwiki_links.ml
+++ b/scripts/cocanwiki_links.ml
@@ -1,7 +1,7 @@
 (* COCANWIKI - a wiki written in Objective CAML.
  * Written by Richard W.M. Jones <rich@merjis.com>.
  * Copyright (C) 2004 Merjis Ltd.
- * $Id: cocanwiki_links.ml,v 1.2 2004/09/28 11:28:39 rich Exp $
+ * $Id: cocanwiki_links.ml,v 1.3 2004/10/07 12:22:11 rich Exp $
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -20,10 +20,13 @@
  *)
 
 open ExtString
+open ExtList
 
 let split_tags_re = Pcre.regexp ~flags:[`DOTALL] "<.*?>|[^<]+"
 let internal_re = Pcre.regexp "class=\"internal\""
+let newpage_re = Pcre.regexp "class=\"newpage\""
 let href_re = Pcre.regexp "href=\"/(.*?)\""
+let title_re = Pcre.regexp "title=\"(.*?)\""
 
 let get_links_from_section dbh hostid content =
   let html = Wikilib.xhtml_of_content dbh hostid content in
@@ -42,22 +45,45 @@ let get_links_from_section dbh hostid content =
   (* Only interested in the <a> tags. *)
   let html = List.filter (fun str -> String.starts_with str "<a ") html in
 
-  (* Only interested in the tags with class="internal". *)
-  let html =
-    List.filter (fun str ->
-                  Pcre.pmatch ~rex:internal_re str
-                  && Pcre.pmatch ~rex:href_re str)
-      html in
-
-  (* Extract the URL names. *)
-  let links = List.map (fun str ->
-                         let subs =
-                           try Pcre.exec ~rex:href_re str
-                           with Not_found -> assert false in
-                         Pcre.get_substring subs 1) html in
-
-  (* Return the list of links. *)
-  links
+  (* Only interested in the tags with class="internal" or class="newpage". *)
+  let internal_links =
+    let html =
+      List.filter (fun str ->
+                    Pcre.pmatch ~rex:internal_re str
+                    && Pcre.pmatch ~rex:href_re str)
+       html in
+
+    (* Extract the URL names. *)
+    List.map (fun str ->
+               let subs =
+                 try Pcre.exec ~rex:href_re str
+                 with Not_found -> assert false in
+               Pcre.get_substring subs 1) html in
+
+  let newpage_links =
+    let html =
+      List.filter (fun str ->
+                    Pcre.pmatch ~rex:newpage_re str
+                    && Pcre.pmatch ~rex:title_re str)
+       html in
+
+    (* Extract the titles. *)
+    let titles =
+      List.map (fun str ->
+                 let subs =
+                   try Pcre.exec ~rex:title_re str
+                   with Not_found -> assert false in
+                 Pcre.get_substring subs 1) html in
+
+    (* Map the titles to URLs. *)
+    List.filter_map
+      (fun title ->
+        match Wikilib.generate_url_of_title dbh hostid title with
+          | Wikilib.GenURL_OK url -> Some url
+          | _ -> None) titles in
+
+  (* Return the complete list of links. *)
+  internal_links @ newpage_links
 
 let insert_link dbh hostid from_url to_url =
   if from_url <> to_url then (