1 (* COCANWIKI - a wiki written in Objective CAML.
2 * Written by Richard W.M. Jones <rich@merjis.com>.
3 * Copyright (C) 2004 Merjis Ltd.
4 * $Id: page.ml,v 1.30 2004/10/10 16:31:52 rich Exp $
6 * This program is free software; you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License as published by
8 * the Free Software Foundation; either version 2 of the License, or
9 * (at your option) any later version.
11 * This program is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License
17 * along with this program; see the file COPYING. If not, write to
18 * the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
19 * Boston, MA 02111-1307, USA.
30 open Cocanwiki_template
33 open Cocanwiki_server_settings
36 type fp_status = FPOK of int * string * string * Dbi.datetime * bool
37 | FPRedirect of string
40 (* Referer strings which help us decide if the user came from
41 * a search engine and highlight terms in the page appropriately.
43 let search_engines = [
44 Pcre.regexp "^http://.*google\\.", [ "q"; "as_q"; "as_epq"; "as_oq" ];
45 Pcre.regexp "^http://.*yahoo\\.", [ "p" ];
46 Pcre.regexp "^http://.*msn\\.", [ "q"; "MT" ]
48 let split_words = Pcre.regexp "\\W+"
50 let split_qs_re = Pcre.regexp "\\?"
52 let xhtml_re = Pcre.regexp "<.*?>|[^<>]+"
54 let run r (q : cgi) (dbh : Dbi.connection) hostid
55 ({ edit_anon = edit_anon;
56 view_anon = view_anon } as host)
58 let template_page = get_template dbh hostid "page.html" in
59 let template_404 = get_template dbh hostid "page_404.html" in
61 let page = q#param "page" in
62 let page = if page = "" then "index" else page in
64 (* Host-specific fields. *)
65 let sth = dbh#prepare_cached "select css is not null,
66 feedback_email is not null,
67 mailing_list, search_box, navigation
68 from hosts where id = ?" in
69 sth#execute [`Int hostid];
70 let has_host_css, has_feedback_email, mailing_list, search_box, navigation =
71 match sth#fetch1 () with
72 | [ `Bool has_host_css; `Bool has_feedback_email; `Bool mailing_list;
73 `Bool search_box; `Bool navigation ] ->
74 has_host_css, has_feedback_email, mailing_list, search_box,
76 | _ -> assert false in
78 (* Can the user edit? Manage users? etc. *)
79 let can_edit = can_edit host user in
80 let can_manage_users = can_manage_users host user in
81 let can_manage_contacts = can_manage_contacts host user in
82 let can_manage_site = can_manage_site host user in
83 let can_edit_global_css = can_edit_global_css host user in
85 (* Do we have a stats page set up? *)
86 let has_stats = server_settings_stats_page dbh <> None in
88 (* Given the referer string, return the list of search terms. If none
89 * can be found, then throws Not_found.
91 let search_terms_from_referer referer =
93 List.find (fun (rex, _) -> Pcre.pmatch ~rex referer) search_engines in
95 match Pcre.split ~rex:split_qs_re ~max:2 referer with
96 | [url] | [url;""] -> url, ""
98 | _ -> assert false in
99 let args = Cgi_args.parse qs in
101 List.find (fun argname -> List.mem_assoc argname args) argnames in
102 let search_string = List.assoc argname args in
103 Pcre.split ~rex:split_words search_string
106 (* Given a full page of XHTML, highlight search terms found in the
107 * <body> part of the page.
109 let highlight_search_terms xhtml search_terms span_class =
110 (* Split the original XHTML into strings and tags. For example if
111 * the original string is: "This is some <b>bold</b> text.<br/>", then
112 * after this step we will have the following list:
113 * [ "This is some "; "<b>"; "bold"; "</b>"; " text."; "<br/>" ]
115 let xhtml = Pcre.extract_all ~rex:xhtml_re xhtml in
116 let xhtml = Array.to_list xhtml in
117 let xhtml = List.map (fun matches -> matches.(0)) xhtml in
119 (* Find the <body> ... </body> tags. We only want to apply
120 * highlighting to tags within this area.
122 let rec list_split f acc = function
123 | [] -> List.rev acc, []
124 | ((x :: _) as xs) when f x -> List.rev acc, xs
126 let acc = x :: acc in
130 list_split (fun str -> String.starts_with str "<body") [] xhtml in
132 list_split ((=) "</body>") [] body in
133 (* NB: Hopefully, xhtml = head @ body @ tail. *)
135 (* The search terms are a list of simple words. Turn into a big
136 * regular expression, because we want to substitute for each. We
137 * end up with a regexp like '(word1|word2|word3)'.
140 Pcre.regexp ~flags:[`CASELESS]
141 ("(" ^ String.concat "|" search_terms ^ ")") in
143 (* Do the substitution, but only on text, not elements! *)
146 "<span class=\"" ^ span_class ^ "\">" ^ text ^ "</span>"
149 if String.length str > 0 && str.[0] != '<' then
150 Pcre.substitute ~rex ~subst str
154 (* Join the XHTML fragments back together again. *)
155 String.concat "" (List.concat [ head ; body ; tail ])
158 (* Check the templates table for extensions. *)
159 let get_extension url =
160 let sth = dbh#prepare_cached "select extension from templates
164 sth#execute [`String url];
167 let name = sth#fetch1string () in
168 Some (List.assoc name !extensions)
173 (* This code generates ordinary pages. *)
174 let make_page title description pageid last_modified_date has_page_css
175 version page page' extension =
176 let t = template_page in
178 t#set "last_modified_date" last_modified_date;
180 (match description with
181 None -> t#conditional "has_description" false
182 | Some description ->
183 t#conditional "has_description" true;
184 t#set "description" description);
186 if page <> page' then (* redirection *) (
188 t#set "original_page" page; (* XXX title - get it from database *)
189 t#conditional "redirected" true
192 t#conditional "redirected" false
195 t#conditional "has_host_css" has_host_css;
196 t#conditional "has_page_css" has_page_css;
198 t#conditional "has_feedback_email" has_feedback_email;
199 t#conditional "mailing_list" mailing_list;
200 t#conditional "search_box" search_box;
201 t#conditional "navigation" navigation;
203 t#conditional "can_edit" can_edit;
204 t#conditional "can_manage_users" can_manage_users;
205 t#conditional "can_manage_contacts" can_manage_contacts;
206 t#conditional "can_manage_site" can_manage_site;
207 t#conditional "can_edit_global_css" can_edit_global_css;
209 t#conditional "has_stats" has_stats;
211 (* Pull out the sections in this page. *)
216 let sth = dbh#prepare_cached
217 "select ordering, sectionname, content, divname
218 from contents where pageid = ? order by ordering" in
219 sth#execute [`Int pageid];
222 (function [`Int ordering;
223 (`Null | `String _) as sectionname;
225 (`Null | `String _) as divname] ->
226 let divname, has_divname =
229 | `String divname -> divname, true in
230 let sectionname, has_sectionname =
231 match sectionname with
233 | `String sectionname -> sectionname, true in
234 let linkname = linkname_of_sectionname sectionname in
235 [ "ordering", Template.VarString (string_of_int ordering);
237 Template.VarConditional has_sectionname;
238 "sectionname", Template.VarString sectionname;
239 "linkname", Template.VarString linkname;
242 (Wikilib.xhtml_of_content dbh hostid content);
243 "has_divname", Template.VarConditional has_divname;
244 "divname", Template.VarString divname ]
245 | _ -> assert false) in
247 (* Call an extension to generate the first section in this page? *)
252 let content = extension dbh hostid page' in
254 "ordering", Template.VarString "0";
255 "has_sectionname", Template.VarConditional false;
256 "linkname", Template.VarString "";
257 "content", Template.VarString content;
258 "has_divname", Template.VarConditional true;
259 "divname", Template.VarString "form_div";
261 section :: sections in
263 t#table "sections" sections;
265 (* Are we showing an old version of the page? If so, warn. *)
268 t#conditional "is_old_version" false
270 t#conditional "is_old_version" true;
271 t#set "old_version" (string_of_int pageid));
276 t#conditional "user_logged_in" false
277 | User (_, username, _) ->
278 t#conditional "user_logged_in" true;
279 t#set "username" username);
281 (* If logged in, we want to update the recently_visited table. *)
282 if pageid <> None then (
284 | User (userid, _, _) ->
285 let sth = dbh#prepare_cached "delete from recently_visited
286 where hostid = ? and userid = ?
288 sth#execute [`Int hostid; `Int userid; `String page'];
289 let sth = dbh#prepare_cached
290 "insert into recently_visited (hostid, userid, url)
292 sth#execute [`Int hostid; `Int userid; `String page'];
297 (* Navigation links. *)
299 let max_links = 18 in (* Show no more links than this. *)
301 (* What links here. *)
302 let wlh = what_links_here dbh hostid page' in
303 let wlh_urls = List.map fst wlh in (* Just the URLs ... *)
307 | User (userid, _, _) ->
308 (* Recently visited URLs, but don't repeat any from the 'what
309 * links here' section, and don't link to self.
311 let not_urls = page' :: wlh_urls in
312 let limit = max_links - List.length wlh_urls in
313 let qs = Dbi.placeholders (List.length not_urls) in
316 ("select rv.url, p.title, rv.visit_time
317 from recently_visited rv, pages p
318 where rv.hostid = ? and rv.userid = ?
319 and rv.url not in " ^ qs ^ "
320 and rv.hostid = p.hostid and rv.url = p.url
323 let args = List.map (fun s -> `String s) not_urls in
325 ([`Int hostid; `Int userid] @ args @ [`Int limit]);
327 (function [`String url; `String title; _] ->
333 let f (page, title) = [ "page", Template.VarString page;
334 "title", Template.VarString title ] in
335 let table = List.map f wlh in
336 t#table "what_links_here" table;
337 t#conditional "has_what_links_here" (wlh <> []);
339 let table = List.map f rv in
340 t#table "recently_visited" table;
341 t#conditional "has_recently_visited" (rv <> []);
344 (* If we are coming from a search engine then we want to highlight
345 * search terms throughout the whole page ...
348 let referer = Table.get (Request.headers_in r) "Referer" in
349 let search_terms = search_terms_from_referer referer in
351 (* Highlight the search terms. *)
352 let xhtml = t#to_string in
353 let xhtml = highlight_search_terms xhtml search_terms "search_term" in
355 (* Deliver the page. *)
360 (* No referer / no search terms / not a search engine referer. *)
364 (* This code generates 404 pages. *)
366 Request.set_status r 404; (* Return a 404 error code. *)
368 let t = template_404 in
374 ('a'..'z' | 'A'..'Z' | '0'..'9') as c -> c
377 t#set "search_terms" search_terms;
379 t#conditional "has_host_css" has_host_css;
381 t#conditional "can_edit" can_edit;
382 t#conditional "can_manage_users" can_manage_users;
383 t#conditional "can_manage_contacts" can_manage_contacts;
384 t#conditional "can_manage_site" can_manage_site;
385 t#conditional "can_edit_global_css" can_edit_global_css;
387 t#conditional "has_stats" has_stats;
392 (* Fetch a page by name. This function can give three answers:
393 * (1) Page fetched OK (fetches some details of the page).
394 * (2) Page is a redirect (fetches the name of the redirect page).
395 * (3) Page not found in database, could be template or 404 error.
397 (* XXX Should do a case-insensitive matching of URLs, and if the URL differs
398 * in case only should redirect to the lowercase version.
400 let fetch_page page version allow_redirect =
403 if allow_redirect then (
406 "select redirect, id, title, description, last_modified_date,
408 from pages where hostid = ? and url = ?" in
409 sth#execute [`Int hostid; `String page];
411 (match sth#fetch1 () with
412 | [ `Null; `Int id; `String title; `String description;
413 `Timestamp last_modified_date; `Bool has_page_css ] ->
414 FPOK (id, title, description, last_modified_date,
416 | `String redirect :: _ ->
420 Not_found -> FPNotFound)
421 ) else (* redirects not allowed ... *) (
424 "select id, title, description, last_modified_date,
426 from pages where hostid = ? and url = ?" in
427 sth#execute [`Int hostid; `String page];
429 (match sth#fetch1 () with
430 | [ `Int id; `String title; `String description;
431 `Timestamp last_modified_date; `Bool has_page_css ] ->
432 FPOK (id, title, description, last_modified_date,
436 Not_found -> FPNotFound)
441 "select id, title, description, last_modified_date,
444 where hostid = ? and id = ? and
445 (url = ? or url_deleted = ?)" in
446 sth#execute [`Int hostid; `Int version;
447 `String page; `String page];
449 (match sth#fetch1 () with
450 | [ `Int id; `String title; `String description;
451 `Timestamp last_modified_date; `Bool has_page_css ] ->
452 FPOK (id, title, description, last_modified_date,
456 Not_found -> FPNotFound)
459 (* Here we deal with the complex business of redirects and versions. *)
460 (* Only allow the no_redirect and version syntax for editors. *)
461 let allow_redirect, version =
463 not (q#param_true "no_redirect"),
464 try Some (int_of_string (q#param "version")) with Not_found -> None
468 let rec loop page' i =
469 if i > max_redirect then (
470 error ~title:"Too many redirections" ~back_button:true
471 q ("Too many redirects between pages. This may happen because " ^
472 "of a cycle of redirections.");
475 match fetch_page page' version allow_redirect with
476 | FPOK (pageid, title, description, last_modified_date, has_page_css)->
477 (* Check if the page is also a template. *)
478 let extension = get_extension page' in
479 make_page title (Some description) (Some pageid)
480 (printable_date last_modified_date) has_page_css
481 version page page' extension
482 | FPRedirect page' ->
485 (* Might be a templated page with no content in it. *)
486 let extension = get_extension page' in
487 (match extension with
488 | (Some _) as extension ->
490 make_page title None None
491 "Now" false None page page'
499 register_script ~restrict:[CanView] run