1 * Why isn't it statically-genereated, but generated dynamically by CGI? It
2 seems like it could be beneficial to have it rendered in the post-commit
3 hook, just like everything else in the wiki.
5 > I hope to statically generate it eventually, currently the problem is
6 > that it takes at least several seconds to generate the recentchanges
7 > page, and adding several seconds to every page edit is not desiriable. If
8 > the time can be reduced it could be done, I'm also not adverse to
9 > adding an optional way to statically render it even at the current
12 * Also, is it planned/desired that recent changes generate the same
13 information in RSS feed format? This seems like it could be a useful way
14 to keep track of the wiki as a whole.
16 > This is used by various interwiki type things, I think, so should be
19 * Lastly, would it be possible to use the recent changes code with a
20 pagespec? I understand this sort of infringes on territory covered by the
21 inline plugin, but the inline plugin only puts a page in the RSS feed
22 once, when it's created, and I imagine some people -- some deranged,
23 obsessive-compulsive people like myself -- would like to know about the
24 changes made to existing pages as well as newly-created pages.
26 > That would work rather well for pages like [[todo]] and [[bugs]], where
27 > you want to know about any updates, not just initial
28 > creation. --[[JoshTriplett]]
30 > Of course you can use email subscriptions for that too.. --[[Joey]]
32 >> I have more thoughts on this topic which I will probably write
33 >> tomorrow. If you thought my other patches were blue-sky, wait until
34 >> you see this. --Ethan
36 OK, so here's how I see the RecentChanges thing. I write blog posts and
37 the inline plugin generates RSS feeds. Readers of RSS feeds are notified
38 of new entries but not changes to old entries. I think it's rude to change
39 something without telling your readers, so I'd like to address this.
40 To tell the user that there have been changes, we can tell the user which
41 page has been changed, the new text, the RCS comment relating to
42 the change, and a diff of the actual changes. The new text probably isn't
43 too useful (I have a very hard time rereading things for differences),
44 so any modifications to inline to re-inline pages probably won't help,
45 even if it were feasible (which I don't think it is). So instead we
46 turn to creating diffs automatically and (maybe) inlining them.
48 I suggest that for every commit, a diff is created automagically
49 but not committed to the RCS. The page containing this diff would be
50 a "virtual page", which cannot be edited and is not committed.
51 (Committing here would be bad, because then it would create a new
52 commit, which would need a new diff, which would need to be committed,
53 etc.) Virtual pages would "expire" and be deleted if they were not
54 depended on in some way.
56 Let's say these pages are created in edits/commit_%d.mdwn. RecentChanges
57 would then be a page which did nothing but inline the last 50 `edits/*`.
58 This would give static generation and RSS/Atom feeds. The inline
59 plugin could be optionally altered to inline pages from `edits/*`
60 that match any pages in its pagespec, and through this we could get
61 a recent-changes+pagespec thing. You could also exclude edits that have
62 "minor" in the commit message (or some other thing that marks them as
65 You could make an argument that I care way too much about what amounts
66 to edits anyhow, but like Josh says, there are use cases for this.
67 While this could be done with mail subscriptions, I can think of sites
68 where you might want to disable all auth so that people can't edit
71 > I really dislike all Wiki engine recentchanges pages. They all tend to be
72 > fairly machine readable, but confusing for non-wiki users to grok. And I've
73 > yet to see an _attractive_ recentchanges implementation. IkiWikis' is no
74 > better or worse than the others.
76 > I really like the frontpage of [Bill
77 > Seitz](http://webseitz.fluxent.com/wiki/FrontPage) as an recentchanges
78 > format. Note how he uses some clever css to show changes in different
79 > sections of the website. I modeled my own
80 > [recentchanges](http://xtermin.us/recentchanges) page page on his ideas. This
81 > probably isn't appropriate for non-WikiLog style setups, but is this
82 > something closer to what you what was requested?
84 > BTW: My recentchanges plugin does not seem to add a lot processing time
85 > to compiling. Then again, I'm not pulling changelog message from the RCS
92 Here's a full design for redoing recentchanges, based on Ethan's ideas:
94 * Add a recentchanges plugin that has a preprocessor directive:
95 \[[!recentchanges num=100 pages=* template=recentchanges.tmpl]]
96 If put on the [[recentchanges]] page, this would result in up to 100
97 recentchanges/change_$id.mdwn files being created.
98 * Which means the plugin has to store state and use a checkconfig hook
99 or the like to create the requested pages (and delete old ones) when
100 the wiki is rebuilt and when the post_commit hook is run.
101 * Then it's a simple matter of using inline on the recentchanges page
102 to display the changes. (With a special template to display nicely.)
103 * Rss/atom comes for free..
104 * So drop mail notifications.
105 * If someone wants to subscribe to notifications for only a subset
106 of pages, they can either filter the recentchanges in their rss
107 aggregator, or they can set up their own page that uses the recentchanges
108 directive for only the pages they want.
109 * The `rcs_notify` functions will be removed.
110 * To add diffs, another plugin can add a pagetemplate hook that calls
111 a `rcs_diff`. (optional)
112 * So to update the changes files, just call `rcs_recentchanges`, create
113 files for each new id, and delete files for each id that is no longer
115 * The cgi support for recentchanges can be dropped, or moved to a different
118 I'm unsure how fast this will all be, but by using regular pages, there's
119 cacheing, at least. The main slowdown might turn out to be the inlining and
120 not the generation of the changes pages. The current cgi recentchanges
121 code saves a tenth of a second or so by memoizing htmllink, an optimisation
122 that won't be available when using the more general inlining code.
124 An obvious optimisation, and one implied by this design, is that each change
125 file is only written once. This assumes that the data in them doesn't ever
126 change, which actually isn't true (svn commit messages can be changed), but
127 is probably close enough to true for our purposes.
129 Another optimisation would be to htmlize the change files when they're
130 written out -- avoids re-rendering a given file each time a new change is
131 made (thus doing 1/100th the work).
133 Links in the change files to the changed pages will need special handling.
134 These links should not generate backlinks. They probably shouldn't be
135 implemented as wikiliks at all. Instead, they should be raw, absolute
136 html links to the pages that were changed.
138 Only problem with this approach is that the links break if the changed
139 page later gets deleted. I think that's acceptable. It could link to
140 `ikiwiki.cgi?do=redir&page=foo`, but that's probably overkill.
144 [[done]] !! (in this branch at least :-)