1 ## Why internal pages? (unresolved)
3 Comments are saved as internal pages, so they can never be edited through the CGI,
4 only by direct committers.
6 > So, why do it this way, instead of using regular wiki pages in a
7 > namespace, such as `$page/comments/*`? Then you could use [[plugins/lockedit]] to
8 > limit editing of comments in more powerful ways. --[[Joey]]
10 >> Er... I suppose so. I'd assumed that these pages ought to only exist as inlines
11 >> rather than as individual pages (same reasoning as aggregated posts), though.
13 >> lockedit is actually somewhat insufficient, since `check_canedit()`
14 >> doesn't distinguish between creation and editing; I'd have to continue to use
15 >> some sort of odd hack to allow creation but not editing.
17 >> I also can't think of any circumstance where you'd want a user other than
18 >> admins (~= git committers) and possibly the commenter (who we can't check for
19 >> at the moment anyway, I don't think?) to be able to edit comments - I think
20 >> user expectations for something that looks like ordinary blog comments are
21 >> likely to include "others can't put words into my mouth".
23 >> My other objection to using a namespace is that I'm not particularly happy about
24 >> plugins consuming arbitrary pieces of the wiki namespace - /discussion is bad
25 >> enough already. Indeed, this very page would accidentally get matched by rules
26 >> aiming to control comment-posting... :-) --[[smcv]]
28 >>> Thinking about it, perhaps one way to address this would be to have the suffix
29 >>> (e.g. whether commenting on Sandbox creates sandbox/comment1 or sandbox/c1 or
30 >>> what) be configurable by the wiki admin, in the same way that recentchanges has
31 >>> recentchangespage => 'recentchanges'? I'd like to see fewer hard-coded page
32 >>> names in general, really - it seems odd to me that shortcuts and smileys
33 >>> hard-code the name of the page to look at. Perhaps I could add
34 >>> discussionpage => 'discussion' too? --[[smcv]]
36 >>> (I've now implemented this in my branch. --[[smcv]])
38 >> The best reason to keep the pages internal seems to me to be that you
39 >> don't want the overhead of every comment spawning its own wiki page. --[[Joey]]
43 The plugin now allows multiple comment formats while still using internal
44 pages; each comment is saved as a page containing one `\[[!comment]]` directive,
45 which has a superset of the functionality of [[ikiwiki/directives/format]].
47 ## Access control (unresolved?)
49 By the way, I think that who can post comments should be controllable by
50 the existing plugins opendiscussion, anonok, signinedit, and lockedit. Allowing
51 posting comments w/o any login, while a nice capability, can lead to
52 spam problems. So, use `check_canedit` as at least a first-level check?
55 > This plugin already uses `check_canedit`, but that function doesn't have a concept
56 > of different actions. The hack I use is that when a user comments on, say, sandbox,
57 > I call `check_canedit` for the pseudo-page "sandbox[postcomment]". The
58 > special `postcomment(glob)` [[ikiwiki/pagespec]] returns true if the page ends with
59 > "[postcomment]" and the part before (e.g. sandbox) matches the glob. So, you can
60 > have postcomment(blog/*) or something. (Perhaps instead of taking a glob, postcomment
61 > should take a pagespec, so you can have postcomment(link(tags/commentable))?)
63 > This is why `anonok_pagespec => 'postcomment(*)'` and `locked_pages => '!postcomment(*)'`
64 > are necessary to allow anonymous and logged-in editing (respectively).
66 >> I changed that to move the flag out of the page name, and into a variable that the `match_postcomment`
67 >> function checks for. Other ugliness still applies. :-) --[[Joey]]
69 > This is ugly - one alternative would be to add `check_permission()` that takes a
70 > page and a verb (create, edit, rename, remove and maybe comment are the ones I
71 > can think of so far), use that, and port the plugins you mentioned to use that
72 > API too. This plugin could either call `check_can("$page/comment1", 'create')` or
73 > call `check_can($page, 'comment')`.
75 > One odd effect of the code structure I've used is that we check for the ability to
76 > create the page before we actually know what page name we're going to use - when
77 > posting the comment I just increment a number until I reach an unused one - so
78 > either the code needs restructuring, or the permission check for 'create' would
79 > always be for 'comment1' and never 'comment123'. --[[smcv]]
81 >> Now resolved, in fact --[[smcv]]
83 > Another possibility is to just check for permission to edit (e.g.) `sandbox/comment1`.
84 > However, this makes the "comments can only be created, not edited" feature completely
85 > reliant on the fact that internal pages can't be edited. Perhaps there should be a
86 > `editable_pages` pagespec, defaulting to `'*'`? --[[smcv]]
88 ## comments directive vs global setting (resolved?)
90 When comments have been enabled generally, you still need to mark which pages
91 can have comments, by including the `\[[!comments]]` directive in them. By default,
92 this directive expands to a "post a comment" link plus an `\[[!inline]]` with
93 the comments. [This requirement has now been removed --[[smcv]]]
95 > I don't like this, because it's hard to explain to someone why they have
96 > to insert this into every post to their blog. Seems that the model used
97 > for discussion pages could work -- if comments are enabled, automatically
98 > add the comment posting form and comments to the end of each page.
101 >> I don't think I'd want comments on *every* page (particularly, not the
102 >> front page). Perhaps a pagespec in the setup file, where the default is "*"?
103 >> Then control freaks like me could use "link(tags/comments)" and tag pages
104 >> as allowing comments.
106 >>> Yes, I think a pagespec is the way to go. --[[Joey]]
108 >>>> Implemented --[[smcv]]
111 >> The model used for discussion pages does require patching the existing
112 >> page template, which I was trying to avoid - I'm not convinced that having
113 >> every possible feature hard-coded there really scales (and obviously it's
114 >> rather annoying while this plugin is on a branch). --[[smcv]]
116 >>> Using the template would allow customising the html around the comments
117 >>> which seems like a good thing? --[[Joey]]
119 >>>> The \[[!comments]] directive is already template-friendly - it expands to
120 >>>> the contents of the template `comments_embed.tmpl`, possibly with the
121 >>>> result of an \[[!inline]] appended. I should change `comments_embed.tmpl`
122 >>>> so it uses a template variable `INLINE` for the inline result rather than
123 >>>> having the perl code concatenate it, which would allow a bit more
124 >>>> customization (whether the "post" link was before or after the inline).
125 >>>> Even if you want comments in page.tmpl, keeping the separate comments_embed.tmpl
126 >>>> and having a `COMMENTS` variable in page.tmpl might be the way forward,
127 >>>> since the smaller each templates is, the easier it will be for users
128 >>>> to maintain a patched set of templates. (I think so, anyway, based on what happens
129 >>>> with dpkg prompts in Debian packages with monolithic vs split
130 >>>> conffiles.) --[[smcv]]
132 >>>>> I've switched my branch to use page.tmpl instead; see what you think? --[[smcv]]
134 ## Raw HTML (resolved?)
136 Raw HTML was not initially allowed by default (this was configurable).
138 > I'm not sure that raw html should be a problem, as long as the
139 > htmlsanitizer and htmlbalanced plugins are enabled. I can see filtering
140 > out directives, as a special case. --[[Joey]]
142 >> Right, if I sanitize each post individually, with htmlscrubber and either htmltidy
143 >> or htmlbalance turned on, then there should be no way the user can forge a comment;
144 >> I was initially wary of allowing meta directives, but I think those are OK, as long
145 >> as the comment template puts the \[[!meta author]] at the *end*. Disallowing
146 >> directives is more a way to avoid commenters causing expensive processing than
147 >> anything else, at this point.
149 >> I've rebased the plugin on master, made it sanitize individual posts' content
150 >> and removed the option to disallow raw HTML. Sanitizing individual posts before
151 >> they've been htmlized required me to preserve whitespace in the htmlbalance
152 >> plugin, so I did that. Alternatively, we could htmlize immediately and always
153 >> save out raw HTML? --[[smcv]]
155 >>> There might be some use cases for other directives, such as img, in
158 >>> I don't know if meta is "safe" (ie, guaranteed to be inexpensive and not
159 >>> allow users to do annoying things) or if it will continue to be in the
160 >>> future. Hard to predict really, all that can be said with certainty is
161 >>> all directives will contine to be inexpensive and safe enough that it's
162 >>> sensible to allow users to (ab)use them on open wikis.
167 I have a test ikiwiki setup somewhere to investigate adopting the comments
168 plugin. It is setup with no auth enabled and I got hammered with a spam attack
169 over the last weekend (predictably). What surprised me was the scale of the
170 attack: ikiwiki eventually triggered OOM and brought the box down. When I got
171 it back up, I checked out a copy of the underlying git repository, and it
172 measured 280M in size after being packed. Of that, about 300K was data prior
173 to the spam attack, so the rest was entirely spam text, compressed via git's
174 efficient delta compression.
176 I had two thoughts about possible improvements to the comments plugin in the
179 * comment pagination - there is a hard-to-define upper limit on the number
180 of comments that can be appended to a wiki page whilst the page remains
181 legible. It would be useful if comments could be paginated into sub-pages.
183 * crude flood control - asides from spam attacks (and I am aware of
184 [[plugins/blogspam]]), people can crap flood or just aggressively flame
185 repeatedly. An interesting prevention measure might be to not let an IP
186 post more than 3 sequential comments to a page, or to the site, without
187 at least one other comment being interleaved. I say 3 rather than 2 since
188 correction follow-ups are common.