1 files changed, 71 insertions, 0 deletions
diff --git a/doc/bugs/Comments_are_not_sorted_by_their_date_attribute.mdwn b/doc/bugs/Comments_are_not_sorted_by_their_date_attribute.mdwn
new file mode 100644
index 000000000..5a4c4b2ae
--- /dev/null
+++ b/doc/bugs/Comments_are_not_sorted_by_their_date_attribute.mdwn
@@ -0,0 +1,71 @@
+(brief, sorry, via Phone. More details to follow)
+
+I  gradually splitting discussion pages into separate comment pages, containing a comment directive.
+
+The "date" attribute is being set to the date output by gig for a commit. (I'd hope this was parseable)
+
+The presentation of the resulting comments is not sorted by this date, which I would hope/expect, but instead by the ctime or mtime of the file at the other end, as best I can tell.
+
+-- [[Jon]]
+
+> Yes, comments are displayed via an inline, and usual [[pagespec/sorting]]
+> (eg, default of when the file was first seen) is used. The comment
+> date only affects the date displayed.
+>
+> > That's not what I intended - it's meant to be more or less just
+> > syntactic sugar for `\[[!meta date=foo]]`, setting the `%pagectime`.
+> > The code looks as though it ought to work, but perhaps it's buggy?
+> > (edited to add: it is, see below) --[[smcv]]
+> 
+> The only time I've seen this be much problem personally is when moving
+> a page, which means moving its comments directory, which tends to
+> jumble the order. (And --gettime does not help, as ikiwiki does not
+> tell git to follow renames for speed reasons.)
+> 
+> I wonder if it wouldn't be best to just get rid of the extra date
+> inside the comment, and rely on the file date as is done for other pages.
+> Thoughts [[smcv]]?
+>
+> Altenatively, since comments tend to be named "comment_N_.....", 
+> adding a new [[pagespec/sorting]] method that sorts by filename, 
+> rather than by title, and using it by default for comments might be 
+> better than the current situation. --[[Joey]] 
+
+>> Since git does not track file time, I tend to prefer to encode date
+>> stuff inside files where possible.  For other pages, I put an explicit
+>> [[plugins/meta]] date into the source when I create the page.  I've
+>> had to reconstruct ordering after moving to a different git checkout
+>> after a server move before, it was painful ☺
+>>
+>> In my current situation, I could live with by-filename ordering. By-title
+>> ordering would also be workable. — [[Jon]]
+
+>>> I agree with Jon's reasons for embedding an explicit date in the file.
+>>> As I said, this is *meant* to work, but it might not.
+>>>
+>>> Sorting by filename would only be useful with
+>>> [[!cpan Sort::Naturally]], since normal `cmp` ordering would break pages
+>>> with more than 9 comments. --s
+
+----
+
+[[!template id=gitbranch author="[[smcv]]" branch=smcv/comments-metadata]]
+
+I thought that, as internal pages, comments were not preprocessed
+(and so their date attributes did not have a chance to take effect) until
+they were already being inlined, by which time they have already been
+sorted by the files' ctimes. Actually, I was wrong about that - internal
+pages have a special case elsewhere - but they did skip the `scan` hook,
+which is also fixed in my branch.
+
+The real bug was that the preprocess hook for comments didn't run
+in the scan phase; my branch fixes that, streamlines that hook a bit
+when run in the scan phase (so it doesn't htmlize, and only runs nested
+directives in scan mode), and adds a regression test. --[[smcv]]
+
+[[!tag patch]]
+
+> Thanks.. I am not 100% sure if I just forgot to scan internal pages
+> or left it out as some kind of optimisation since none needed to be
+> scanned. Anyway, if it was an optimisation it was not much of one
+> since they were preprocessed. All applied, [[done]]. --[[Joey]]