The following bug applied to 1.x, not sure if it still
exists:
I'm not sure how to do patches so I'm messaging this one to you.
Let me know if there's another way you'd like me to send this to
you.
I chatted with you on #nitrogen about my my app and the fact that
the comet implementation was reversing the order of updates. I
determined that there was a bug in the comet implementation, and
it's supported here:
http://stackoverflow.com/questions/2002487/two-quick-consecutive-ca...
Here's my patch (formatted with git format-patch):
From 7dc693492765978cac82bb8abb43fd60def8bb9f Mon Sep 17
00:00:00 2001
From: root root@rohan-laptop.(none)
Date: Wed, 17 Feb 2010 18:12:07 -0500
Subject: [PATCH] Fixed a bug in comet requests where
Javascript updates happen in reverse order
src/lib/wf_comet.erl | 4 ++-- 1 files changed, 2 insertions(+),
2 deletions(-)
diff --git a/src/lib/wf_comet.erl b/src/lib/wf_comet.erl
index 2f86ad8..03efe18 100644
--- a/src/lib/wf_comet.erl +++ b/src/lib/wf_comet.erl @@ -154,10
+154,10 @@ comet_loop() ->
end.
%% collect_content/0 - Gather all content sent to this comet_loop.
-collect_content() -> lists:reverse(inner_collect_content()).
+collect_content() -> inner_collect_content().
inner_collect_content() ->
receive
{content, C} -> [C|inner_collect_content()]
after
0 -> []
end. \ No newline at end of file
+ end.
1.6.3.3
hanrelan said about 1 month ago:
Sorry, the formatting destroyed the patch. Basically it's just
removing the reverse of inner_collect_content in collect_content in
wf_comet.erl since inner_collect_content is already in the right
order
