Paul J. Davis
2dd7fb0798
CSS tweaks for notes markup.
2010-03-26 16:00:31 -04:00
Paul J. Davis
4dbde5c2ef
Some edits to the new docs.
2010-03-26 14:50:00 -04:00
benoitc
8d52e1e77f
sync
2010-03-26 18:01:21 +01:00
benoitc
59396b2557
typo
2010-03-26 18:00:47 +01:00
benoitc
7b02420ed7
sync
2010-03-26 16:26:59 +01:00
benoitc
eb91338852
more doc edit. fix README. spotted by temoto on irc
2010-03-26 16:26:19 +01:00
benoitc
4f54c06121
make specific case for websockets
2010-03-26 15:53:53 +01:00
benoitc
8c6266866e
fix example
2010-03-26 15:19:46 +01:00
benoitc
3b2788e3a2
update README for eventlet/Gevent
2010-03-26 12:16:57 +01:00
benoitc
f3a8c224ab
resync
2010-03-26 12:15:48 +01:00
benoitc
dfb715b0cc
s/Installing/Installation
2010-03-26 12:15:21 +01:00
benoitc
de9c0728aa
typo
2010-03-26 12:14:38 +01:00
benoitc
f5ab129997
sync
2010-03-26 12:11:57 +01:00
benoitc
7d24a7597d
update doc
2010-03-26 12:10:14 +01:00
benoitc
6863bba84b
Ignore SIGWINCH in worker. Fix crash on OpenBSD
2010-03-26 11:08:24 +01:00
Paul J. Davis
d20fd9b411
Fix example config in docs.
2010-03-25 20:02:18 -04:00
Paul J. Davis
fe3f7db88c
Wrong license listed in the setup() vars.
2010-03-24 14:40:17 -04:00
benoitc
c1aec22097
fix response
2010-03-23 21:09:15 +01:00
benoitc
d192b1e653
no need to close the socket here
2010-03-23 21:02:24 +01:00
Paul Davis
1d326f88c4
Fixed async responses.
2010-03-23 15:54:22 -04:00
Paul Davis
e4b7263314
Fix async req/resp handlers.
2010-03-23 15:40:14 -04:00
Paul J. Davis
2242be27cf
Silly typo prevented response header parsing.
2010-03-23 15:10:52 -04:00
Paul J. Davis
014d711bd7
Fix PEP 333 compliance for the write callable.
2010-03-23 14:58:10 -04:00
benoitc
e37bfa3c4e
fix requirement
2010-03-22 22:32:08 +01:00
benoitc
9f572bc8ea
add headers_sent guard.
2010-03-21 10:30:57 +01:00
benoitc
125392d054
next version will be 0.7.0
2010-03-20 15:11:39 +01:00
benoitc
514a0ba0e3
change the way we handle last chunk, on't break but instead test if last
...
was an empty string. So we allows idiot content to send empty string at
first.
2010-03-19 10:00:15 +01:00
benoitc
771d44902a
use formatted title so we make sure all browsers will correctly read
...
them
2010-03-18 08:58:05 +01:00
benoitc
6845d10460
fix worker. status is a response thing. instead test if we got a status
...
line or if environ isn't empty.
2010-03-17 23:06:45 +01:00
Jonas Nockert
20c0f03843
Example application's module and WSGI callable is actually test:app
2010-03-17 21:36:59 +01:00
benoitc
d63b6015ef
useless line, and handle only ValueError
2010-03-17 01:10:10 +01:00
benoitc
49e57d36bb
remove useless examples
2010-03-17 01:06:17 +01:00
benoitc
40f1c43ccf
iadd --arbiter option to django run command
2010-03-17 00:53:57 +01:00
benoitc
8dbc10be8c
only patch select and socket with eventlet
2010-03-17 00:37:19 +01:00
benoitc
451647d5ad
remove arbiter import from main
2010-03-17 00:32:02 +01:00
benoitc
0781caadab
merge grainbows & gunicorn
2010-03-17 00:18:16 +01:00
benoitc
06d58ae1c4
oups revert this change in util
2010-03-16 22:15:32 +01:00
benoitc
23927f39b1
allows custom arbiter
2010-03-16 22:12:26 +01:00
Paul J. Davis
8b9cb09b6b
Be explicit about blocking status.
...
Apparently the assumption that sockets don't inherit their blocking
status from the server socket is invalid. Or python sets the flags
to match. Either way, we need clients to use blocking sockets so
we're now explicitly setting the blocking status.
2010-03-16 11:46:01 -04:00
Paul J. Davis
61a84eece3
Be more considerate to future me.
...
Sergey points out that its more proper to |= with the current value of
the fd flags. I concur.
2010-03-15 22:59:26 -04:00
Paul J. Davis
09b5dd2ad1
Fix close_on_exec bug.
...
Reported by Sergey Shepelev. I changed his patch slightly because of the
man page language. As I read things, the close on exec flag is separate
from the F_SETFL flags.
2010-03-15 22:12:54 -04:00
benoitc
562c9a2d70
revert previous change
2010-03-13 23:25:29 +01:00
benoitc
d21254edfc
update readme
2010-03-13 23:21:42 +01:00
benoitc
33cf5dab01
with statement for py25 + bump release number
2010-03-13 22:16:58 +01:00
benoitc
6709a2c781
fix message
2010-03-12 23:59:25 +01:00
benoitc
f53532a92e
always use TeeInput
2010-03-12 14:15:21 +01:00
benoitc
1ee01152b3
fix chuynked response
2010-03-12 00:52:21 +01:00
benoitc
5bb0734532
put creation of worker in its own class so we can pass custom workers to
...
it.
2010-03-11 14:29:20 +01:00
benoitc
f75a413773
some fixes to config and pidfile
2010-03-11 13:53:10 +01:00
benoitc
7280ad20f8
seem like the arbiter was reverted in my previous commit . commit it
...
again
2010-03-11 13:45:45 +01:00