Patrick Gansterer [Wed, 26 Feb 2014 20:56:59 +0000 (21:56 +0100)]
Rename leave to cleanup
The name ‘leave’ can not be used on some Windows CE platforms, so rename it.
Patrick Gansterer [Wed, 26 Feb 2014 20:37:31 +0000 (21:37 +0100)]
Avoid calls to gettimeofday()
Add a new function to get the current time in microseconds, since gettimeofday() does not exist on Windows.
Keep the current implementation for the test applications.
Patrick Gansterer [Wed, 26 Feb 2014 18:52:11 +0000 (19:52 +0100)]
Replace perror() with lwsl_warn()
Use the general logging function for printing error messages.
Patrick Gansterer [Wed, 26 Feb 2014 18:21:40 +0000 (19:21 +0100)]
Fix endian detection on non-linux systems
f975f736408f32e7384cf38270a7806e68d5e112 introduced inclusion of endian.h,
which is not a standard header. Only include it on Linux machines and
add similar includes for Apple and FreeBSD.
Patrick Gansterer [Wed, 26 Feb 2014 17:51:05 +0000 (18:51 +0100)]
Add missing LWS_VISIBLE to libwebsocket_set_timeout()
This aligns it with the style of all other function declarations.
Andy Green [Wed, 26 Feb 2014 23:28:45 +0000 (07:28 +0800)]
unconfuse quoting in CMakeLists.txtt
via Marjan Tomas https://github.com/warmcat/libwebsockets/issues/67
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Wed, 26 Feb 2014 23:19:21 +0000 (07:19 +0800)]
revert wrong fix visibility of return_http_status
Signed-off-by: Andy Green <andy.green@linaro.org>
Joakim Soderberg [Mon, 24 Feb 2014 17:23:00 +0000 (18:23 +0100)]
Fix windows compilation.
Gabriel Gritsch [Sat, 22 Feb 2014 10:00:28 +0000 (18:00 +0800)]
use SO_REUSEADDR on windows since supposedly ok now
Andy Green [Fri, 21 Feb 2014 10:51:38 +0000 (18:51 +0800)]
fix visibility of return_http_status
From "DevYourWorld" on github
https://github.com/warmcat/libwebsockets/issues/64#issuecomment-
35720037
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Fri, 21 Feb 2014 10:48:05 +0000 (18:48 +0800)]
fix WITHOUT_CLIENT install in CMake
Via "DevYourWorld" on github
https://github.com/warmcat/libwebsockets/issues/64#issuecomment-
35719084
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Fri, 21 Feb 2014 10:45:45 +0000 (18:45 +0800)]
update cmake opts in README.build
via "DevYourWorld" on github
https://github.com/warmcat/libwebsockets/issues/64
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Fri, 21 Feb 2014 10:43:42 +0000 (18:43 +0800)]
interface_to_sa no longer server only
Via "DevYourWorld" on github
https://github.com/warmcat/libwebsockets/issues/64
Signed-off-by: Andy Green <andy.green@linaro.org>
Craig McQueen [Wed, 19 Feb 2014 23:27:11 +0000 (10:27 +1100)]
Call libwebsocket_set_timeout() before callback LWS_CALLBACK_HTTP.
This allows the LWS_CALLBACK_HTTP callback to override the timeout
with libwebsocket_set_timeout() if it wants.
Mattias Lundberg [Tue, 18 Feb 2014 09:06:57 +0000 (10:06 +0100)]
Add support to binding client to interface.
Joakim Soderberg [Tue, 18 Feb 2014 12:38:14 +0000 (13:38 +0100)]
Fix compilation on Windows.
C89 which Microsofts compiler only support does not allow variable
declarations anywhere but at the start of a scope.
Also, only try to copy the test-server files if they actually exists. For
instance the OpenSSL cert generation fails if you run cmake under the git
bash shell on Windows (won't work with the unix method either) so that
file will be missing... This would result in a compilation error when
tests where turned on.
Craig McQueen [Tue, 18 Feb 2014 22:24:17 +0000 (09:24 +1100)]
Add LWS_VISIBLE to libwebsocket_set_timeout() so it can be used externally.
Andy Green [Sat, 15 Feb 2014 17:52:50 +0000 (01:52 +0800)]
add missing CMake pieces
Signed-off-by: Andy Green <andy.green@linaro.org>
Michael Haberler [Sat, 15 Feb 2014 12:33:46 +0000 (20:33 +0800)]
document new extpoll apis
Michael Haberler [Sat, 15 Feb 2014 12:21:02 +0000 (20:21 +0800)]
update test server for new extpoll
Michael Haberler [Sat, 15 Feb 2014 12:19:26 +0000 (20:19 +0800)]
change extpoll in lib to new way
Michael Haberler [Sat, 15 Feb 2014 12:18:24 +0000 (20:18 +0800)]
replace old extpoll callbacks with LWS_CALLBACK_CHANGE_MODE_POLL_FD
Michael Haberler [Sat, 15 Feb 2014 12:15:55 +0000 (20:15 +0800)]
define struct libwebsocket_pollargs
Andy Green [Sat, 15 Feb 2014 11:25:50 +0000 (19:25 +0800)]
add outermost wsi lifetime callbacks
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 15 Feb 2014 08:49:41 +0000 (16:49 +0800)]
LWS_CALLBACK_USER
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 15 Feb 2014 08:36:38 +0000 (16:36 +0800)]
libwebsocket_callback_all_protocol
Signed-off-by: Andy Green <andy.green@linaro.org>
Gabriel Gritsch [Sat, 15 Feb 2014 08:20:25 +0000 (16:20 +0800)]
add const http write helper
Since libwebsocket_write doesn't change the source buffer in HTTP protocol,
we can have a little helper to hide the harmless cast
Andy Green [Sat, 15 Feb 2014 08:00:37 +0000 (16:00 +0800)]
fix zero length ping pong
Signed-off-by: Andy Green <andy.green@linaro.org>
Daniel Ludwig [Thu, 13 Feb 2014 23:17:46 +0000 (00:17 +0100)]
missing endian preprocessor defines on Linux environments
Andy Green [Sat, 15 Feb 2014 06:45:14 +0000 (14:45 +0800)]
update version in libwebsockets.pc
Signed-off-by: Andy Green <andy.green@linaro.org>
arnaudviala [Sat, 15 Feb 2014 06:39:40 +0000 (14:39 +0800)]
fix memory leak
Fix a possible memory leak in
libwebsocket_context_destroy() when compiled without extension support.
Andy Green [Sat, 15 Feb 2014 06:36:02 +0000 (14:36 +0800)]
add timeout between accept and negotiation
Signed-off-by: Andy Green <andy.green@linaro.org>
Alexandre Erwin Ittner [Fri, 7 Feb 2014 01:15:51 +0000 (23:15 -0200)]
Fix documentation for LWS_CALLBACK_FILTER_NETWORK_CONNECTION
At the time callback LWS_CALLBACK_FILTER_NETWORK_CONNECTION is called,
there is no client connection information yet, so the parameter wsi
still pointing to the main server connection. Add an description of
this behavior to the documentation.
Jakob Flierl [Mon, 27 Jan 2014 11:37:47 +0000 (12:37 +0100)]
make force_exit volatile.
Instruct the compiler to always access force_exit in memory.
Joakim Soderberg [Sat, 15 Feb 2014 05:48:56 +0000 (13:48 +0800)]
cmake and other fixes
martell [Wed, 22 Jan 2014 19:01:14 +0000 (19:01 +0000)]
Removed needless VS projects now that cmake is fixed
martell [Sat, 15 Feb 2014 05:37:08 +0000 (13:37 +0800)]
additional mingw fixes
Signed-off-by: martell <martellmalone@gmail.com>
martell [Wed, 22 Jan 2014 18:17:03 +0000 (18:17 +0000)]
fixed windows build
Andy Green [Sat, 15 Feb 2014 04:46:33 +0000 (12:46 +0800)]
trac 53 mingw ssize_t redef
Signed-off-by: Andy Green <andy.green@linaro.org>
John Clark [Sun, 9 Feb 2014 23:24:29 +0000 (07:24 +0800)]
fix LWS_NO_SERVER compile
Signed-off-by: John Clark <inidev@gmail.com>
Michael Haberler [Fri, 10 Jan 2014 22:40:14 +0000 (23:40 +0100)]
Subject: [PATCH] libwebsockets.h: use _GNU_SOURCE instead of __USE_GNU
this collides with use of features.h which says '#define __USE_GNU 1'
see also second answer:
http://stackoverflow.com/questions/7296963/gnu-source-and-use-gnu?answertab=active#tab-top
nononame [Sat, 11 Jan 2014 05:12:34 +0000 (13:12 +0800)]
fix lws_lookup off by one
Signed-off-by: nononame <nononame@le-huit.fr>
Andy Green [Sat, 11 Jan 2014 04:37:07 +0000 (12:37 +0800)]
ppoll fix signal mask sense
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Wed, 25 Dec 2013 08:34:37 +0000 (16:34 +0800)]
Use ppoll to allow signals to wake service loop early on change of pollfd event
To enable this code you need to force LWS_HAS_PPOLL to de defined.
#defining it at the top of libwebsockets.c is enough.
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 21 Dec 2013 03:18:34 +0000 (11:18 +0800)]
unify all pollfd lock management
This provides a single place for pollfd event changing,
external locking for that and extpoll management.
It saves about 85 lines of duplication and simplifies the callers.
Signed-off-by: Andy Green <andy.green@linaro.org>
tmu [Sat, 21 Dec 2013 02:33:33 +0000 (10:33 +0800)]
openwrt add build makefile
http://libwebsockets.org/trac/libwebsockets/ticket/50
Signed-off-by: tmu <tmu-INVALID@invalid.com>
Nikolay Dimitrov [Sat, 21 Dec 2013 02:22:17 +0000 (10:22 +0800)]
fix dropmask dereference even when NULL
Signed-off-by: Nikolay Dimitrov <ndimitrov@setelis.com>
pmcdona [Wed, 18 Dec 2013 02:17:25 +0000 (10:17 +0800)]
Trac 53 minw update
http://libwebsockets.org/trac/libwebsockets/ticket/53
Andy Green [Wed, 18 Dec 2013 01:48:26 +0000 (09:48 +0800)]
add locking callback for fds
This adds two new callbacks in protocols[0] that are optional for allowing limited thread
access to libwebsockets, LWS_CALLBACK_LOCK_POLL and LWS_CALLBACK_UNLOCK_POLL.
If you use them, they protect internal and external poll list changes, but if you want to use
external thread access to libwebsocket_callback_on_writable() you have to implement your
locking here even if you don't use external poll support.
If you will use another thread for this, take a lot of care about managing your list of
live wsi by doing it from ESTABLISHED and CLOSED callbacks (with your own locking).
Signed-off-by: Andy Green <andy.green@linaro.org>
James Devine [Sat, 14 Dec 2013 03:41:29 +0000 (11:41 +0800)]
Optionally allow non-SSL connections on same port as SSL
If enabled one listening socket will accept both SSL and plain HTTP connections.
Do not enable if you regard SSL handshake as some kind of security, eg, use
client-side certs to restrict access.
AG: changed flag names, added extra comments, changelog, add -a in test server
Signed-off-by: James Devine <fxmulder@gmail.com>
Signed-off-by: Andy Green <andy@warmcat.com>
James Chen [Sat, 14 Dec 2013 03:35:01 +0000 (11:35 +0800)]
windows use msvc baestd and ssize_t from there
Signed-off-by: James Chen <jianhua.chen@cocos2d-x.org>
Christian Wyss [Tue, 10 Dec 2013 13:34:30 +0000 (21:34 +0800)]
distinguish error from truncation
Signed-off-by: Christian Wyss <christian.wyss@gfk.com>
John Harrison [Tue, 10 Dec 2013 13:20:03 +0000 (21:20 +0800)]
ie10 lws url workaround
Signed-off-by: John Harrison <john@h-agtec.com>
mroszko [Tue, 10 Dec 2013 13:15:00 +0000 (21:15 +0800)]
NO_EXTENSIONS fixes
Trac #52
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Mon, 9 Dec 2013 06:16:17 +0000 (14:16 +0800)]
handle EAGAIN during send
This patch deploys the truncated send work to buffer output in case
either send() or the SSL send return a temporary "unable to send"
condition even though they signalled as writeable.
I added a by-default #if 0 test jig which enforces only half of what
you want to send is sendable, this is working when enabled.
One subtle change is that the pipe reports choked if there is any
pending remaining truncated send. Otherwise it should be transparent.
Hopefully...
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Mon, 9 Dec 2013 03:27:07 +0000 (11:27 +0800)]
solve broken partial file sends
We can't force the wsi state to HTTP_BODY without considering the callback
may already have set the state to sending a file.
This fixes the bug that we can get stalled in the test app at
"choked before able to send whole file"
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sun, 8 Dec 2013 13:26:52 +0000 (21:26 +0800)]
interpret zero rx buffer size as default in http
Reported by pystub
https://github.com/warmcat/libwebsockets/pull/33#issuecomment-
29578258
Signed-off-by: Andy Green <andy.green@linaro.org>
Andreas Pakulat [Mon, 4 Nov 2013 11:51:50 +0000 (12:51 +0100)]
protect the label with an ifdef
The only part that actually goes to this label is inside such an ifdef,
so building without extension support makes gcc bail out since an unused
label is considered an error in this project.
Jonti [Fri, 22 Nov 2013 11:14:26 +0000 (13:14 +0200)]
Subject: [PATCH] Fixed the logging of OpenSSL errors to report the correct
details for the current OpenSSL error
kapejod [Tue, 19 Nov 2013 12:38:16 +0000 (13:38 +0100)]
Support for HTTP POST.
Rewritten by Andy Green to use chunks, add timeout, use extra states in handshake not parser.
Signed-off-by: Andy Green <andy@warmcat.com>
prasannateamf1 [Wed, 13 Nov 2013 01:21:01 +0000 (17:21 -0800)]
support for client authentication based on certs
Andy Green [Wed, 13 Nov 2013 00:03:05 +0000 (08:03 +0800)]
case insensitive http headers
Svetlin wrote on github
According to RFC2616, all header field names in both HTTP requests and HTTP responses are case-insensitive. But libwebsockets uses a case-sensitive compare.
Reproduce:
Run libwebsockets against a server that sends all of its HTTP header field names in lower-case (for example: https://github.com/extend/cowboy). libwebsockets reports an error. The expected behavior is no errors reported and a successful handshake procedure.
This changes the parser reference table in minilex to all lower case.
The code to walk the parser tables then just forces a tolower on the incoming chars.
This (and minilex tables) only applies to header names.
Reported-by: svetlin-mladnov <?@github>
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Tue, 12 Nov 2013 23:45:17 +0000 (07:45 +0800)]
introduce uri args
If the URI coming from the client contains '?' then
- the URI part is terminated with a '\0'
- the remainder of the URI goes in a new header WSI_TOKEN_HTTP_URI_ARGS
- the remainder of the URI is not subject to path sanitization measures (it
still has %xx processing done on it)
In the test server, http requests now also dump header information to stderr.
The attack.sh script is simplified and can now parse the test server header dumps.
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Tue, 12 Nov 2013 22:53:21 +0000 (06:53 +0800)]
back up directory paths properly
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sun, 10 Nov 2013 23:30:33 +0000 (07:30 +0800)]
real http status codes update attack.sh
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sun, 10 Nov 2013 22:14:52 +0000 (06:14 +0800)]
uri santitation fixes deal with single dot update attack.sh
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sun, 10 Nov 2013 07:15:21 +0000 (15:15 +0800)]
translate and protect uri test sever use uri path
This translates %xx in the GET uri and removes /.. and /... type sequences along with
translating // or /// etc to /.
Since the result is hopefully secure, it also changes the test server to actually use
the uri path pasted on a resource directory without whitelisting.
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 9 Nov 2013 03:59:56 +0000 (11:59 +0800)]
introduce LWS_CALLBACK_FILTER_HTTP_CONNECTION
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 9 Nov 2013 03:06:29 +0000 (11:06 +0800)]
add cookie example to test server
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 9 Nov 2013 03:04:35 +0000 (11:04 +0800)]
allow other headers in http send file
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 9 Nov 2013 03:40:32 +0000 (11:40 +0800)]
make ah available in http callback properly
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 9 Nov 2013 02:09:09 +0000 (10:09 +0800)]
improve minilex use external parsing header
Clean up minilex
Move the header output to stdout
Introduce lexfile.h as the header output
Use lexfile.h in both minilex itself and lws
Add the following header support
"Accept:",
"If-Modified-Since:",
"Accept-Encoding:",
"Accept-Language:",
"Pragma:",
"Cache-Control:",
"Authorization:",
"Cookie:",
"Content-Type:",
"Date:",
"Range:",
"Referer:"
Signed-off-by: Andy Green <andy.green@linaro.org>
Graham Newton [Sat, 9 Nov 2013 00:11:41 +0000 (08:11 +0800)]
windows workaround unsigned fd in emulated poll
A common practise to temporarily disable a socket in a poll call is to negate the socket fd. poll should then ignore the socket.
emulated_poll does this with the following code:
if (poll_fd->fd < 0 || !poll_fd->events)
goto skip1;
However on Windows the fd field in WSAPOLLFD is unsigned int!!! So the check for a negative fd value always fails.
This results in select returning an error with an error code of 10038 -- Socket operation on nonsocket.
The fix is to type cast fd like so:
if ((int)poll_fd->fd < 0 || !poll_fd->events)
This may be the cause of some high CPU load reports. I noticed the load being 50% with my application running on Windows XP.
Signed-off-by: Graham Newton <gnewton@peavey-eu.com>
Andy Green [Sat, 9 Nov 2013 00:07:38 +0000 (08:07 +0800)]
interface improve visibility fail if not found
Signed-off-by: Andy Green <andy.green@linaro.org>
Andreas Pakulat [Mon, 28 Oct 2013 14:00:16 +0000 (15:00 +0100)]
Windows Fix usage in Windows XP targetting projects
This is important since this stuff is in publicly included headers
and even though the usage of the WSAPoll API is a runtime decision
the public headers may be used in code that needs to build with
_WIN32_WINNT=0x0501 to support WinXP.
When building a project using libwebsockets with that define set
to 0x0501 winsock.h will not define the WSAPOLLFD struct causing that
project to fail to compile.
Andy Green [Mon, 4 Nov 2013 02:17:50 +0000 (10:17 +0800)]
unixify websock w32.h line endings
Signed-off-by: Andy Green <andy.green@linaro.org>
Andreas Pakulat [Wed, 16 Oct 2013 15:57:17 +0000 (17:57 +0200)]
Windows Fix include for win32helpers
The installed headers are all in the same directory, so no relative
path possible there. The cmake files already add the win32helpers
directory to the include paths, so this builds just fine.
Andreas Pakulat [Mon, 28 Oct 2013 15:02:12 +0000 (16:02 +0100)]
Windows Fix installing the dll
Without a runtime install directory, cmake won't install the dll at
all. Only the import library was installed. So fix that.
Andreas Pakulat [Mon, 28 Oct 2013 14:18:04 +0000 (15:18 +0100)]
Support _WIN32 in addition to WIN32 define
MSVC8 and MSVC9 set only _WIN32, so support that define as well to be able
to compile libwebsockets with those compilers.
Andreas Pakulat [Mon, 28 Oct 2013 14:15:44 +0000 (15:15 +0100)]
Drop wsockcompat.h header its not part of msvc
I don't see a wsockcompat.h anywhere in MSVC9 or MSVC8 and their
corresponding sdk's. It does not seem like this is a standard windows
header, so better drop that and add the compat-defines to the same
place that already has other WSA compat defines.
Graham Newton [Mon, 4 Nov 2013 01:47:06 +0000 (09:47 +0800)]
handle HUP on any mode socket
I am using libwebsockets on Windows 7 in external poll mode.
I am finding that if I present a socket fd on a normal HTTP connection
(LWS_CONNMODE_HTTP_SERVING:) to libwebsocket_service_fd with just a HUP
event, the event doesn't get handled but revents gets cleared indicating
that the event has been handled. Should it be handled in the same way
as LWS_CONNMODE_WS_SERVING?
(Modified by AG to apply to all sockets)
Signed-off-by: Andy Green <andy.green@linaro.org>
Signed-off-by: Graham Newton <gnewton@peavey-eu.com>
Andy Green [Sat, 26 Oct 2013 13:18:44 +0000 (21:18 +0800)]
trac41 WIN32 add wsockcompat.h
Using Windows 7 64 bit, cloned repo on
20130926. Using Qt creator and Microsoft Visual C++ Compiler 9.0 (x86).
Result (errors from compile output): D:\Projects\CDPStudioAPI\libwebsockets_orig\lib\client-handshake.c:87: error: C2065: 'EALREADY' : undeclared identifier
D:\Projects\CDPStudioAPI\libwebsockets_orig\lib\client-handshake.c:87: error: C2065: 'EINPROGRESS' : undeclared identifier
Possible solution is to use wsockcompat.h (compatibility header for using EALREADY, EINPROGRESS etc in older versions of Windows SDK). Compiled fine when I #included wsockcompat.h into client-handshake.c
Reported-by: mart22n via Trac 41
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Sat, 26 Oct 2013 12:36:08 +0000 (20:36 +0800)]
v1.2.3 changelog update
Signed-off-by: Andy Green <andy.green@linaro.org>
Markus Elfring [Sat, 26 Oct 2013 12:23:00 +0000 (20:23 +0800)]
make header protection unique remove __prefix
Signed-off-by: Markus Elfring <markus.elfring@web.de>
Markus Elfring [Sat, 26 Oct 2013 12:03:43 +0000 (20:03 +0800)]
fix cmake quoting
A few parameters (like "${PROJECT_BINARY_DIR}/lws_config.h" and "${PROJECT_SOURCE_DIR}/lib") are passed to CMake commands in your build script without enclosing them by quotation marks. I see that these places will result in build difficulties if the contents of the used variables will contain special characters like spaces.
Signed-off-by: Markus Elfring <markus.elfring@web.de>
shys [Fri, 25 Oct 2013 13:50:21 +0000 (15:50 +0200)]
cover-proxy-read-EAGAIN.patch
Subject: [PATCH] Because of nonblocking sockets reading from proxy might not
work in one turn. We should check for EAGAIN.
shys [Fri, 25 Oct 2013 13:49:11 +0000 (15:49 +0200)]
Subject: [PATCH] Additional check for EISONN. On iOS "if (connect()) == -1 ||
errno == EISCONN)" doesn't work.
shys [Fri, 25 Oct 2013 13:52:47 +0000 (15:52 +0200)]
Subject: [PATCH] On iOS with nonblocking sockets I experience SSL error which
is not really an error (
http://lists.freeradius.org/pipermail/freeradius-users/2009-August/040155.html
). Added check for that. Additional fixes for comments and logging.
u0u0 [Fri, 25 Oct 2013 09:13:11 +0000 (17:13 +0800)]
fix origin
And fix client header of Origin, add prefix "http://" to make a valid URI, to make golang websocket server happy.
Andy Green [Fri, 25 Oct 2013 14:07:57 +0000 (22:07 +0800)]
defer changing to proxy uri until connected
Reported-by: shyswork
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Thu, 24 Oct 2013 14:07:04 +0000 (22:07 +0800)]
add EISCONN handling to client connect attempt
Signed-off-by: Andy Green <andy.green@linaro.org>
shys [Thu, 24 Oct 2013 14:27:08 +0000 (22:27 +0800)]
wait ssl connect completion on client proxy path
Subject: [PATCH] We can ran into situation (at least on iOS) when with openssl
nonblocking BIO and http proxy we don't perform ssl_connect straight away so
we need to retry until we finish ssl_connect. If we don't do that we will
fail in LWS_CONNMODE_WS_CLIENT_WAITING_PROXY_REPLY when testing for "HTTP/1.0
200" successful connection.
Signed-off-by: shys <shyswork@zoho.com>
shys [Thu, 24 Oct 2013 14:12:03 +0000 (22:12 +0800)]
manual proxy forcing api
add function to manually setup proxy. Useful on iOS where
getenv doesn't return proxy settings
Simplified by AG
Signed-off-by: shys <shyswork@zoho.com>
Signed-off-by: Andy Green <andy@warmcat.com>
Arokux X [Thu, 24 Oct 2013 13:53:39 +0000 (21:53 +0800)]
CMAKE help if cwd not git dir
Hi all,
please consider applying the patch below. The git is called in some
different directory and that is why fails. I'm not sure why it was
called in different directory, maybe because I build libwebsockets
with CMake's ExternalProject.
Best,
Arokux
Signed-off-by: Arokux X <arokux@gmail.com>
Andy Green [Thu, 24 Oct 2013 13:47:06 +0000 (21:47 +0800)]
fix typo in NO_SERVER case
Reported-by: shyswork via github
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Mon, 21 Oct 2013 22:49:30 +0000 (06:49 +0800)]
use compatible_close for sockets
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Fri, 18 Oct 2013 11:23:06 +0000 (19:23 +0800)]
fix clean return path user code handle
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Thu, 17 Oct 2013 00:09:19 +0000 (08:09 +0800)]
add explicit error for partial send
This patch adds code to handle the situation that a prepared user buffer could not all be sent on the
socket at once. There are two kinds of situation to handle
1) User code handles it: The connection only has extensions active that do not rewrite the buffer.
In this case, the patch caused libwebsocket_write() to simply return the amount of user buffer that
was consumed (this is specifically the amount of user buffer used in sending what was accepted,
nothing else). So user code can just advance its buffer that much and resume sending when the socket
is writable again. This continues the frame rather than starting a new one or new fragment.
2) The connections has extensions active which actually send something quite different than what the
user buffer contains, for example a compression extension. In this case, libwebsockets will dynamically
malloc a buffer to contain a copy of the remaining unsent data, request notifiction when writeable again,
and automatically spill and free this buffer with the highest priority before passing on the writable
notification to anything else. For this situation, the call to write will return that it used the
whole user buffer, even though part is still rebuffered.
This patch should enable libwebsockets to detect the two cases and take the appropriate action.
There are also two choices for user code to deal with partial sends.
1) Leave the no_buffer_all_partial_tx member in the protocol struct at zero. The library will dyamically
buffer anything you send that did not get completely written to the socket, and automatically spill it next
time the socket is writable. You can use this method if your sent frames are relatvely small and unlikely to get
truncated anyway.
2) Set the no_buffer_all_partial_tx member in the protocol struct. User code now needs to take care of the
return value from libwebsocket_write() and deal with resending the remainder if not all of the requested amount
got sent. You should use this method if you are sending large messages and want to maximize throughput and efficiency.
Since the new member no_buffer_all_partial_tx will be zero by default, this patch will auto-rebuffer any
partial sends by default. That's good for most cases but if you attempt to send large blocks, make sure you
follow option 2) above.
Signed-off-by: Andy Green <andy.green@linaro.org>
Andy Green [Fri, 20 Sep 2013 12:26:12 +0000 (20:26 +0800)]
trac 40 client connection properly nonblocking
As spotted by JM on Trac#40
http://libwebsockets.org/trac/libwebsockets/ticket/40
client connect didn't do anything about being truly nonblocking. This patch
should hopefully solve that.
Signed-off-by: Andy Green <andy.green@linaro.org>
Fujii Bunichiroh [Wed, 18 Sep 2013 13:01:02 +0000 (21:01 +0800)]
force failed header parse to act
''...Websocket servers return failure responses other than HTTP Status 101 in the case of
mismatches of WebSocket version or additional header data etc...
It seems that libwebsockets shows "WARN: problems parsing header" error in such cases without
calling any callbacks or returning error code. Is this right?
I modified lib/client.c to handle this.''
Signed-off-by: Fujii Bunichiroh <Bunichiroh.Fujii@jp.sony.com>