Dear Forrest,
Your question is both off-topic for libmicrohttpd, and impossible to
answer as there are complex trade-offs that depend too much on the
context in which your solution is used, so one cannot even make a good
recommendation without understanding your requirements a lot better. And
Hi Silvio,
Well, we didn't try to break anything with the update, so you shouldn't
have to change anything on your side either. Can you send us a stack
trace/valgrind output and/or test code to reproduce? That'll expedite a
fix...
Happy hacking!
-Christian
On 2/26/24 19:44, silvioprog wrot
Dear all,
We're happy to announce reaching the first milestone for the
STF-funded MHD 2.0 project, which is completing the MHD HTTP header
and thus the design for the next generation API. We have now
spent several months iterating between discussing, designing,
optimizing, editing, and testing d
Hi MrDave,
You're not exactly doing anything wrong, there is simply no guarantee
(by the MHD API) as to when the ContentReaderFreeCallback will be called
other than "after MHD is done with the response". So MHD _could_ call it
before or after the NOTIFY_COMPLETED callback, so your code should
hat
case, but a SIGPIPE), unless you're on GNU/Linux. If this is not the issue,
try using valgrind to find out more about your segfault...
Happy hacking,
Christian
>
> On Thu, Sep 15, 2011 at 2:57 PM, Christian Grothoff
>
> wrote:
> > Eh, MHD_connection_close is NOT
On 09/16/2011 01:53 PM, Frank Meier wrote:
hi
I tried to use libmicrohttpd on solaris 10, and my simple programm only
starting and stopping the daemon (see below) did hang in.
after some investigation (see debug code) I found, that the shutdown
call in daemon.c fails, since the daemon socket is
Dear Peter,
I suspect you're simply queueing your response too early. The API may
not be 100%-intuitive here: if you have an upload (and did not replace
the 100-CONTINUE with an error), you need to wait for MHD to call you
with an '*upload_size' of zero before giving your response. From what
ility to fix it).
Happy hacking!
Christian
On 09/18/2011 01:17 PM, Peter Ross wrote:
On Sun, Sep 18, 2011 at 08:03:00AM +0200, Christian Grothoff wrote:
Dear Peter,
I suspect you're simply queueing your response too early. The API
may not be 100%-intuitive here: if you have an uploa
I see. But given that on Linux it works without a pipe and that a pipe
also "consumes" two extra FDs, I think the answer should be slightly
different: revert 14584 *conditionally* for non-Linux, non-FreeRTOS
OSes. So we'd have pipes on Solaris/*BSD/OS X, and no-pipes on
Linux/FreeRTOS. That
On Monday, September 19, 2011 10:29:24 AM Frank Meier wrote:
> hi
>
> > Well, that would seem to solve the problem, but would introduce a
> > race. With 'close', another thread might open a file or socket and
> > MHD's might then select/poll/accept on that unrelated socket.
>
> I'm not sure if I
24566==by 0x626CD8B: start_thread (pthread_create.c:304)
> ==24566==by 0x5A9D04C: clone (clone.S:112)
> ==24566== Address 0x50 is not stack'd, malloc'd or (recently) free'd
>
>
>
>
> On Fri, Sep 16, 2011 at 1:41 PM, Christian Grothoff
>
> wr
Yes, we should have some meta-#define and set it via configure.
-Christian
On Sunday, September 18, 2011 02:25:40 PM Will Bryant wrote:
> That sounds workable.
>
> I'd be a little reluctant to sprinkle lists of platforms in ifdefs
> throughout the project - we'd be too likely to miss one bit of
On Monday, September 19, 2011 03:00:28 AM Will Bryant wrote:
> I have a patch for this almost working on OS X, but I'm getting hangs
> running "make check" in daemontest_timeout. My patch is causing an error,
> which I'll look into, but it's hanging because the current error handling
> code tries
et_response_cleanup.c:70
#4 0x0805157d in main (argc=1, argv=0x6b) at
daemontest_get_response_cleanup.c:141
On 20/09/2011, at 00:08 , Christian Grothoff wrote:
Yes, we should have some meta-#define and set it via configure.
-Christian
Thanks, the patch seems fine (minor fixes and configure-integration are in SVN
16982), and are commited (SVN 16980, 16981).
On the OpenIndiana performance, do you have something like 'strace' where you
could monitor what's going on? Most likely the code hangs blocking in some
syscall for long
I guess we could create a channel #libmicrohttpd on freenode --
sometimes these days there is a lot of GNUnet-talk on #gnunet. I can
also add a bot to archive IRC chat to #libmicrohttpd. But not tonight ;-).
Happy hacking!
-Christian
On 09/24/2011 08:37 PM, Nils Durner wrote:
Hi,
Is ther
Hi!
This is a good question, I didn't think this kind of HTTP request was
even legal... But I think we could make it work. First of all, the
current code just throws the "wsdl" away, but we can fix that:
$ svn diff
Index: src/daemon/connection.c
=
Dear all,
I'm pleased to announce the release of GNU libmicrohttpd 0.9.15.
GNU libmicrohttpd is a small C library that is supposed to make it easy to run
an HTTP server as part of another application. GNU libmicrohttpd is fully
HTTP 1.1 compliant and supports IPv6. Finally, GNU libmicrohttpd i
You need to wait until your callback is called with an *upload_size of 0
(MHD does *incremental* post-processing!). Then you can queue the reply.
Happy hacking!
Christian
On 10/27/2011 07:55 PM, Avner Flesch wrote:
Hi,
I am using the post process to handle POST message.
After that I want to
On Thursday, October 27, 2011 09:29:52 PM Sven Geggus wrote:
> Hello,
>
> I try to use libmicrohttpd in conjunction with a main function which
> acquires data in an endless loop (currently every 4 seconds).
>
> What I want to achieve is that all the current client connections
> with an active req
why the connections remain "ESTABLISHED" (as
reported by netstat -atn). Perhaps the browser keeps the connections open
in order to speed up subsequent requests?
I would appreciate any feedback as to what I may be overlooking and how I
can resolve this issue.
Regards,
Jon Nalley
/*
T
nable to do so. I am wondering if my issue may be
related to uClibc on the PowerPC 405.
Thanks,
Jon Nalley
On Fri, Oct 28, 2011 at 11:14 AM, Christian Grothoff
wrote:
Dear Jon,
I've tried (and failed) to reproduce your problem. I've attached my
modified version of 'minimal_ex
On 10/28/2011 10:49 PM, Sven Geggus wrote:
Christian Grothoff schrieb am Freitag, den 28. Oktober um 15:01 Uhr:
Well, that's technically not required. You can simply integrate your data-
polling loop into the external select loop of MHD and in the response
callbacks for data-aquisition r
It's not a race (no threads=> no races!), I had forgotten to initialize
the size of the response. I've added a global flag for "terminate after
one round" as well now.
-Christian
On 10/29/2011 04:16 PM, Sven Geggus wrote:
Christian Grothoff schrieb am Samstag, den 2
On 10/30/2011 05:16 PM, Avner Flesch wrote:
Hi,
I am developing simple secured HTTP (HTTPS) server for embedded system.
It seems that the delay is rather long - for example ~230 KB file delay
is almost 6 seconds.
Is there a way to speed it up?
On an embedded system, I suspect the main contribu
sure that you compiled libgcrypt with optimizations (-O3).
Best regards,
Christian
-Original Message-
From: Christian Grothoff
Reply-to: libmicrohttpd development and user mailinglist
To: libmicrohttpd@gnu.org
Subject: Re: [libmicrohttpd] HTTPS delay
Date: Sun, 30 Oct 2011 17:56:01 +0100
vate key (for
example, 1024 bit -- you can try less, but it won't be really secure anymore).
Happy hacking,
Christian
> Thanks
>
> Avner
>
> -----Original Message-
> From: Christian Grothoff
> Reply-to: libmicrohttpd development and user mailinglist
>
&
Hi Dirk,
Please let me know if the attached patch fixes the problem...
Happy hacking,
-Christian
On 11/11/2011 02:52 PM, dirk@gmail.com wrote:
Hello together,
I observed after problems with libmicrohttpd after updating to 0.9.16.
The problems seem to be related to the patch 17892.
I hav
Hi Dirk,
Please let me know if the attached patch fixes the problem...
Happy hacking,
-Christian
On 11/11/2011 02:52 PM, dirk@gmail.com wrote:
Hello together,
I observed after problems with libmicrohttpd after updating to 0.9.16.
The problems seem to be related to the patch 17892.
I hav
If you look at MHD_connection_handle_idle for the case of
MHD_CONNECTION_FOOTERS_RECEIVED, you will see that MHD tests if you (or
the browser) added 'Connection: close' and if so, MHD *will* close the
TCP connection.
So adding a 'Connection: Close' header not only tells the browser to not
re-
Dear Kendrick,
I see your point, and after re-reading the respective standard it is
clear that for systems without a clock it is actually permitted to
exclude the DATE header. Now, I don't think this should be done via
configure -- wouldn't a flag given to MHD_start_daemon do just as well?
(
This is now implemented in SVN 18242.
Happy hacking,
Christian
On 11/21/2011 04:54 PM, Kendrick Hamilton wrote:
Sure, that would work
Thank you for your fast response.
Kendrick
On 11-11-21 08:50 AM, Christian Grothoff wrote:
Dear Kendrick,
I see your point, and after re-reading the
On 11/24/2011 11:16 AM, Andreas Wehrmann wrote:
I noticed that MHD_create_post_processor() refuses to create a processor
when the
MIME type is not "application/x-www-form-urlencoded" or
"multipart/form-data".
Right, that is the documented and intended behavior.
However, I'd like to post proce
Well, if creating the PP failed, why would you try to destroy it? ;-).
I don't think that any function has to work nicely when being called
with NULL (unless specified that it should accept NULL, of course).
Happy hacking,
Christian
On 11/24/2011 01:35 PM, Andreas Wehrmann wrote:
I ran into
On 11/24/2011 03:10 PM, Andreas Wehrmann wrote:
On 11/24/2011 03:01 PM, Christian Grothoff wrote:
Well, if creating the PP failed, why would you try to destroy it? ;-).
I don't think that any function has to work nicely when being called
with NULL (unless specified that it should accept
Yeah, that would be a very good assumption, as it is certainly correct.
Patch applied as SVN 18423. Thanks!
Happy hacking!
Christian
On 12/01/2011 02:32 PM, Sven Geggus wrote:
Hi there,
am I right in the assumption, that the documentation at
http://www.gnu.org/s/libmicrohttpd/tutorial.html
On 12/13/2011 07:26 PM, Sven Geggus wrote:
Hello,
to support dual stack I just changed my code from:
MHD_start_daemon(MHD_NO,...
to
MHD_start_daemon(MHD_USE_IPv6,...
Unfortunately the daemon seems to be ipv6 _only_ now.
How ist this supposed to work?
From generic socket-programming I usuall
On 12/13/2011 08:09 PM, Sven Geggus wrote:
Christian Grothoff wrote:
The correct way to get dual-stack is to call MHD_start_daemon twice,
once for IPv4 and once for IPv6.
Hm so what would I do if I want to server both IPv4 and IPv6 clients
in one select loop? Calling MHD_get_fdset twice
On 12/19/2011 10:20 PM, Keith Mendoza wrote:
I'm just wondering if it's possible when a client is uploading a file
that the content arrives out of sequence. For example is it possible
to get let's say a sequence of offset values: 0, 50, 10, 20, 30, 40?
I'm under the impression that this will not
The offset comes from MHD counting bytes, not from the client.
Happy hacking,
Christian
On 12/20/2011 05:47 PM, Keith Mendoza wrote:
I figured that much. A follow-up question: is the offset provided by the
client as part of multi-part or is this a libmicrohttpd thing? I just
want to know if I
On 01/15/2012 04:41 PM, Steinar H. Gunderson wrote:
Hi,
Apologies for not using the bug tracker, but I really couldn't get it to
work; on one of my email addresses, it never sent a notification, and in
another one, I got a notification but as soon as I had created an account and
set a password,
On 01/16/2012 10:31 AM, Steinar H. Gunderson wrote:
On Sun, Jan 15, 2012 at 05:11:14PM +0100, Christian Grothoff wrote:
0.4.6? You got to be kidding me. We're at 0.9.17, 0.4.6 is about 2
years old.
That's the joy of being on Debian stable, I guess :-)
More importantly, the patch
Dear all,
After a recent update of libcurl / libgnutls on my Debian unstable
system, the fully automated tests of GNU libmicrohttpd for HTTPS started
to fail. These tests start an HTTPS server using libgnutls and GNU
libmicrohttpd and then try downloading a site using libcurl.
Here is the k
Daniel Stenberg wrote:
On Thu, 19 Jan 2012, Christian Grothoff wrote:
One of our tests also provokes a failure by selecting incompatible
versions of the SSL protocol. With older versions, that test produces
ONCE:
curl version: libcurl/7.21.3 GnuTLS/2.8.6 zlib/1.2.3.4 libidn/1.18
curl_easy_pe
This has actually been on my "potential features" list for a little
while. The implementation is quite trivial; all that is really needed
is some minor modification to the MHD_response internals / API to give
the application a way to create a "websocket" response; that response
would then send
On 01/23/2012 07:12 PM, Simon Newton wrote:
I need to add server triggered notifications to my app so I'd love to
see this as well.
Ok, I guess I should propose an API then. Given that MHD is really
supposed to "just" do HTTP and that (a) WebSocket standardization
doesn't seem to be complete
On 01/23/2012 11:14 PM, Daniel Stenberg wrote:
On Mon, 23 Jan 2012, Nikos Mavrogiannopoulos wrote:
It doesn't look right. I'd change "-VERS-TLS-ALL:+VERS-SSL3.0" with
"NORMAL:-VERS-TLS-ALL:+VERS-SSL3.0".
However your priority string seem quite radical. You only allow SSL 3.0.
That particular
On 01/24/2012 10:44 AM, Sven Geggus wrote:
Christian Grothoff wrote:
The one big major roadblock for implementing this in MHD is that this
would be tricky to automatically (!) test any such implementation.
Hm how about this one:
http://www.tavendo.de/autobahn/testsuite.html
Great, that
The real issue with this kind of performance monitoring is that for
multi-threaded applications it is very hard to give you a "sound"
number. Also, I don't see how this would be helpful for the application
that is running at the time.
If the system operator wants this information, it is prett
On 02/22/2012 07:13 PM, Glenn wrote:
I've been getting familiar with libmicrohttpd and the more more I understand
it, the more I like it. One thing I haven't been able to figure out, though, is
why the binaries I've been able to build are so large.
Executing './configure --disable-https --disa
Dear all,
I just wanted to let you know that Daniel Baumann orphaned the
libmicrohttpd Debian package, so if one of you is a Debian Developer (or
wants to become one...), this might be a good chance to pick up a new
package ;-).
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=660445
Danie
On 02/27/2012 09:41 PM, Erik Smith wrote:
The arguments don't come back in the order they appear. Is that a bug?
Good question. The API does not specify that they have to come back in
the order that they appear, so one could argue that no. The reason is
also clear: during parsing, we prepe
e how much time is needed, if there are no immediate objections from
you I could contact Daniel and we could work something out.
yours,
pg
P.S.
sorry for misprints I do not have a physical keyboard on this device.
On Feb 27, 2012 8:52 PM, "Christian Grothoff" mailto:christ...@grothoff
Dear Martin,
The first call to the "test_send_response_helper" is made *before* the
upload data is available (after we just got the headers) so that you can
choose not to send 100 CONTINUE for http/1.1. You need to change your
function to return an 'int' and return 'MHD_YES' to get the next call
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 04/26/2012 05:50 AM, Martin Dluhos wrote:
> On 04/25/2012 09:22 AM, Christian Grothoff wrote:
>> The first call to the "test_send_response_helper" is made *before*
>> the upload data is available (after we just got the he
On 05/22/2012 10:45 PM, s...@orbitalfox.com wrote:
Hello,
Im using libmhd from a not so old svn version. When I do the
following:
if(0 == strcmp(method, MHD_HTTP_METHOD_POST)){
const char* content_t = MHD_lookup_connection_value(c,
MHD_HEADER_KIND, "Content-Type"
Dear all,
I'm pleased to announce the release of GNU libmicrohttpd 0.9.20.
GNU libmicrohttpd is a small C library that is supposed to make it easy
to run an HTTP server as part of another application. GNU libmicrohttpd
is fully HTTP 1.1 compliant and supports IPv6. Finally, GNU
libmicrohttpd is
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 06/06/2012 12:27 PM, Cristian Constantin wrote:
> hi!
>
> the file:
>
> libmicrohttpd/src/testcurl/https/tls_multi_thread_mode_test.c
>
> tests multithreading capabilities of libmicrohttpd(+tls) with multiple threads
> both on the client and on t
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 06/05/2012 03:40 PM, simon h wrote:
> Dear all,
> I'm looking to use libmicrohttpd to implement some webservices style
> interfaces on modules build for a windows environment.
>
> I've downloaded the last windows binaries (0.4.5 ?) and notice the
>
On 06/08/2012 05:32 PM, Cristian Constantin wrote:
> On Thu, Jun 07, 2012 at 02:20:26PM +0200, Christian Grothoff wrote:
> On 06/06/2012 12:27 PM, Cristian Constantin wrote:
> >>> hi!
> >>>
> >>> the file:
> >>>
> >>> libmicrohttpd
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 06/12/2012 04:05 PM, Enrico Migliore - FATTI srl wrote:
> Hi,
>
> I'm reading the source code and I got a question.
>
>
> Let's say that a browser crashes after opening and using some sockets.
>
> The browser can't close the sockets since it jus
Dear Mike,
Thanks for your patches. I've committed all 3 of them in SVN 22116
(first two) and SVN 22117 (this one) respectively, so they will be part
of the next release.
Happy hacking!
Christian
On 06/19/2012 06:50 PM, Mike Crowe wrote:
> Change various uses of time(NULL) to new MHD_monotoni
On 06/20/2012 08:27 AM, Enrico Migliore - FATTI srl wrote:
> Hello,
>
> I got MHD up and running and I ran successfully many tests.
>
> I can't find an example of file upload via the POST method.
>
> Are there any?
Not AFAIK. There is an example for 'POST', but I do not think we have
one for uplo
On 07/29/2012 07:46 PM, Markus Becker wrote:
> Hi,
>
> I was wondering whether it is possible to not call any MHD functions in my
> MHD_AccessHandlerCallback function, but do that later? I am trying to
> implement a proxy and have to fetch the data to return to the client from the
> server firs
Short answer: nothing has happened since. IIRC the proposed API still
had some issues (but I don't recall which ones right now), and nothing
has been implemented by me (and I have received no patches or specific
feedback on the API). It's still on my agenda, but I'm currently
focused on othe
Dear Mike,
I've added the missing #includes to make all examples compile without
warnings on my system.
I've also been able to reproduce your error with the largepost.c
example. I'll try to look into it more later tonight (as it is not
obvious to me what is wrong).
Happy hacking!
Christ
think the error has to do with how sprintf is being used. If you pass a
const char* to send_page all goes well. I created a C++ version using
std::string and that works fine too.
Cheers,
MS
On 14 août 2012, at 19:18, Christian Grothoff wrote:
Dear Mike,
I've added the missing #includ
yway, I have done another sample test and the tcpdump capture looks OK.
Regards,
Bertrand
2012/8/8 Christian Grothoff <mailto:christ...@grothoff.org>>
Short answer: nothing has happened since. IIRC the proposed API
still had some issues (but I don't recall which ones right now
On 08/20/2012 08:29 AM, Martin Velek wrote:
Hello,
I am trying to use this library(libmicrohttpd-0.9.9.tar.gz) with LWIP
and FreeRTOS on LPC2478. Due to limited resources I have enabled an
option "-fshort-enums" (use as small as possible size to hold values
from enum). Similar post has been alre
Yes, this should be fine.
-Christian
On 08/21/2012 08:07 AM, Martin Velek wrote:
Hello,
is it possible to decrease the XBUF_SIZE (line #34, postprocessor.c)
from 1024 to e.g. 512? Even if this could bring some performance
issues. On embedded system with RTOS the stack size is a limiting
factor
On 08/21/2012 08:44 AM, Martin Velek wrote:
Hello,
Would it not be simpler (and possibly more efficient) to just use 'int'
instead of 'enum MHD_OPTION' as suggested by the compiler's warning? I'd
certainly be willing to do that instead as well...
Yes, it would be. I hope it will not break code
04:02 PM, Keith Mendoza wrote:
Maybe this can be made into a configuration item.
Sent from my Android phone
On Aug 20, 2012 11:22 PM, "Christian Grothoff" <mailto:christ...@grothoff.org>> wrote:
Yes, this should be fine.
-Christian
On 08/21/2012 08:07 AM,
Fixed in SVN 23381. -Christian
On 08/23/2012 01:41 AM, m...@mikesolomon.org wrote:
Hey Christian,
Thanks for fixing that! One other little thing about largepost.c:
In answer_to_connection for the if if (0 != *upload_data_size) {...} else
{...}, the file pointer in con_info is still open when
/AFAIK, t/he only difference between PUT and POST is that PUT must be
idempotent. That is a semantic distinction of relevance to your
application, it changes nothing for the wire protocol or the MHD API.
MHD internally makes no distinction between PUT and POST and will pass
you all of the dat
On 08/25/2012 10:37 AM, Martin Velek wrote:
Hello,
may I ask you, how to correctly implement SSI
(http://en.wikipedia.org/wiki/Server_Side_Includes) pages into
libmicrohttpd?
E.g. I have a page
#define PAGE "libmicrohttpd demo"\
"Temeprature:"
The SSI parser knows that he should
Technically, whatever amount of memory is left in the memory pool
will be used for the buffer. You can use the
MHD_OPTION_CONNECTION_MEMORY_LIMIT
to set the initial size of the memory pool (default: 32k). Some
of that memory will go to storing the HTTP headers, the rest will be used as
the buf
; only 4 bytes overhead.
Martin
On Sat, Aug 25, 2012 at 2:21 PM, Christian Grothoff
wrote:
Technically, whatever amount of memory is left in the memory pool
will be used for the buffer. You can use the
MHD_OPTION_CONNECTION_MEMORY_LIMIT
to set the initial size of the memory pool (defau
Dear Martin,
You can totally do it later (the disadvantage being that then the client
will have started the upload, and if you're then out-of-memory and
cannot handle the request, bandwidth will be wasted). So the question
is if it is worse to put the test on filepost1 vs. filepost2 into the
or is implemented on many server, e.g.
wget www.google.cz/dasfasdf --post-file=/tmp/very_big_file.
Any ideas how to handle POST requests?
Thank you
Martin
On Wed, Aug 29, 2012 at 11:51 AM, Christian Grothoff
wrote:
Dear Martin,
You can totally do it later (the disadvantage being that then the cl
On 09/19/2012 07:54 PM, Eivind Sarto wrote:
> Hi,
>
> Most (all?) of the examples in libmicrohttpd will queue the response from the
> second call to the client handler.
>
> If you already know that you want to respond with an error during the initial
> call to the client handler (client context
On 09/19/2012 10:42 PM, Eivind Sarto wrote:
> Thanks.
> Great.
> With GET/HEAD will keep-alive work, too?
Yes.
-Christian
On 09/20/2012 07:03 PM, Glenn wrote:
>
> Following up on today's digest ---
>
> I'm going through the libmicrohttpd tutorial and working through the
> examples. One question that comes to my mind is why the second invocation of
> the connection handler is necessary/desirable. In the tutorial it
On 09/21/2012 04:46 PM, Enrico Migliore - FATTI srl wrote:
Hi,
I'm playing with MHD and I got 2 questions.
Question 1
--
I noticed that MHD calls shutdown() and then close() after creating
each HTTP response. Is that correct?
Yes.
If yes, I got another question:
Since HTTP 1.1 allow
On 10/05/2012 11:04 AM, Chunxi Chen wrote:
> Dear Sir/Madam,
>
> I want to use libmicrohttpd in my project. But I have a doubt: if I
> run the server in the HTTPS mode, does it still support HTTP access? I
> really appreciate it if you could reply me. Many thanks.
>
> Best regards
> Chunxi
If you w
Dear Luis,
Thanks for the report, should be fixed in SVN 24333 by moving all
WINDOWS macro tests to #ifdef/#ifndef.
Happy hacking!
Christian
On 10/16/2012 12:19 AM, luis garcia wrote:
> Hi,
>
> When I compile the library with the c compiler instead of c++ i get errors
> due to mix use of #if a
Dear Dilyan,
Not every race is a bug. In all of the cases you mention, the code
*expects* "undefined" results (as in, a race where either thread may
win) and is written to work fine in either case (to the very best of my
knowledge --- the races you mention were all known when this part was
d
Dear Jan,
Totally reasonable patches, except for your de-capitalization of the
TSEARCH (and related calls):
/* Search for the IP address */
- if (NULL == (nodep = TSEARCH (key,
+ if (NULL == (nodep = tsearch (key,
This would break our W32 build.
I'll try to find some time later tonight
Dear Helin,
Sending data with a GET request is, eh, unusual. Have you considered
using PUT? In any case, I suspect it should kind-of work nevertheless
(but of course I didn't try this). The PostProcessor not working is most
likely because it didn't understand your data encoding. You can eit
On 11/18/2012 01:21 AM, Helin Wang wrote:
Hi Christian,
Thanks for your response.
I did more tests, seems post processor can be created but won't be called
if I did not specify encoding. post processor can't be created if I use
GET. Event more, with GET (curl -d 'data' -G), *upload_data_size wil
Dear all,
I've finally made up my mind on a decent API for the "HTTP UPGRADE"
implementation. HTTP UPGRADE is used for websockets (and possibly other
HTTP-extensions) to switch a TCP socket from HTTP-mode to some other,
application-specific bi-directional protocol. As mentioned here a few
m
On 11/25/2012 05:07 PM, Simon Newton wrote:
I use microhttpd with the Open Lighting Architecture. As I've posted
before we have a C++ wrapper around libmicrohttpd (HTTPServer), which
then integrates with OLA's event management system.
While this works, it's rather messy. The wrapper sets up a ca
On 12/04/2012 07:59 PM, Sebastiano Merlino wrote:
Hi all,
I am a big fan of the library and I realized several projects using it.
I am now trying to write a proxy using libmicrohttpd and curl; at the
moment, I simply do a curl_easy_perform to the real service when I
receive a request. I am usi
Dear Matt,
Thanks for the patch! That one must have been hard to reproduce & find.
Fixed as suggested in SVN 25267; MHD_VERSION >= 00091701 / MHD > 0.9.23.
Happy hacking!
Christian
On 12/05/2012 07:10 PM, Matthew Mundell wrote:
Hi
We've been having some mysterious parameter loss of POST p
As far as I see it, the proposed API would allow you to "return
asynchronous results" (as MHD does in general for HTTP), so
unless I am misunderstanding your request, you're saying that
"having this would be useful" (to you)? If you see a problem
with the proposed API or need a clarification, pl
eing in 'select'); once your callback returns non-zero, MHD will put
the socket back into the FD set.
Happy hacking!
Christian
Thanks,
Sebastiano
2012/12/5 Christian Grothoff <mailto:christ...@grothoff.org>>
On 12/04/2012 07:59 PM, Sebastiano Merlino wrote:
On 12/06/2012 02:51 AM, Eivind Sarto wrote:
> We are using libmicrohttpd (0.9.19) for a http video server.
> One of our customers reported that a video client they were testing failed
> because the client was setting
> "Connection: close" in the request headers and was expecting a "Connection:
>
On 12/18/2012 08:33 PM, Eivind Sarto wrote:
> I was was having problems receiving data from a client using POST with
> chunked encoding.
> It turns out this client is violating the HTTP spec by setting the
> "Transfer-Encoding: Chunked"
> as well as "Content-Length: 0"
>
> Here are the client he
I think the real issue is that you're trying to create a post processor
for a data format (XML) that is not supported by the post processor.
For XML data, you should just pass the 'upload_data' directly to your
XML parser (and update *data_size to reflect the number of bytes you did
(not) proce
On 01/05/2013 12:33 AM, Eivind Sarto wrote:
> I am running into problems with my MHD server. I may be confusing the order
> of the various callback functions.
> As a result, my server sometimes will sometimes de-reference and allocated
> structure after it has been freed.
>
> I allocate a priva
On 01/06/2013 08:57 PM, Sascha Swiercy wrote:
> I'd like to know if it is possible to use
> MHD_create_response_from_callback even if the content length is 0. I
> tried to pass 0 as first argument to MHD_create_response_from_callback
> and also returned 0 from the callback, but this made the librar
301 - 400 of 803 matches
Mail list logo
