Re: [libmicrohttpd] MHD in an environment with many requests

ell, I've > given it a 2nd shot now :-). -Christian > > On 11/09/2015 07:59 AM, silvioprog wrote: > > Very nice. I'll try it this afternoon. :-) > > > > Is the 0.9.46 available only in SVN? I saw only the 0.9.45 at FTP link. > > > > Thank you! > > &

Re: [libmicrohttpd] MHD in an environment with many requests

Oops ... :-) On Sun, Nov 8, 2015 at 11:20 PM, silvioprog <silviop...@gmail.com> wrote: > ... and them worked fine. > I meant: "and they". -- Silvio Clécio

Re: [libmicrohttpd] MHD in an environment with many requests

d in 0.9.45 > could cause connections to be dropped. Please try 0.9.46. > > Happy hacking! > > Christian > > On 11/09/2015 03:20 AM, silvioprog wrote: > > Hello, > > > > I'm testing MHD in an environment with a lot of simultaneous requests, > and > &g

[libmicrohttpd] Small upgrade for MHD home page

Hello, I took a look at the MHD home page, and it shows the only following versions: Linux libmicrohttpd-0.9.38.tar.gz Windows libmicrohttpd-0.9.17-w32.zip This versions are so outdated. :-( It should be (like NodeJS: https://nodejs.org/en/download/): Windows Binary 32 / 64 bits (.zip) Mac OS

Re: [libmicrohttpd] Why MHD latency is so high on Windows?

/hgmloofddffdnphfgcellkdfbfbjeloo On Sun, Nov 15, 2015 at 3:06 PM, silvioprog <silviop...@gmail.com> wrote: > Hello, > > I'm using MHD in production, so I noticed a small problem in a customer: > the MHD latency is very high. It is about 250ms. (Apache24 - 15ms; Nginx - > 2ms; NodeJS - 5ms;

[libmicrohttpd] Why MHD latency is so high on Windows?

Hello, I'm using MHD in production, so I noticed a small problem in a customer: the MHD latency is very high. It is about 250ms. (Apache24 - 15ms; Nginx - 2ms; NodeJS - 5ms; Jetty - 5ms; libevent - 3ms) You can't reproduce it testing in localhost, so you need two machines. If you want to test

Re: [libmicrohttpd] Small upgrade for MHD home page

Hm... about Windows package, the problem is just to keep an old version on home page, because there is the first place that an user find the lib, and this very old version miss a lot of important fixes. :-/ I have two sugestions to distribute the Windows package on home page: 1. keep only the

[libmicrohttpd] Keep-alive and POST method

Hello, Is the keep-alive feature compatible only with GET method? See this log below, it compares the headers from MHD and NodeJS: === NodeJS === ... GET ... $ curl -I http://127.0.0.1:8080 HTTP/1.1 200 OK Content-Type: text/html Date: Thu, 05 Nov 2015 22:34:34 GMT *Connection: keep-alive*

Re: [libmicrohttpd] Keep-alive and POST method

pload data that > we're now just dropping). > > Try handling the POST data "for real" (RTFM, and/or look at actual > examples!), and then you'll get your keep-alive from MHD as well. > > Happy hacking! > > Christian > > > On 11/05/2015 11:43 PM, silvio

Re: [libmicrohttpd] Why MHD latency is so high on Windows?

On Wed, Nov 18, 2015 at 10:20 AM, Evgeny Grin wrote: > Hi! > > libmicrohttpd was ported to use Windows API directly, so PlibC wrapper is > not used anymore. > Very nice. > According to MSDN docs there is no TCP_FASTOPEN flag on Windows: >

Re: [libmicrohttpd] Why MHD latency is so high on Windows?

Hello Nils, I can't found any `select.c` file and PlibC entry. :-/ Is this file available in the latest MHD release? But, after your tip, I tried to debug the MHD sources, so I found a build flag called `TCP_FASTOPEN`, and after enable it in my cmake file, rebuilt the DLL and used the

Re: [libmicrohttpd] Small upgrade for MHD home page

up a binary. Note that you should sign it with *your* GnuPG key, I do not sign binaries created by other people. On 11/15/2015 06:42 PM, silvioprog wrote: > Hm... about Windows package, the problem is just to keep an old version on > home page, because there is the first place that an use

[libmicrohttpd] Is there any C example showing how to use the MHD_USE_SUSPEND_RESUME flag?

Hello, So, I did many searches on Google but I couldn't find any sample showing how to use this flag. =/ Can anyone send a small example writen in C? Thanks in advance! -- Silvio Clécio

Re: [libmicrohttpd] Is there any C example showing how to use the MHD_USE_SUSPEND_RESUME flag?

Hello Flavio, Firstly, thanks for share it! =) On Fri, Sep 25, 2015 at 6:44 PM, Flavio Ceolin wrote: [...] > I've used it on the project i'm working. Just take a look: > > >

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

`, and it is 0,00% to MHD too! :-) On Tue, Dec 8, 2015 at 9:05 AM, Evgeny Grin <k...@yandex.ru> wrote: > > 07.12.2015, 22:58, "silvioprog" <silviop...@gmail.com>: > > Hello, I have good news in the test 2, however in test 3 ... > > @Evg, do I need to change th

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

On Tue, Dec 8, 2015 at 3:33 PM, silvioprog <silviop...@gmail.com> wrote: > [...] > And Evg, sorry if I'm being boring, but, can you send me the compiled 32 > and 64 binaries? I can compile that, but I don't know if my environment is > the better to compile MHD. The 32 bit bina

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

I forgot other information... I use thread in my own app, so I disable the MHD threads, but MHD still depending on `libwinpthread-1.dll` lib. Is this a normal behaviour or I need to pass some flag before compile it? On Tue, Dec 8, 2015 at 5:29 PM, silvioprog <silviop...@gmail.com> wrote:

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

Hm... so I did something wrong there. The commands that I did was (built for production): *$ ./configure --disable-doc --disable-examples --disable-messages --enable-https=no --disable-bauth --disable-dauth --disable-curl* $ make $ make install Ie, without HTTPS support, and I used the latest

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

, "silvioprog" <silviop...@gmail.com>: Very nice Evgeny! :-) Let me know when this feature was implemented so that we can release the Windows package on MHD home page. :-) On Mon, Nov 30, 2015 at 1:30 PM, Evgeny Grin <k...@yandex.ru> wrote: Patch for *BSD and Darwin platform

Re: [libmicrohttpd] libmicrohttpd 0.9.47 released

Very good news!  I'm going to send the Windows binaries to update it at home page. I just need to confirm Evgeny if this task will be done by him or me (he seems more indicated because his expansive knowledge about MHD). On Fri, Dec 4, 2015 at 10:11 AM, Christian Grothoff

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

Hm... My test is from Windows client to Windows server. I'll retest it using some binary tool and back to show the result here ... On Fri, Dec 4, 2015 at 8:36 AM, Evgeny Grin wrote: > Did test it using MinGW-w64 and VS builds. Windows machine was used as > server, FreeBSD,

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

Hello bro, Yes, it killed the problem definitively! 3:) I did many tests using different machines and now it works like a charm, thanks a lot for this fix. However ( ^^' ), I saw a small problem, but it is related only to the keep-alive feature, even using previous versions like 0.9.46: when I

Re: [libmicrohttpd] libmicrohttpd 0.9.47 released

It is OK Evgeny! I did a test using a HTTP client written in C and now MHD latency is 1-4ms. :-) On Fri, Dec 4, 2015 at 2:05 PM, silvioprog <silviop...@gmail.com> wrote: > Very good news!  > > I'm going to send the Windows binaries to update it at home page. I just > need

[libmicrohttpd] Compiling MHD on MinGW-w64

Hello, Now MHD implements important features for Windows (thanks Evgeny!), so it is time to update the latest Windows package (libmicrohttpd-0.9.17-w32.zip) at MHD home page. This topic is just to confirm if the used environment is OK: - original Windows Professional 7 64 bits; - latest

[libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

Hello, I'm trying to use Keep-Alive in the `MHD_USE_THREAD_PER_CONNECTION` mode with the `MHD_OPTION_CONNECTION_TIMEOUT` option, however, when timeout is ends, it locks the client that can't get more responses from my MHD server. Is very easy to reproduce that, you just: . compile and run this

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

Strange... when I call the `MHD_get_daemon_info()` function it unlock the clients. :-/ This[1] example doesn't lock the clients. Do I need to trigger the `MHD_get_daemon_info()` always? [1] http://pastebin.com/dfykB1US On Fri, Dec 4, 2015 at 9:21 PM, silvioprog <silviop...@gmail.com>

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

You did an excelent job. I compiled it with the latest mingw-w64 release and now MHD latency is very low, about 25ms. Are you using the TCP_NODELAY? If so, is there some define to enable it in my CMake file? The only note is that using it I could get a latency about 5ms. Yes, 25ms is very good,

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

Hello, I have good news in the test 2, however in test 3 ... @Christ, on Linux I got 0.00% erros. MHD is so fast on Linux that I believe that it can support C10K problem. :-) @Evg, do I need to change the `FD_SETSIZE` and the MHD source

Re: [libmicrohttpd] Problem with Keep-Alive and MHD_OPTION_CONNECTION_TIMEOUT in MHD_USE_THREAD_PER_CONNECTION mode

On Mon, Dec 7, 2015 at 5:27 PM, Christian Grothoff wrote: > [...] > Well, on Linux C10k is easy these days. We had people using MHD that in > practice had what you might call the C100k problem, and that was doable > ;-). > Very good! Currently I'm using it only as embedded

Re: [libmicrohttpd] Fwd: Maintaining libmicrohttpd

He worked a lot to make the new versions of MHD compatible and stable on Windows, and his code is very good, I'm using it in production and it is working like a charm. Now that Evgeny is co-maintainer of MHD, can I help him to update the packages on MHD home page? It would be nice because I'm

Re: [libmicrohttpd] Fwd: Maintaining libmicrohttpd

On Wed, Dec 23, 2015 at 5:10 AM, Andreas Wehrmann wrote: > On 12/22/2015 06:42 PM, Hein-Pieter van Braam wrote: > >> Hi, >> >> Sorry, it was just a silly joke. Gopher is an (arguably) obsolete >> protocol that predates the web and HTTP. >> >>

Re: [libmicrohttpd] Fwd: Maintaining libmicrohttpd

I like it! :-D I did not know that this protocol existed. :o Thx for share it. On Tue, Dec 22, 2015 at 2:42 PM, Hein-Pieter van Braam wrote: > Hi, > > Sorry, it was just a silly joke. Gopher is an (arguably) obsolete > protocol that predates the web and HTTP. > >

Re: [libmicrohttpd] Fwd: Maintaining libmicrohttpd

Very nice news, and thanks for the time dedicated to MHD project! :-) On Mon, Dec 21, 2015 at 3:56 PM, Evgeny Grin <k...@yandex.ru> wrote: > > 21.12.2015, 21:25, "silvioprog" <silviop...@gmail.com>: > > He worked a lot to make the new versions of MHD compatibl

Re: [libmicrohttpd] Fwd: Maintaining libmicrohttpd

Sorry for my ignorance, but, what is Gopher module? On Mon, Dec 21, 2015 at 9:15 PM, Hein-Pieter van Braam wrote: > On Mon, 2015-12-21 at 21:37 +0300, Evgeny Grin wrote: > > Hurray! I'm co-maintainer now! > > Hopefully we will not drive the project in opposite directions. :) > >

Re: [libmicrohttpd] Fwd: Maintaining libmicrohttpd

2/21/2015 07:24 PM, silvioprog wrote: > > Now that Evgeny is co-maintainer of MHD, can I help him to update the > > packages on MHD home page? It would be nice because I'm going to > > indicate the new MHD in many groups and forums that I'm participating. > > You could have

Re: [libmicrohttpd] libmicrohttpd 0.9.48 released

Noce job! :-) I did many tests in this version and it is very stable on Windows. The 0.9.48 should replace the 0.9.46.tar.gz and 0.9.17-w32.zip on MHD home page. On Sat, Dec 19, 2015 at 6:10 AM, Evgeny Grin wrote: > Dear all, > > I'm glad to announce the release of

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

Hello Evgeny, Any news about this patch? Do you need help? If you are busy I can do it for you. :-) On Fri, Nov 20, 2015 at 6:30 AM, Evgeny Grin wrote: > Hi, Silvio > > You patch will turn on TCP_NODELAY globally upon request, but it should be > used at specific moments as

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

Very nice Evgeny! :-) Let me know when this feature was implemented so that we can release the Windows package on MHD home page. :-) On Mon, Nov 30, 2015 at 1:30 PM, Evgeny Grin wrote: > Patch for *BSD and Darwin platforms is already added. > We are developing solution for

Re: [libmicrohttpd] Why MHD latency is so high on Windows?

Is there someone that needs performance in MHD on Windows besides me? :-D I need to test it, I'm going to do it these days. For now, I disabled the `TCP_FASTOPEN` flag and left it with high latency, until I figure out how this works on Windows, once it doesn't offer TFO support. On Wed, Nov 18,

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

Very nice Evgeny, I'm expecting anxiously this feature! :-D On Fri, Nov 20, 2015 at 6:30 AM, Evgeny Grin wrote: > Hi, Silvio > > You patch will turn on TCP_NODELAY globally upon request, but it should be > used at specific moments as Martin Bonner pointed in >

[libmicrohttpd] [PATCH] Added support for TCP NODELAY

Hello, About the `TCP NODELAY` feature, it reduces the network latency on many operating systems, and this patch allows MHD to disable it. Popular servers like nginx allows to enable the `TCP NODELAY` feature[1]. You can see more details about this patch here in this topic[2]. Feel free to

Re: [libmicrohttpd] [PATCH] Added support for TCP NODELAY

If you apply this patch, please update the MHD version to 0x00094603! :-) On Fri, Nov 20, 2015 at 12:43 AM, silvioprog <silviop...@gmail.com> wrote: > Hello, > > About the `TCP NODELAY` feature, it reduces the network latency on many > operating systems, and this patch all

Re: [libmicrohttpd] Small upgrade for MHD home page

Hello again, I'm trying to solve a MHD latency problem on Windows system, and I'm very happy by see the community helping in this task. (this is the best think of MHD!) After solve it I'll send a Widnows package ASAP! On Wed, Nov 18, 2015 at 12:08 AM, silvioprog <silviop...@gmail.com>

Re: [libmicrohttpd] libmicrohttpd 0.9.50 released

not worth it. In fact, the page > should work without JavaScript. -Christian > > On 06/04/2016 07:41 PM, silvioprog wrote: > > Awesome. Can we use something like jQuery on there? > -- Silvio Clécio

Re: [libmicrohttpd] libmicrohttpd 0.9.50 released

Awesome. Can we use something like jQuery on there? On Sat, Jun 4, 2016 at 8:52 AM, Christian Grothoff <groth...@gnunet.org> wrote: > On 06/04/2016 01:50 PM, silvioprog wrote: > > Lol I haven't seen, thanks for remind me. I'm going to download it. > > > > So, what lan

Re: [libmicrohttpd] libmicrohttpd 0.9.50 released

recreated to each new release. :-) On Sat, Jun 11, 2016 at 9:56 PM, Christian Grothoff <groth...@gnunet.org> wrote: > Should be fixed now. No need for code, a symlink does the trick ;-). > -Christian > > On 06/04/2016 09:23 PM, silvioprog wrote: > > No JS, no some language

Re: [libmicrohttpd] Windows binary with SSL3/TLS

Hello Peter, I have a bat that makes MHD and you can choose between the smallest (just the HTTP server, no HTTPS, messages, auth etc.) and fullest version (full HTTP server, HTTPS, messages, auth etc.). I'm away from my development PC, but I'll send it for you ASAP ... On Fri, Jun 10, 2016 at

[libmicrohttpd] MHD in "worker processes" mode

Hello, Please take a look at this question and the first answer: http://askubuntu.com/questions/399431/how-many-nginx-processes-should-there-be It raise a question: can I use the worker processes approach with MHD? If I understood, it need to up two or more process, but when I try that: "Failed

Re: [libmicrohttpd] libmicrohttpd 0.9.50 released

Lol I haven't seen, thanks for remind me. I'm going to download it. So, what language was used to build MHD home page? I can send a code that prints the latest released packages on it. On Fri, Jun 3, 2016 at 7:59 AM, Evgeny Grin wrote: > W32 binaries are available on ftp

Re: [libmicrohttpd] Status 304 and insistent Content-Length header

responses? On Wed, Jun 1, 2016 at 2:20 AM, silvioprog <silviop...@gmail.com> wrote: > Hello, > > In some cases, we need to remove the Content-Length header, for example, > when the status code is 304 (not modified), you don't need to send some > content-* headers again. See

[libmicrohttpd] Status 304 and insistent Content-Length header

Hello, In some cases, we need to remove the Content-Length header, for example, when the status code is 304 (not modified), you don't need to send some content-* headers again. See this code: https://github.com/expressjs/express/blob/master/lib/response.js#L192 Great, it makes sense, but

Re: [libmicrohttpd] Windows binary with SSL3/TLS

On Windows, the MinGW pacman has an old MHD version: mingw-w64-i686-libmicrohttpd 0.9.42-2. But on Manjaro Linux (I have a copy just for tests) it has the latest version. @offtopic ON Can I upgrade my Windows MinGW repo to get the latest MHD version like pacman on Manjaro? @offtopic OFF. On Fri,

Re: [libmicrohttpd] Windows binary with SSL3/TLS

Awesome. I found this link showing many tools, so I tried one of them: $ ldd build/src/microhttpd/.libs/libmicrohttpd-12.dll | grep mingw | awk '{ print $1 }' libgcrypt-20.dll libgnutls-30.dll libgpg-error-0.dll

[libmicrohttpd] Broken image in the MHD manual

Hello, The broken picture: https://www.gnu.org/software/libmicrohttpd/manual/libmicrohttpd.html#fig_003aperformance Thanks! :-) -- Silvio Clécio

Re: [libmicrohttpd] How do I check if the request was done from HTTP or HTTPS?

Eureka! It seems that starting two daemons is better way instead of MHD_CONNECTION_INFO_GNUTLS_SESSION: just redirect all requests from daemon:80 to daemon:443. Great idea CG, thanks very much! :-) On Tue, Feb 23, 2016 at 10:31 PM, Christian Grothoff wrote: > Hi! > > Well,

Re: [libmicrohttpd] Can I get the entire MHD log without depends on C function like sprintf()?

Complementing: It would be very useful to the ones that need to get the entire log without depend on C format functions or providing some built-in MHD format function. On Sat, Feb 27, 2016 at 5:14 PM, silvioprog <silviop...@gmail.com> wrote: > Hello, > > I'm using MHD in a sm

Re: [libmicrohttpd] Can I get the entire MHD log without depends on C function like sprintf()?

cause it's not C-compatible is not a good trade-off in my mind. Still, I think sharing this here is useful, as others may have the same question in the future, and I actually think your solution is great. Happy hacking! Christian On 02/27/2016 09:14 PM, silvioprog wrote: > Hello, > > I'm u

Re: [libmicrohttpd] How to disable the connection keep-alive header?

ndex.ru> wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA256 > > On 22.01.2016 20:27, silvioprog wrote: > > Is there some link where I can get it? I need this one to document it in > > my framework that uses MHD. > https://tools.ietf.org/html/rfc7230#page-5

Re: [libmicrohttpd] How to disable the connection keep-alive header?

alive" is to explicitly set "Connection: close". > > On 01/22/2016 04:17 AM, silvioprog wrote: > > Hello, > > > > Sometimes we need to disable the keep-alive header. For example, see this > > headers sent from the Express framework: > > > > X-Pow

[libmicrohttpd] How can I decode the form-data fields with MHD_http_unescape() function?

Hello, I have following form-data: Content-Type: multipart-form-data id=10=Silvio+Cl%C3%A9cio ... some uploads here ... But when I decode it with MHD_http_unescape() function: ..., char *key, ... MHD_http_unescape(key); I got: Silvio+Clécio But I need the full unescaped value, that is

Re: [libmicrohttpd] How can I decode the form-data fields with MHD_http_unescape() function?

On Tue, Feb 16, 2016 at 9:29 AM, Evgeny Grin wrote: > There is a conflict between standards. > HTML states that "+" must be decoded to space in url-encoding. > https://www.w3.org/TR/html/forms.html#url-encoded-form-data > RFC 3986 doesn't assume any special treatment of "+". >

[libmicrohttpd] What is the `transfer_encoding` param in the MHD_PostDataIterator callback?

Hello listers, The MHD_PostDataIterator documentation says: *transfer_encoding* * encoding of the data, NULL if not known;* Awesome. But, I don't know how to use this feature, so I have two questions: . do I need to send some header in the x-www-form-urlencoded POST? . do I need to send some

Re: [libmicrohttpd] Doubt about digest auth algorithm

Hello Christian, So my logic is correct. I'll document my code using this informations. Thank you very much! :-) On Sun, Mar 13, 2016 at 10:36 AM, Christian Grothoff wrote: > Dear Silvio, > > Yes, the difference is that with basic_auth, MHD "returns" you the > password

[libmicrohttpd] [VIDEO] Introduction to GNU libmicrohttpd

Hello, Awesome video with an introduction to GNU libmicrohttpd: https://youtu.be/jmkYbftonLk Enjoy! :-D Speaker: Christian Grothoff Language: English Time: ~27 min -- Silvio Clécio

Re: [libmicrohttpd] [VIDEO] Introduction to GNU libmicrohttpd

Awesome! :-) On Mon, Mar 14, 2016 at 4:22 AM, Christian Grothoff wrote: > Same talk, without advertising and without using the Google: > > https://entropia.de/GPN13:Introduction_to_GNU_libmicrohttpd -- Silvio Clécio

[libmicrohttpd] How to avoid the "Internal application error, closing connection" message when AHC = NO?

Hello, Sometimes, an application needs to refuse the client connection, something like this: static int ahc_echo(void *cls, ...) { if (!logged()) return MHD_NO; } However, when I pass the MHD_NO to the AHC, I always get the "Internal application error, closing connection" in my log.

Re: [libmicrohttpd] Demo "largepost.c" with a file about 6GB

On Tue, Mar 29, 2016 at 3:32 PM, silvioprog <silviop...@gmail.com> wrote: [...] > I've noticed that the progress stay a long time (~1min) in 99% even on > Linux, but it receive the entire upload. > Does anyone else notice the same behavior? -- Silvio Clécio

[libmicrohttpd] Some questions about the example "digest_auth_example.c"

Hello, I'm making some high level classes for handling a12n with MHD, but I have some questions about the example digest_auth_example.c , that implements the digest a12n and I'm unsing this one to understand how MHD handles

[libmicrohttpd] Is the "digest_auth_example.c" example working on Firefox?

Hello, Can anybody check if the "digest_auth_example.c " example is still working on Firefox? It works fine with Chrome and Internet Explorer, but when I try using Firefox, it shows "Access denied", and when I refresh the

[libmicrohttpd] [FEATURE REQUEST] Declares all log and errors messages in constants

Hello, Please take a look at issue #4614 . Thank you! -- Silvio Clécio

Re: [libmicrohttpd] Creating instances in the option MHD_OPTION_URI_LOG_CALLBACK

On Wed, Aug 10, 2016 at 6:45 AM, Christian Grothoff <groth...@gnunet.org> wrote: > On 08/08/2016 02:26 AM, silvioprog wrote: > > Anyway, I have some doubts. When I set a routine to the > > MHD_OPTION_URI_LOG_CALLBACK option, is this one always called once time > for

[libmicrohttpd] Question about the new `mhd_threads.h` header

Hello, I've looked at last commits and found a new header: mhd_threads.h. It has functions like MHD_create_thread(), MHD_create_named_thread() and MHD_join_thread(), however it seems only to MHD's internal using. BTW these functions could be also very useful for MHD consumers. WDYT? -- Silvio

Re: [libmicrohttpd] Question about the new `mhd_threads.h` header

The same question to the `mhd_locks.h`. Useful cross platform thread/locks functions. On Thu, Aug 11, 2016 at 4:06 PM, silvioprog <silviop...@gmail.com> wrote: > Hello, > > I've looked at last commits and found a new header: mhd_threads.h. It has > functions like

Re: [libmicrohttpd] Question about the new `mhd_threads.h` header

On Thu, Aug 11, 2016 at 4:31 PM, Christian Grothoff wrote: > They do not relate to providing an HTTP server, and they thus should > remain an internal API. [...] +1 Fully agreed! :-) -- Silvio Clécio

Re: [libmicrohttpd] Doubt regarding to MHD_UpgradeAction

On Sat, Aug 13, 2016 at 1:44 PM, Christian Grothoff wrote: > You should not enable it, as it has not yet been implemented and is > likely to change before it is stable. -Christian Awesome, removed from my header! :-) Thank you! -- Silvio Clécio

[libmicrohttpd] Doubt regarding to MHD_UpgradeAction

Hello, I'm upgrading my Pascal header to the MHD version 0.9.50, however, I found the following comment in the MHD_UpgradeAction enum: Enumeration for actions MHD should perform on the underlying socket > of the upgrade. This API is not finalized, and in particular > the final set of actions is

Re: [libmicrohttpd] Query on using openssl with microhttpd

It would be a nice feature to be added in a wishlist. :-) On Fri, Aug 5, 2016 at 12:38 PM, Christian Grothoff wrote: > Not at this time. -Christian > > On 08/05/2016 04:49 PM, Geetha Shanmugam wrote: > > Hi, > > Is there a way we can use openssl library (instead of gnutls)

[libmicrohttpd] Creating instances in the option MHD_OPTION_URI_LOG_CALLBACK

Hello listers, I have a Pascal object representing the request so it is created for each browser request, and MHD maintains the life cycle of that one. So far so good, but the only problem is about my approach choice, I used the option MHD_OPTION_URI_LOG_CALLBACK to create the object and the

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

E_EPOLL_INTERNALLY can be used. On Tue, Jan 31, 2017 at 8:58 AM, Christian Grothoff <groth...@gnunet.org> wrote: > On 01/31/2017 05:27 AM, silvioprog wrote: > > On Mon, Jan 30, 2017 at 5:38 PM, Evgeny Grin <k...@yandex.ru > > <mailto:k...@yandex.ru>> wrote: &g

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

e version? > Many new flags were added in git and are not yet released. > > -- > Best Wishes, > Evgeny Grin > > 31.01.2017, 07:28, "silvioprog" <silviop...@gmail.com>: > > On Mon, Jan 30, 2017 at 5:38 PM, Evgeny Grin <k...@yandex.ru> wrote:

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

Hello guys, Problem solved. The problem was I'm using features available only in trunk version but local installed version is 0.9.52. Thanks Grothoff and Grin for the help! :-) On Tue, Jan 31, 2017 at 1:59 PM, silvioprog <silviop...@gmail.com> wrote: > Complementing: > > $

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

Thanks a lot for the tips! :-) On Tue, Jan 31, 2017 at 3:13 PM, Christian Grothoff <groth...@gnunet.org> wrote: > On 01/31/2017 05:59 PM, silvioprog wrote: > > Complementing: > > > > $ cat /usr/local/include/microhttpd.h | grep 'MHD_VERSION' > > #define M

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

t; made no sense linguistically, so we decided to not > rename these flags, also to minimize disruptions. > > On 01/29/2017 10:35 PM, silvioprog wrote: > > Hello masters, > > > > I've noticed all flags for `USE` scope was prefixed with `MHD_USE_*`, > > except these two fla

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

ressED" instead of "suppress" and "end" instead of "no") On Mon, Jan 30, 2017 at 3:59 PM, Christian Grothoff <groth...@gnunet.org> wrote: > Hmm. This should not be the case, as we kept the old names around for > now to ensure backwards-compatibility. Which flag

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

Oops, was these flags removed?: MHD_USE_AUTO | MHD_USE_INTERNAL_POLLING_THREAD | MHD_USE_ERROR_LOG On Mon, Jan 30, 2017 at 4:35 PM, silvioprog <silviop...@gmail.com> wrote: > I can't remember at this moment, but it was fixed with a message talking > something like "X is deprec

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

On Mon, Jan 30, 2017 at 4:48 PM, silvioprog <silviop...@gmail.com> wrote: > Oops, was these flags removed? > Damn typo. I meant "... were these flags ...". -.-' -- Silvio Clécio

Re: [libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

On Mon, Jan 30, 2017 at 5:38 PM, Evgeny Grin wrote: > No. Those flags are not removed. Do you have problems with it? > > -- > Best Wishes, > Evgeny Grin > Yes, I do. My MHD version: 0.9.52. Problem: "error: ‘MHD_USE_ERROR_LOG’ undeclared". But it compiles fine with other flags

Re: [libmicrohttpd] Bench-marking issue

ee if you compiled MHD (or the other servers) with any > special CFLAGS, i.e. O2/O3/Os or even O0/O1. So that might also make > some small difference. > > Happy benchmarking! > > Christian > > On 01/28/2017 09:05 PM, silvioprog wrote: > > Hello masters, > > > > I'

[libmicrohttpd] Typo?: MHD_SUPPRESS_DATE_NO_CLOCK and MHD_NO_FLAG

Hello masters, I've noticed all flags for `USE` scope was prefixed with `MHD_USE_*`, except these two flags. So, it was a typo or is that OK? Anyway MHD_USE_SUPPRESS_DATE and MHD_USE_NO_FLAG makes more sense, following the same other FLAGs pattern. Thanks! -- Silvio Clécio

Re: [libmicrohttpd] How to compile MHD for ARMv7?

yes Digest auth.: yes Postproc: yes HTTPS support: no (disabled) poll support: yes epoll support: yes build docs:yes build examples:yes configure: License : LGPL or eCos On Thu, Feb 16, 2017 at 2:42 AM, silvioprog <silviop...@

Re: [libmicrohttpd] RPM package for MHD v0.9.50

Hm... so you need the 32 bits package: ftp://ftp.gwdg.de/pub/opensuse/repositories/filesharing/openSUSE_Leap_42.1/i586/libmicrohttpd12-0.9.50-58.1.i586.rpm . But I'm not sure if it is compatible with CentOS. :-/ Specifically for CentOS I found only the version 0.9.33:

Re: [libmicrohttpd] How to compile MHD for ARMv7?

Done. In attachment (from 0.9.52). On Fri, Feb 17, 2017 at 11:09 AM, silvioprog <silviop...@gmail.com> wrote: > Yes, I can! :-) > > I'm going to get it from the stable package 0.9.52 ... > > On Fri, Feb 17, 2017 at 5:46 AM, Evgeny Grin <k...@yandex.ru> wrote: >

Re: [libmicrohttpd] How to compile MHD for ARMv7?

Oh, sorry my delay. I updated my GIT copy to SHA 82d65c48 and rebuild my Android app, but same problem when I try to compile it. :-/ config.log in attachment. On Thu, Feb 16, 2017 at 7:22 AM, Christian Grothoff <groth...@gnunet.org> wrote: > On 02/16/2017 07:48 AM, silviop

Re: [libmicrohttpd] How to compile MHD for ARMv7?

Please apply this new patch and delete sent config.logs. :-) This patch fix the problem after latest Grothoff's commit. (SHA 82d65c48) -- Silvio Clécio From ffeb64478ea9affd2a65b0ab60e11c1f54436030 Mon Sep 17 00:00:00 2001 From: silvioprog <silviop...@gmail.com> Date: Sun, 19 Feb 2017 02

Re: [libmicrohttpd] MHD can't find MO files

On Fri, Feb 24, 2017 at 9:46 PM, Christian Grothoff <groth...@gnunet.org> wrote: > n 02/24/2017 11:41 PM, silvioprog wrote: > > P.S: Looking at mhd_options.h I found the following comment: > > > > /** > > * Macro to make it easy to mark text for translation. No

Re: [libmicrohttpd] MHD can't find MO files

at 7:45 PM, silvioprog <silviop...@gmail.com> wrote: > Complementing, the output translated message (for pt_BR) should be "Falha > ao vincular à porta: Permission denied" instead of "Failed to bind to > port 80: Permission denied". > > On Fri, Feb 24, 2017

[libmicrohttpd] fix tipo: `MHD_FEATURE_MESSGES` -> `MHD_FEATURE_MESSAGES`. #2

Hello dudes, Please apply this PR: https://github.com/Karlson2k/libmicrohttpd/pull/2 . Thank you! :-) -- Silvio Clécio

Re: [libmicrohttpd] MHD can't find MO files

oth...@gnunet.org> wrote: > On 02/25/2017 03:05 AM, silvioprog wrote: > > > > Anyway I'm curious, is there any special reason to MHD not to call > > bindtextdomain() internally? 樂 (it seems locale.h/libintl.h is present > in > > all SOs supported by MHD ...) >

Re: [libmicrohttpd] MHD can't find MO files

Complementing, the output translated message (for pt_BR) should be "Falha ao vincular à porta: Permission denied" instead of "Failed to bind to port 80: Permission denied". On Fri, Feb 24, 2017 at 7:41 PM, silvioprog <silviop...@gmail.com> wrote: > Hello masters

Re: [libmicrohttpd] How to compile MHD for ARMv7?

Perfect dude! :-) I'm going to retest it (from trunk) and back feedback ... On Thu, Feb 16, 2017 at 7:22 AM, Christian Grothoff <groth...@gnunet.org> wrote: > On 02/16/2017 07:48 AM, silvioprog wrote: > > The attached patch fix the compilation. > > Maybe, but as

  1   2   3   >