<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://purl.org/rss/1.0/" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:syn="http://purl.org/rss/1.0/modules/syndication/" xmlns:admin="http://webns.net/mvcb/">
  <channel about="http://blog.gmane.org/gmane.comp.web.curl.library">
    <title>gmane.comp.web.curl.library</title>
    <link>http://blog.gmane.org/gmane.comp.web.curl.library</link>
    <description/>
    <syn:updatePeriod>hourly</syn:updatePeriod>
    <syn:updateFrequency>1</syn:updateFrequency>
    <syn:updateBase>1901-01-01T00:00+00:00</syn:updateBase>
    <items>
      <rdf:Seq>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21553"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21550"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21546"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21542"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21535"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21532"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21527"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21524"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21523"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21516"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21513"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21511"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21501"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21498"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21495"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21492"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21484"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21457"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21442"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.comp.web.curl.library/21440"/>
      </rdf:Seq>
    </items>
    <image rdf:resource="http://gmane.org/img/gmane-25t.png"/>
    <textinput rdf:resource=""/>
  </channel>
  <image rdf:about="http://gmane.org/img/gmane-25t.png">
    <title>Gmane</title>
    <url>http://gmane.org/img/gmane-25t.png</url>
    <link>http://gmane.org</link>
  </image>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21553">
    <title>Bug or not Bug</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21553</link>
    <description>
Fedore Core 8curl and libcurl 7.18.2-6.fc8  + relative devel pack          ( RPM pack downloaded with yum )curl -V = curl 7.18.2 (i386-redhat-linux-gnu) libcurl/7.18.2 NSS/3.12.0.3 zlib/1.2.3 libidn/0.6.14 libssh2/0.18Protocols: tftp ftp telnet dict http file https ftps scp sftpFeatures: GSS-Negotiate IDN IPv6 Largefile SSL libzThis code don' t do anything but generate a SIGABRT after 8 - 12 hours#include &lt;iostream&gt;#include &lt;curl/curl.h&gt;#include &lt;gtk/gtk.h&gt;size_t NoProcess(void *buffer, size_t size, size_t nmemb, void *userp){ return size * nmemb;}gpointer Test(gpointer data){ CURL * Fin; Fin = curl_easy_init();//if (curl_easy_setopt(Fin, CURLOPT_NOSIGNAL, 1) != CURLE_OK)return NULL; while (1) { if (curl_easy_setopt(Fin, CURLOPT_WRITEDATA, NULL) != CURLE_OK)return NULL; if (curl_easy_setopt(Fin, CURLOPT_WRITEFUNCTION, NoProcess) != CURLE_OK)return NULL; if (curl_easy_setopt(Fin, CURLOPT_URL, "https://www.fineco.it/fineco/jsp/login/content.jsp") != CURLE_OK)return NULL; if (curl_easy_perform(Fin) !=</description>
    <dc:creator>Pietro Incardona</dc:creator>
    <dc:date>2008-11-29T15:52:34</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21550">
    <title>libCurl not sending basic authentication in 1 out of 300 systems...</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21550</link>
    <description>I wrote c/c++ software that uses libCurl 7.17.1 and runs on a Linux computer
for periodically archiving images from Panasonic cameras.  It's really
simple, we just retrieve a given url passing in the username/password for
basic authentication.  I've sold 300 systems, and it has run fine on all of
them, until now.  On one system, which is otherwise identical to the all the
others, it appears that libCurl is *not* sending the basic authentication
username/password for no apparent reason.  The command line utility works
fine, though.  I even created a sample app with hardcoded values to be 100%
sure everything is correct.  So I'm totally baffled.

This is the command line that works: curl -v -u "dceadmin:dcepass" -o
foo2.jpg "
http://192.168.0.82/low=/SnapshotJPEG?Resolution=320x240&amp;Quality=Standard"
and returns:

OpenSSL/0.9.8h zlib/1.2.3
&lt; HTTP/1.0 200 OK
  % Total    % Received % Xferd  Average Speed   Time    Time     Time
Current
                                 Dload  Upload   Total   Spent    Left
Speed
</description>
    <dc:creator>Paul Bergen</dc:creator>
    <dc:date>2008-11-29T01:31:29</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21546">
    <title>CURLOPT_WRITEDATA required in HTTP post?</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21546</link>
    <description>Not sure if this is a bug or just me making a mistake.

I was using libcurl to do a multi part form upload using HTTP post.

It worked 95% of the time.

Then occassionally it would fail with the error "Failed writing body".

It was occurring when libcurl was parsing the last received HTTP response
(the "HTTP OIK" response).

I tracked it down to sendf.c. When it was pushing the data out, whatever it
was writing to, didn't accept all the bytes.

Then, I realised that I hadn't actually set CURLOPT_WRITEFUNCTION, so it was
set to the default (which I think writes to a file). However, I also hadn't
set CURLOPT_WRITEDATA.

So, anyway, I explicitely set CURLOPT_WRITEFUNCTION to point to a dummy
function which just discarded the data, and my problem went away.

So my questions/comments  are:

1. Do you explicitely need to set either/both CURLOPT_WRITEFUNCTION  and
CURLOPT_WRITEDATA when using HTTP POST? I hadn't set them because I wasn't
interested in the response from the server (beyond whether it returned HTTP
20</description>
    <dc:creator>John Wood</dc:creator>
    <dc:date>2008-11-28T16:08:18</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21542">
    <title>implicit SSL with FileZilla server Unknown SSL protocol error1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21542</link>
    <description>There have been reports of this error when using curl with FileZilla server:
  error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number
or, more generally,
  Unknown SSL protocol error

(e.g. http://curl.haxx.se/mail/lib-2008-01/0338.html and following).

I'd like to clarify that this is the same problem that has been reported
for FileZilla server when using implicit SSL.  FileZilla server assumes
that the data channels are, by default, in clear mode.

One workaround is to use explicit SSL on the server. Another workaround
is to force curl to use a 'PROT P' command--for example, using
   -Q 'PROT P'
on the command line. This does
   curl_easy_setopt(curl, CURLOPT_SOURCE_QUOTE, ...);
in libcurl.  For the record, -Q '+PROT P' did not work; apparently the
command is sent too late.

In 2007 there was some talk on this list of always sending a PROT
command after a PBSZ, even in implicit SSL mode. I think this would be a
good idea.  We communicate with 4 other customers that use FTPS and our
communication </description>
    <dc:creator>Ken Hirsch</dc:creator>
    <dc:date>2008-11-26T22:11:34</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21535">
    <title>Reducing the number of DNS queries in the multi handle.</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21535</link>
    <description>Hi,
(I am using "libcurl 7.19.1" with c-ares 1.5.3 (over linux)).
I am writing a program that uses the libcurl to fetch a bundle of urls
using the multi interface.

Now, many of the urls I am fetching belongs to the same domain.
Example: www.aaa.com ,  www.aaa.com/1/ ,  www.aaa.com/2/

I can see that all DNS requests are sent at the beginning (probably
right after calling curl_multi_perform())  that's b/c  I am using
asynchronous name resolves.
I have noticed that the libcurl issues a different dns request for
every URL in the multi handle although many of them share the same
domain.
This means that for the example above I get to see 3 DNS requests for
www.aaa.com.

I am wondering if there is an option that I can use to tell the
libcurl to issue only one DNS request for www.aaa.com and only when
response arrives perform the HTTP requests for all the URL's sharing
the same domain ?


Thanks,
Tsachi

</description>
    <dc:creator>Tsachi</dc:creator>
    <dc:date>2008-11-26T10:41:01</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21532">
    <title>libcurl to tunnel VNC as SSL ?</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21532</link>
    <description>Hi,

Background (for sake of context) -

I am trying to create a VNC solution which uses tightvnc. The idea is
basically that the TightVNC server does a reverse connect to a VNC Reflector
(so that the server doesn't have to listen on any port). Then a viewer can
do a connect to the VNC Reflector and then basically see the desktop where
TightVNC server is running. VNC Reflector acts as a *client* to the TightVNC
server and as a *server* to a VNC Client. Basically, this lets us have a
solution where the reflector can be on a remote server and the users don't
have to listen to any port.

I got this working against a squid proxy (TCP tunnel).

Issue - Proxies :

Libcurl is used in a modified TightVnc Server code, where curl does the
initial handshakes - reverse connection - , tunneling
(CURLOPT_HTTPPROXYTUNNEL) and gives back the raw socket for the TightVNC
server to use.
The issue is this, there seems to be quite a few good proxy servers /
firewalls out there, which can figure out if the tunnel is actually carr</description>
    <dc:creator>Bharat Varma</dc:creator>
    <dc:date>2008-11-26T08:25:46</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21527">
    <title>-o Ignore Non-Existent Files</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21527</link>
    <description>I'm using curl via the cmd line/batch file in Windows XP.  I have it pulling a series of png files from a web server via the HTTP command:  "curl http://webserver/file.png -o c:\local\file.png" etc.  

The files are dynamically created and sometimes there is 1 file and other times there are up to 20.  Right now I have curl pulling all 20 files, but if the file does not exist on the server, curl creates an empty (1 kb) png file on the local server.  

How can tell curl to ignore these empty files, or do something within the batch file to ignore or delete these files.

Thanks.


      

</description>
    <dc:creator>Jake Abramson</dc:creator>
    <dc:date>2008-11-25T15:48:12</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21524">
    <title>Segmentation fault???</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21524</link>
    <description>I am using the following code to test:

#include &lt;stdio.h&gt;

#include &lt;curl/curl.h&gt;
#include &lt;curl/types.h&gt;
#include &lt;curl/easy.h&gt;

#define LOCAL_FILE      "/tmp/uploadthis.txt"
#define UPLOAD_FILE_AS  "while-uploading.txt"
#define REMOTE_URL      "ftp://user:pass&lt; at &gt;localhost/"  UPLOAD_FILE_AS


int main(int argc, char **argv)
{
CURL* curl;
CURLcode res;
FILE* err;
FILE* hd_src;

curl_global_init(CURL_GLOBAL_ALL);

curl = curl_easy_init();
if( curl )
{
err = fopen("/tmp/stderr.log", "a");
hd_src = fopen(LOCAL_FILE, "r");

curl_easy_setopt(curl, CURLOPT_UPLOAD, 1L);
curl_easy_setopt(curl, CURLOPT_TRANSFERTEXT, 1L);
curl_easy_setopt(curl, CURLOPT_READDATA, hd_src);
curl_easy_setopt(curl, CURLOPT_URL, REMOTE_URL);
curl_easy_setopt(curl, CURLOPT_STDERR, err);
curl_easy_setopt(curl, CURLOPT_VERBOSE, 1L);

res = curl_easy_perform(curl);

if( err )
{
fclose(err);
err = NULL;
}

if( hd_src )
{
fclose(hd_src);
hd_src = NULL;
}

curl_easy_cleanup(curl);

fprintf(stder</description>
    <dc:creator>Wei Weng</dc:creator>
    <dc:date>2008-11-25T04:35:40</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21523">
    <title>Get one filename-- C++</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21523</link>
    <description>Hi, I am doing some work for an open-source game.  I am working on an  
addon manage, i need to be able to get each files name, and print out  
each one that ends with 'tbz'.



    1. for (int num; !end-of-ftp-directory-list; num++)
    2. {
    3.      filename = ftpfile[num];
    4.      if (filename.extension== "tbz")
    5.      {
    6.           cout &gt;&gt; filname &gt;&gt; "\n";
    7.      }
    8. )

That is just the beginning of what I will be doing, but I need this  
function first

</description>
    <dc:creator>Josh Dye</dc:creator>
    <dc:date>2008-11-25T02:16:44</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21516">
    <title>Unknown SSL protocol error in connection</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21516</link>
    <description>
I'm new to OpenSLL so I apologize if I ask something trivial of the list.

I have libcurl and OpenSSL built on the CellOS but I'm having problem when
I try to use SSL (normal HTTP works).

command line:  --trace-ascii -k --cacert /app_home/mycert.pem --url
https://www.fortify.net/sslcheck.html

tty:
== Info:   Trying 64.202.169.234... == Info: connected
== Info: Connected to www.fortify.net (64.202.169.234) port 443 (#0)
== Info: libcurl is now using a weak random seed!
== Info: successfully set certificate verify locations:
== Info:   CAfile: /app_home/cakey.pem
  CApath: none
== Info: SSLv3, TLS handshake, Client hello (1):
=&gt; Send SSL data, 95 bytes (0x5f)
0000: ...[..D.s.......6..5W.7.b.....K..~)..\..4.9.8.5.............3.2.
0040: /.E.D.A........................
== Info: Unknown SSL protocol error in connection to www.fortify.net:443
== Info: Closing connection #0
curl: (35) Unknown SSL protocol error in connection to www.fortify.net:443


additionally, I notice if I set a break point the function int
s</description>
    <dc:creator>Jeff_Curley&lt; at &gt;playstation.sony.com</dc:creator>
    <dc:date>2008-11-24T23:49:32</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21513">
    <title>Need --enable-libcares</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21513</link>
    <description>Hello all,

I'm working on a new debian package (isn't official yet, but
the idea it will be) for the open source 'Second Life' viewer
(Open Meta Verse viewer, or omvviewer).

Linden Lab made their viewer open source and provides binary
downloads for several platforms, including linux 32bit.

I'm working with 64bit myself and that works perfectly,
compiled from source.

The binary distributed by LL links with a curl library
that is configured with --enable-libcares. They also
distribute this shared library (along with a lot of
others) and have people link with those libraries
when they start the viewer.

However, the debian team of omvviewer wants to use as
much as possible the shared libraries that are provided
by debian, of course. Several packages have already been
added to debian for this reason.

The problem at the top of our TODO list is currently:
the libcurl package in debian is not using libcares.
This results in the viewer freezing for a whole second,
sometimes even more, several times per minute. </description>
    <dc:creator>Carlo Wood</dc:creator>
    <dc:date>2008-11-24T19:58:52</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21511">
    <title>https connection over proxy</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21511</link>
    <description>hi all,i would like to know if there are any special settings/curl options
to make secure connections work over a http proxy. iam able to browser
non-secure sites by setting the http_proxy env. code reference below
(webkit+gtk port)

http://svn.webkit.org/repository/webkit/trunk/WebCore/platform/network/curl/ResourceHandleManager.cpp

would appreciate any inputs on this.

thanks,
Zaheer
</description>
    <dc:creator>zaheer ahmad</dc:creator>
    <dc:date>2008-11-24T19:03:31</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21501">
    <title>[BUG]Possible Bug? iphlpapi.dll causing libcurl(curllib) fail toreslove domain name</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21501</link>
    <description>I've been use curlLib for a while. Its great.

But One of my program just wont work with curlLib. I compile it with 
curl.lib (static link win32 program), curl always report can’t reslove 
domain name.

It confuse me for several days. Soon I found out, it is because the 
iphlpapi.dll which is in the same directory with my program.

After I remove iphlpapi.dll, my program with curlLib works fine. ~~~ Why?

Is it "work as intended"?

Thanks
Tomasen 


</description>
    <dc:creator>Tomasen</dc:creator>
    <dc:date>2008-11-22T17:53:37</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21498">
    <title>Curl with Asynchronous sockets</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21498</link>
    <description>Hi All,
Currently I am porting curllib to a proprietary ARM platform. In this
platform I have only asynchronous sockets in which all the socket events
(Connect, Read, Write and Close) are notified via events. As for my
understanding the the current curllib implementation expects the socket to
be synchronous or blocking at certain levels (At least for connecting). Can
anyone suggest me some idea or specific places/modules/functions where I can
start customizing curllib to support asynchronous sockets.


Regards
Kulanth
</description>
    <dc:creator>Kulanthaivel</dc:creator>
    <dc:date>2008-11-22T09:16:56</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21495">
    <title>Corrupted test cases</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21495</link>
    <description>Looks like something goofy happened to test cases 35, 262, 545 and 552. The
data files were all replaced by the text "Binary file (standard input)
matches" in the Pragma: no-cache check-in.

</description>
    <dc:creator>Dan Fandrich</dc:creator>
    <dc:date>2008-11-20T17:27:27</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21492">
    <title>data-&gt;set.str[STRING_PASSWORD] &lt;=&gt; conn-&gt;passwd</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21492</link>
    <description>Hello,

I am trying to understand the difference between

  conn-&gt;user       &lt;=&gt; data-&gt;set.str[STRING_USERNAME]
  conn-&gt;passwd     &lt;=&gt; data-&gt;set.str[STRING_PASSWORD]
  conn-&gt;proxyuser  &lt;=&gt; data-&gt;set.str[STRING_PROXYUSERNAME]
  conn-&gt;passwd     &lt;=&gt; data-&gt;set.str[STRING_PROXYPASSWORD]

On a first glance, the main difference seems to be that the
data-&gt;set.str[] entries are set by curl_easy_setopt() while the conn
entries are parsed from the url/.netrc.

</description>
    <dc:creator>Josef Wolf</dc:creator>
    <dc:date>2008-11-19T22:54:17</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21484">
    <title>curl talk on video</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21484</link>
    <description>Hey

I just wanted to mention that if you want to see my 38 minute presentation of 
curl and libcurl I did in December 2007, the video of my talk is now available 
online here:

  http://daniel.haxx.se/blog/2008/11/19/curl-presentation-video/

</description>
    <dc:creator>Daniel Stenberg</dc:creator>
    <dc:date>2008-11-19T14:23:16</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21457">
    <title>CWD Definitive Help Please</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21457</link>
    <description>First an outline of what I'm trying to do.

1. FTP Access to browse by GUI interface the ftp site in question.
2. Get files from the FTP site and download them.

I'm using PureBasic, I can understand C &amp; VB so if you have code
examples of this following problem I'd be grateful if you could post
them in those languages, if not I'll take anything I can get.


I'm stuck on step 1, no matter what I throw at libcurl it won't CWD and
won't list the new directory, could someone show me a clear example of a
persistent (constant just in case I used the wrong word) connection
changing Up and Down the directory structure, without having to
disconnect and log back in with the new path appended to the url.

I've seen a few people having problems in this area over my 24+ hours
trying to figure this out, the only solutions I've found is diconnecting
re-logging in with the CURLOPT_URL appended with the folder, this is
unacceptable I don't want to disconnect and reconnect for each folder, I
would like the connection to stay.</description>
    <dc:creator>Timothy Roughton</dc:creator>
    <dc:date>2008-11-17T04:34:16</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21442">
    <title>Bug: curl 7.18.2/7.19.2 looses timeouts</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21442</link>
    <description>Hi,

I'm having issues with curl 7.18.2 on ubuntu x86_64 smp

curl 7.18.2 (x86_64-pc-linux-gnu) libcurl/7.18.2 OpenSSL/0.9.8g 
zlib/1.2.3.3 libidn/1.8
Protocols: tftp ftp telnet dict ldap ldaps http file https ftps
Features: GSS-Negotiate IDN IPv6 Largefile NTLM SSL libz

(I tried 7.19.2 build from source as well, same problem)


I'm using the curl_multi_socket interface, to download via http.


The hiperfifo.c code from doc/examples is outdated as it does not use 
curl_multi_socket_action, and uses curl_multi_socket_all for no obvious 
reason, but it fits perfectly to describe my problem.

Extending the hiperfifo code by:

   curl_easy_setopt(conn-&gt;easy, CURLOPT_LOW_SPEED_TIME, 3L);
   curl_easy_setopt(conn-&gt;easy, CURLOPT_LOW_SPEED_LIMIT, 10L);

around line 315,

and compiling&amp;starting it.


starting netcat6 as a dumb listener which does nothing but accepting 
connections:

nc6 --continuous --exec /dev/null -l -p 8087

start a *single* download by writing the url to our netcat listener to 
the fifo
echo "ht</description>
    <dc:creator>koettermarkus&lt; at &gt;gmx.de</dc:creator>
    <dc:date>2008-11-15T06:07:07</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21440">
    <title>ioctl includes missing #ifdefs</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21440</link>
    <description>
In porting over to Cell I noticed a few includes missing ifdef's around
them like the following:

lib\dict.c
lib\file.c
lib\telnet.c
lib\tftp.c

are all missing checks around ioctl.h
#include &lt;sys/ioctl.h&gt;

should be
#ifdef HAVE_SYS_IOCTL_H
#include &lt;sys/ioctl.h&gt;
#endif

lib\transfer.c is missing a check around
#include &lt;signal.h&gt;

should be
#ifdef HAVE_SIGNAL_H
#include &lt;signal.h&gt;
#endif

Lastly I got stuck on a line of code that *should* compile but for some
reason using ProDG's GCC compiler it's not compiling and I was looking for
some help.

In connect.c the function singleipconnect()

has the following line:
#if defined(EAGAIN) &amp;&amp; EAGAIN != EWOULDBLOCK

that's checking to see if EAGAIN and EWOULDBLOCK are different to see if it
can add the EAGAIN case.

On Cell these expand out to:

sys/return_code.h
#define EAGAIN          -2147418111 /* 0x80010001 */

netex/errno.h
#define     EWOULDBLOCK                         SYS_NET_ERROR_EWOULDBLOCK
#define     SYS_NET_ERROR_EWOULDBLOCK           SYS_NET_ERROR_E</description>
    <dc:creator>Jeff_Curley&lt; at &gt;playstation.sony.com</dc:creator>
    <dc:date>2008-11-14T20:02:33</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.comp.web.curl.library/21439">
    <title>CURL and Java Hashtable</title>
    <link>http://comments.gmane.org/gmane.comp.web.curl.library/21439</link>
    <description>Hello,
I found a java applet on the internet, which retrieves a hashtable from 
the server.

Did someone of you ever try to read this kind of 
client-server-communication with a curl c++ programme and could give me 
some advice?

The response of the server looks like this:
java.util.Hashtable»%!J
loadFactorI
javaVendort
Windows XPx

Thanks,
Michael

</description>
    <dc:creator>Michael Hecker</dc:creator>
    <dc:date>2008-11-14T18:55:50</dc:date>
  </item>
  <textinput about="http://search.gmane.org/?group=$group=gmane.comp.web.curl.library">
    <title>Search Engine</title>
    <description>Search the mailing list at Gmane</description>
    <name>query</name>
    <link>http://search.gmane.org/?group=$group=gmane.comp.web.curl.library</link>
  </textinput>
</rdf:RDF>
