but the infrastructure is there.

to add more method in the future.

added.

funky logic in order to make re-tries work fine with OpenSSL. This corrects
the problem David Orrell noticed.

libcurl will stop reading when the number of bytes have arrived and not wait
for a closed socket.

#745122.

returned

data to the initial request buffer, if the total post data is less than
100K.

two separate send() calls, first the headers and then the data. I've now made
a fix that for static and known content that isn't to be chunked-encoded,
everything is now sent in one single system call again. This is also better
for network performance reasons.

same as using a httpproxy so we must make sure to better check for http
proxies before we do HTTP proxy stuff. This included authorization and
URI usage in the request etc.

contents. Test cases 144 and 145 were added to verify the fix.

Now we deal with return code 450 properly and other codes also do proper
cleanup.

SOCKS5 fix as suggested by Jis in bugreport #741841.

we unescape a string. We now check and decode only valid %XX strings.

in memory, only add it when we save the cookie. This makes all tailmatching
and domain string matching internally a lot easier.

This was also the reason for a remaining bug I introduced in my overhaul.

for FTP command responses take >1 second.

the define accordingly in the hostip.h header to work nicely all over.

it is.

name isn't an ip address

 set to "".  This frees the application from having to know which encodings
 the library supports.

the new socket to non-blocking

DATA

 EAGAIN but only EWOULDBLOCK, which caused badness on HPUX. We also check for
 and act the same on EINTR errors as well now.

extract the correct set of cookies to send. This functionality is verified
by test case 62.

portability. Also use Andy Cedilnik's compiler warning fixes.