Fixes #3436
Closes #3448

 Problem 1

After LOTS of scratching my head, I eventually realized that even when doing
10 uploads in parallel, sometimes the socket callback to the application that
tells it what to wait for on the socket, looked like it would reflect the
status of just the single transfer that just changed state.

Digging into the code revealed that this was indeed the truth. When multiple
transfers are using the same connection, the application did not correctly get
the *combined* flags for all transfers which then could make it switch to READ
(only) when in fact most transfers wanted to get told when the socket was
WRITEABLE.

 Problem 1b

A separate but related regression had also been introduced by me when I
cleared connection/transfer association better a while ago, as now the logic
couldn't find the connection and see if that was marked as used by more
transfers and then it would also prematurely remove the socket from the socket
hash table even in times other transfers were still using it!

 Fix 1

Make sure that each socket stored in the socket hash has a "combined" action
field of what to ask the application to wait for, that is potentially the ORed
action of multiple parallel transfers. And remove that socket hash entry only
if there are no transfers left using it.

 Problem 2

The socket hash entry stored an association to a single transfer using that
socket - and when curl_multi_socket_action() was called to tell libcurl about
activities on that specific socket only that transfer was "handled".

This was WRONG, as a single socket/connection can be used by numerous parallel
transfers and not necessarily a single one.

 Fix 2

We now store a list of handles in the socket hashtable entry and when libcurl
is told there's traffic for a particular socket, it now iterates over all
known transfers using that single socket.

[skip ci]

Closes #3417

[skip ci]

Bug: https://curl.haxx.se/mail/lib-2018-12/0088.html

Closes #3449

similar to commit f508d29f

Closes #3443

Added Curl_resolver_kill() for all three resolver modes, which only
blocks when necessary, along with test 1592 to confirm
curl_multi_remove_handle() doesn't block unless it must.

Closes #3428
Fixes #3371

This reverts commit 07ebaf83.

This also reopens PR #3275 which brought the change now reverted.

Fixes #3384
Closes #3439

This allows programs to include curl/urlapi.h directly.

Reviewed-by: Daniel Gustafsson
Reported-by: Ben Kohler
Fixes #3438
Closes #3441

Starting with Visual Studio 2017 Update 9, Visual Studio doesn't like
the MinimalRebuild option anymore and warns:

cl : Command line warning D9035: option 'Gm' has been deprecated and
will be removed in a future release

The option can be safely removed so that the default is used.

Closes https://github.com/curl/curl/pull/3425

When building with Unicode on MSVC, the compiler warns about freeing a
pointer to const in Curl_unicodefree. Fix this by declaring it as
non-const and casting the argument to Curl_convert_UTF8_to_tchar to
non-const too, like we do in all other places.

Closes https://github.com/curl/curl/pull/3435

Closes #3426

... since it doesn't link with libcurl.

Reverts the commit dcd6f810 changes from this file.

Bug: https://curl.haxx.se/mail/lib-2019-01/0000.html
Reported-by: Shlomi Fish
Reviewed-by: Daniel Gustafsson
Reviewed-by: Kamil Dudka

Closes #3434

Closes #3431

Reviewed-by: Daniel Gustafsson
Closes #3432

Add a few missing examples to make `make examples` not leave the
workspace in a dirty state.

Closes #3427
Reviewed-by: Daniel Gustafsson <daniel@yesql.se>

Add Adrian Burcea who made the artwork for the curl://up 2018 event
which was held in Stockholm, Sweden.

When a non-empty list is appended to, and used as the returnvalue,
the list pointer can leak in case of an allocation failure in the
curl_slist_append() call. This is correctly handled in curl code
usage but we weren't explicitly pointing it out in the API call
documentation. Fix by extending the RETURNVALUE manpage section
and example code.

Closes #3424
Reported-by: dnivras on github
Reviewed-by: Daniel Stenberg <daniel@haxx.se>

MinGW-w64 defaults to targeting Windows 7 now, so GetTickCount64 is
used and the milliseconds are represented as unsigned long long,
leading to a compiler warning when implicitly converting them to long.

Researched-by: Tae Wong

NTLM2 did not work i.e. no NTLMv2 response was created. Changing the
check seems to work.

Ref: https://winprotocoldoc.blob.core.windows.net/productionwindowsarchives/MS-NLMP/[MS-NLMP].pdf

Fixes https://github.com/curl/curl/issues/3286
Closes https://github.com/curl/curl/pull/3287
Closes https://github.com/curl/curl/pull/3415

Due to a report of a missing name in THANKS I manually went through an
old CHANGES.0 file and added many previously missing names here.

The previous fix for parsing IPv6 URLs with a zone index was a paddle
short for URLs without an explicit port. This patch fixes that case
and adds a unit test case.

This bug was highlighted by issue #3408, and while it's not the full
fix for the problem there it is an isolated bug that should be fixed
regardless.

Closes #3411
Reported-by: GitYuanQu on github
Reviewed-by: Daniel Stenberg <daniel@haxx.se>

Reported-by: Tae Wong

This adds support for wildcard hosts in CURLOPT_RESOLVE. These are
try-last so any non-wildcard entry is resolved first. If specified,
any host not matched by another CURLOPT_RESOLVE config will use this
as fallback.

Example send a.com to 10.0.0.1 and everything else to 10.0.0.2:
  curl --resolve *:443:10.0.0.2 --resolve a.com:443:10.0.0.1 \
       https://a.com https://b.com



This is probably quite similar to using:
  --connect-to a.com:443:10.0.0.1:443 --connect-to :443:10.0.0.2:443

Closes #3406
Reviewed-by: Daniel Stenberg <daniel@haxx.se>

- Trailer function support.
- http 0.9 option.
- curl_easy_upkeep.

The project bug tracker is no longer hosted at sourceforge but is now
hosted on the curl Github page. Update the FAQ to reflect.

Closes #3410
Reviewed-by: Daniel Stenberg <daniel@haxx.se>

The DCL code had a typo in one of the commands which would make the
OpenSSL discovery on VAX fail. The correct syntax is F$ENVIRONMENT.

Closes #3407
Reviewed-by: Viktor Szakats <commit@vszakats.net>

Reviewed-by: Sergei Nikulov

closes #3196

This reverts commit c98ee5f6 since commit f3ce3873 fixed the
problem in a more generic way.

Follow-up to fb445a1e: Set conn->data explicitly to point out the
current transfer when invoking the protocol-specific disconnect function
so that it can work correctly.

Bug: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=12173

- Use QueryPerformanceCounter on Windows Vista+

There is confusing info floating around that QueryPerformanceCounter
can leap etc, which might have been true long time ago, but no longer
the case nowadays (perhaps starting from WinXP?). Also, boost and
std::chrono::steady_clock use QueryPerformanceCounter in a similar way.

Prior to this change GetTickCount or GetTickCount64 was used, which has
lower resolution. That is still the case for <= XP.

Fixes https://github.com/curl/curl/issues/3309
Closes https://github.com/curl/curl/pull/3318