The first argument to select is an int, while curl_socket_t is
unsigned long long when using WinSock. It's ignored anyway [1].

[1] https://msdn.microsoft.com/en-us/library/windows/desktop/ms740141.aspx

Don't convert string literal to char * before assigning it to
const char *.

Don't do anything in this file if CURLRES_THREADED is not defined.

The cases this warns about are handled elsewhere, so just use an
intermediate variable to silence the warning.

Apparently, /usr/bin/m4 ignores the --version parameter and waits for
input from stdin.

Fixes #1471

...to render properly nroff.

... and USE_ENVIRONMENT and --environment. It was once added for RISC OS
support and its platform specific behavior has been annoying ever
since. Added in commit c3c8bbd3, mostly unchanged since
then. Most probably not actually used for years.

Closes #1463

Also, document that numbered datacheck sections are possible.

Avoid casting string literals to non-const char *.

calloc and ai_addrlen expect different (usually unsigned) types.

Suggested-by: Dan Fandrich

... instead of numerical order.

Closes #1466

assign string literals to const char * instead of char * in order to
avoid a lot of these warnings:
cast from 'const char *' to 'char *' drops const qualifier
[-Wcast-qual]

Windows does not allow setting the locale with environment variables (as
the test attempted to do), so the test failed when run with a user
locale that has a comma as radixchar. Changed the test to call
setlocale() explicitly to ensure that a known working locale is set even
on Windows.

SEND_QUAL_ARG2 had to be set, but was never used. Use it in swrite to
avoid warnings about casting away low-level const.

Closes https://github.com/curl/curl/pull/1464

Fix the following warnings when building the tests by using the correct
types:
cast from 'const char *' to 'void *' drops const qualifier
[-Wcast-qual]
implicit conversion changes signedness [-Wsign-conversion]

Closes https://github.com/curl/curl/pull/1452

Closes https://github.com/curl/curl/pull/1452

Previous TODO wanting to write in chunks. We should support writing more
at once since some TELNET servers may respond immediately upon first
byte written such as WHOIS servers.

Closes #1389

... and instead properly respond with an error message to the user
instead of silently ignoring.

Fixes #1453
Closes #1458

Support for _MPRINTF_REPLACE in mprintf.h was removed in
55452ebd, replaced with curl_printf.h.

Closes #1449

The buffer can have other sizes.