Stefan Agner [Fri, 6 Apr 2018 19:00:02 +0000 (21:00 +0200)]
tool_operate: Fix retry on FTP 4xx to ignore other protocols
Only treat response code as FTP response codes in case the
protocol type is FTP.
This fixes an issue where an HTTP download was treated as FTP
in case libcurl returned with 33. This happens when the
download has already finished and the server responses 416:
HTTP/1.1 416 Requested Range Not Satisfiable
winbuild: make the clean target work without build-type
Due to the check in Makefile.vc and MakefileBuild.vc, no make call can
be invoked unless a build-type was specified. However, a clean target
only existed when a build type was specified. As a result, the clean
target was unreachable. Made clean target unconditional.
FTP: allow PASV on IPv6 connections when a proxy is being used
In the situation of a client connecting to an FTP server using an IPv6
tunnel proxy, the connection info will indicate that the connection is
IPv6. However, because the server behing the proxy is IPv4, it is
permissable to attempt PSV mode. In the case of the FTP server being
IPv4 only, EPSV will always fail, and with the current logic curl will
be unable to connect to the server, as the IPv6 fwdproxy causes curl to
think that EPSV is impossible.
Jon DeVree [Thu, 29 Mar 2018 22:39:56 +0000 (18:39 -0400)]
file: restore old behavior for file:////foo/bar URLs
curl 7.57.0 and up interpret this according to Appendix E.3.2 of RFC
8089 but then returns an error saying this is unimplemented. This is
actually a regression in behavior on both Windows and Unix.
Before curl 7.57.0 this URL was treated as a path of "//foo/bar" and
then passed to the relevant OS API. This means that the behavior of this
case is actually OS dependent.
The Unix path resolution rules say that the OS must handle swallowing
the extra "/" and so this path is the same as "/foo/bar"
The Windows path resolution rules say that this is a UNC path and
automatically handles the SMB access for the program. So curl on Windows
was already doing Appendix E.3.2 without any special code in curl.
libcurl (with the OpenSSL backend) performs server certificate verification
even if verifypeer == 0 and the verification result is available using
CURLINFO_SSL_VERIFYRESULT. The commit that is being reverted caused the
CURLINFO_SSL_VERIFYRESULT to not have useful information for the
verifypeer == 0 use case (it would always have
X509_V_ERR_UNABLE_TO_GET_ISSUER_CERT_LOCALLY).
Jay Satiro [Mon, 26 Mar 2018 22:29:30 +0000 (18:29 -0400)]
travis: enable apt retry on fail
This is a workaround for an unsolved travis issue that is causing CI
instances to sporadically fail due to 'unable to connect' issues during
apt stage.
Daniel Stenberg [Thu, 22 Mar 2018 16:59:58 +0000 (17:59 +0100)]
threaded resolver: track resolver time and set suitable timeout values
In order to make curl_multi_timeout() return suitable "sleep" times even
when there's no socket to wait for while the name is being resolved in a
helper thread.
Daniel Stenberg [Mon, 19 Mar 2018 15:28:05 +0000 (16:28 +0100)]
http2: read pending frames (including GOAWAY) in connection-check
If a connection has received a GOAWAY frame while not being used, the
function now reads frames off the connection before trying to reuse it
to avoid reusing connections the server has told us not to use.
Don [Fri, 16 Mar 2018 19:49:13 +0000 (12:49 -0700)]
cmake: add support for brotli
Currently CMake cannot detect Brotli support. This adds detection of the
libraries and associated header files. It also adds this to the
generated config.
Rick Deist [Sat, 17 Mar 2018 19:10:04 +0000 (20:10 +0100)]
resolve: add CURLOPT_DNS_SHUFFLE_ADDRESSES
This patch adds CURLOPT_DNS_SHUFFLE_ADDRESSES to explicitly request
shuffling of IP addresses returned for a hostname when there is more
than one. This is useful when the application knows that a round robin
approach is appropriate and is willing to accept the consequences of
potentially discarding some preference order returned by the system's
implementation.
When a transfer is requested to get done and it is put in the pending
queue when limited by number of connections, total or per-host, libcurl
would previously very aggressively retry *ALL* pending transfers to get
them transferring. That was very time consuming.
By reducing the aggressiveness in how pending are being retried, we
waste MUCH less time on putting transfers back into pending again.
Some test cases got a factor 30(!) speed improvement with this change.
Daniel Stenberg [Fri, 16 Mar 2018 11:03:52 +0000 (12:03 +0100)]
pause: when changing pause state, update socket state
Especially unpausing a transfer might have to move the socket back to the
"currently used sockets" hash to get monitored. Otherwise it would never get
any more data and get stuck. Easily triggered with pausing using the
multi_socket API.
Reported-by: Philip Prindeville
Bug: https://curl.haxx.se/mail/lib-2018-03/0048.html
Fixes #2393
Closes #2391
Daniel Stenberg [Thu, 15 Mar 2018 15:43:00 +0000 (16:43 +0100)]
rate-limit: use three second window to better handle high speeds
Due to very frequent updates of the rate limit "window", it could
attempt to rate limit within the same milliseconds and that then made
the calculations wrong, leading to it not behaving correctly on very
fast transfers.
This new logic updates the rate limit "window" to be no shorter than the
last three seconds and only updating the timestamps for this when
switching between the states TOOFAST/PERFORM.
Kamil Dudka [Tue, 13 Mar 2018 11:00:54 +0000 (12:00 +0100)]
tests/.../spnego.py: fix identifier typo
Detected by Coverity Analysis:
Error: IDENTIFIER_TYPO:
curl-7.58.0/tests/python_dependencies/impacket/spnego.py:229: identifier_typo: Using "SuportedMech" appears to be a typo:
* Identifier "SuportedMech" is only known to be referenced here, or in copies of this code.
* Identifier "SupportedMech" is referenced elsewhere at least 4 times.
curl-7.58.0/tests/python_dependencies/impacket/smbserver.py:2651: identifier_use: Example 1: Using identifier "SupportedMech".
curl-7.58.0/tests/python_dependencies/impacket/smbserver.py:2308: identifier_use: Example 2: Using identifier "SupportedMech".
curl-7.58.0/tests/python_dependencies/impacket/spnego.py:252: identifier_use: Example 3: Using identifier "SupportedMech" (2 total uses in this function).
curl-7.58.0/tests/python_dependencies/impacket/spnego.py:229: remediation: Should identifier "SuportedMech" be replaced by "SupportedMech"?
Daniel Stenberg [Wed, 31 Jan 2018 07:40:11 +0000 (08:40 +0100)]
FTP: reject path components with control codes
Refuse to operate when given path components featuring byte values lower
than 32.
Previously, inserting a %00 sequence early in the directory part when
using the 'singlecwd' ftp method could make curl write a zero byte
outside of the allocated buffer.
richardthe3rd [Fri, 2 Mar 2018 22:13:42 +0000 (22:13 +0000)]
winbuild: prefer documented zlib library names
Check for existence of import and static libraries with documented names
and use them if they do. Fallback to previous names.
According to
https://github.com/madler/zlib/blob/master/win32/README-WIN32.txt on
Windows, the names of the import library is "zdll.lib" and static
library is "zlib.lib".
Marcel Raad [Sat, 3 Mar 2018 19:25:24 +0000 (20:25 +0100)]
krb5: use nondeprecated functions
gss_seal/gss_unseal have been deprecated in favor of
gss_wrap/gss_unwrap with GSS-API v2 from January 1997 [1]. The first
version of "The Kerberos Version 5 GSS-API Mechanism" [2] from June
1996 already says "GSS_Wrap() (formerly GSS_Seal())" and
"GSS_Unwrap() (formerly GSS_Unseal())".
Use the nondeprecated functions to avoid deprecation warnings.
Marcel Raad [Sat, 3 Mar 2018 18:52:43 +0000 (19:52 +0100)]
curl_ctype: fix macro redefinition warnings
On MinGW and Cygwin, GCC and clang have been complaining about macro
redefinitions since 4272a0b0fc49a1ac0ceab5c4a365c9f6ab8bf8e2. Fix this
by undefining the macros before redefining them as suggested in
https://github.com/curl/curl/pull/2269.
Marcel Raad [Sun, 25 Feb 2018 20:16:32 +0000 (21:16 +0100)]
unit1309: fix warning on Windows x64
When targeting x64, MinGW-w64 complains about conversions between
32-bit long and 64-bit pointers. Fix this by reusing the
GNUTLS_POINTER_TO_SOCKET_CAST / GNUTLS_SOCKET_TO_POINTER_CAST logic
from gtls.c, moving it to warnless.h as CURLX_POINTER_TO_INTEGER_CAST /
CURLX_INTEGER_TO_POINTER_CAST.