| Age | Commit message (Collapse) | Author |
|---|
|
Test-Information:
Without this building failed under Gentoo Linux distribution.
Verified this on Sabayon, a desktop Linux distribution based
on Gentoo, by adding it to our cross distribution test.
With this fix it builds on all platforms, including Sabayon.
Change-Id: I23aec19b9755244dd3dbe6f581cf170eaf334dfc
|
|
Test-Information:
Swift builds with test=all on macOS 10.12 and all tests pass.
Change-Id: I9d7e8c49e65ac272a7ee672a95da5e24609d0a22
|
|
In the process converted try/catch use to Boost's error code
for more explicit error checking and added log output.
Test-Information:
Swift builds with test=all on macOS 10.12 and all tests pass.
Change-Id: I6e465354dbb38c7178b0406b9a544ebd3be62ddd
|
|
All our pimpl idiom usage used std::shared_ptr due to being
written in C++03. Now we use C++11 and const std::unique_ptr
is more sensible.
Test-Information:
Builds on macOS 10.12 and unit tests pass.
Change-Id: I1b9b3fbb22e337d53ae71e5a5e03118998cc3376
|
|
Test-Information:
Build successfully on openSUSE Tumbleweed which only has
libminiupnpc 2.0 as package. Unit tests passed.
Change-Id: I650e1224014162b55cdf64df7e72466b52dc3ade
|
|
Move std::cout/cerr logging to SWIFT_LOG.
Test-Information:
Builds and all tests pass on OS X 10.11.5.
Change-Id: I0a0c24654a8b3abf3244a79fd6d970eee90559d0
|
|
Test-Information:
Test that it still builds and unit test pass on OS X 10.11.4.
Change-Id: I2eb4a0b707991aee553db36a8cd1ae28b813acab
|
|
This cuts down our dependency on Boost further. Another
benefit is that mutex classes of C++11 standard library are
recognized by TSAN.
Test-Information:
Unit and integration tests pass on OS X 10.11.4.
Change-Id: Id4dcdb42e3d5155e107ce1d7618acbf26f913b6f
|
|
Drop executable rights on Swiften/Network/UnboundDomainNameResolver.*
License:
This patch is BSD-licensed, see Documentation/Licenses/BSD-simplified.txt for details.
Change-Id: Ie6e6eab48b96dff64e82e76615c51863baf37816
|
|
Test-Information:
Builds on OS X 10.11.4 with Apple clang and clang master.
Change-Id: I012577e29c6fcf2fb452b4f13912aaeb37250fb5
|
|
Boost.Signals was deprecated and is not improved further.
This patch removes Boost.Signals from 3rdParty and adds
Boost.Signals2 and its dependencies.
Also removed the Qt signals compatibility file
Swiften/Base/boost_bsignals.h.
Test-Information:
Build and ran unit tests on OS X 10.11.4. Confirmed successful
login using Swift client.
Change-Id: Ie6e3b2d15aac2462cda95401582f5287a479fb54
|
|
Run 'clang-tidy -fix -checks=modernize-loop-convert' on all
source code files on OS X. This does not modernize platform
specific code on Linux and Windows
Test-Information:
Code builds and unit tests pass on OS X 10.11.4.
Change-Id: I65b99e0978cfab8ca6de2a3e5342e7a81416c12c
|
|
Test-Information:
Builds on OS X 10.11.4 and unit tests pass.
Change-Id: I8775e8d1e3addbc88b220c1cc618637f706daca2
|
|
This change was done by applying the following 'gsed'
replacement calls to all source files:
's/\#include <boost\/shared_ptr\.hpp>/\#include <memory>/g'
's/\#include <boost\/enable_shared_from_this\.hpp>/\#include <memory>/g'
's/\#include <boost\/smart_ptr\/make_shared\.hpp>/\#include <memory>/g'
's/\#include <boost\/make_shared\.hpp>/\#include <memory>/g'
's/\#include <boost\/weak_ptr\.hpp>/\#include <memory>/g'
's/boost::make_shared/std::make_shared/g'
's/boost::dynamic_pointer_cast/std::dynamic_pointer_cast/g'
's/boost::shared_ptr/std::shared_ptr/g'
's/boost::weak_ptr/std::weak_ptr/g'
's/boost::enable_shared_from_this/std::enable_shared_from_this/g'
The remaining issues have been fixed manually.
Test-Information:
Code builds on OS X 10.11.4 and unit tests pass.
Change-Id: Ia7ae34eab869fb9ad6387a1348426b71ae4acd5f
|
|
Run 'clang-tidy -fix -checks=modernize-use-nullptr' on all
source code files on OS X. This does not modernize platform
specific code on Linux and Windows
Test-Information:
Code builds and unit tests pass on OS X 10.11.4.
Change-Id: Ic43ffeb1b76c1a933a55af03db3c54977f5f60dd
|
|
Removed trailing spaces and whitespace on empty lines
in the process.
Changed CheckTabs.py tool to disallow hard tabs in source
files.
Test-Information:
Manually checked 30 random files that the conversion worked
as expected.
Change-Id: I874f99d617bd3d2bb55f02d58f22f58f9b094480
|
|
Changed "" style includes to <> style.
Test-Information:
Build with Clang 3.9.0 and ran all tests on OS X 10.11.4.
Change-Id: Ic05e53f2e5dba39cc1307b116fc5f17b62ab9eb8
|
|
The issue occurs with ProxiedConnection that started
connecting but do not have an external reference anymore.
As soon as the handlers of the ProxiedConnection are
disconnected from the signals of the connection_ object,
the remaining references to a shared ProxiedConnection
vanish and the ProxiedConnection is deleted, while it still
requires access to its members in
ProxiedConnection::handleConnectFinished().
Test-Information:
All tests pass on OS X 10.11.3. No TSAN reports on Debian 8
in a scenario with randomly connecting/disconnecting Client
instances that use a HTTP proxy.
Change-Id: I4d6d2c85013e066d9ed298aa9b913afc83949e35
|
|
ThreadSanitizer reported a data-race between Boost's socket
close() and async_read_some().
Test-Information:
Verified all tests still pass and that TSAN does not report
an error anymore in a scenario where a Client connects/dis-
connects randomly.
Tested on Debian 8 and OS X 10.11.3.
Change-Id: I5e705efb15bee767dd5a55539854b5e488b3bf64
|
|
Code review makes me think that if the timer fires at the same time as
stop() is called, it would be possible for the clearing of events from
the loop to happen before the event is put into the loop, leading to
the sort of crashes we've seen with timers firing and access exceptions.
I can't reproduce those to know if this fixes them, and I'm not even sure
I'm not being dense thinking this patch fixes a real issue.
Test-Information:
Unit tests pass
Change-Id: I76337d5556a9c3902d5c2f4da754ae657810d436
|
|
Adds missing virtual keyword to destructors of classes which
have a non-empty destructor and are inheriting from other
classes and implement virtual functions.
Test-Information:
Compiles and unit tests pass on Windows 8 with VS 2013.
Change-Id: I172b5de8eda63eb8057113fbc979444abde3e0a7
|
|
This should enable IPv4/IPv6 dual-stack support for Swift(-en)
Jingle file-transfer support.
Add Connection::getRemoteAddress() method.
Test-Information:
Tested IPv6 file-transfer and IPv4 file-transfer between two
Swift instances.
Added integration test verifying IPv4 only, IPv6 only and
IPv4/IPv6 dual-stack support on the running platform.
Additionally added test to verify remote addresses on dual-stack
server.
Change-Id: Ie384a71833eacca554f69e6f12a1c8330d0d747f
|
|
Test-Information:
Builds and unit and system tests succeed on OS X 10.11.2.
Change-Id: I9e4805f31c4ba63c64d73dbfc2ee6b6423d949a8
|
|
This allows execution of events inside an existing io_service
if an application is already using Boost ASIO for other things
and can share the io_service.
Test-Information:
Builds on OS X 10.11.2.
Change-Id: I092ed7a25b24ef95d4664bae98ed84cc0f149073
|
|
Test-Information:
Unit tests pass on OS X 10.10.5.
Change-Id: I7b505513b4342001596ee8518bfdcf9e77c91479
|
|
This allows to authenticate using SASL EXTERNAL over BOSH
using a client TLS certificate for the HTTPS connection of
the BOSH channel.
The implementation also enforces the HTTPS server certificate
of subsequent BOSH connections not to change.
This commit also removes TLSConnection and TLSConnectionFactory
as no code is using them.
Test-Information:
Tested against M-Link 16.3v6-0 on Debian 7.9 and Swift on
OS X 10.10.5. Verified working client certificate authentication.
Verified Swift not falling back to password-based authentication,
in case EXTERNAL is not allowed by the server over BOSH or
the client certificate is invalid.
Change-Id: Ia96bcac27cac9fc9261ed847c82c6328307bfbd1
|
|
Test-Information:
Still builds and tests pass on OS X 10.10.5.
Change-Id: Ic616e7b9de443ce34b46de63ac1bb0cca34b08ee
|
|
This patch fixes the code to handle HTTP/1.0 proxies which
do not keep the connection alive after a single request.
If a HTTPTrafficFilter returns a new header reply, the HTTP
CONNECT proxy code will issue the request over a new connection.
The final connection is kept alive, as it used for the
persistent connection forwarding.
In addition, the response status line is now passed to the
HTTPTrafficFilter handling method to provide ability to act
upon the response status code.
Missing passthrough of the HTTPTrafficFilter object the down
the stack to the HTTPConnectProxiedConnection is added.
Test-Information:
Added a unit test following an NTLM HTTP proxy authentication
with a static HTTPTrafficFilter. This and other unit tests
still pass.
Change-Id: Ida0d1aec08a60f10c1480e1eecaecbd3f87b0dca
|
|
Added integration tests for certificate validation and
revocation behavior checking.
Test-Information:
Tested client login over TLS against Prosody and M-Link.
Verified client certificate authentication works against
M-Link.
Change-Id: I6ad870f17adbf279f3bac913a3076909308a0021
|
|
Test-Informations:
Code compiles without warnings and tests still pass.
Change-Id: If74c615706b8125c3c5186f0d940c103749ddb80
|
|
According to boost doucmentation shared deadline_timers are not
thread-safe. Adding a mutext to protect access to
boost::asio::deadline_timer instance in Swift::BoostTimer.
This fixes a data-race reported by TSAN when running
Swiften/QA/ClientTest/ClientTest.
Test-Information:
Verified that the data-race report is gone with this fix.
Change-Id: I62c8c3a07d6ea16fe6e2d24c879340040406699b
|
|
As the correct API usage might not be obvious
Change-Id: I1d4887d7101ce59eb8f3a3dc13d94befe4a58948
|
|
Explicitly disconnect from remaining resources in the
ChainedConnected and ClientSessionStanzaChannel destructors, so
the event loop will not call the signal handler methods on a
freed object.
Test-Information:
Repeating the test case of creating a Swift::Client instance,
connecting it and then deleting it after a random time below one
second.
On Mac OS X 10.9.5 running this test case causes two ASAN
heap-use-after-free errors and with this patch the errors are
gone.
Change-Id: I3e48150c3633f4076ca9172aad9e85ba389df950
|
|
Some servers have very restrictive TLS stacks that respond badly
to a bug in the SChannel TLS implementation, meaning that TLS
has to be limited to 1.0.
Add ClientOptions.tlsOptions. This is a method of passing options into
the TLS stack. It's currently
only used for the TLS 1.0 workaround in SChannel, but we might reasonably
expose other options in the future, such as limiting cypher suites.
Disables use of SSLv3 for SChannel
Also updates the coding style in SchannelContext a bit.
Test-Information:
Compiles on both OS X and Windows(SChannel). OS X doesn't show the new
option. Windows shows it, and remembers it between logins. Not tested
against a server requiring 1.0 only, but a previous hack with the
same approach was tested.
Change-Id: I1e7854d43811fd173f21f98d4dc3915fc7a4b322
|
|
Test-Information:
Tested build on Windows 8 with VS 2014 and ran unit tests.
Change-Id: I3d8096df4801be6901f22564e36eecba0e7310c4
|
|
Test-Information:
Verified that SCons still runs on OS X.
Change-Id: I7e9b97f90ee5581a691a959b6f2c999d93e0be53
|
|
released early
This patch also disconnects possible slots on stop() from the signals.
Test-Infromation:
This fixes a crash that occured during testing with
ConcurrentFileTransferTest.
Change-Id: I2b443f40ac3359c8142fcffc084647312b9a1e76
|
|
Removed dead debugging code and unnecessary includes as well.
Test-Information:
Tested with Swiften/QA/ClientTest, adjusted to have a manual hostname
set in the ClientOptions.
Tested on OS X 10.9.5, with NDK API level 14 on an emulated Android 4.0
ARM instance.
Without this patch the connector timed out during name lookup trying
to resolve the IP set in the ClientOptions::manualHostname. With this
patch it skips lookup and connects successfully.
Tested and verified unchanged behavior with this patch and without the
manual address override set.
Change-Id: I737327b2e66c9da78a1963e754bcf201b7d40626
|
|
Test-Information:
Compiles without the warning, unit tests pass.
Change-Id: I31bce47e4d6595b4ef405a87b56cf71db004e6a0
|
|
This fix lets HTTPConnectProxiedConnection buffer response data in
pieces until the end of the HTTP header is reached. Only then it will
try to parse the HTTP header.
This is *not* the HTTP chunked transfer encoding.
Test-Information:
Adjusted one test to respond in pieces and added a new test case that
verifies that response data is buffered.
Change-Id: Icfb987bdf2fc5771401a8a9c6979fa9ad1eebdca
|
|
This patch adds HTTPTrafficFilter and integrates it into
HTTPConnectProxiedConnection. This allows the HTTP CONNECT proxy
initialization process to be customized.
Test-Information:
Added a unit test that verifies the new functionality.
Change-Id: I0b93c319fb205487b8be65717276cd0dd38851a3
|
|
Test-Information:
Builds.
Change-Id: Ib5f1b4e9de814ab46723487e4d15974f0f5bd658
|
|
Test-Information:
Tested interoperability with Swiften using FileTransferTest.
Change-Id: Ic13a68a91cad199be0bfc8852ff43c25c7085f12
|
|
Test-Information:
Run on ClientTest as TSAN enabled build on Linux. Without the patch
it reports a data race for the stopRequested member.
Change-Id: I98373966c45d4c6e247f69fa8850278d885556f7
|
|
Test-Information:
All tests still succeed on OS X and Linux.
Change-Id: I2c02252b646b539b960cad18b6f88807afd84d32
|
|
Test-Information:
Verified using Port Map.app tool and router supporting UPnP.
Change-Id: I8d43a812f136979a70ef4dc62ff45448c6603b8b
|
|
libnatpmp has a high (over 2 minutes) timeout for discovery of NAT-PMP
support. This leads to a bad user experience when there is no NAT-PMP
support in the network.
Test-Information:
Tested this in an university network where neither UPnP nor NAT-PMP is
available. File-transfers waited more than 2 minutes for an answer of
a NAT-PMP device before starting the file-transfer.
Change-Id: I3ceaac7f89c982ffa683ee367d5ea3c3d4a2c5ca
|
|
urn:xmpp:jingle:apps:file-transfer:4.
Test-Information:
Adjusted unit tests and successfully build/run them on OS X 10.9.5.
Change-Id: I63789e3fb351999f719157b54fa9fcf95f40fb07
|
|
Boost query objects are setup with the flag address_configured by
default which only does IPv6 resoltuion if a non-loopback device is
found. This detection mechanism does not seem to work. At least in my
test VM there was a non-loopback device with IPv6 address and it still
reported no IPv6 addresses. Setting the flag to passive will
have it always send queries for AAAA records.
Test-Informatioon:
Tested on Mac OS X 10.9.5 and Elementary OS 0.2.1 (Linux).
Change-Id: I346bcb7982dee465aa7847a7bae8981a16f2d3bc
|
|
Change-Id: I94ab4bbb68c603fe872abeb8090575de042f5cb4
|
Swift