This brings us to almost parity with the C++11 Random Engine API
requirements (see chapter 26.5.1.4 [rand.req.eng]). We don't implement
the templated Sseq requirements because it would require moving the
implementation details to the public API. And we don't implement the
<iostreams> code because we don't want to.
Change-Id: Icaa86fc7b54d4b368c0efffd14f05ff813ebd759
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Since we don't document how many bytes one needs (it's 2496), it's
difficult for the caller to provide just enough data in the seed
sequence. Moreover, since std::mt19937 doesn't make it easy to provide
the ideal size either, we can't actually write code that operates
optimally given a quint32 range either -- we only provide it via
std::seed_seq, which is inefficient.
However, we can do it internally by passing QRandomGenerator to the
std::mersenne_twister_engine constructor, as it's designed to work.
Change-Id: Icaa86fc7b54d4b368c0efffd14f0613c10998321
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Now only QRandomGenerator::system() will access the system-wide RNG,
which we document to be cryptographically-safe and possibly backed by a
true HWRNG. Everything else just wraps a Mersenne Twister.
Change-Id: I0a103569c81b4711a649fffd14ec8cd3469425df
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Since we're adding a deterministic generator that inherently does not
use syscalls, and people should really use that one by default, there is
no point in optimizing the secure generator wrt syscalls. Besides,
keeping the random data in memory for longer than needed is likely
inadviseable.
Change-Id: Ib17dde1a1dbb49a7bba8fffd14ed0871117fe930
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
To obtain "proper" directory behavior, we have to check against the
extracted "resources" directory instead of its qrc counterpart.
Change-Id: I4996ba74419945f78d356ad953a5b826ff663687
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
To obtain the file's handle, we need to obtain it from the extracted
test data instead of qrc.
Change-Id: I89c5c3f3a7da7e36205a439581a6d83efffdc07c
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Resource files are extracted to m_dataDir in tst_QFile::initTestCase.
Instead of trying to access the file from the resource on systems that
use qrc for bundling the test data, we have to use the files that were
extracted at the beginning of the test.
Change-Id: I35453fbdeb27e317d1342ff1cb7bbea9cebea14d
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
On some platforms (like UWP) files that are copied during
qfile auto tests are not writable by default. The cleanup will fail for
these files if the permissions are not set accordingly.
Change-Id: Id925dcadfc6b505c87f1f55d5ea05e286b60a5a5
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
The vast majority is actually switched to QRandomGenerator::bounded(),
which gives a mostly uniform distribution over the [0, bound)
range. There are very few floating point cases left, as many of those
that did use floating point did not need to, after all. (I did leave
some that were too ugly for me to understand)
This commit also found a couple of calls to rand() instead of qrand().
This commit does not include changes to SSL code that continues to use
qrand() (job for someone else):
src/network/ssl/qsslkey_qt.cpp
src/network/ssl/qsslsocket_mac.cpp
tests/auto/network/ssl/qsslsocket/tst_qsslsocket.cpp
Change-Id: Icd0e0d4b27cb4e5eb892fffd14b5285d43f4afbf
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
... by moving them in QTestPrivate namespace (qtesthelpers_p.h). This
header file is a convenient staging area for helper APIs, eventually
some could be moved to public QTest API.
This header file utilizes the same pattern as other qtestlib header
files - wrapping functions with QT_${LIBNAME}_LIB to automatically
enable certain APIs based on what is in the projects dependencies,
e.g. QT += widgets.
Change-Id: Ic0266429939c1f3788912ad8b84fc6e0d5edd68b
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Discussed during Qt Contributor Summit 2017. We concluded that we don't
want to make these functions public, as they do not follow Qt coding
style API. Specifically,
qStartsWith(a, b)
is not easily understood which argument is the needle and which argument
is the haystack (same problem memcpy() has). Compare that to
a.startsWith(b)
which can clearly be read in English as a subject-verb-object sentence.
This commit removes the unit tests that called compare().
Discussed-on: http://lists.qt-project.org/pipermail/development/2017-October/031060.html
Change-Id: Icaa86fc7b54d4b368c0efffd14ee6205eb9043fb
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Right now,this does really nothing. This commit is just to allow us to
transition the other modules (besides qtbase) to use the syntax that
will become the API.
I've marked three places to use the system CSPRNG:
1) the QHash seed
2) QUuid
3) QAuthenticator
I didn't think the HTTP multipart boundary needed to be
cryptographically safe, so I changed that one to the global generator.
Change-Id: Ib17dde1a1dbb49a7bba8fffd14ecf1938bd8ff61
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Open failures due to sharing violations have been observed in Coin.
Change-Id: If7fbe01a454b3c343c0b87f73db50c28eae901c3
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@qt.io>
Guard the FILE * obtained by fopen() by a RAI class ensuring
the file is closed on destruction.
Change-Id: I9297f91ca2120238f3a44bad92bca5f920e01aa8
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@qt.io>
There were a couple of corner cases where doing setPort() would result
in QUrl thinking that an authority was not present. Since the full URL
parsing implies that a host is always present if the authority is
present, then we also imply that setting the port number makes the host
be present too.
Change-Id: I69f37f9304f24709a823fffd14e67c12da18d69f
Reviewed-by: David Faure <david.faure@kdab.com>
The previous version was good, just not optimal. Because the input was
an unsigned 64-bit number, compilers needed to generate extra code to
deal with HW instructions that only convert 64-bit signed input. And
that was useless because a double uniformly distributed from 0 to 1 can
only have 53 bits of randomness.
The previous implementation did exactly what the Microsoft libstdc++ and
libc++ implementations do. In my opinion, those implementations have an
imperfect distribution, which is corrected in this commit. In those, all
random input bigger than 0x20000000000000 has a different frequency
compared to input below that mark. For example, both 0x20000000000000
and 0x20000000000001 produce the same result (4.8828125e-4).
What's more, for the libc++ and MSVC implementations, input between
0xfffffffffffff001 and 0xffffffffffffffff results in 1.0 (probability 1
in 2⁵³), even though the Standard is very clear that the result should
be strictly less than 1. GCC 7's libstdc++ doesn't have this issue,
whereas the versions before would enter an infinite loop.
Change-Id: Ib17dde1a1dbb49a7bba8fffd14eced3c375dd2ec
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Error out if it's missing or broken (Mersenne Twister not present).
This ensures that we never have a low-quality random generator in Qt.
Change-Id: I0a103569c81b4711a649fffd14ec80649df7087e
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Auto-reset events are automatically reset to non-signaled when we get
notified about the signaled state. This implies that we cannot check the
event state again in activateEventNotifiers. Instead, store the signaled
state in a QAtomicInt and test and decrement that.
This amends commit 85403d0a.
Task-number: QTBUG-63555
Change-Id: I0adee3d2929783f98d91060f9106c8b5266d72fa
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
We're not going to crash to avoid core dumps getting recorded by
systemd, filling up filesystems or showing dialog boxes. We just need to
exit without running destructors. Now, exit() isn't supposed to run
function-scope destructors, so just in case we use _exit().
Change-Id: I0b48fc8e90304e0dacc3fffd14e909ff248ce1a1
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
<future> is needed by QThread::create. Instead of a fragile series
of preprocessor tests, move its detection to a configure test.
This dramatically simplifies the code, but on the other hand ties
the availability of QThread::create() to the system used to compile
Qt (rather the one used to compile an application).
Change-Id: If1b06363379bf29126cfa68f2a0651cbb78a67f7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Commits aba336c2b4 (in Qt 5.2) and
aba336c2b4 (in 5.6) both tried to deal
with this problem, with different levels of success. This is the third
attempt (and hopefully the charm).
Instead of modifying the path that the user provides, go straight ahead
and declare it invalid. This is supported by RFC 3986, which declares
this expansion impossible:
relative-part = "//" authority path-abempty
/ path-absolute
/ path-noscheme
/ path-empty
path-abempty = *( "/" segment )
path-absolute = "/" [ segment-nz *( "/" segment ) ]
path-noscheme = segment-nz-nc *( "/" segment )
The "path-abempty" and "path-noscheme" cases are the two issues we
already handle. This commit adds the third one: path-absolute, which
requires that the first segment of the path be of non-zero length.
That is, it is now possible again to have http://example.com//path
constructed piece-wise, without it producing http://example.com/path.
Additionally, it catches the case of http://example.com//path parsed
from full URL then followed by setAuthority("").
Change-Id: I69f37f9304f24709a823fffd14e67a5e7212ddcd
Reviewed-by: David Faure <david.faure@kdab.com>
The qWaitFor functions themselves can not trigger a test failure, as that
will not result in the test function exiting early, so every single call
to qWaitFor needs to be wrapped in a QVERIFY.
Change-Id: Id15a1549f31d06cdbf788e1d84ea431c28636ec8
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
When many runnables are executed, this improves the
performance by not resizing the queue for each runnable,
which was the case in the previous version, because of
many calls to QVector::takeFirst().
Also add a test that makes sure tryTake() is safe to
call and does not leave the queue in a bad state that
tries to use nullptr entries.
Change-Id: I608134ecfa9cfc03db4878dcbd6f9c1107e13e90
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Convert QSysInfo/QOperatingSystemVersion to __builtin_available where
required or possible, or to QOperatingSystemVersion where
__builtin_available cannot be used and is not needed (such as negated
conditions, which are not supported by that construct).
Change-Id: I83c0e7e777605b99ff4d24598bfcccf22126fdda
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This avoids looping prior to the main 16-byte loop, by performing one
load that may include bytes prior to the start of the string. This is
guaranteed not to fault, since str points to a valid character, but it
may cause Valgrind to print warnings.
Change-Id: I6e9274c1e7444ad48c81fffd14dcae854bba24b2
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
"generate" is better than "get", and we already have "generate(it, it)"
which uses std::generate(). This changes:
- get32() → generate()
- get64() → generate64() and QRandomGenerator64::generate()
- getReal() → generateDouble()
Change-Id: I6e1fe42ae4b742a7b811fffd14e5d7bd69abcdb3
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
One example is VxWorks.
Change-Id: I253df715a9417c1f9cede79b1e1860924e0da8a9
Reviewed-by: Tuomas Heimonen <tuomas.heimonen@qt.io>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Performance is more important in this case than the theoretical benefit
of constexpr. This commit implements the SSE2 search for 16-bit null and
it might be possible to implement the equivalent for AArch64
(investigation required). It also adds a fallback to wcslen() for
systems where wchar_t is short (non-x86 Windows or 32-bit x86 build with
-no-sse2).
We can re-add the constexpr loop once the C++ language has a way of
overloading constexpr and non-constexpr. GCC has a non-standard way to
do that with __builtin_constant_p, which is also implemented in this
commit, but note that the inline function is still not constexpr.
Change-Id: I6e9274c1e7444ad48c81fffd14dcaacafda5ebdc
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QMacAutoReleasePool is backed by an NSAutoreleasePool, which documents that
"you should always drain an autorelease pool in the same context (invocation
of a method or function, or body of a loop) that it was created".
This means allocating QMacAutoReleasePool on the heap is not a supported
use-case, but unfortunately we can't detect it on construction time.
Instead we detect whether or not the associated NSAutoreleasePool has been
drained, and prevent a double-drain of the pool.
Change-Id: Ifd7380a06152e9e742d2e199476ed3adab326d9c
Reviewed-by: Simon Hausmann <simon.hausmann@qt.io>
[ChangeLog][QtCore][QFileInfo] Relative symbolic links on Windows are
now resolved to their absolute path by symLinkTarget().
Task-number: QTBUG-62802
Change-Id: I5826517130bd389aef994bf3f4b6d99b2a91b409
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Since MSVC doesn't have <chrono> (according to QT_HAS_INCLUDE), the QSKIP
in the test was printed for every line in the table. Instead, add the
skip in the _data() function.
Change-Id: I6e9274c1e7444ad48c81fffd14dbcee5e5a322aa
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Olivier Goffart (Woboq GmbH) <ogoffart@woboq.com>
Conflicts:
examples/examples.pro
qmake/library/qmakebuiltins.cpp
src/corelib/global/qglobal.cpp
Re-apply b525ec2 to qrandom.cpp(code movement in 030782e)
src/corelib/global/qnamespace.qdoc
src/corelib/global/qrandom.cpp
src/gui/kernel/qwindow.cpp
Re-apply a3d59c7 to QWindowPrivate::setVisible() (code movement in d7a9e08)
src/network/ssl/qsslkey_openssl.cpp
src/plugins/platforms/android/androidjniinput.cpp
src/plugins/platforms/xcb/qxcbconnection.cpp
src/plugins/platforms/xcb/qxcbconnection_xi2.cpp
src/widgets/widgets/qmenu.cpp
tests/auto/widgets/kernel/qwidget_window/tst_qwidget_window.cpp
Change-Id: If7ab427804408877a93cbe02079fca58e568bfd3
Consider the following:
/root/target - a file
/root/path/link -> ../target
/root/path/other/exe - executable
Running from /root/path/other.
exe is:
#include <QDebug>
#include <QFileInfo>
int main()
{
qDebug() << QFileInfo("../link").symLinkTarget()
return 0;
}
The link references /root/target, but the current output is
/root/path/target.
The link doesn't depend on the PWD. It depends on its own directory.
Change-Id: I61e95018154a75e0e0d795ee801068e18870a5df
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
