| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
See http://www.unicode.org/reports/tr14/#CB
and http://www.unicode.org/reports/tr14/#LB20 for details
Change-Id: Ice0aa2b2ce81f6e39839a353240420436eddd754
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
| |
Change-Id: I8362663454e4c6604ecb6289ae8009d47c78aeb1
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
to make it conformant to the Unicode 6.1 specifications #14 and #29.
The most important changes are:
* The implementation has been reworked from scratch to fix all known bugs;
* Separate-out the grapheme and the line breaking implementation to eliminate
an overhead due to calculating unnecessary breaks;
* Stop using deprecated SG class in favor of resolving pairs of surrogates;
* A proper support for SMP code points;
* Support for extended grapheme clusters (a drop-in replacement for the legacy
grapheme clusters as of Unicode 5.1);
* The hardcoded tailoring of UBA has been eliminated which breaks the 7 years-old
lineBreaking test. Some later, we'll investigate if such a tailoring is still needed.
Change-Id: I9f5867b3cec753b4fc120bc5a7e20f9a73d89370
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
| |
Change-Id: If22018ff83cfc6b9c984f689648da038fce11d84
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This header can be included at any time on x86 systems and is present
since the GCC versiosn that also support AVX. It contains intrinsics
for instructions that have been present in x86 CPUs since the dawn of
time.
Change-Id: I9adb066c2c0b56ce8fd5ed7366716038f1254502
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
|
|
|
|
|
|
|
|
|
| |
According to ISO 8601 (section 4.2.2.3), seconds can be omitted
from a string representing time.
Task-number: QTBUG-2813
Change-Id: I2578f290845e46a8f49be489f1d7427984ae7f08
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
to keep them consistent with positions for all other flags.
This changes the internal behavior so that attributes[0].lineBreakType now means
"break opportunity at start of the text (before the first character in the string)"
and is always assigned with HB_NoBreak to conform rule LB2
(see http://www.unicode.org/reports/tr14/#LB2).
The current implementation is based on the sample implementation from tr14
that aimed to be as simple as possible rather than to be optimal.
From now, we can use pieces of the attributes array "as is"
without having to adjust some positions. Or we can analize some long text
by chunks (e.g. paragraph by paragraph) and consume less memory.
This introduces a minor overhead that will be eliminated shortly.
Change-Id: Ic873a05a9d5203b1c3d5aff2e4445a3f034c4bd2
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
| |
SoftHyphen enum value was added to specify such a boundary reason
Change-Id: I4248909eed6ab8cbca419de4dcf9fe917620a158
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The white spaces determination doesn't belong to the text breaking algorithm.
A proper breaking implementation shouldn't assume spaces are
break opportunities (actually, space is allowed to be a grapheme base);
However, the whiteSpace flag should never be checked alone while iterating
over the text to find the space sequence; the grapheme boundaries should always
be taken into account. This covers the SMP code points in UTF-16 text and
graphemes that consist of a space followed with one or more grapheme extenders.
This introduces a minor overhead that would be eliminated some later.
Change-Id: Ic2cc7f485631fd0b436fc256ce112ded5f94fc07
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The main problem with using QWeakPointer to track QObjects is that it
has API to convert to QSharedPointer, which is undefined behavior. There
is no need to overload the meaning of QWeakPointer with QObject
tracking.
QPointer is more suitable and should be used instead. Because QPointer
is implemented in terms of QWeakPointer, we need to add some overloads
which are not deprecated.
Change-Id: If0333e4800c9fd277629cc69185c9ca3e4e7e81d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
| |
Suggested by Oswald Buddenhagen. This function is about the (main)
UI language, not about other locale features like number formatting etc.
It not in the public API anymore in Qt 5.0 so it can be renamed.
Change-Id: I2a23892c67e5813de4c0e57330749568777e9ee6
Reviewed-by: Oswald Buddenhagen
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Because QSystemLocale::fallbackLocale() is about UI languages,
it makes sense to check LANGUAGE as well if appropriate.
Adapt tst_qlocale.cpp accordingly.
Suggested by Oswald Buddenhagen.
Change-Id: Ib2c9674081809e3251be4e34456b05210eebc010
Reviewed-by: Oswald Buddenhagen
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit b9790a04eeba664ecdf9ace24911094a71b5f0bd
(https://codereview.qt-project.org/24304) introduced the problem that
the country gets initialized from LC_NUMERIC and the language from
LC_MESSAGES. For example, if LC_NUMERIC=ru_RU and LC_MESSAGE=fr_FR,
then QLocale::system().name() returns "fr_RU". It is not nice to mix
the values of two LC_ variables there. Therefore, revert this change
and use LC_MESSAGES instead of LC_NUMERIC in
QSystemLocale::fallbackLocale(). This was also suggested in the
changelog of b9790a04 and it looks like a better way to fix the
problem.
Change-Id: I8fa6fec2b33e9f1f5a31c4b288503a658dad6d30
Reviewed-by: Denis Dzyubenko
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This class does not exist, is not needed and has never been published
in a released version of Qt.
It existed during the development of QSharedPointer, when internal
reference counting (also known as intrusive counting) was a goal. That
goal was abandoned when use with forward-declared classes was
preferred.
Change-Id: If3a5a29c07fc71e2001d6ba64b90ddd241ab8ae3
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Currently, QString::toFloat() returns 0 (and sets ok to false) if you
try to convert "inf". This is because inf is greater than QT_MAX_FLOAT
and there is currently no check to handle inf.
Task-number: QTBUG-8629
Change-Id: I498daf4a7a6f880f928461fca628fcaf7d1d6d08
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
|
|
|
|
|
|
|
|
| |
These names were confusing and conflicting with standard C memory
management functions.
Change-Id: I6efe20665d2ec7ad3e00f3a806cc1843a57374d4
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
| |
ISO 8601 section 4.2.3 states that "The end of one calendar day [24:00]
coincides with [00:00] at the start of the next calendar day", so
fromString() was updated to account for this.
Task-number: QTBUG-25387
Change-Id: I391db0da755dbc822ba0820c302a2c10391e1f3b
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
| |
QLocale::textDirection() was missing Pashto as a
right to left language.
Change-Id: I1623abf711597a26f283a86708dc756696790b7d
Reviewed-by: Lars Knoll
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Chars that have a case conversion that converts
them into several characters can't be handled
by QChar::toUpper() etc and should get ignored. The code
didn't do that correctly.
Change-Id: I281d122e90bf49187b6449088d2fccef2ef75e86
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Initialiser lists were not tested before in the QVector rewrite, so
the older malloc call was left behind.
Also, std::initializer_list has const iterators returning const data
and broke the build in a few places where const qualifiers were
missing.
Change-Id: I3c04e58361989aa7438621cda63c7df457d7dad8
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move the iterator classes into QArrayTypedData and add constBegin()
and constEnd() to that class.
I also had to add an operator T*() to the strict iterators, since
there are many places that expect the iterator to behave like a
pointer (including in QVector itself).
Change-Id: Icc5ed56ad47b013664a48eef9d31b5273aecb4e3
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@nokia.com>
|
|
|
|
|
|
|
| |
even if QT_NO_CAST_FROM_ASCII is defined.
Change-Id: I8c4deceedb6f3e3cd5bdf72d6e9d189c509c9ff3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
to QLatin1String's compare operators that takes const char *s or QByteArray.
Such comparison leads to a potential misuse since QByteArray could contain any arbitrary data
in any arbitrary encoding and QLatin1String is used to only contain strings in UTF-8 -
they are just a different beasts aimed for different purposes, and since QT_NO_CAST_*_ASCII
disallow indirect conversions and require the user to know what he's doing,
let's be consistent here too.
Change-Id: I9bf5f326495157db8a6af064d6154961b7861a7e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
| |
Change-Id: I39b849721f3ba790c4a9197d69ac48e98cc2f5bd
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
|
|
|
|
|
|
|
| |
It was depending on internals of Q_ASSERT
Change-Id: I3dfc0ae0438135a30961f36808dbfc3e663a5538
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
| |
Previously, the append functions in QConcatenable in the QStringBuilder
dereferenced the data() pointer of the argument QLatin1String without
performing null check.
Change-Id: I629f19fbce3113f1f80f4272fa7ae34e1dbc6bee
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
|
|
|
|
|
|
|
|
|
|
| |
The mysterious comment was just suggesting that they weren't
necessary. Turns out the comment is wrong: they are necessary for
non-strict iterator modes.
Task-number: QTBUG-25063
Change-Id: I20ada17035642ee656c431d6bf2152a5243cecdb
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The QT_ALWAYS_HAVE_xxx macros are gone: they were hard to use and not
defined properly. It indicated that the compiler was producing code
that required that particular instruction set, so we could use it in
our code unconditionally. Instead, let's use the GCC-style __SSE2__
and __ARM_NEON__. MSVC does not generate the __SSE2__ macro, so let's
do it for the compiler. Also, define __AVX__ and the macros for the
technologies leading to it when we manage to detect an /arch:AVX build
(currently not possible, see note in the header).
ICC and MSVC allow one to use the intrinsics anywhere, but for Qt all
uses of the intrinsics are either in specially-built files, protected
by runtime checks, or they are unconditional (qstring.cpp). So we only
use the intrinsics when the compiler was instructed to generate code
for that instruction set anyway.
Change-Id: If8382f30422cee0e5831d051b003acf036824abf
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The QT_HAVE_xxx macros are replaced with QT_COMPILER_SUPPORTS_xxx.
They indicate that the compiler supports those intrinsics, but not
necessarily that they can be used right now.
ICC and MSVC allow one to use the intrinsics anywhere, but for Qt all
uses of the intrinsics are either in specially-built files, protected
by runtime checks, or they are unconditional (qstring.cpp). So we only
use the intrinsics when the compiler was instructed to generate code
for that instruction set anyway.
Change-Id: Ie58eebbc0518ad1d5420a85174fd84153bb9abaa
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
|
|
|
|
| |
Change-Id: I109f46892aed2f6024459812d24922b12358814d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
| |
e.g. for the text "Aaa bbb ccc.\r\nDdd eee fff." the \r\n wasn't treated
as a hard line break or as a line break opportunity, ever.
Quite ancient bug...
Change-Id: I8d8497c55a3a4d51c27de99ccfe1e31f3bf4de77
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add QUnicodeTools namespace and rename qGetCharAttributes to initCharAttributes;
Make it possible to disable tailoring globally by overriding
qt_initcharattributes_default_algorithm_only value
(useful for i.e. running the specification conformance tests);
This is mostly a preparation step for the upcoming patches.
Change-Id: I783879fd17b63b52d7983e25dad5b820f0515e7f
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
| |
qiterator.h doesn't use std::bidirectional_iterator_tag or
std::random_access_iterator_tag, so remove the forward declarations.
(These actually confuse clang when building with C++11 support enabled).
Change-Id: Idd0daa0840a0995e6e5b0ffc01ddcdf06f048149
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
| |
Change-Id: I0de764b51a972de0b6eb2bf3c04d2b190f581f52
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
|
|
|
|
|
|
|
| |
QTBUG-25438
Change-Id: I8cf9bfb295195548b6f7d4410682e4d675181a65
Reviewed-by: Denis Dzyubenko <denis.dzyubenko@nokia.com>
|
|
|
|
|
| |
Change-Id: I05e0f866c632f2a7e966e6bae9c73eeb77d99217
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
| |
in order of consistency with QChar
Change-Id: I8a7cf8960eb64ef177113d4569f1c49ae31c828e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
| |
Task-Id: QTBUG-24502
Change-Id: I360dee4dc68c165de0631ce4cf34e76fd873080e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
| |
This is needed for the change that follows
Change-Id: I05611defe422fa4bbb5be27b102e39b1f61a1cbc
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
| |
QString::from{Ascii|Latin1|Utf8|Local8Bit} does the string length calculation for us,
so let's use that and don't repeat the copy-paste bugs like the previous commit has fixed.
Change-Id: If0bced3ebaf75b56dde6be1266c47c3fbf89dab0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Unify all it's overloads into a single private helper functionand use this new helper
where possible - so we could optimize all those operators in one step some later
(this also fixes `QBytArray("a\0b") < "a"` didn't respect the \0 while operator==
handles nul(s) correctly);
Add operators <,>,<=,>=(const char*) to QStringRef so that they doesn't create a temporary QString object;
Add missing QT_NO_CAST_FROM_ASCII
Change-Id: I8b6562a92fdb96e67aadee181f23f823d206f5fd
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|
|
|
|
|
| |
Change-Id: Ic74e8e2471e92aa2014735f6ab0bb4f3b88de206
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Right now, valgrind reports that there is some reachable memory in
QLibrary because we don't call unload() in the libraries we loaded. So
do unload() them.
Unfortunately, ICU seems to have some global statics it doesn't free. If
we really unload the libraries, valgrind will report a leak. So use the
PreventUnloadHint, which causes libdl to not actually unload the
libraries.
Change-Id: I273f09627e27b9116366ddc427e1f3f53ea0f61a
Reviewed-by: Denis Dzyubenko <denis.dzyubenko@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
| |
This macro is now left defined from qglobal.h, so we don't need to
redefine it everywhere (and risk getting it wrong).
Change-Id: I2a11d10fe0434b85e79d0dda5f11fa90e2edc431
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Denis Dzyubenko <denis.dzyubenko@nokia.com>
|
|
|
|
|
|
|
|
|
| |
The new randomization of QHash is enabled by default. There may be cases
where you need deterministic behavior, e.g. for debugging or regression
testing. This patch disables randomization if QT_HASH_SEED is defined.
Change-Id: Idfad55ea7aba830add0a36334f0f763c62fdce13
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Make them call exactly their Latin 1 counterparts.
For the QString functions that take a single char, also use fromAscii
directly.
Change-Id: I87645aba6ab9cde34c1df3cbc3a979fbd9e91f9d
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The fallback implementation of QStringLiteral did not (up to now)
enforce the need to use a literal. So it was possible to write:
const char *foo = "Hello";
QString s = QStringLiteral(foo);
Which would do the wrong thing and create s == "Hel" on 32-bit
platforms (sizeof(foo) == 4) or, wrose, s == "Hello\0XY" on 64-bit
platforms (sizeof(foo) == 8, X and Y are garbage).
This change enforces the need for a literal by producing errors on the
above cases, as well as when foo is a char array variable.
GCC:
error: expected ‘)’ before ‘foo’
Clang (abbreviated):
error: expected ')'
namespace X { QString x() { const char foo[42] = "Hello"; return QStringLiteral(foo); } }
^
note: to match this '('
^
ICC:
error: expected a ")"
namespace X { QString x() { const char foo[42] = "Hello"; return QStringLiteral(foo); } }
^
The first C++11 error currently is:
error: expected primary-expression before ‘enum’ (GCC)
error: expected a ")" (ICC)
Change-Id: I317173421dbd7404987601230456471c93b122ed
Reviewed-by: hjk <qthjk@ovi.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
|
|
|
|
|
| |
Change-Id: I2325bcab9bb80e5507f53887b282a859d0fdb58c
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The patch that broke this was assuming the signature of xgetbv was
_xgetbv(int). This lead to that there were no exact match for the
function resolver, thus the ambiguity.
Apparently, the signature of _xgetbv is _xgetbv(unsigned int).
Changing the static xgetbv to uint makes the match exact, thus no more
ambiguity.
Change-Id: I8db95e00a9fef264d7a1f84d02bb929db84e6e5a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Miikka Heikkinen <miikka.heikkinen@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Since its hard to detect if __cpuidex() is actually available at
compile time, we'll add a function overload that will be chosen if the
intrinsic __cpuidex() is not available.
Note that the QtXgetbvHack that was used for _xgetbv did not really
work (MS compiler will bail out because of ambiguous overloads if the
intrinsic _xgetbv existed).
Therefore, we apply the same workaround for _xgetbv.
Change-Id: Iee3bf8bc6352ba0861b05d779f1f001d4eb013ff
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
|