This replaces QtAndroidPrivate::runOnAndroidThread{Sync} calls.
This also now allows passing std::function<> that can return values,
and not only an std::function<void()>.
This adds some tests for this calls as well.
Fixes: QTBUG-90501
Change-Id: I138d2aae64be17347f7ff712d8a86edb49ea8350
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
When adding an offset from UTC, arithmetic may overflow. Likewise
when combining a date and time (that have been offset for UTC). Also
check the return from epochMSecsToLocalTime(), as it can fail; and pay
attention to the status stored by setDateTime(), to notice when it
hits an overflow. Fixed some tests that only passed because we
neglected these checks. Extended a test to check we detect overflow in
a couple of cases close to the extremes.
Change-Id: I127a670302f94a07bb9b087b1b9c608b7c08785c
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Øystein Heskestad <oystein.heskestad@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QDateTime has long followed a convention of ignoring what it knows
about time-zone transitions before the epoch. This produces unhelpful
artefacts (such as an ahistorical spring-forward skipping the first
hour of 1970 in Europe/London, which was in permanent DST at the time)
and complicates the code. It documented that DST transitions were
ignored, but in fact ignored all transitions prior to 1970 and simply
assumed that the current time-zone properties (half a century later)
applied to all times before 1970.
This appears to be based on the fact that the MS APIs using time_t all
limit their range to after 1970. Given that we have to resort to
"other means" to deal with times after the end of time_t, when it's
only 32-bit (and after year 3000, on MS systems), we have the means in
place to handle times outside the range supported by the system APIs,
so have no need to mimic this restriction. (Those means are not as
robust as we might want, but they are less bad than assuming that the
present zone properites were always in effect prior to 1970.) On
macOS, the time_t functions only reach back to the start of 1900; it
reaches to the end of its time_t range and Linux covers the whole
range. Given this variety, the range is now auto-detected the first
time it is needed (based on some quick and dirty heuristics).
Various CET-specific tests now need adjustments in tests of times
before the introduction of time-zones (when they are in fact on LMT,
not CET). The systemZone() test of QTimeZone can now restore its
pre-zone test cases. Various comments on tests needed updates.
[ChangeLog][QtCore][QDateTime] Available time-zone information is now
used to its full extent, where previously QDateTime used LocalTime's
current standard time for all dates before 1970. Where we have
time-zone information, it is considered reliable, so we use it. This
changes the "best efforts" used for times outside the range supported
by the system APIs, in most cases giving less misleading results.
Fixes: QTBUG-80421
Change-Id: I7b1df7622dd9be244b0238ed9c08845fb5b32215
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Output object name and class in QCOMPARE(). This should help
to debug flaky QWidget tests that for example check on focusWidget().
[ChangeLog][QtTestLib] QCOMPARE() now reports QObject * values by class and objectName().
Task-number: QTBUG-64446
Change-Id: Ife04e89bba04fc78d077c8f0f07af17a17c9cf8c
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
There were two problems here: first, qHash(std::pair) must be declared
before qHashMulti that might call back to qHash(std::pair) (i.e., a pair
with one element that is also a pair). But moving the declaration above
causes the second problem: the noexcept expression can't refer to qHash
functions that aren't declared yet. So we forward-declare a constexpr
function for that result, but implement it far below.
Fixes: QTBUG-92910
Change-Id: Ia8e48103a54446509e3bfffd16767ed2e29b026c
Reviewed-by: Christian Kandeler <christian.kandeler@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Commit 37e0953613 added a to-do, but we
can actually change the type, since we've documented since Qt 5.10 that
setting a non-zero value (aside from -1) with qSetGlobalQHashSeed was
not allowed. Storing a value to be reset later is simply not supported.
Change-Id: Id2983978ad544ff79911fffd1671f7b5de284bab
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
The check is over-arching. is_move_constructible is sufficient;
we don't have to support "ridiculous" types that are copiable but
have deleted move operations (such types are fundamentally broken).
This is in line with the Move* (legacy) named requirements.
Change-Id: Idc7116b39013501b9be39628a4e7afd35fe15530
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
There was previously no way to compare QByteArrayView to with another
QByteArrayView case-insensitively without allocating memory.
[ChangeLog][QtCore][QByteArrayView] Added compare(), enabling case
sensitive and insensitive comparison with other QByteArrayViews.
Change-Id: I7582cc414563ddbde26da35a568421edcc649f93
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
After commit 3966b571 the function was kinda broken already, though
this got unnoticed since it was not covered by an the auto-test.
This commit adds another test case with Windows native separators
and removes the use of QDir::fromNativeSeparators. Instead use the
original code from QDir::fromNativeSeparators to replace the backslashes.
Pick-to: 5.15 6.0 6.1
Change-Id: I190560d0e75cb8c177d63b142aa4be5b01498da2
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
The BASE argument of qt_add_resources now denotes the root point of the
alias of the file. Before, BASE was merely prepended to every file that
got passed to qt_add_resources.
Old behavior:
qt_add_resources(app "images"
PREFIX "/"
BASE "../shared"
FILES "images/button.png")
Alias is "../shared/images/button.png", and pro2cmake generated
QT_RESOURCE_ALIAS assignments to fix this.
New behavior:
qt_add_resources(app "images"
PREFIX "/"
BASE "../shared"
FILES "../shared/images/button.png")
The alias is "images/button.png". No extra QT_RESOURCE_ALIAS assignment
is needed.
The new behavior is in effect for user projects and for Qt repositories
that define QT_USE_FIXED_QT_ADD_RESOURCE_BASE. Qt repositories will be
ported one by one to this new behavior. Then the old code path can be
removed.
Pick-to: 6.1
Task-number: QTBUG-86726
Change-Id: Ib895edd4df8e97b54badadd9a1c34408beff131f
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
QItemSelectionModel has the property selectedIndexes with
the notification signal selectionChanged. When a row is deleted
or inserted above the current selection, the row number of the
current selection changes and thus the return value of
selectedIndexes changes. This should trigger its notification
signal.
This signal was not emitted. This patch fixes this and
adds a unit test to verify this.
[ChangeLog][Important Behavior Changes][QtCore]
QItemSelectionModel now emits the selectionChanged signal
if only the indexes of the selected items change.
Fixes: QTBUG-93305
Change-Id: Ia5fb5ca32d658c9c0e1d7093c57cc08a966b9402
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: David Faure <david.faure@kdab.com>
"Cherry-pick" of C++2b's std::to_underlying.
[ChangeLog][QtCore][QtGlobal] The qToUnderlying function has been
added, to convert an value of enumeration type to its underlying
value.
Change-Id: Ia46bd8e4496e55174171ac2f0799eacbcca02cf9
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Not just a compile-time one (that the macros compile and don't
raise warnings).
Change-Id: I5642bf242a6c26a33730708f3c1710237fc107a2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
There's no reason to be storing `int` in the array data header and
then using it as a QFlags. Just store the QFlags.
Change-Id: I78f489550d74d15a560dacf338110d80a7ddfdd2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QFlags was lacking a named function for testing whether a QFlags object
contains _any_ of the bits set by a given enumerator/other QFlags.
Drive-by, add testFlags taking a QFlags, not just an object of the
enumeration; and simplify the implementation of testFlags to more
closely follow what its documentation says it does.
[ChangeLog][QtCore][QFlags] The testFlags, testAnyFlag and
testAnyFlags functions have been added.
Change-Id: Ie8688c8b0dd393d34d32bc7786cdcee3eba24d1c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
While at it, move tst_android under corelib/platform/android.
Change-Id: Icf91cd75cb5e04d03fe6a81d52fba52a485ca41f
Reviewed-by: Ville Voutilainen <ville.voutilainen@qt.io>
Since we changed binding evaluation to be always eager, we notify and
evaluate all bindings as soon as any dependency changes. This includes
bindings which have been initially installed on a property, but which
were later removed.
With lazy evaluation, we would only notify those bindings and mark them
as dirty, which is unproblematic. With eager evalution, we attempt to
evaluate the binding, though. While that part is still fine, afterwards
we would attempt to write the new value into the property. However,
there is no property at that point, as the binding is not installed.
Instead of adding a check whether the propertydataptr is null, we skip
the reevaluation completely by removing the bindings observers - and
thus the cause for the binding function's reevaluation. As soon as the
binding is set, we reevaluate the function anyway, at which point we
also capture the observers again.
Task-number: QTBUG-89505
Change-Id: Ie1885ccd8be519fb96f6fde658275810b54f445a
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
This removes traces of QPropertyAlias which is internal API which is
a) not really working even before this change (no compatibility with
QBindableInterface due to QPropertyAlias not being derived from
QUntypedPropertyData)
b) not used anywhere
For BIC reasons, we need to keep some methods still around until Qt 7,
though.
Change-Id: I5bb4735a4c88cba275dc2cc6e29a46ca09622059
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
This takes care of the last two QSFPM properties. They were postponed
because of the tricky relation between them and some bugs in handling
the changes.
The bug was fixed in bcbbbdb2d6 and
now the behavior is well-determined: updating filter regexp does
trigger case sensitivity change and vice versa. However updating
only regexp pattern (via QString overload or setFilterFixedString
or setFilterWildcard) does not change case sensitivity.
Task-number: QTBUG-85520
Change-Id: Idc41cf817de9e6263ed65a80fa40fc8415c8c856
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
filterRegularExpression and filterCaseSensitivity
will be handled in a separate commit
Task-number: QTBUG-85520
Change-Id: I848c5c6cbe8642efa156f4f5d33467976bbf0351
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
The commit amends commit 3966b571 to take UNC prefix into account as
well. Fixes the weird file name output as reported in QTBUG-74291 and
QTBUG-83365. Replace manual separator normalizing in qt_cleanPath(),
this is another spot where UNC prefix handling needs to be applied.
Also make QTemporaryFile operate on '/' as file separators to fix
creating both file types with native path separators on network shares.
Fixes: QTBUG-74291
Fixes: QTBUG-76228
Fixes: QTBUG-83365
Pick-to: 5.15 6.0 6.1
Change-Id: Iff8d26b994bf4194c074cd5c996cda3934297fa5
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
[ChangeLog][QtCore][QMetaMethod] It is now possible to query the
constness of a method with QMetaMethod::isConst.
Change-Id: I8a94480b8074ef5b30555aeccd64937c4c6d97d4
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Reviewed-by: Ulf Hermann <ulf.hermann@qt.io>
The libs and symlinks are in the same directory, no need
to have absolute paths.
Change-Id: I22dab933b1f3bdf244b0953c6bb7caaeedef5697
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
CLDR v39 has no locales for Abkhazian, so the locale_index[] entry for
it actually indexes the last entry before the next language up the
enum. This has m_language_id less than Abkhazian.
Change-Id: If8b88f30476a981b3ee00ff8760a46ede0b7aab7
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
The calls were trying to pass a JNIEnv* from a QJniEnvironment using
conversion operator which was removed, and weren't detected since they
are templates and were missing tests. This fix that and add test cases
for setField() and setStaticField() calls.
Pick-to: 6.1
Change-Id: I6e1e6b7f557bbc664248ad364c48d63f58b70756
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The API allows users to request and check the status of various
permissions. A predefined enum class of the common permission types
on different platforms is used to allow requesting permission with
a common code. Platform specific permissions are defined only on their
relevant platform. For permissions that are not predefined, they can
be requested via a string variant of this API.
This adds the Android implementation only.
[ChangeLog][QtCore] Add new API for handling app permissions with an
initial implementation for Android.
Task-number: QTBUG-90498
Change-Id: I3bc98c6ab2dceeea3ee8edec20a332ed8f56ad4f
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The biggest trick here is the getter (QAbstractProxyModel::sourceModel),
which is returning nullptr, while internally using a global
staticEmptyModel() instance.
This lead to inconsistency while binding to a proxy model without
source model. The bound object would point to staticEmptyModel()
instance, while sourceModel() getter returns nullptr.
To solve this issue a custom QBindableInterface is implemented.
Task-number: QTBUG-85520
Change-Id: I597df891c7e425d51b55f50ccbacabdfe935cbac
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Previously, if multiplication overflowed when trying to set the date
and time of a formerly short-form QDateTime, its status didn't get set
to reflect the failed validity check. Added a test that now correctly
detects that it's produced an invalid date-time on overflow, where
previously it produced a wrong valid date-time.
Change-Id: Id46ca34d1e32e9b9b0630f3723cefd1c13b5761e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The QCOMPARE made dead code of a more verbose QVERIFY2 below
Change-Id: I26b8286f61534f88b649fffd166b67d8603280a7
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Instead of silently failing, we now print an explanatory warning to aid
with debugging.
Task-number: QTBUG-89512
Change-Id: I36dd2ce452af12d0523c19286919095e366bd390
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
This patch extends the QJniObject::callStatic[Object]Method functions
with the overload which accepts a jmethodID parameter. This can be
convenient when the method id is already cached and you do not want
to query the method by its name and signature.
Task-number: QTBUG-92952
Change-Id: Ib0852a5a27da2a244ac63112784751ef9e32cfa5
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
This patch adds some convenience methods to QJniEnvironment API:
* an overload of registerNativeMethods() that accepts jclass instead
of const char *className.
* a findMethod() function is added to query a methodID of a static
or nonstatic method by its name and signature.
Task-number: QTBUG-92952
Change-Id: Ib1bc892decea97e625c4822888b6183af6edd6dc
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
Reviewed-by: Alex Blasche <alexander.blasche@qt.io>
Skipped startValue/endValue properties, since they are computed and
writable, which is not supported at the moment.
Skipped currentValue, since its setter might be called inside the
getter, which is not recommended.
Task-number: QTBUG-85520
Change-Id: I1f872b4fcc7227ed91b6915891bbc66019151826
Reviewed-by: Ivan Solovev <ivan.solovev@qt.io>
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
This patch amends bcbbbdb2d6.
It fixes the logic of filterCaseSensitivityChanged signal emission.
The call to QRegularExpression overload of setFilterRegularExpression
could change the filterCaseSensitivity, but the signal was never
emitted.
[ChangeLog][QtCore][QSortFilterProxyModel] A call to QRegularExpression
overload of setFilterRegularExpression now emits a
filterCaseSensitivityChanged signal, if required.
Pick-to: 6.0
Pick-to: 6.1
Change-Id: Id4ef04227c1f8ed98153fa5107ec3fbe4c0c77fb
Reviewed-by: Samuel Gaist <samuel.gaist@idiap.ch>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Analysis of problems with the new test for qFuzzyIsNull() revealed
that, where its version for double uses approximately 4500 * epsilon
and for float used 84 * epsilon as threshold, the qfloat16 version's
value was barely more than epsilon, with the result that the test had
to use a different value than the threshold to pass. (Converting the
threshold from float to qfloat16 and back made it bigger; in effect,
the threshold value was not <= itself.)
Furthermore, comparison with qFuzzyCompare() implied a value of
1/102.5 should be used, roughly 10 * epsilon, for consistency. When
1/102.5 is rounded to three significant digits (the precision we use
in QTest::toString(), for example), to give 0.00976f as threshold, we
get a value that, after conversion to qfloat16 and back to float, does
give a result <= what we started with. So change qFuzzyIsNull() and
its test to use this as qfloat16's threshold value.
[ChangeLog][QtCore][QFloat16] The qfloat16 threshold value for
qFuzzyIsNull() has changed from 1e-3 to 9.76e-3, almost a factor of
ten increase, for consistency with qFuzzyCompare()'s tolerance. Values
between these would previously have had qFuzzyIsNull(f) false despite
qFuzzyCompre(f, 1+f) being true.
Change-Id: I35816dce78da34a3e2339c8bc42d5bd03714a3f6
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Reviewed-by: Andreas Buhr <andreas.buhr@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
They were missing.
I also wanted to verify that it's true for sub-normal values.
At the same time, relocate qfloat16's implementation of qFuzzyIsNull()
to between those of qFuzzyCompare() and qIsNull(), since its apparent
absence initially confused me.
Change-Id: I9637c0070e754d16744c76fc9f846596257c6a63
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This reverts commit ec8808c302 but
retains its test, as the problem it fixed is now solved by having the
TZ backend validate the ID it's passed, so that it now only accepts
valid POSIX zone-descriptions and valid IANA IDs. The former were
being excluded by this check.
Amended a POSIX test to fail with the check in place; it passes now.
Change-Id: I0d5e8c6e0a315ac2509f3d23bebb52aede8f79d0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Robert Löhning <robert.loehning@qt.io>
With C++20 standard, src/corelib/kernel/qproperty.h of Qt Base cannot be
compiled at line 100:
QPropertyBindingSourceLocation(
const std::experimental::source_location &cppLocation
)
The reason is that source_location has been merged into namespace std
since C++20, and the header file has also been change from
<experimental/source_location> to <source_location>.
The problem can be avoided by define a constant.
Fixes: QTBUG-93270
Change-Id: I46b4daac6ea20f9623b43746880500d41396afb2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Cast consistently to quint8, which is the type we are streaming out
in the test.
Change-Id: I44d360ca6b75f14e7a2b80962ad249a6f6b1cb31
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The comparison between nullptr and QWeakPointer was just bogus
and ill-formed. The INTEGRITY compiler catches that even if
nothing tries to use the comparison. It is an ill-formed, no
diagnostic required case of a function template never being
able to produce a valid specialization. And while we're at
it, this patch makes the result of comparing a nullptr to
a QWeakPointer or vice versa the same as asking .isNull() from
the weak pointer, because it seems mind-boggling if those
are not the same operation.
Task-number: QTBUG-93093
Change-Id: I0cc80e795c9af2be1b76de05157aa458ef260f2e
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This is one of the more complicated ports. The target object
was represented by two variables in the past: A raw pointer and a
QPointer. The QPointer is checked in some cases to check whether
the target object still exists.
This patch introduces a targetObjectDestroyed() slot and connects
it to the destroyed(QObject*) signal of the target object.
In this slot, the animation is stopped. The checks become
obsolete thereby and it is sufficient to represent the target
Object in one raw pointer.
This raw pointer becomes a bindable property.
Fixes: QTBUG-92992
Change-Id: I7e2ddb5d8aed007400fe74bea1becf7bdfbf2563
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Windows wants ISO C++ _fileno and _unlink instead of posix fileno and
unlink. For fileno we have the QT_FILENO macro in qplatformdefs.h, so use
it. For unlink we don't have a macro in Qt, so declare one in the test.
Change-Id: I56c5c3fb4e500769c744132c46107816f89bb2c3
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The property 'model' is ported to a bindable property.
The properties hasSelection, selection, selectedIndexes,
and currentIndex are left for later patches.
Task-number: QTBUG-85520
Change-Id: Ia424ce99fc80c3d807c634c21d161a3ad94b27d2
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
Previously, an apparent POSIX rule would be saved and any defects in
it would only be discovered when trying to use it to generate
transitions. Instead, check that it has the right form during the
initial parsing of its data.
In the process, since checking for DST in the process is trivial,
implement a long-standing TODO to cache hasDaylightTime()'s
answer. The array it scanned was in any case being scanned during
construction, so detecting DST in init()'s scan is trivial; and its
failure to check the POSIX rule mean it failed to notice when zones
entirely specified by a POSIX rule have DST.
Adapt a test using a POSIX-only rule to verify it does know the zone
has DST; it did not, before this change.
Change-Id: I690c013d3331600f7348dae61c35d41e5599da70
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The description necessarily has fake transitions at start and end of
the year, potentially outside the year. These transitions should not
be reported by QTzTimeZonePrivate as transitions, although its data()
must find a "transition" whose data it can use (as in the permanent
standard time case, which could potentially be represented the same
way, although there's a saner way to do so, that the code already
handles) to report the zone's properties.
In the process, fix (and make more straightforward) the convoluted
decision-making code that was deciding which transitions to include in
the returned list. It was assuming invalidMSecs() would be set as the
atMSecsSinceEpoch of a transition, although this is computed in a way
that makes that value most unlikely, even when the result is invalid.
It also rather confusingly mixed < 0 tests as tests for overflow with
the one < 0 test that's about ignoring DST before 1970. Also added
comments to clarify some of what's going on there.
Expanded a recently-added test of a permanent DST zone to verify this
now works correctly.
Change-Id: Ia8d98f433fb1e479dba5479220a62196c30f0244
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
