| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Change-Id: I3d1822457efd740e7c9d91747e978d787c0dd4b8
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Iacbc9cc7f89aeacbf1ce918eb28a1a5c7db7040d
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
When insertItem() was called for the first time as a result of
"statically" declared SwipeView children, the currentIndex would be
changed to 0. We should only do this if the currentIndex is invalid
(-1), otherwise we disrespect the user's settings.
Change-Id: I692b60581e706edcfa4939bbe59a815ff3e019e5
Task-number: QTBUG-48086
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
| |
When the font of a QQuickControl subclass is set, it will
propagate it to all children that are also derived from
QQuickControl, as well as TextField and TextArea.
Change-Id: I7c851f84b89609094d9a81d239ade0f0ac212985
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
| |
...like in all other tests.
Change-Id: Ia713226f535bca3b6ab731d9441bb5d1b7f9c6f2
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QTBUG-47317 revealed that dynamically created items remained visible
after popping, so they were not only kept visible but also weren't
properly destroyed. The visibility issue had to be fixed so that
"indestructible" (not owned by StackView) were hid.
This change tackles the destruction part, and marks pop-exit items
for removal so they get destroyed when transitions are finished.
Now that the transitionable items are actually destroyed, a new
problem was releaved. We must not delete transitionable items until
both enter & exit transitions have been finished. If we did that,
the finished() callback would not get called for the pop-exit
transition and tst_StackView::test_transitions() would fail (yay).
Change-Id: Ie2eedf99d0366820260776591ad1f8b5b68e2ec3
Task-number: QTBUG-47317
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
| |
Change-Id: I1ceb0d2bd269c66a5663dab258698e1eddbd7d25
Task-number: QTBUG-47963
Reviewed-by: Liang Qi <liang.qi@theqtcompany.com>
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
| |
Only items that were created from pushed components were initialized.
This change fixes item initialization for StackView::initialItem and
pushed items.
Change-Id: I6fe1afa233e63320180dfe8eac3874fed69cbb3b
Task-number: QTBUG-47318
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Passing null view bounds for the transitionable exit item helps to kick
off the transition as appropriate. Otherwise the view transition item
determines in QQuickItemViewTransitionableItem::prepareTransition()
that no transition is necessary.
Change-Id: Icb664725c4393ef3bb894b71a5912652f82c9da0
Task-number: QTBUG-47949
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a partial revert of 6ee9612. The idea was to keep items visible
until both transitions had finished.
StackView {
pushExit: null
pushEnter: Transition { ... }
}
When both transitions had finished, all stack elements (except the
current one) were hid. The problem was that in pop & replace operations
the exiting item was no longer part of the stack (because it had been
already popped out) and therefore was never hid.
This change restores the original (intended) behavior that a null
transition would equal to an immediate operation. This keeps the code
simpler, and makes the behavior consistent in a sense that all items
are hidden as soon the respective transition finishes. PauseAnimation
can be used to wait if desired.
Change-Id: Ia9e78506754bdc18949676f061dbfff28a7128b7
Task-number: QTBUG-47317
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
| |
Change-Id: I479704184ceff878a239c2ee72d0c0929859b97b
Task-number: QTBUG-47963
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
We just extend the wait interval. This is to make sure
the control's press-and-hold timer will timeout before
the wait. Since we're probably not using the most accurate
timers, adding 20% to something that's around one second
remains acceptable.
Change-Id: I1d10c88a168e882c9069f9345e896c1aab2f2bda
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I78d580c28287fecaf287c4634a166c44926c48fb
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
| |
This enables creation of non-wrapping Tumblers.
Change-Id: I0e21b860b84c456c0651923e87217cafc42c69b7
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
We factorize the feature code identical to TextField's
into a helper class.
Change-Id: I4496f2d192a023f6a6c03ed81d81bca4c08a59d0
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I94275e10681629d679228d6a086b3d6a3856acef
Reviewed-by: Topi Reiniƶ <topi.reinio@digia.com>
|
|
|
|
|
|
|
|
|
| |
The default style uses QQuickItem's transform property, so filling
the handle with a Rectangle, for instance, will not show it in the
correct position, even though it (the handle) is.
Change-Id: I85af897d1bcc5e7f343cb9814da6d896088720b8
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
| |
Change-Id: I9c4801ac91c52d487e98a99ff4d38e133f0679a3
Reviewed-by: Liang Qi <liang.qi@theqtcompany.com>
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I634da6568a456a5f21e30e6295d139a4e1a89e75
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
These were copied from Qt Quick Controls.
There are some issues with regards to QQuickThemeData, so some are
commented out for now (see tests for details).
I also added a test for QTBUG-47318.
Change-Id: Iaee0f4531b8d7bc9b4984146aaf00b76afa77793
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Ie80fac70141ef72bb68f7c9e91827e834dc1fc4a
Reviewed-by: Mitch Curtis <mitch.curtis@theqtcompany.com>
|
|
|
|
|
|
|
| |
Not so much to test there, but test that at least creation succeeds.
Change-Id: I81aa3e43602b511d12ca9bb017e8df7452f9eaf3
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I42f5a2f8d2c2f59d64bd2fcff4433de3fc5b9ad8
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Iaa62f9ba650ff4f60e6370a770fb0af2645eafb5
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I94d079ad0789215768e43652075ccf6ad7fbd48d
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
|
|
|
|
|
| |
Change-Id: I6fb8e1b238a087f8775df87e63ffa515b626226d
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
Better start with a minimal set. It will be also easier to transition
from Theme to Palette if it gets done.
Change-Id: Ic9951ac5b913f99a84e2b8efb9ae7f8d61177aee
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
|
|
|
|
|
|
|
| |
Use Theme.disabledColor instead.
Change-Id: I76311d6c6b573b01a26ee2b8d434a841c4d86292
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add support for toggled() and toggled(bool) signals to bring
ExclusiveGroup 2.0 on par with ExclusiveGroup 1.0. Furthermore,
the attached property is now "ExclusiveGroup.group" to be consistent
with all other attached properties like StackView.view, TabView.view,
and others. It looks clumsy to repeat, but it's better to keep the
type names to the minimum, and the documentation also stays in one
place.
Change-Id: Ic46f8d2ec4b58c5056e8b2bbca5c9f88968bfb80
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Ie2e5e09e906b2886525a986bccbd1b2722a0a795
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
| |
- always verify component creation to catch errors early
- remove empty init() and cleanup() functions
- remove redundant "defaults" checks
Change-Id: I39f16404fa2e3d3e38348742afd8f51ea36b2f05
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is going to include all snippets used in the documentation,
which will be then automatically tested for vality, and at the
same time the respective screenshots are automatically generated.
Notice that copying the screenshots over to src/doc/images is still
a manual step. This is fully intentional because the screenshot
results may vary quite a bit across different platforms due to
different fonts and so on. FWIW, my personal favorite for generating
the screenshots is OS X.
Change-Id: I72b299f15b498abc1afd0a686c56d43159bee775
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Idf7ca1850b41d882d039092e7d900a144faeb0c8
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I79a974c64f6ec57ebfa83cea0857b2a1590af61f
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
| |
TextField and TextArea have different semantics for contentWidth
and contentHeight. In addition, it looks like we need to restore
contentWidth and contentHeight for Frame & GroupBox.
This reverts commit 6941b934bdf5f3c7b65f682c8dfc5c6562292351.
Change-Id: Ia115025a67b472ac67fbbc915b80312e4b4a7431
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I61d077758c83027cbfc1b99d32cb68efc0829ffd
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Ia39a52e7e8d968b05fd7a2d64ab3b3030a9f528e
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I8f883dd0fd06172032dd78ff60437c906c2fe8dd
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Rename the former tst_declarative to tst_sanity and
extend it with tests to prevent:
a) use of anchors, and
b) multiple attached object instances of the same type.
Anchors are not allowed for two reasons:
- performance (QQuickAnchors is a QObject), and
- to let users relayout delegates (x/y/width/height bindings
can be overridden, anchors not so easily)
Multiple attached object instances (eg. Theme) can easily happen by
accident. "Theme.fooColor" can be used in the control root, but in all
delegate items, "control.Theme.fooColor" must be used instead.
Change-Id: I4045d5bd717fa21db79d1c3bd618fc450e292fa4
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Id0d3489c0c047c837bf2aab5bafcd75f9db3b966
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Iee61e89d8519e7e3d27f1bf1c35af210f48bc7c9
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
|
| |
Not allowed:
- .js files
- signal handlers
- function declarations
Change-Id: I31247f154b96718ca95835120c9e7326e3c97968
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I1bf29fab227f1f2642994f9167c9a6957abcbb65
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
|
| |
The recently introduced contentWidth/Height was ambiguous. It had even
different semantics than in Containers, TextXxx & Flickable etc. Thus,
replace with availableWidth/Height.
Change-Id: Iaeef25ec82b395c602724d222ca73666460ffd29
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
- width follows only if x and width are both not specified
- height follows only if y and height are both not specified
Change-Id: Ib881daa9da339baac53f7f9d4de6e5bdad7723e2
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
|
|
|
| |
It was emitted only when LayoutMirroring.enabled changed, but
it also needs to be emitted when layoutDirection changes...
Change-Id: I9424fdf232c11669780c9abace5c162a0df69dbe
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: Icbb39aa1c3ac6f4b00adbc59c09a6c79e99bafd3
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: I44df09c17d6d20cfd929af4e331e15f48b19587e
Reviewed-by: J-P Nurmi <jpnurmi@theqtcompany.com>
|
|
|
|
|
| |
Change-Id: If10ef6f111b4a91216bbb2d0b52bc182533edaa8
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This API is cleaner than having to connect to each signal
handler namually.
We solve the 'pressed' signal problem by connecting to the
other signal accessor, namely 'onPressed'. And we do so for
all signals.
We also lay foundation to capture the signal parameters as
each signal is emitted. What we may do with those values is
not clear yet.
Auto-tests updated accordingly.
Change-Id: Ieda96a43072ca7dff2b183e07f6f67aa49c8cf5d
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
|