If the visible roles of the details-view are equal to the visible roles
of other views, then switching to the details-view will trigger an
assert because the invisible roles don't get updated.
Thanks to Frank Reininghaus for the detailed analyses!
In the context of this fix optimizations have been done when switching
view-modes: The "don't-animate-workaround" could be removed.
BUG: 294531
FIXED-IN: 4.8.1
Up to now KItemListView has not been aware to which column or row
an item belongs to. This has been handled internally in
KItemListViewLayouter. But for deciding whether a move-animation
might result in overlapping items it is mandatory to be aware about
the column/row. The current approach to guess a column/row change
by comparing the source- and target-rectangle works very well but
breaks on some corner-cases when e.g. zooming.
Now the layouter allows to access the information. This assures
that in the vertical alignment no move-animation is done between
rows and the in the horizontal alignment no move-animation is
done between the columns.
As no rubberband-selection is enabled for the Folders Panel it does
not make sense to keep the selection region as small as possible.
BUG: 294111
FIXED-IN: 4.8.1
When the filter has been enabled, the sibblings got not updated. Explicitely
update the sibblings when inserting, removing or moving items. For layouting
implicitely update the sibblings only for recycled widgets.
Up to now no indicator-branches have been drawn when showing a tree. The patch
fixes this so that that the style-dependent branches are drawn.
The main part of the patch is the implementation of
KItemListView::updateSiblingsInformation(). Most of the other changes are
related due to an internal renaming of the expansionsLevel-role to
expandedParentsCount and some related cleanups.
BUG: 290276
FIXED-IN: 4.8.1
The ViewState struct gets a new bool member to indicate if the selected
items should have been activated after the last event or not. One could
argue that adding a bool parameter to the constructor is not optimal,
but I think adding an enum is not really worth the effort because the
struct is not used outside this unit test. Moreover, I could not think
of a good name for the enum type and its values.
(cherry picked from commit 307cca7b31)
Since commit 66ce4db487, enabling or
disabling groupling may change the number of columns in the view.
Therefore, the test should first change the "grouping" setting and then
change the view geometry to make sure that the column count is correct.
(cherry picked from commit 61390b201a)
- Don't use a margin for the first group header
- Provide margins between the groups
- Use a more subtle coloring
- Prevent animation glitches when resizing a window
- No padding should be applied to the icon-rectangle
- Base the selection-toggle size on the icon-size of the style not
the actual pixmap content, as this might lead to smaller selection-toggles
when having pixmap-previews with a small height.
- Differ internally between margins and paddings
- Add a small padding in the icons-view and the compact-views
- Optimize the width of the icons-view if previews are shown
- Use a simpler background and colors as suggested by Martin Zilz.
This is just an early draft, I need to post some comparison screenshots to
Martin for review.
- Fixes of some layout issues that have been revealed because of adjusting
the group header heights.
- More clever animation/no-animation detection when doing listview-property
changes in parallel.
Usecase:
- No scrollbar is shown
- Window size will be decreased so that a scrollbar gets necessary
The decreased window size minus the space required for the scrollbar
results in a relayout of the items. In 99 % of all
cases a decreased window size won't result in showing more items in parallel in
comparison to a larger window size. However in the remaining 1 % this can
happen (e.g. see bug 293318 for a sample). This results in an endless loop as
now no scrollbar is required anymore, the layout changes again, the scrollbar
is required again, ...
BUG: 293318
FIXED-IN: 4.8.1
The use case of this function (Folders Panel) requires the expansion of
the parent items of a single URL, so it's not needed to handle a full
set of URLs in this function. Moreover, the issue that not only the
parents, but also the URLs themselves were expanded is fixed by this
commit.
(cherry picked from commit 89082ca391)
- Assure that landscape-previews use the whole available width of the icon-area
instead of only using the square width. This waste of space occured if the
text-width was larger than the icon-width.
- Only use one margin instead of two between the icons and the text to improve the
visual appearance (thanks to Martin Zilz for the hint)
BUG: 293268
FIXED-IN: 4.8.1
CCMAIL: martin@kreativkonzentrat.de
During determining the item-count for directories just show an empty string
until either the item-count has been calculated or if the item-count is unknown.
Thanks to Nikita Skovoroda for the initial proof-of-concept patch.
BUG: 291823
FIXED-IN: 4.8.1
When counting the number of sub-directories consider unknown file types
as potential directories. In the "worst case" an expansion toggle will be
shown although the directory contains only files, however from a performance
point of view this seems to be an acceptable compromise.
BUG: 292642
FIXED-IN: 4.8.1
At the moment, only key press events are tested, and the current item
and selection after the event are verified.
Moreover, this commit makes sure that
KItemListController::keyPressEvent() really does not select anything
if the selection mode is NoSelection.
(cherry picked from commit 7457f4868c)
At the moment, only key press events are tested, and the current item
and selection after the event are verified.
Moreover, this commit makes sure that
KItemListController::keyPressEvent() really does not select anything
if the selection mode is NoSelection.
The asynchronous resolving to bypass performance bottlenecks is not necessary
anymore as multiple ranges can be inserted in one step now. This solves the issue
that e.g. opening a tree resulted in temporary unknown icons for a short period
of time.
The asynchronous resolving to bypass performance bottlenecks is not necessary
anymore as multiple ranges can be inserted in one step now. This solves the issue
that e.g. opening a tree resulted in temporary unknown icons for a short period
of time.
Having animated items is useful when e.g. deleting or inserting one
item or when the size or position slightly changes by increasing the
zoom level or the window size. However if the number of rows or columns
change, the animation gets obtrusive and in this case lets just apply
the final layout.
Having animated items is useful when e.g. deleting or inserting one
item or when the size or position slightly changes by increasing the
zoom level or the window size. However if the number of rows or columns
change, the animation gets obtrusive and in this case lets just apply
the final layout.
The problem was that m_keyboardAnchorIndex and m_keyboardAnchorPos were
not updated when Home or End are pressed. This causes the following
unexpected behavior in Icons View: Press Home, then Right, then Down,
then Home, finally Down. One would expect that the first item in the
second row is the current item then, but before this commit, it was the
second one because the controller still remembered that the second
column was active before Home was pressed.
(cherry picked from commit 999234a94a)
The problem was that m_keyboardAnchorIndex and m_keyboardAnchorPos were
not updated when Home or End are pressed. This causes the following
unexpected behavior in Icons View: Press Home, then Right, then Down,
then Home, finally Down. One would expect that the first item in the
second row is the current item then, but before this commit, it was the
second one because the controller still remembered that the second
column was active before Home was pressed.
FIXED-IN: 4.8.1
