The icons of items which are "cut" are shown faded in the view. The
"is cut" state of the widget representing an item is updated if the
clipboard contents change. Before this commit, if the sort order was
changed however, then each widget kept the "is cut" state of the item
which had been shown previously at its position.
The solution is to update the "is cut" state also if any of the
widget's data change.
KDirectoryContentsCounter: do not delete currently active worker objects
Before this patch, the destructor of KDirectoryContentsCounter might
delete the worker object, which lives in another thread, while one of
its methods was still being executed. This could cause a crash. Only if
the destroyed KDirectoryContentsCounter was the last one, the worker
thread was stopped, and the destructor waited until all workers are
done.
When you open a new tab while the search mode is enabled, the
newly opened tab also starts the same search (Because new tab is opened with
the current view url), but the search box is in read-only mode. So you cannot
close the search bar nor edit the search text.
This patch fixes this by parsing the search url. The value of the "search"
parameter is used as search text and the value of the "url" parameter is used
for the search path ("root" folder for the search when "Search from here" mode
is enabled).
In case of Baloo search urls, we use Baloo::Query::fromSearchUrl.
Removed everything related to read only mode in DolphinSearchBox, not needed
anymore.
Vishesh Handa [Mon, 24 Mar 2014 09:48:53 +0000 (10:48 +0100)]
Dolphin Facet Widget: Convert type selection into radio buttons
They were previously checkboxes. Most files cannot have more than 1 type
considering that the exposed types were "Document", "Image", "Video" and
"Audio".
Also, it is not very simple in baloo to search through ORs for types.
Not through the exposed API anyway.
Vishesh Handa [Wed, 19 Mar 2014 14:44:35 +0000 (15:44 +0100)]
Dolphin Facet Widgets: Implement date based filtering
This required changes in Baloo which require reindexing. Since the
storage location has changed for Beta 3 everyone should be re-indexing
everything, so it won't be a problem.
Use MIME type inheritance to find out if a file contains text
Before this commit, filenamesearch checked if a file's MIME type begins
with "text/" to determine if its content can be searched. This prevented
that text inside shell scripts is found.
Use QMutableHashIterator for deleting items from a QHash
KItemListViewAnimation::slotFinished() used a QHashIterator to iterate
over a QHash, and then removes an item from the hash using
QHash::remove() inside the loop.
This is quite unusual - the recommended way is to use a
QMutableHashIterator (or std-style iterators and then QHash::erase(it)).
This might be related to the cause of a crash in this function.
Always go back/forward in history when pressing the respective buttons
Before this patch, pressing one of these buttons while an item is
hovered selected this item. The motivation for this behavior was to
provide a fast way to select items. However, this was counter-intuitive
and confusing for many users.
Make the handling of the "maximum text lines" setting more robust
If the user sets a maximum number of text lines in the settings, this
number was translated into a maximum height in pixels using
QFontMetrics::lineSpacing() before this commit.
In KStandardItemListWidgetInformant::itemSizeHint(), this maximum height
limited the size that is reserved for the item.
However, in KStandardItemListWidget::updateIconsLayoutTextCache(), the
maximum height was translated back into a maximum number of lines,
which limits the number of lines that are created using the QTextLayout.
This approach could lead to problems if the real height of the layouted
text is 1 pixel more or less than QFontMetrics::lineSpacing() times
"number of lines".
Now we do not store a "maximum height" inside the "maximum size"
explicitly, but store a maximum number of lines and a maximum with (for
Compact View) separately, and then use the number of lines also to
calculate the required size in
KStandardItemListWidgetInformant::itemSizeHint(). This should make sure
that the correct height is reserved for each item.
Thanks to Christoph Feck and Emmanuel Pescosta for helping to debug this
problem and testing the patch.
Ensure that KItemListViewLayouter always has a size hint resolver
KItemListViewLayouter uses a KItemListSizeHintResolver to find out how
much space the items will need in the view.
Before this commit, the size hint resolver object could be changed at
runtime, and it could also be null. However, we never made use of these
possibilities, so all the code that checks if m_sizeHintResolver is
null is actually not needed at all.
Only initialize the hash m_items in KFileItemModel if it is needed
Moreover, clear the entire hash if items are added or removed.
This saves time and memory when loading a directory, and it fixes
problems that might occur if the model is in an inconsistent state, such
as crashes that can happen when we try to remove individual items from
m_items.
Restore the URLs of both views correctly when restoring a session
The problem was that we restored the URL of the right view while the
left one is still active. When we received the signal
urlChanged(KUrl& url) from the right URL navigator, we then set the URL
of the active (i.e., left) view to 'url', such that both views showed
the same URL.
Always enable the "Create New..." menu if the URL is writable
This commit works around the problem that KDirLister may not provide a
"rootItem" for some kioslaves by setting up a KFileItem with the view
URL and using this to find out if the URL is writable.
Use only one "directory contents counting" thread per process
The pointer to QThread object is stored in a global variable, and each
view increments/decrements a reference count when it starts/stops using
the thread. If this thread reaches zero, the thread is stopped.
Note that we cannot just use a smart pointer, like QSharedPointer, to
manage the thread, because we must make sure that the thread is not
running any more before the QThread is deleted.
Remove the "retrieved items" code from UpdateItemStatesThread and VersionControlObserver.
Showing an error message, makes no sense in this case - the user can see it when all items are "unversioned".
The plugins still have the ability to show error/warning messages on real errors. (and only where it makes sense ;)
The speed up is really small, but theses changes are mostly straightforward and make the code a bit nicer - break
the KStandardItemListWidgetInformant::itemSizeHint beast into three smaller functions.
Kill any running preview jobs before starting a new one
If loading a preview takes long, it was possible before this commit
that a preview for a new item was requested before the first preview
was shown. In that case, there was a race condition, and the first
preview to arrive stayed in the Information Panel.
This commit fixes this by keeping a pointer to the preview job and
killing it before starting a new one.
Avoid calling KFileItemModel::index() in KFileItemModelRolesUpdater
KFileItemModel allows to find out the index of a KFileItem with its
index(const KFileItem&) method. Calling this method is not extremely
expensive, but it's also not free (it looks up the URL of the KFileItem
in a QHash, i.e., it has to call qHash(QString) for the full URL).
In KFileItemModelRolesUpdater, we sometimes converted the known index to
a KFileItem and then back to an index in applyResolvedRoles(KFileItem).
This patch fixes this by modifying applyResolvedRoles such that it takes
the index directly as its argument.
Remove redundant data from KItemListViewLayouter's ItemInfo struct
It is not necessary to save the position of each item as a QPointF
because all items in a row will have the same y-coordinate, and all
items in a column will have the same x-coordinate. Therefore, we can
reduce the number of doubles that we store from
Fix crash when dragging children of expanded folders in Details View
The problem was that the loop in KFileItemModel::createMimeData() which
is supposed to find out if any parent of an item has been added to the
QMimeData already (because it is not necessary to add the item in that
case) did not loop through the parents of the item, but incorrectly
replaced the local variable 'itemData' by its parent.
Add unit test for KFileItemModel::createMimeData().
The test verifies that creating the mime data for a child of an expanced
folder does not cause a crash. The regression happenened in the master
branch, but it doesn't hurt to have the test also in KDE/4.12.
Since Dolphin 4.11, we store not only KFileItems, but also the
corresponding ItemData struct for filtered items. This is required for
keeping track of the parent-child relationships, and has the nice side
effect that the ItemData need not be re-determined when the items are
shown again.
However, this can become a problem if the visible roles or the sort role
change while some items are filtered.
This is fixed by is fixed by clearing the QHash "values" for the
filtered items if the visible roles change. The hash will be
re-populated with all requested data as soon as the items are shown
again and the data(int) method of the model is called.
Moreover, before the items are inserted into the model after filtering,
we have to make sure that the sort role "Permissions"/"User"/etc. is
present in the hash "values". This is achieved by factoring out the code
that currently does this job for new items in createItemDataList() into
a new function, and calling this in insertItems(), because the same
treatment is required for the previously filtered files.
Moreover, this commit ensures that the order of the URLs in the
QMimeData object is the same as the order of the items in the view.
Selecting many items and copying them to the clipboard could take quite
a bit of time. This is because we used
KDirModel::simplifiedUrlList(urls) to remove child items from the list
of URLs, and this function sorts the URLs internally to make it easier
to find out which of them are child URLs.
However, since commit 5c5d87fec44e7c5934e4b24060200173153f0ff4, the
selected indices are already stored in ascending order, and this makes
it easy to detect if an item is a child of the last item that has been
added to the QMimeData.
The recent changes which prevent that all data for each item are saved
in a QHash already when loading the folder (see
https://git.reviewboard.kde.org/r/112725/), which save both memory and
time, do not work yet in Compact View, because
KItemListWidgetInformant::itemSizeHint() calls the model's data(int)
method for every item, which then initializes the hash.
This patch prevents that by accessing the file name directly if only
the "Name" is shown in the view, just like it's done in Icons View.
Yichao Yu [Wed, 20 Nov 2013 23:47:23 +0000 (00:47 +0100)]
Fix incorrect geometry updates in KItemListContainer
When resizing the window and when KItemListContainer::updateGeometries
is called before the scrollbar visibility is updated, a relayout is
triggered in `m_controller->view()->setGeometry` which updates the
scrollbar visibility and calls back to
`KItemListContainer::updateGeometries` again. Since the first call,
which has the wrong geometry (due to the incorrect scrollbar states),
updates the geometries of the scene and viewport after the second call
(which has the right geometry!!) returns, the final result is a size
that corresponded to the old scrollbar state before this commit.
This patch uses the new geometry of the view after updating it (since
it might not be the size we put in) and therefore makes the sizes
consistent.
Make the "Create New..." menu in the "item context menu" work again
Since https://git.reviewboard.kde.org/r/111989/, the "Create New..."
menu in the context menu that appears when right-clicking a folder was
not a child of the DolphinMainWindow any more, but of the context menu
itself. This is the reason why the dialog that asks for a file name
when choosing one of the "Create New..." options disappeared
immediately.
This patch makes sure that the main window is the parent of the
"Create New..." menu again.
Yichao Yu [Mon, 18 Nov 2013 22:40:40 +0000 (23:40 +0100)]
Fix scrollbar spacing when no scrollbar is visible in dolphin.
Before this commit, Dolphin reserved space for the scrollbar spacing
even when no scrollbar is visible resulting in a ugly gap in the view
when:
1. the theme uses QStyle::SH_ScrollView_FrameOnlyAroundContents and
2. the theme has a positive PM_ScrollView_ScrollBarSpacing.
QtCurve can have both while Oxygen have 1 but not 2.
To reproduce the problem with Oxygen style. Replace the
`width += ....` (which returns -2 or 0 for Oxygen) with `width += 2`.
See more info here:
https://github.com/QtCurve/qtcurve-qt4/issues/9#issuecomment-28630517
Update the Places Panel entries when switching the language
Before this commit, we stored the translated "Places" in
.kde4/share/apps/kfileplaces/bookmarks.xml. This was not intentional -
it only happened because
PlacesItem::updateBookmarkForRole(const QByteArray& role) always stored
the translated text (returned by PlacesItem::text()) in the KBookmark.
This is be fixed by first checking if the text() is equal to the
translation of the text that is already stored in the KBookmark, and
not updating it in that case.
Note that we have to make sure that all "Places"-related use the same
context "KFile System Bookmarks" to make that work.
Thanks to Burkhard Lück and Albert Astals Cid for helping to fix this
problem!
David Rosca [Thu, 14 Nov 2013 08:08:29 +0000 (09:08 +0100)]
Do not select items when navigating back/forward with the mouse
If we detect that the user pressed the back/forward buttons while
hovering the empty space of the view, such that
DolphinView::slotMouseButtonPressed(int, Qt::MouseButtons) will request
that Dolphin navigates back/forward in the history, handling the mouse
press event should stop. This prevents the possible unexpected selection
of items in the new directory.
Only do a fast pre-sorting when "Sort by Name" is used
This is a follow-up to commit 0e9f4a398735cfc19ae783d2ab054d2400d95416,
which tries to speed up sorting the items naturally by their name using
the idea that a fast non-natural pre-sorting already sorts the items
mostly correctly and thus reduces the number of expensive natural
comparisons.
This change only makes sense if the view is really sorted by "Name". In
other cases, the pre-sorting will most likely not be useful.
Since Dolphin 2.0, we have stored the selected items in a QSet<int>,
which is neither space-efficient nor particularly fast when inserting
many items which are in a consecutive range.
This commit replaces the QSet<int> by a new class "KItemSet", which
stores the items in a sorted list of ranges. For each range, we only
store the first index and the length of the range, so we need a lot
less memory for most common selection patterns, and we also save quite
a few CPU cycles in many situations, because adding an item to the
KItemSet will in many cases not need a memory allocation at all, and
it's particularly easy when inserting sorted items into the KItemSet in
a row.
KItemSet contains a minimal subset of QSet's API which makes it
suitable as a drop-in replacement for our needs. It also has iterators,
such that the items can be iterated through easily, also with foreach.
One advantage of KItemSet compared to QSet<int> is that the items are
always iterated through in ascending order.
Sort in items in two stages to speed up natural sorting
Sort the items in a folder first according to their name, without doing
a natural/locale-aware sorting. This is very fast, but the order of the
items is then already close to the final order in most cases.
The number of expensive natural comparisons required to sort the items
is thus greatly reduced.
In my experiments with a folder with 100,000 items, the time required
to sort the files was reduced by 63% with this patch.
Prevent storing some redundant data in KItemListViewLayouter
In KItemListViewLayouter, we have always stored a QRectF for each item,
which is "the area that the item occupies". However, the size of the
QRectF is already stored in the size hint resolver.
Therefore, it is sufficient to store the position of the top left
corner of the QRectF in a QPointF and construct the QRectF on demand.
This patch reduces the memory usage by 16 bytes for each item in the
view:
* a QRectF is 4 doubles -> 32 byes
* a QPointF contains only 2 doubles -> 16 bytes
The fix for bug 161385 (which was about Dolphin still showing an empty
view if a device that had been unmounted earlier was clicked in the
Places Panel) caused a regression: the view state (current item, scroll
position, Details View expansion state) was not restored any more when
going "Back".
The reason is that "m_view->reload()" in
DolphinViewContainer::setUrl(const KUrl& newUrl) was always executed
just after entering a directory, and that command overwrites this
information.
Distinguishing between "change URL" and "reload the view" works better
if it's done in DolphinMainWindow instead of DolphinViewContainer.
Two small simplifications in KItemListViewLayouter
1. Remove the unneeded variable rowCount.
2. Simplify the calculation of the member m_maximumScrollOffset. We can
just use the current value of "y" because this is the offset that
the next row would have.
Fix crash when triggereing the "Compare files" action via D-Bus
If the number of selected items is not two, Dolphin disables this
action. However, it is still possible to trigger it via D-Bus, and this
could cause a crash in DolphinMainWindow::compareFiles() because this
function did not test at all if there are really two items selected.
This patch adds such a check and simplifies the code in that function.
Erik Hahn [Sun, 13 Oct 2013 08:44:13 +0000 (10:44 +0200)]
Files passed as arguments: Ignore unsupported files
With this patch, Dolphin ignores all files passed to it that it can't
Also, archives are now opened inside Dolphin so it can be used as an
archive manager at least for local files. If the user tries to open a
remote archive Dolphin still opens it externally; I have observed that
if it receives one as an argument, it will display a pseudo-folder that
contains only said archive. So having it set as the archive handler is
still broken, but in a less annoying way.
Make the code that removes items from KFileItemModel more robust
When we remove items from the model, we called the function
KFileItemModel::removeItems(const KFileItemList&, RemoveItemsBehavior).
This function then looked up the indexes of the items using the hash
m_items. This is wasteful in the situations when the indexes of the
removed items are known in advance (like when an expanded folder is
collapsed in Details View), and it can cause trouble if one item is
contained in the model multiple times (can happen when searching, and a
file both matches the search and is a child of a folder that matches
the search). Even if expanding folders in the search results list might
not be particularly useful most of the time, it makes sense to make the
model more robust to prevent crashes and other unexpected behavior in
such situations.
This patch makes the following changes to achieve that goal:
* Change the argument of removeItems() from KFileItemList to
KItemRangeList. To make this work, the "look the indexes up in
m_items" code is moved from that function to slotItemsDeleted(). In
the other places where removeItems() is called, the indexes are
calculated directly (which is not more difficult than determining the
removed items as a KFileItemList, if one considers that we needed the
function childItems(KFileItem) for that, which is not needed any more
with this patch).
* Also removeFilteredChildren() takes a KItemRangeList now. Rather than
putting the parent KFileItems into a QSet for O(1) lookup (which
prevents O(N^2) worst case behavior for the entire function), it uses
a QSet<ItemData*> now, which should even be more efficient (hashing a
pointer is cheaper than hashing a KFileItem/KUrl).
Reload the view if a previously unmounted device is mounted again
The problem was that DolphinViewContainer::setUrl(newUrl) was ignored
if newUrl is equal to the URL which is shown in the view already.
The new approach is to reload the view in that method if it is empty, to
make sure that we do not miss that a previously unmounted device has
been re-mounted.
Thanks to Grigoriadis Grigoris for analyzing the root cause of this
issue!
Before this commit, we only added pressed keys to the search string if
they have no other meaning. This means that files containing a Space in
their name could not be searched because Ctrl+Space toggles the
selection state of the current item, and Space alone selects the
current item.
After this commit, Space is added to the search string if
(a) the key press did not have any other effect, i.e., if Ctrl was not
pressed, and the current item is selected already, and
(b) a keyboard search has been started already (to prevent unexpected
effects when pressing Space accidentally - I think that it's rather
uncommon to have files whose names start with a Space - and to make
the unit test simpler).
I modified the unit test of KItemListController, which did not test
keyboard search yet. This uncovered a small problem in
KItemListController::slotChangeCurrentItem() when NoSelection mode is
used. It's not really relevant for anything that is executed inside
Dolphin, but I still fixed it to make the unit test happy.
Save memory and time in KFileItemModel by lazy-loading the "ItemData"
To reduce unnecessary memory comsumption and CPU usage, we only fill the
QHash<QByteArray, QVariant> if the methods data(int) or setData(int) are
called for the corresponding index, or the data is necessary for sorting
the model.
According to my tests, this patch reduces the memory usage when loading
a folder with 100,000 items by 17% in Icons View, and by 26% in Details
View.
Add unit test for the calculation of "name" groups with expanded items
This prevents a possible regression that would have happened with the
first version of https://git.reviewboard.kde.org/r/112725/
The problem was that isChildItem(int index) would return "false"
incorrectly when the QHash for that item was not initialized yet. The
grouping code would then try to read the "text" from the empty QHash,
which yielded an empty QString, and then accessing the first character
of that string caused a crash.
Also factor out the code that transforms a sorted list of ints to a
KItemRangeList. This removes some duplicated code from KFileItemModel.
Note that overriding operator<<() in KItemRangeList was necessary
because it's not a typedef for QList<KItemRange>, but a class derived
from that now, and some code fails to compile if the return type of
that function is QList<KItemRange> and not KItemRangeList.
projects / dolphin.git / log