g_array_new() doesn't reserve any size by default, so during the initial population
of the file array, we'll do more reallocs than strictly needed. We'll start with
a reasonable preallocated size, in this case the number of files which we can
get in a single chunk out of GIO.
Signed-off-by: Federico Mena Quintero <federico@novell.com>
There's no point in running a GtkFileSystemModel with invalid column types.
This way we can also avoid clearing the memory of the column_types array.
Signed-off-by: Federico Mena Quintero <federico@novell.com>
The previous function enumerated the whole directory and used a lot of
outdated API to decide how to show files.
The new code queries the filesystem model to decide about this.
The now unused old functions were removed.
Previously information about file sizes was not available for search
results and recent files, so the column was always hidden. As this
information is now available, we can stop the special handling and use
the same setting as in browse mode.
Use the faster _gtk_file_system_model_get_value() function instead of
gtk_tree_model_get() inside the sort functions. This gives a significant
speed-up when sorting large lists.
In a test case with 40.000 files, the sorting time went from ~5 seconds
to less than 0.5 seconds for my test case. There is 2 significant
problems with gtk_tree_model_get() that cause this:
1) The value is copied, which takes quite a bit of time for strings.
~25% of excessive time or ~1 second in my test
2) The tree model functions need to lookup the interface vfunc. And
gtk_tree_model_get() doesn't do that only once, but multiple times
(verifying column id, getting the actual value, ...)
~75% of excessive time or ~3 seconds in my test
Replace the list model code with the file system model and use all the
file system model API niceties we get from that.
Also adds the function _gtk_file_system_model_add_and_query_file() which
g_file_query_info()'s the file before adding it, so it gets added with
the right information.
Since the time taken by g_file_enumerate_children() depends a lot on the
attributes that are queried, we query the minimum attributes that need
to be queired to display the file chooser.
In particular, the attributes for loading the icon are ignored, as icons
are loaded on demand (see previous commit).
Because loading icons takes a noticable performance, this code loads the
pixbuf on demand and only loads icons for rows that are visible. There
is a few caveats to this:
- The pixbuf cell renderer must report the proer size even if the icon
is not yet loaded. This is achieved by setting a fixed size.
- On theme changes the cahced pixbufs and the cell renderer must be
updated to conform to the new theme.
Previously, there was a GtkTreeSortModel wrapped around the filesystem
model to make it sortable. As the new implementation implements the
GtkTreeSortable interface, we can use this instead.
A lot of code special cases accesses to the tree view for the different
browse modes, which was previously necessary, because the models were
different. Now that they are identical in the first columns, there is no
such need anymore, and the functions don't need to be special cased.
Previously custom functions were used to compute the data passed to the
cell renderers. Now that all this data is saved by the tree models with
compatible nodes, the usual default attribute-to-column mapping can be
used.
With this, caching of the values can happen in the tree model, which
avoids costly lookups of icons or computation of strings. Last but not
least it avoids spurious bugs that could happen when strings changed
without anyone noticing, like the mtime when a new day begins.
All tree models in browse mode now share the first 10 column types
containing all the necessary information to display the model on screen.
Therefor it is now easy to just operate on the tree model associated
with the file tree view and in most cases it isn't necessary anymore to
special case the browse modes.
The new model is mostly API-compatible with the old model (minimal
changes were required), but is a lot faster and has a lot of very
desirable features.
- the model does no longer support a tree, just a list of files in a
given directory
- the storage has been moved to a GArray as opposed to a tree
- no more dependency on GtkFileSystem
- columns are managed by the creator of the model, so any number of
nodes can be added as needed. This also makes the API more similar
to GtkListStore.
- Values are filled on demand using a function given when creating the
model.
- The function can decide to let the model cache returned values or
decide to be called again the next time the value is queried.
- implements GtkTreeSortable
- _gtk_file_system_model_get_value() was added to significantly speed
up value access, which is necessary when sorting large models.
Add API for GtkDialog to return widgets by response ID.
Added gtk_dialog_get_widget_for_response() to access to all kinds
of buttons with all kinds of responses.
Zero width/height is unsupported and will magically be turned into one.
For instance, gtk_widget_size_allocate() will eventually do this magic
on the value stored in widget->allocation.
However, if we don't do this magic conversion early, then the value
returned from gtk_window_compute_configure_request() will not be
comparable with whats stored in widget->allocation. (I.E. they will
differ if width or height are zero).
This is dangerous, as we do such a comparison in gtk_window_move_resize().
Currently a change from e.g. 10x1 (current allocation) to 10x0 (new size)
will be expected to produce a ConfigureNotify, when it actually won't,
thus never thawing the frozen toplevel.
Fixes bug #588059
The convention for landscape pages, and required by PostScript
Language Reference Manual, is for landscape printing to rotate user
space 90 degrees counterclockwise. Part of bug 596423
Like the other GtkSettings already registered, these are influenced from
the outside and cannot be properly tested. The defaultvalues test
passes again now.
gtk_tree_model_build_level() always needs to emit row-inserted when
requested, this should not depend on whether the level has a parent
level or a virtual root, which is a check whether or not we need to
reference the node in the child model. Furthermore, we also need
to emit row-has-child-toggled after row-inserted when appropriate.
When gtk_tree_model_filter_row_changed() pulls in the root level, it
must request build_level() to emit signals for this. The refilter
function uses row_changed to process the changes, so build_level() in
the first call to row_changed() might pull in multiple new nodes in this
scenario, for all of these signals need to be emitted. Of course,
build_level() will then also emit the signals for the node row_changed()
is processing, we should not emit a duplicate signal, this is now
accounted for.
Add a unit test for this. For this small functionality to block the
row-changed signal has been implemented, so that we can simulate calls
to the refilter function using the current visible column setup.
_gtk_tooltip_handle_event, which is called for many events in the GTK+
main loop, calls gtk_tooltip_set_last_window, which keeps a weak
reference to the last window we passed through. If the window being
set is the same than the last one there's really no need to update our
weak reference, so add a check for that and exit early.
Check for 'auth-info-required' attribute from printer attributes to
find out whether an authentization of user is needed.
Change password dialog of print backend to be able to require informations
requested thru 'auth-info-required' (#566522).
This patch tests availability of remote host before getting ppd file
for selected printer (#586207). It also adds a state message for
failure of getting details.
progress-border and invisible-char style properties had wrong amount
of colons for gtk-doc (signal markup instead of properties), so the
Since tags didn't seem to get picked up.
GtkEntry:invisible-char also had a wrong Since: 2.22, fixed to 2.18.
With Bug 302240, a check was added to avoid keynav to unrealized
radio buttons in the group, but that's not the right check - the
check should be on MAPPED instead, since a widget can be realized
even if it, or a parent, is hidden. Bug 595599
Applications using non-double-buffered drawing using cairo (e.g. Abiword)
can draw directly to the window using cairo and thus manage to avoid
the automatic flushing of outstanding moves. This can cause redraw
inconsistencies like bug 593507.
We fix this by always flushing when exposing non-double-buffered widgets.
Fixed the button-sensivity patch done to GtkComboBox to account for
changes in appearance (changes to the appears-as-list style property).
Also, in list-mode, the event box that has been created below the cell
view also needs to have its sensitivity updated.
Commit 3f306a4042 made it possible for
gtk_tree_view_top_row_to_dy() to be called recursively. (In a different
way than was already guarded for). This caused a single test case in
the scrolling test suite to fail. We now also guard for recursive calls at
the beginning of gtk_tree_view_top_row_to_dy().
Rework the sort and filter models to store their reference to the parent
element as an array index instead of a pointer to an array element.
These pointers could become invalid with any array modification, whereas
indices do not.
If we have a level with zero visible nodes and the first node becomes
visible in that level, then parent has just become a "real" parent node.
In such a case we need to emit row-has-child-toggled. This only applies
to non-root levels that have a parent. This problem was also found when
writing the unit test, the respective cases in the unit test have been
corrected.
This fixes bugs:
Bug 372010 - Filtering not working properly
Bug 525965 - Filtered and sorted GtkTreeView is missing rows
When a filter function is used that determines the visibility property
of a node using whether or it it has children, the state of this node
very likely changes when the filter model receives a
row-has-child-toggled signal. Therefore, we need to handle state
changes in the row-has-child-toggled handler.
Add a function to obtain the effective context id, and reset the slave
only when the effective context id is different from the current context
id, when setting a client window and on focus in. This might fix
bug 593868 and bug 567124.
Also use a different means to check that the given mark is in the text
view's buffer. This will also play nice with anonymous and NULL marks.
Suggested by Paolo Borelli.
When unsetting the old buffer always set the buffer on the layout to
NULL. More importantly, clear the pending scroll. (The scroll is
handled in an idle, when not cleared an idle handler might touch the
layout later on, possibly corrupting the BTree). Unref the buffer after
removing the selection from the clipboard, not before. Patch merged
from maemo-gtk.
In gtk_text_view_queue_scroll() we need to verify if the given mark
exists in the text view's current buffer. When not done, this can
result in corruption of the BTree data structure. Patch merged from
maemo-gtk.
Updating the IM spot location in gtk_text_view_value_changed() might
invalidate the layout, so we need to make sure that we update it before
validating the layout again. Otherwise, the layout will be invalidated
right after validating it (possibly resulting in a failed
onscreen_validated assertion). Patch merged from maemo-gtk.
Likewise to other cell renderers, GtkCellRendererToggle now properly
follows the insensitive state of the widget requesting rendering. It
does this by checking the state of the widget pointer, not by using
GtkCellRendererFlags (there exists an insensitive flags). Later on, I
think we should move to using GtkCellRendererFlags and not the state of
the widget requesting rendering. Patch merged from maemo-gtk.
If the cell view is insensitive, it needs to properly propagate this
insensitive state to the cell renderers using cell renderer flags.
Merged from maemo-gtk.
This was removed long ago by the patch in bug 101235. On a hindsight,
all other scrollable widgets are processing exposes at this point, which
completely makes sense to me, so I am putting it back.
Special case \r\n in gtk_text_buffer_backspace since \r should not be
reinserted even if we are deleting one char at a time. Also add
corresponding unit test. Fixes bug #544724.
Such paths (eg. a child node that is collapsed) should be ignored. This
is fixed by checking the return value of _gtk_tree_view_find_node(),
which returns a partial return value (the parent node) when it returns
TRUE. Also added a unit test.
Reorder drawing in gtk_tree_view_bin_expose() so that the tree lines are
drawn after the cells have been drawn. This is because cell-background
is handled in the cell renderer, so the tree lines need to be drawn
after this.
Contrary to what was believed before, do_validate_rows() does need to
queue a normal resize (including a redraw) when it has to. The redraw
is required because of the size of the tree has changed because new
row(s) have been validated.
Block 2BUTTON and 3BUTTON press events using a button press event
handler in GtkCellRendererSpin, so that they won't be eaten and
processed by tree view.
Previously the rightmost button with a proper response id was activated,
this is now only done if there is no default button set.
With this patch the right widget gets activated when there are multiple
widgets wth response ids conforming to is_stock_accept_response_id() as
the selected widget is made the new default widget before showing the
overwrite confirmation dialog.
In validate_visible_area() it was assumed that gtk_tree_path_prev()
would always return the correct path of the preceding node. This is
obviously not true. The if-clause has been removed so that we now
always use _gtk_tree_view_find_path() to get the path from the tree,
node.
GtkCellRendererAccel also needs to acquire the GTK+ grab in addition to
a GDK keyboard grab. With the GDK keyboard grab, KeyPress and
KeyRelease events are delivered as usual, although we only want to
receive them for our grab widget.
Make the tooltip code a bit more robust for a case that only occurs when
GTK+ is used from a language binding. It looks like this case appears
because the memory management / ref counting is handled differently in
some of the language bindings. Instead of asserting, we will fail
silently. Also fix a think-o in gtk_tooltip_start_delay(). Patch from
O. Andrieu.
The previous code would trigger on an unmodified 'S' key for a
'Super+S' accel. The current code avoids that at the cost of breaking
combinations like 'Mod4+Super+S' which are too exotic to worry about...
Bug 591526
Shift-click inside an existing selection reduces the selection to the
range from the insert mark to the clicked point instead of removing the
selection. This makes GtkTextView more consistent with GtkEntry.
Previously, this flag wasn't cleared properly when the notebook lost
focus, but only when the notebook was focussed again later.
As this flag is only used to advance correctly focus pages by grabbing
focus to the new page after switching the page, this is not good.
In particular, it can cause a focus grab when programmatically switching
the notebook page and the focus is inside a completely different widget.
A previous attempt at this fix in
6e0af6c252 has been reverted in
dfe0c8c0ca because it didn't work
correctly.
Print when Enter pressed in Name entry or Command Line entry in the
print dialog (#564695).
Add gtk_printer_option_set_activates_default() function and
gtk_printer_option_get_activates_default() function to control
behaviour of GtkPrinterOptionWidget.
When printing, increment page sequence before rendering of the page,
not after (#590084).
In opposite case it runs "end_page" function with wrong parameters when
drawing a page in another thread.
Add gtk_widget_get_visible(). For symmetry reasons and for convenience
when a widget's visibility state is available as a boolean condition,
also add gtk_widget_set_visible() (which simply calls show()/hide()).
Previously, this flag wasn't cleared properly when the notebook lost
focus, but only when the notebook was focussed again later.
As this flag is only used to advance correctly focus pages by grabbing
focus to the new page after switching the page, this is not good.
In particular, it can cause a focus grab when programmatically switching
the notebook page and the focus is inside a completely different widget.
Checks for details request before sending a job to a printer.
Shows busy state indication (cursor + status field).
This avoids printing to wrong printer (bug #576601).
Properly invalidate iterators in iter_next, iter_children,
iter_nth_child and iter_parent when FALSE is returned. ListStore was
actually already mostly in order as the GSequence pointers are very
strictly checked, but to be a little stricter we've added stamp = 0
lines. TreeStore did not invalidate its iterators, so stamp = 0 lines
were added where appropriate. Unit tests have been added to check this.
Fixes bug 586374, reported by Mike Gemunde.
gtk_tree_view_enter_notify() now "ignores" the synthesized crossing
events. The synthesized crossing events always have (0, 0) as
coordinates, which messes things up. This patch does not fix all issues,
at least it makes the behavior much more reasonable again. Watch bug
555109 for further discussion on the issue.
Added checks to see whether the passed iterators are not the same in
convert_child_iter_to_iter() and iter_to_child_iter() functions in both
the sort and filter models.
Fixes bug 573321, suggestion by Vadim Godunko.
Adapt gtk_tree_view_get_path_at_pos() and
gtk_tree_view_get_drag_dest_row() to just return FALSE when bin_window
is NULL and not hit a warning. Makes this case consistent with the tree
view not having a model. Documentation has been updated to clarify
this, unit test has been added.
Fixes bug 539377, based on a patch by Bjorn Lindqvist.
Instead of failing with warning on !tree_view->priv->tree, return
silently when tree_view->priv->model is NULL. Clarified in the
documentation that for invalid paths (and every path is invalid when no
model is set), the function will fail silently although the current
cursor will be unset.
Fixes bugs 498010 and 546005.
Includes a test case for bug 546005 to start with, logic provided by
Paul Pogonyshev and Bjorn Lindqvist. In the future, we should maybe
merge treeview-scrolling.c with this one to create one large monolithic
tree view tester.
Instead of refiltering the recently used files we should rebuild
the list altogether when the current GtkFileFilter changes; this
allows us to keep showing the maximum amount of recently used
files.
Fixes bug:
http://bugzilla.gnome.org/show_bug.cgi?id=514260
gtk_tree_view_discover_dirty_iter() and gtk_tree_view_discover_iter()
have been in the tree forever, but unused since 1.3.8 or so. I guess we
no longer need them, so removed them. Fixes a compiler warning as well.
