| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| |
|
|
|
|
|
|
|
|
| |
This is a regression from commit
0b9ddf8826d68325f3070723e73177f134bc8b1f, which broke dragging multiple
rows.
https://bugzilla.gnome.org/show_bug.cgi?id=683751
|
|
|
|
| |
We were still using non-symbolic icons for mount roots.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
Since we use the vertical layout only for the desktop these days, remove
NautilusCanvasLayoutMode completely, and just return TRUE from
nautilus_canvas_container_layout_is_vertical() if we're a desktop
container.
This allows to remove a lot of unused layouting code.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This ensures desktop icons are pinned to the primary monitor, instead of
moving around when plugging in a new display.
There's also a slight change in behavior introduced by this patch - that
is for multihead configurations where the first monitor in visible order
is not the primary one, it won't be possible to move desktop icons to
monitors to the left, or to the top, of the primary monitor - in other
words, desktop icons layout starts from the primary monitor and expands
rightwards/downwards.
https://bugzilla.gnome.org/show_bug.cgi?id=420624
|
|
|
|
|
|
|
| |
If the renaming_file member sticks around, we will ignore activation of
new files. Instead, make sure to clear it out when the model is cleared.
https://bugzilla.gnome.org/show_bug.cgi?id=685109
|
|
|
|
|
| |
Factor out two functions, to show and hide the query editor, and use
them from nautilus_window_slot_set_query_editor_visible()
|
|
|
|
|
|
| |
Or we'll segfault calling nautilus_directory_get() on a NULL location.
https://bugzilla.gnome.org/show_bug.cgi?id=685293
|
|
|
|
|
| |
This makes the code more robust in case it's called e.g. with visible =
FALSE on a slot that already has the query editor hidden.
|
|
|
|
|
|
|
|
| |
Since we always return TRUE from the nautilus_window_go_to_full()
callback, we fail to show the error dialog in case connecting fails, and
we print a warning to command line.
In case there's an error, remove the warning and return FALSE instead,
to report the error to the user.
|
|
|
|
|
|
| |
nautilus_directory_new() is now the only place that calls g_object_new()
to create a new directory. Replace the _from_saved_search() constructor
with a setter.
|
|
|
|
|
| |
Since we now always override the file type when creating the
NautilusFile, there's no need for a separate, empty, subclass.
|
|
|
|
|
|
|
| |
And not only when first creating it from a GFileInfo, or we'll lose such
information when e.g. browsing the history list.
https://bugzilla.gnome.org/show_bug.cgi?id=685155
|
|
|
|
|
|
| |
Since in that case we don't have a NautilusSearchDirectoryFile.
https://bugzilla.gnome.org/show_bug.cgi?id=685155
|
|
|
|
|
|
|
|
|
| |
Since nautilus_search_directory_new_from_saved_search() expects the
location field to be already populated after g_object_new() is called
(and it's a reasonable expectation), make it a construct only property.
This will also allow further cleanups in the directory code later.
https://bugzilla.gnome.org/show_bug.cgi?id=685155
|
| |
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=685189
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
Since the WM itself spawns us, I think the API should pass the timestamp
down; since it doesn't, get a timestamp ourselves.
https://bugzilla.gnome.org/show_bug.cgi?id=674816
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Opening the default display when parsing options means we'll do it in
non-primary instances of the application.
When the default display is opened, _gdk_x11_display_make_default() will
be called, which reads and removed DESKTOP_STARTUP_ID from the
environment - but we need to keep DESKTOP_STARTUP_ID around until
add_platform_data() is called in the non-primary instance, which takes
care of forwarding it to the primary instance.
This fixes the busy spinning cursor being stuck when opening multiple
Nautilus windows.
https://bugzilla.gnome.org/show_bug.cgi?id=664147
|
|
|
|
|
|
|
| |
Instead of returning a NULL engine, which will cause criticals later,
since NautilusSearchEngine will still try to call start() on it.
https://bugzilla.gnome.org/show_bug.cgi?id=685014
|
|
|
|
|
|
| |
Assume we have a valid query object passed to
nautilus_search_engine_set_query(), as that is always true in the
current code.
|
| |
|
|
|
|
|
|
| |
Or the pattern matching will fail for non-ASCII characters.
https://bugzilla.gnome.org/show_bug.cgi?id=684852
|
|
|
|
|
|
|
| |
Otherwise we'll return results from the wrong base directory for
searches that are saved, or from the back/forward history.
https://bugzilla.gnome.org/show_bug.cgi?id=684769
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
Instead of immediately adding hits to the hash table for every type
we're interested in, and loop through the terms array every time,
first build a list of builtin match candidates, and then check in a
single loop if they're valid search results.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=684697
|
|
|
|
|
| |
This ensures we also return the right icon in the shell search provider
when an XDG directory is in the results.
|
|
|
|
| |
Will be useful to get the icon for XDG bookmarks.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Since g_mount_get_name() and nautilus_bookmark_get_name() can possibly
return NULL, warn against it before normalizing the string and trying to
compare it with the search terms. This fixes a crasher in the shell
search provider.
https://bugzilla.gnome.org/show_bug.cgi?id=684807
|
| |
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=684354
|
|
|
|
|
|
| |
That includes PNG tEXt chunk information.
https://bugzilla.gnome.org/show_bug.cgi?id=613288
|