| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We removed the search action with a rework on the whole handling
of views/slots/search-query handling to decouple the code better
and not expose API that is not needed outside.
The problem is that we actually needed a way to search from the
application, since gnome-shell search provider communicates in that way.
However we missed this since it was just an action in the application,
which made us don't catch this.
Now we allow a search in the whole stack but in a cleaner and direct
way to not be in the same situation in the future.
This patch use that to make the shell search provider work again.
https://bugzilla.gnome.org/show_bug.cgi?id=762076
|
|
|
|
|
|
|
|
| |
Expose the search function on the whole stack.
We will need it for searches from the gnome-shell provider.
https://bugzilla.gnome.org/show_bug.cgi?id=762076
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=762076
|
|
|
|
|
|
| |
We will need it in an upcoming patch.
https://bugzilla.gnome.org/show_bug.cgi?id=762076
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=762076
|
|
|
|
|
|
|
|
|
| |
Some backends doesn't support g_file_get_path, like the trash backends,
and we were crashing trying to create a description with a NULL path.
For those, just use the uri as a description.
https://bugzilla.gnome.org/show_bug.cgi?id=762076
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=762851
|
|
|
|
|
|
|
|
|
|
| |
We don't need to use N_() since it's not an array, but just a const
char.
This was causing problems with the date format not being translated,
since N_() only marks for translation, but doesn't allow translator to
actually translate.
https://bugzilla.gnome.org/show_bug.cgi?id=750917
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We were having a critical when a tab was searching and we closed the
window due to the view reporting it's not loading anymore when the
notebook already removed the slot.
This is due to the slot working independently to the notebook, and the
view reporting to stop loading on dispose, which is fine since the view
needs to report its new status to the window as well. However the window
also update the notebook when is the notebook that removes the slot
and therefore dispose the view.
To avoid that, check on the window if the notebook still contains
the slot or not, and update only in that case.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently we were waiting until the view was fully loaded to select the
first file.
This is fine when only navigating.
However, for search, we actually want to select the first file in order
to allow the user to open right away the most relevant file.
Let's see if this brings problems on icons jumping in large directories,
but I hope it doesn't in the most common cases.
https://bugzilla.gnome.org/show_bug.cgi?id=762595
|
|
|
|
|
|
|
|
|
|
|
|
| |
Avoid to select the first file if there is a selection already.
Although we were checking for the pending selection, it could be that
there is a selection already because the user selected a file.
This will be needed also for an upcoming patch, where we always select
the first file on displaying new files, even if the directory is not
fully loaded.
https://bugzilla.gnome.org/show_bug.cgi?id=762595
|
|
|
|
|
|
| |
It shouldn't happen, but better to be correct here.
https://bugzilla.gnome.org/show_bug.cgi?id=762595
|
| |
|
| |
|
|
|
|
|
|
| |
First, Second, Third were lacking context.
https://bugzilla.gnome.org/show_bug.cgi?id=762851
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Nautilus file operations are implemented as asynchronous jobs scheduled using
g_io_scheduler. Since g_io_scheduler has been deprecated, these operations
should be using the simpler GTask API. The helper functions used in the
operations have been changed in a previous patch so it is now possible to port
the jobs themselves to the new API.
The job structures are now data for tasks, which are handled by the existing
functions in separate threads. For finalizing the operations, the existing
"job_done" functions are now used as callbacks.
https://bugzilla.gnome.org/show_bug.cgi?id=761549
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Nautilus file operations are implemented as asynchronous jobs scheduled using
g_io_scheduler. Since g_io_scheduler has been deprecated, these operations
should be using the simpler GTask API. In order to make this change possible, a
first step would be to replace the scheduler in the helper functions called
during the jobs.
Replace g_io_scheduler_send_to_mainloop with g_main_context_invoke in helper
functions. Since g_main_context_invoke is not blocking, add a mutex and a
condition so the current thread is blocked until the operation is completed in
the main loop.
https://bugzilla.gnome.org/show_bug.cgi?id=761549
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We were not renaming the file at all when the user changed the
file name to a dot file i.e. .whatever.
This was confusing. A better solution would be to show the new
file name until the user changes the directory, but I'm not sure
how much better is it versus hidding the file straig away, since
it can be even more confusing trying to hide your file and nautilus
not hiding it straig away.
For now hide instantaneously a file when renamed to a hidden file.
https://bugzilla.gnome.org/show_bug.cgi?id=342819
|
| |
|
|
|
|
|
|
|
|
|
|
| |
The problem was that when the theme was switched from Adwaita light to Adwaita
dark, the background was kept on white.
The solution was to switch from the white background color, that was always
set for Adwaita to the corresponding theme_base_color of the theme.
https://bugzilla.gnome.org/show_bug.cgi?id=762228
|
|
|
|
|
|
|
|
|
|
|
|
| |
We weren't revealing the folder on creation, since when we moved from
inline renaming to a dialog the folder was not revealed on creation
but instead on renaming.
To fix it, reveal the folder on creation.
Patch made by Nelson Benitez.
https://bugzilla.gnome.org/show_bug.cgi?id=748746
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=689248
|
|
|
|
|
| |
We prefer the dialogs to be modal so they are centered and attached
to the parent.
|
|
|
|
|
|
|
| |
So they don't appear as separate application windows in the GNOME Shell
overview, and to avoid warnings from GTK+.
https://bugzilla.gnome.org/show_bug.cgi?id=753996
|
| |
|
| |
|
|
|
|
| |
patch_2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
desktop files are managed as a special case, and changes on the file
name also affect the contents. But when the content are not a desktop
valid format, it generates an error.
The only thing that comes to my mind to keep maintaining this special
handling of desktop files and also not being in a inconsistent state is
to just explain better why we failed to rename.
https://bugzilla.gnome.org/show_bug.cgi?id=721411
https://bugzilla.gnome.org/show_bug.cgi?id=734401
|
|
|
|
|
|
|
| |
The "Sort" label in the toolbar view menu is slightly to the right of the sort
options. In order to fix this, adjust its right margin.
https://bugzilla.gnome.org/show_bug.cgi?id=762886
|
|
|
|
|
|
|
|
|
|
| |
In Nautilus, thumbnail creation is handled on a separate thread. This thread is
created and managed using the POSIX threads API. For better threading support,
GTask and GThread should be used instead.
Replace the pthread with a GTask. Replace pthread_mutex with GMutex.
https://bugzilla.gnome.org/show_bug.cgi?id=762234
|
|
|
|
|
| |
The query is shared between threads, so make sure we access private data
in a thread safe way.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
For the sidebar fixes.
|
| |
|