summaryrefslogtreecommitdiff
path: root/libre/vim/0001-Fix-widget-sizing-on-current-Gtk-3.22-branch.patch
diff options
context:
space:
mode:
Diffstat (limited to 'libre/vim/0001-Fix-widget-sizing-on-current-Gtk-3.22-branch.patch')
-rw-r--r--libre/vim/0001-Fix-widget-sizing-on-current-Gtk-3.22-branch.patch39
1 files changed, 39 insertions, 0 deletions
diff --git a/libre/vim/0001-Fix-widget-sizing-on-current-Gtk-3.22-branch.patch b/libre/vim/0001-Fix-widget-sizing-on-current-Gtk-3.22-branch.patch
new file mode 100644
index 000000000..ea5df7ac3
--- /dev/null
+++ b/libre/vim/0001-Fix-widget-sizing-on-current-Gtk-3.22-branch.patch
@@ -0,0 +1,39 @@
+From 0060136ccbe78ff791e5bd732510cff12094bec9 Mon Sep 17 00:00:00 2001
+From: "Jan Alexander Steffens (heftig)" <jan.steffens@gmail.com>
+Date: Sun, 23 Oct 2016 21:12:15 +0200
+Subject: [PATCH 1/2] Fix widget sizing on current Gtk+ 3.22 branch
+
+Gtk recently started emitting configure events when repositioning child
+windows[1]. Events for the form's child windows bubble up the widget
+hierarchy, get caught by our configure event handler, and treated as if
+they were reporting the size of the form window. As a result, the shell
+was sized using far too small dimensions[2][3]. This change will be part
+of Gtk+ 3.22.2.
+
+Solve this by checking whether the configure event we get is really for
+the form widget.
+
+[1]: https://git.gnome.org/browse/gtk+/commit/?h=gtk-3-22&id=12579fe71b3b8f79eb9c1b80e429443bcc437dd0
+[2]: https://bugs.archlinux.org/task/51509
+[3]: https://bugzilla.gnome.org/show_bug.cgi?id=773387
+---
+ src/gui_gtk_x11.c | 3 +++
+ 1 file changed, 3 insertions(+)
+
+diff --git a/src/gui_gtk_x11.c b/src/gui_gtk_x11.c
+index 388464215fadc3b5..bf3c99ad9a043ad9 100644
+--- a/src/gui_gtk_x11.c
++++ b/src/gui_gtk_x11.c
+@@ -4429,6 +4429,9 @@ form_configure_event(GtkWidget *widget UNUSED,
+ {
+ int usable_height = event->height;
+
++ if (event->window != gtk_widget_get_window (gui.formwin))
++ return TRUE;
++
+ /* When in a GtkPlug, we can't guarantee valid heights (as a round
+ * no. of char-heights), so we have to manually sanitise them.
+ * Widths seem to sort themselves out, don't ask me why.
+--
+2.10.1
+