Support Wayland in desktop manager #4
Conversation
wb9688
force-pushed
the
gtk-layer-shell
branch
5 times, most recently
from
f039615
to
b5ca268
Compare
configure.ac
Outdated
| [enable_x11="${enableval}"], | ||
| [enable_x11=yes]) | ||
|
|
||
| if test "x$enable_wayland" = "xno" && test "x&$enable_x11" = "xno"; then |
There was a problem hiding this comment.
if test "x$enable_wayland" = "xno" -a "x&$enable_x11" = "xno"; then
is shorter
|
|
||
| gio_modules="gthread-2.0 gio-unix-2.0 >= 2.18.0" | ||
| PKG_CHECK_MODULES(GIO, [$gio_modules]) | ||
| AC_SUBST(GIO_CFLAGS) |
There was a problem hiding this comment.
Why do you think it's better to not separate GIO_FLAGS from the rest?
There was a problem hiding this comment.
Currently we were calling e.g. PKG_CHECK_MODULES twice for those Gio packages and we aren't using just GIO_CFLAGS in a Makefile.am, but always as part of FM_CFLAGS, so I thought it might as well just be part of the rest.
| dest_w, dest_h); | ||
| #endif | ||
| } | ||
| else |
There was a problem hiding this comment.
This else does not make sense if no wayland enabled, it newer happens in that case because IS_X11 will be TRUE.
There was a problem hiding this comment.
Yes, and the compiler will the optimize it out and #ifdef HAVE_WAYLAND isn't needed, because it's not some Wayland-specific API
There was a problem hiding this comment.
I wouldn't be so sure about compiler optimization. And even if that is the case then using #if improves the code readability not enforcing the reader to make all the logic calculations, that really helps to not spend extra time on that.
| dest_w, dest_h); | ||
| if(IS_X11()) | ||
| { | ||
| #ifdef HAVE_X11 |
There was a problem hiding this comment.
Makes no sense at all. If HAVE_X11 undefined then IS_X11 is FALSE, i.e. at this point HAVE_X11 is 100% defined.
There was a problem hiding this comment.
The #ifdef's inside the if(IS_X11()) make sense, because else you get compiler errors due to it using X11-specific APIs that you won't have.
There was a problem hiding this comment.
Well, in such case enclose all if...else into #ifdef please.
| g_signal_connect(screen, "monitors-changed", G_CALLBACK(on_screen_size_changed), self); | ||
| if(IS_X11()) | ||
| { | ||
| #ifdef HAVE_X11 |
There was a problem hiding this comment.
See my comment above, please. This #ifdef makes no sense.
|
Everything should work now, though I want to do a bit more testing (and of course finish lxde/libfm#70). Here's a new screenshot:
|
|
@LStranger When will this patch be merged in? |
|
Please make it possible to build with wayland-only gtk+ build. Now the build fails with error |
This PR adds optional Wayland support to PCManFM's desktop manager through the GTK Layer Shell library.
To do:
This PR depends on lxde/libfm#70 for right-click menus to work properly.
I set the dependency on v0.5.2 of GTK Layer Shell even though it would successfully compile with earlier versions, because that's the version some issue that would make the
search_windowunusable got fixed.