| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
* mono/ProxyBuilder.cs (BuildFinalizer): Fix some invalid IL when
generating the finalizer. Fixes from Ben Maurer.
|
|
|
|
|
|
|
|
|
| |
* mono/ProxyBuilder.cs (BuildConstructor): Decalre a local
variable for storing our callback in. Update the allocation
comments.
* mono/Service.cs (Module): Revert some accidentally
committed code.
|
|
|
|
|
| |
* mono/ProxyBuilder.cs (BuildFinalizer): Need to load arg 0
onto the eval stack when removing the delegate.
|
|
|
|
|
|
|
|
| |
* mono/ProxyBuilder.cs: Add a finalizer to the generated proxy
classes that disconnects the signal handler delegate from the
service object. Fixes a big leak of proxy objects on the
client side of things. Patch from Ben Maurer
<bmaurer@ximian.com>
|
|
|
|
|
|
| |
* NEWS: Update for 0.23.4
* configure.in: Release 0.23.4
|
|
|
|
|
|
|
| |
* dbus/dbus-connection.c (dbus_connection_send_with_reply): Remove
this unref; it doesn't match up evenly in some codepaths.
(_dbus_connection_block_pending_call): Unref at every exitpoint;
this evenly matches with the ref near the top of this function.
|
|
|
|
|
|
| |
* NEWS: Update for 0.23.3
* configure.in: Release 0.23.3
|
|
|
|
|
|
|
| |
* dbus/dbus-object-tree.c
(_dbus_object_tree_unregister_and_unlock): If checks are enabled
and we try to unregister a path that's not registered, still go
through the process of unlocking and don't just return.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix a bunch of lifecycle and memory management problems
in the mono bindings.
* mono/Arguments.cs (Arguments): Implement IDisposable
* mono/Bus.cs (Bus): Don't allow public instantiation. This is
strictly a static class.
* mono/Connection.cs: Move the DBusObjectPathVTable and associated
delegates into this file.
(Connection): Implement IDisposable.
(Dispose): Disconnect the connection and set the raw connection
pointer to IntPtr.Zero.
(~Connection): Call Dispose().
(RegisterObjectPath): Added. Manages the registration of object
paths so we can cleanly disconnect them at dispose/finalize time.
(UnregisterObjectPath): Ditto.
(set_RawConnection): Unregister all of the object paths when
changing the underlying DBusConnection. Add them back onto the
new connection, if any.
* mono/Handler.cs: Don't implement IDisposable; it doesn't use any
more unmanaged resources anymore, so it's not necessary. Move all
the DBusObjectPathVTable stuff out of here.
(Handler): Save references to our delegates so that they don't get
finalized. Call Connection.RegisterObjectPath() instead of
dbus_connection_register_object_path() directly.
(Message_Called): Dispose the message after we're finished with
it.
* mono/Message.cs (Message): Implement IDisposable.
(Dispose): Dispose the Arguments, and set the RawMessage to
IntPtr.Zero.
(SendWithReplyAndBlock): We own the ref to the reply that comes
back from dbus_connection_send_with_reply_and_block() so add a
comment about that and unref it after we've constructed a managed
MethodReturn class around it. Fixes a big, big leak.
* mono/ProxyBuilder.cs: Reflect into Message to get the Dispose
method.
(BuildSignalHandler): After we've sent the Signal message, dispose
of it.
(BuildMethod): Dispose of the method call and reply messages after
we've sent the message and extracted the data we want from the
reply.
* mono/Service.cs (UnregisterObject): Don't call handler.Dispose()
anymore.
(Service_FilterCalled): Dispose of the message after we're
finished with it.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-connection.c (dbus_connection_send_with_reply):
After we attach our pending call to the connection, unref
it. Fixes a leak.
* dbus/dbus-pending-call.c (dbus_pending_call_get_reply):
Add this back. Accidentally broke API/ABI compat.
* mono/Connection.cs (set_RawConnection): Disconnect our
filter and match callbacks from the old connection and
reconnect them to the new connection, if any.
* mono/DBusType/ObjectPath.cs (ctor): Don't leak the
object path that we get back from unmanaged code.
(Append): Don't leak the object path that we pass
into unmanaged code.
* mono/DBusType/String.cs (ctor): Don't leak the string
that we get back from unmanaged code.
(Append): Don't leak the string that we pass into
unmanaged code.
|
|
|
|
|
|
|
| |
* python/dbus_bindings.pyx.in: API changed, so call
dbus_pending_call_steal_reply() rather than
dbus_pending_call_get_reply(). Bug and patch from
Jan de Groot.
|
|
|
|
|
|
|
|
| |
* NEWS: Update for 0.23.2
* configure.in: Release 0.23.2
2005-02-16 Joe Shaw <joeshaw@novell.com>
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-connection-internal.h, dbus/dbus-connection.[ch],
dbus/dbus-pending-call.[ch], dbus/dbus-server-protected.h,
dbus/dbus-server-unix.c, dbus/dbus-server.c,
dbus/dbus-transport-unix.c, dbus/dbus-transport.c: Backport
a ton of thread-related fixes from HEAD to this branch.
* glib/dbus-gproxy.c: Update dbus_pending_call_get_reply() to
dbus_pending_call_steal_reply().
|
|
|
|
|
|
|
|
|
|
| |
* mono/Connection.cs: The unpredictability of finalizers in mono
prevents us from deterministically disconnecting the filters from
the Service class's finalizer, so move tracking of filters and
matches here. Add API for that.
* mono/Service.cs: Remove the code, add code which calls the
methods now on the Connection class.
|
|
|
|
|
|
| |
* NEWS: Update for 0.23.1
* configure.in: Release 0.23.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* dbus/dbus-connection.c
(_dbus_connection_queue_received_message_link,
_dbus_connection_message_sent): Add the path to
the verbose output.
(_dbus_connection_send_preallocated_and_unlock): Added. Calls
_dbus_connection_send_preallocated_unlocked(), updated the
dispatch status, and unlocks. Fixes a bug where certain
situations (like a broken pipe) could cause a Disconnect message
to not be sent, tricking the bus into thinking a service was still
there when the process had quit.
(_dbus_connection_send_preallocated): Call
_dbus_connection_send_preallocated_and_unlock().
(_dbus_connection_send_and_unlock): Added. Calls
_dbus_connection_send_preallocated_and_unlock().
(dbus_connection_send): Call _dbus_connection_send_and_unlock().
(dbus_connection_send_with_reply): Update the dispatch status and
unlock.
* mono/Service.cs (~Service): Added. Removes the filter so that
we don't get unmanaged code calling back into a GCed delegate.
(RemoveFilter); Added.
|
| |
|
|
|
|
|
|
| |
* NEWS: Update for 0.23.
* configure.in: Release 0.23.
|
|
|
|
|
|
| |
* mono/Makefile.am, mono/example/Makefile.am: Always build the
dbus DLL with --debug. Clean up after the .mdb files this leaves
behind.
|
|
|
|
|
|
| |
* mono/Makefile.am, mono/example/Makefile.am: Always build the
dbus DLL with --debug. Clean up after the .mdb files this leaves
behind.
|
|
|
|
|
| |
* mono/doc/Makefile.am: Need to uninstall the docs on "make
uninstall"
|
|
|
|
|
| |
* mono/Makefile.am, mono/example/Makefile.am: Always build the
dbus DLL with --debug.
|
|
|
|
|
|
|
|
| |
* mono/Makefile.am: Always build the dbus DLL with --debug.
* mono/Arguments.cs (GetDBusTypeConstructor): If the type
is an enum, get the enum's underlying type. Another mono
1.1.3 fix.
|
| |
|
|
|
|
|
|
|
| |
Patch from Sjoerd Simons <sjoerd@luon.net>
* mono/Makefile.am, mono/example/Makefile.am: Don't redefine
DESTDIR. It breaks stuff.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Patch from Tambet Ingo <tambet@ximian.com>
* mono/DBusType/Array.cs (Get): Get the underlying element type by
calling type.GetElementType(). The code previously depended on
broken Mono behavior, which was fixed in Mono 1.1.3.
* mono/DBusType/Dict.cs (constructor): Fix the parameters for
Activator.CreateInstance() so that the class's constructor is
called with the right parameters.
|
|
|
|
|
|
|
|
|
|
|
| |
Patch from Timo Teräs <ext-timo.teras@nokia.com>
* dbus/dbus-connection.c
(_dbus_connection_queue_received_message_link): Call
_dbus_connection_remove_timeout() instead of the _locked()
variant, since it's always called from
_dbus_connection_handle_watch(), which handles the locking.
Removed the _locked() variant since it's no longer used.
|
|
|
|
|
| |
* dbus/dbus-internals.h: I'm an idiot, _dbus_assert certainly can
return
|
|
|
|
| |
* dbus/dbus-internals.h: add _DBUS_GNUC_NORETURN to _dbus_assert
|
| |
|
|
|
|
|
|
|
| |
* dbus/dbus-sysdeps.c (_dbus_sysdeps_test): fix using == on
floating point
* dbus/dbus-string.c (_dbus_string_insert_alignment): new function
|
|
|
|
| |
step is to do so.
|
|
|
|
| |
fixup array lengths after setting a value somewhere within the array.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
and work on the set value stuff some
|
| |
|
|
|
|
|
| |
which is the first step toward a sane reimplementation of all that delete/modify
and realign nonsense in dbus-message.c
|
|
|
|
| |
(address of the value passed in, so char**)
|
| |
|
|
|
|
| |
appending in all cases
|
| |
|
|
|
|
| |
* configure.in: add -Wfloat-equal
|
|
|
|
| |
append based
|
|
|
|
|
| |
* dbus/dbus-sysdeps.h: add _DBUS_DOUBLES_BITWISE_EQUAL macro,
for a variety of reasons '==' doesn't do this.
|
| |
|
| |
|