jeanparpaillon / erlang-dbus Goto Github PK

the project looks interresting but it lacks of a user documentation and/or examples. What is the status about that? Any ETA? Do you have any example I could look at?

Hi @jeanparpaillon
Just to get an understanding of what would this entail, is the idea to implement all the SASL mechanisms (https://en.wikipedia.org/wiki/Simple_Authentication_and_Security_Layer#SASL_mechanisms)?

Hi,

I'm the creator of pydbus, DBus bindings for Python. I've stumbled upon the case of bytestrings - ay - which are commonly used to send non-Unicode data over DBus.

Because of the GDBus's API, apps commonly require these strings to be '\0'-terminated. And that's completely unintuitive for high-level languages, and something that requires the user to read API docs.

I'm thinking about solving the problem automatically, and I see two solutions:

1. Auto-append and auto-strip the last character of "ay"s; [EDIT: impossible, some APIs use not-terminated strings]
2. Decree (how?) that GDBus API is wrong, and strings sent over the wire should not be \0-terminated.
3. [EDIT:] Create a new DBus annotation meaning "All "ay"s in the arguments of this method/signal are \0-terminated bytestrings" - and auto-append/auto-strip \0 if it's present.

Whatever we choose, the solution should be agreed upon by authors of bindings for all languages. That's why I'm here.

Do you have any experience with "ay"-encoded bytestrings?

Relevant discussion: LEW21/pydbus#27 (please reply there, I've opened similar issues in 8 different projects to get attention of bindings authors)

The following java code is implementing erocci backend D-Bus interface (see https://github.com/erocci/erocci_backend_dbus):
https://github.com/Scalair-OpenSource/eroccijavabackendsample

When calling Introspect method, unmarshaling fail in decoding headers.
String length is incorrect, see:
https://github.com/lizenn/erlang-dbus/blob/master/src/dbus_marshaller.erl#L692

Commit b0bde84 'temporarily' disabled the gen_dbus module, rendering the demo example impossible to run. Sadly, renaming gen_dbus.erl.bak to gen_dbus.erl and re-compiling does not help, since at least one macro is missing (HEADER_MEMBER). Even after defining such macro, and fixing some other issues, the furthers one gets when starting the dbus_demo application is to a badarg error on the binary:split/3 call on dbus_auth_cookie_sha1 module (line 24).

Same behaviour is observed by simply starting dbus (dbus:start()) and then running dbus:connect(session).

The http fetch is not working right now. I would go ahead and include rebar, or let the user provide it: REBAR=rebar

release it on hex.pm so we can use it as a dependency easily in rebar3 or elixir projects

In introspection XML format, 'name' attribute is optional. But actually, absence of 'name' causes a crash

This bug looks like string size limit:
erocci/erocci_backend_dbus#10

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

DBUS_COOKIE_SHA1 auth is broken

Hi @jeanparpaillon As you said I have opened it as an issue to discuss about proceedings for the project idea of D-Bus Service_Behaviour. Can you please provide me with some information, may be to assign me some bugs as a starting point. Also provide links to the relevant stuff (project specific).
Thanks :)

Executing 'make test' some tests fail due to missing DBUS service 'net.lizenn.dbus.SampleService'

Either the README should state that the following Python script should be started before running 'make test' or the corresponding Makefile rule in file erlang.mk should start/stop the script, eg:

diff --git a/erlang.mk b/erlang.mk
index 98db78b..39ed906 100644
--- a/erlang.mk
+++ b/erlang.mk
@@ -890,7 +890,9 @@ ct:
 else
 ct: test-build
        @mkdir -p logs/
+       ./test/dbus_client_SUITE_data/example-service.py & echo $$! > /tmp/example-service.pid
        $(gen_verbose) $(CT_RUN) -suite $(addsuffix _SUITE,$(CT_SUITES)) $(CT_OPTS)
+       kill `cat /tmp/example-service.pid`
 endif

BTW: even if the example-service is available, I see the following error reports, but the test-report shows 0 failures.

=ERROR REPORT==== 13-Aug-2015::06:55:18 ===
Bad input fd in erts_poll()! fd=18, port=#Port<0.2238>, driver=inert_drv, name=inert_drv

Hi. I use mac os x. I can try run dbus on mac, but I don't think that it's good idea. And i think about vagrantfile for dev/tests. What do you think?

If we look at the code for dbus_service_reg:handle_call({export_service,ServiceName}...) we see a call to dbus_bus_reg:export_service(undefined, ServiceName), which then exports to each bus.

The reason we have separate busses in dbus is separation of concerns, to separate system messages from session messages for example. Exporting a service to every bus is simply the wrong thing to do.

A grep of the source shows that dbus_bus_reg:export_service and dbus_bus_reg:unexport service are only called from dbus_service_reg.

I suggest therefore that dbus_service_reg be removed from the repository, its associated start command removed from dbus_sup and export and unexport be removed from dbus_bus_reg.

Tony

When available, use UDS support from standard OTP.

erlang/otp#612

Hi there,

I have a silly question, how can I disable logging in this project?

Is it possible to add an option to to main app and configure the log level?

gen_fsm module has been replaced by gen_statem in OTP

Initially had a call to dbus_dbus:env_to_bus_id/0 which was not exported.'
Tried changing that to dbus_dbus_connection:get_bus_id(session) but then failed at dbus_dbus_reg:get_bus where dbus_bus:init/1 called dbus_connection:start_link which does not exist.

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

dbus_peer_connection uses ad-hoc call queing when not authenticated.
We can leverage the postpone action of gen_statem for this

I am trying to add dbus support to my phoenix app

$ mix deps.compile dbus
WARN:  Missing plugins: [rebar3_hex]
==> dbus (compile)
Compiling src/dbus_transport_unix.erl failed:
src/dbus_transport_unix.erl:15: can't find include lib "procket/include/procket.hrl"
src/dbus_transport_unix.erl:56: undefined macro 'PF_LOCAL'
src/dbus_transport_unix.erl:154: undefined macro 'UNIX_PATH_MAX'
src/dbus_transport_unix.erl:21: function init/1 undefined
ERROR: compile failed while processing /vagrant/deps/dbus: rebar_abort

I've removed src/dbus_transport_unix.erl and it compiles properly.

Erlang/OTP 18 [erts-7.0] [source] [64-bit] [smp:4:4] [ds:4:4:10] [async-threads:10] [hipe] [kernel-poll:false]
Elixir 1.2.0

@jeanparpaillon I have questions. And I want start from this "use OTP-style Makefiles". What do you mean? I look at https://github.com/richcarl/merl/blob/master/Makefile (was include to otp) and need make similar?