Thanks. I'm still interested because we're working on a full Haskell binding for PulseAudio (in a very, very slow pace, though). I assume you're talking about an interface similar to pa_stream_set_state_callback(). If so, it looks more consistent with the current PulseAudio API. In either way we can deal with it much more efficiently than the old API. Favonia On Thu, Jan 3, 2013 at 9:04 AM, Tanu Kaskinen <tanuk at iki.fi> wrote: > On Sat, 2012-05-19 at 20:03 -0400, Favonia wrote: >> - Add new API `pa_operation_set_cancel_callback` >> for users to set cancelation callbacks. >> --- >> src/pulse/context.c | 9 +++++++-- >> src/pulse/internal.h | 3 +++ >> src/pulse/operation.c | 25 ++++++++++++++++++++++++- >> src/pulse/operation.h | 7 +++++++ >> 4 files changed, 41 insertions(+), 3 deletions(-) > > Sorry for not replying earlier. Is this something that you are still > interested in? > > I think this would be pretty harmless addition to the API, so I'm not > opposed (comments from others would be very welcome too). What would you > think about changing the semantics from reporting just canceling to > reporting any operation state changes, though? So the added function > would be pa_operation_set_state_callback(). > > -- > Tanu >
