[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Gnash-dev] RTMP + deleteStream

From: Red Flesh
Subject: [Gnash-dev] RTMP + deleteStream
Date: Fri, 30 Jan 2009 11:56:26 +0100

Hi all,
first post here so first of all let me thank you for the beatiful project.

I'm trying to understand the relation between streamID/SourceCaller and channelID/ObjectID, and how to instruct the
FM$ to close the created stream after use.

As long this is the result of my limited (mis)understanding of how a createStream/publish exchange is done.

chID = channelID, called ObjectD by wireshark/libpcap, defined in the header
stID = streamID, called SourceCaller by wireshark/libpcap, defined in the header

Invokes and corresponding _result/_error are bound because they
contain a sequence number, the sequence number of an invoke is
contained in the response (_result or _error).

C: send a createStream invoke on channel 3, with a sequence number of N in the body.
   Syntax: string(createStream) number(sequence) null()

S: reply with a _result on channel 3, containing the stID to be used
   for the publish.
   Syntax: string(_result) number(sequence) null() number(streamID)

C: send a publish using in the header the streamID defined above (which channel?, any > 3 seems to work)
   Syntax: string(publish) number(0) null() string(callName) string(live)

What I still didn't managed to understand is how the deleteStream is supposed to work, what I noticed is that if I
don't clean up the used streams after use when I reach a certain limit I got a disconnection from the FMS (which is a problem for me).

I grepped the gnash source and I didn't found any occurrence of "deleteStream", can you say if it is documented somewhere
and provide me some pointer/any guess of how it works?

Yes I already  read:

and linked pages, but I couldn't find any relevant info.
I'll gladly contribute to the docs if possible.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]