Very goood. I neither searched the specs for this. But I think having something like an RFC document, with an generic implementation at it's end, would be a great thing to have for the wave protocol. Maybe someday something like the RFC for FTP or Telnet. :)
Greetings, Sascha On Oct 23, 9:56 am, Torben <[email protected]> wrote: > Hi, > > On 23 Okt., 06:33, Sascha Wedler <[email protected]> wrote: > > > You might also want to have a look at the Wave-API group > > (athttp://groups.google.com/group/google-wave-api). > > > In the discussion "Google Wave Inbox Notification tool > > for the Windows Desktop" (post 15) pamela noted that > > reverse engineering parts of the service might not be > > permitted. > > I know this thread. Very entertaining :-) > However, this dude scanned the HTML of the Google preview which is not > open source > and not permitted by the terms of use. > I just inspected the FedOne C/S protocol which is open source. In fact > it is not > even reverse engineering. I was just too lazy to step through the > heaps of java code. > > Greetings > Torben > > > > > I don't know if the same thing applies here. The protocol > > should be part of the open source specification. But i see > > similarities (since you wrote "Reverse engineering".) > > > Greetings, > > Sascha > > > On Oct 22, 9:48 pm, Torben Weis <[email protected]> wrote: > > > > Hi, > > > > yes, I will publish the C++ code as open source. > > > > In the meantime I succeeded (hopefully) in uncovering the binary format > > > used > > > in the RPC. > > > Seems that every message starts with a 4 byte integer in little endian > > > notation. > > > It describes the size of the message. > > > Then follows a varint serial number (meaning not totally clear to me > > > currently) as a varint > > > (see protocol buffer definition). Next is a varint which describes the > > > length of the function-name. > > > Next is the function-name. Next is a varint which determines the size of > > > the > > > payload. > > > The payload itself is a normal protocol buffer. That's it. The next > > > message > > > follows suit. > > > > However, I did not implement this, so it might still be wrong. At least > > > the > > > format > > > seems to be rather straight forward. > > > > Greetings > > > Torben > > > > 2009/10/22 Sascha <[email protected]> > > > > > I am interested in this as well. With an C++ implementation, a PHP > > > > implementation would become quite easy to make too. > > > > > Do you plan, to publish a C++ library? > > > > > On Oct 21, 5:24 pm, Torben <[email protected]> wrote: > > > > > Hi, > > > > > > I am working on a C++ wave client (written from scratch, no port of > > > > > the java stuff). > > > > > I have the GUI and basic OT stuff in place already. > > > > > Now I want to connect to FedOne. I got the proto files and created the > > > > > C++ code using protoc. > > > > > However, Google does not ship a RPC implementation. While I can write > > > > > one on my own, > > > > > I need to know how the google one is encoded. > > > > > > I tried to capture the TCP communication and decode it using protoc, > > > > > but this seems not to > > > > > work. Can anybody tell me how the RPC message is encoded? It seems to > > > > > contain the name > > > > > of the RPC being called, but what is the meaning of the other bytes? > > > > > > Any hint is appreciated. Reverse engineering binary messages is not > > > > > really fun. > > > > > > Greetings > > > > > Torben > > > > -- > > > --------------------------- > > > Prof. Torben Weis > > > Universitaet Duisburg-Essen > > > [email protected] --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Wave Protocol" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/wave-protocol?hl=en -~----------~----~----~----~------~----~------~--~---
