On Jul 23, 2010, at 1:42 PM, Ted Lemon wrote: > I will take a look at it when I get a chance--I definitely have some ideas > for how to improve libssh2, some of which you might like--but for now I will > use the larger buffer size to get over this obstacle.
It turns out that my problem was something much lamer than I thought. I would get a single read available event every time new data came from the server. But since I wasn't reading all the data, and since the ssh library had read and cached all the data, I would get *some* data, but not the rest. This became obvious once I got the keyboard working--every time I hit return, I'd get 1024 bytes of output. :) So in fact there is no bug in the library--I'm just not as smart as I thought. _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
