From: Raghuram D. <dra...@gm...> - 2007-08-01 13:36:34
|
On 8/1/07, Charlie Groves <cha...@gm...> wrote: > The problem was blindingly obvious once I realized what it was: the > test was attempting to read the amount written by the server socket > from the server socket, and the amount written on the client socket > from the client socket. If they both wrote the same amount it would > pass, but if they differed at all the one that wrote more would be > unable to read all it was expected to. I swapped those numbers to > their correct values(along with restructuring the test to be more like > a regular unit test) in r3373. Let me know if it passes for you two > now. It does. Nice work. |