AsynchronousByteCharChannel and Timeouts

[cowwoc]

  Hi Alan,

     Replies below.

On 14/07/2010 11:09 AM, Alan Bateman wrote:
> The intended behavior is that a read operation completes when some 
> bytes (up to r) are read. The result of the operation is the number of 
> bytes read. It just wouldn't be useful otherwise as there are many 
> protocols where you don't know the number of bytes in advance.

     Please make sure that this is spelled out explicitly in the 
specification. I did not think of this and I suspect that others might 
not as well. Either way, I prefer that the specification spell this out 
explicitly so users don't have to make assumptions.

>> How does this (substantially) increase the burden on the implementer? I
>> assume you're not talking about the extra burden of implementing
>> read-forever on top of read-with-timeout because this just requires 
>> you to
>> wrap your existing code with a single while() loop.
> The extra burden is that it would require every implementation to 
> include support for timeouts.

     How can users develop robust software on top of sockets, serial 
ports, and other potentially-blocking streams without timeout support? 
Don't all operating systems provide timeout support for them anyway?

Thank you,
Gili