VW socket question

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

VW socket question

David Farber
OK, i've got ALL the test running in Squeak now. unfortunately, i can't
bring up the swazoo.org site. (i can in VW). i think the problem is that
i've not totally captured VW's socket semantics. so here is my question:

in HTTPServer there is a method

newSocket: aSocket
        ^[aSocket accept] on: OSErrorHolder transientErrorSignal
                do: [:ex | ex restart]

is aSocket accept supposed to block (ie. not return) until there is data
from an incoming connection?

again, the tests run fine, which makes me think that data gets sent to the
socket before it times out in Squeak.

any pointers/suggestions welcome.

david

--
David Farber
[hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: VW socket question

Alan Knight-2
Yes, accept is blocking. See the comment

accept
         "Accept a connection (receiver must be stream-based, have already
         been bound to an address and listening)."
         "Accept extracts the first connection on the queue of pending
         connections, and returns a new socketAccessor instance with the
         same properties as the receiver. It also returns the SocketAddress
         of the peer (which this code throws away).
         If there are no outstanding connections, the calling process will
         block indefinitely."

At 12:49 PM 11/3/00 -0700, David Farber wrote:

>OK, i've got ALL the test running in Squeak now. unfortunately, i can't
>bring up the swazoo.org site. (i can in VW). i think the problem is that
>i've not totally captured VW's socket semantics. so here is my question:
>
>in HTTPServer there is a method
>
>newSocket: aSocket
>         ^[aSocket accept] on: OSErrorHolder transientErrorSignal
>                 do: [:ex | ex restart]
>
>is aSocket accept supposed to block (ie. not return) until there is data
>from an incoming connection?
>
>again, the tests run fine, which makes me think that data gets sent to the
>socket before it times out in Squeak.
>
>any pointers/suggestions welcome.
>
>david
>
>--
>David Farber
>[hidden email]
>_______________________________________________
>Swazoo-devel mailing list
>[hidden email]
>http://lists.sourceforge.net/mailman/listinfo/swazoo-devel