Sure.
I have used an FTP connection for Monticello in several different
versions of the vm: 3.9, 3.10.2, and now the closure-vm with Cassou's
dev-image from last week with no updates -- i cut-n-pasted the mouse
fixes for the mac after that first note. Images used for 3.9 and 3.10
included both Pharo and straight-up Squeak. (There is a chance I used
it under 3.8, but... I am not certain.)
Oh, my Mac work has all be in the last three weeks, starting with a
Core image 3 weeks ago.
Let me get into the debugger...
--
Socket>>connectTo: hostAddress port: port
"Initiate a connection to the given port at the given host address.
Waits until the connection is established or time outs."
NetNameResolver useOldNetwork
ifTrue: [self connectTo: hostAddress port: port
waitForConnectionFor: Socket standardTimeout]
ifFalse: [
hostAddress port: port.
self connectTo: hostAddress]
--
Host address is an instance of SocketAddress for me.
The stack is:
...
Socket>>connectTo:port:
SocketStream class>>openConnectionToHost:port:
FTPClient(ProtocolClient)>>ensureConnection
FTPClient(ProtocolClient)>>openOnHost:port:
FTPClient class(ProtocolClient class)>>openOnHost:port:
FTPClient class(ProtocolClient class)>>openOnHostNamed:port:
...
And the code for the latter is...
FTPClient class(ProtocolClient class)>>
openOnHostNamed: hostName port: portNumber
| serverIP |
serverIP := NetNameResolver addressForName: hostName timeout: 20.
^self openOnHost: serverIP port: portNumber
So... I would test that message to NetNameResolve in a workspace.
Also, SocketStream openConnectionToHost: self host port: self port
(see FTPClient>>ensureConnection).
Good luck.
Cam
_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
