| Age | Commit message (Collapse) | Author |
|---|
|
'client address' so we can be sure the client address is always resolved
|
|
|
|
this ID' from 'some' strings and replace the string name with something more sensible.
|
|
to mathematical overflow.
|
|
time the window gets repainted
|
|
|
|
fallback even when freeaddrinfo is available
|
|
|
|
fixed value.
|
|
delayed so much that the network stuff was already closed and the packet would never reach the master server causing the server to appear online longer than necessary.
|
|
[network]:server_bind_ip doesn't exist anymore. Add the IPs/hostnames to [server_bind_addresses]
|
|
the master socket do that too; too bad getaddrinfo can't handle binding to any address on any port at the same time :(
|
|
you've got an unresolved address, the hostname cannot be resolved either so return it as-is.
|
|
to register an IPv6 address (only if it could ofcourse)
|
|
|
|
|
|
|
|
connect to them in any case
|
|
|
|
|
|
NETWORK_CLIENT_NAME_LENGTH for the length of client names.
|
|
is bound to.
|
|
of returned server lists.
|
|
passing the hostname and the IP into a string.
|
|
to the config file.
|
|
quite pointless.
|
|
messages when binding
|
|
|
|
address binding fails
|
|
masterserver
|
|
|
|
|
|
|
|
to multiple sockets.
|
|
multiple sockets.
|
|
unlimited.
|
|
passing it around is easier.
|
|
|
|
keys' from the masterserver so servers can register with multiple IPs as the same server.
|
|
duplicates in the game list
|
|
not provide a nice 'interface', so we have to do it the ugly way.
|
|
|
|
|
|
client_address.
|
|
some support for IPv6 netmask matching.
|
|
|
|
default value and not all OSes actually support IPv4 over IPv6 so making it the same on all OSes eases debugging and such
|
|
server) ;)
|
|
|
|
|
