float32?

[FUBAR]

im working on a program for a game called half-life my prog is written in vb, but i need help because the half-life server protocol is for C++ i can already recive player info from server, like name, and frags, i also want to be able to recive the player time in game, but the packet is float32 it says, and i dont know what to do with it...here is the server protocal thing

look down under "players" see how it says the packet is float32 for the player time in game, i get 3 ascii characters back with each player, can somebody help me, asm me questions if u have to..here is protocol....



Game servers will answer the following messages:
Messages are sent to the server by sending 4 consecutive bytes of 255 (32-bit integer -1) and then the string command followed by a zero byte to terminate it

"ping"
Server responds with a single byte code ASCII 'j'

"info"
Server responds with the following packet:
(int32) -1
(byte) ASCII 'C' (info response, S2A_INFO)
(string) net address of server
(string) name of the host / server
(string) name of the map
(string) game directory (i.e. valve/)
(string) Game description (e.g. "half-life multiplay")
(byte) active client count
(byte) maximum clients allowed
(byte) protocol version (currently 7)

"players"
Server responds with the following packet:
(int32) -1
(byte) ASCII 'D' (players response, S2A_PLAYER)
(byte) active client count

for each active client
(byte) client number / index
(string) player name
(int32) client's frag total
(float32) client's total time in-game

"rules"
Server responds with the following packet:
(int32) -1
(byte) ASCII 'E' (rules response, S2A_RULES)
(int16) number of rules

for each rule
(string) rule name
(string) rule value


-------------------

Master server query protocol:

Sent to master:

(byte) ASCII 'c' ( A2M_GET_SERVERS )

Response from master:

(int32) -1
(byte) ASCII 'd' ( M2A_SERVERS )
(byte) unused

// This is followed by as many 6 byte ip addresses as will fit in the message. The master currently feeds about 2K worth of IP address ( 300+ or so ), but we may up that # significantly in the near future.

(4 x byte) ip address
(int16) port #



The old protocol is still valid, but there is a new, extended protocol:

Here are both the old and new protocols ( the new one is the "details" message ):

"info"
Server responds with the following packet:
(int32) -1
(byte) ASCII 'C' (info response, S2A_INFO)
(string) net address of server
(string) name of the host / server
(string) name of the map
(string) game directory (i.e. valve/)
(string) Game description (e.g. "half-life multiplay")
(byte) active client count
(byte) maximum clients allowed
(byte) protocol version (currently 37)

"details"
(int32) -1
(byte) ASCII 'm' ( S2A_INFO_DETAILED )
(string) net address of server
(string) name of the host / server
(string) name of the map
(string) game directory (i.e. valve/)
(string) Game description (e.g. "half-life multiplay")
(byte) active client count
(byte) maximum clients allowed
(byte) protocol version (currently 37)
(byte) type of server == 'l' for listen or 'd' for dedicated
(byte) os of server == 'w' for win32 or 'l' for linux
(byte) password on server == 1 or yes, 0, for no
(byte) is server running a mod? == 1 for yes, 0 for no

IFF the server is running mod byte was 1:

(string) URL for mod's "info" website
(string) URL for mod's download ftp server
(int32) mod version #
(int32) mod download size ( in bytes, approx. )
(byte) is the mod a server side only mod? 1 == yes, 0 == no
(byte) does this server require you to have a custom client side .dll ( client.dll )? 1 == yes, 0 == no.



Here is some code that uses the new protocol. The first function starts the protocol and the second services the message and requests more ip addresses:

/*
==================
Host_GetBatchServerList_f

Request full server list from HL master server.
==================
*/
void Host_GetBatchServerList_f( void )
{
unsigned char c[10];
adrlist_t *p;
int i = 0;

// Request a server list from the master servers.
NET_Config( true ); // Initialize networking

if ( gfNoMasterServer )
return;

if ( !valvemaster_adr )
return;

c[0] = A2M_GET_SERVERS_BATCH; // ascii 'e'
i++;

*(int *)&c[1] = 0; // Request first batch.
i += sizeof( int );

p = valvemaster_adr;

while ( p )
{
// send to valve master
Con_Printf ("Requesting batch server list from %s\n", NET_AdrToString ( p->adr) );
NET_SendPacket (NS_CLIENT, i, c, p->adr );
p = p->next;
}
}

// Upon receiving ( 5 bytes, 4 for the -1 and 1 byte for the 'f' character ):
int32 -1
M2A_SERVER_BATCH // Ascii 'f', the response to the above

void CL_ParseBatchServerList( void )
{
char szAddress[128] = "";
unsigned char cIP[4];
int i;
unsigned short iIPPort;
int nNumAddresses;
int count = 1;
int unique = 0;

MSG_ReadByte(); // Skip \r

unique = MSG_ReadLong();

// So far we have read 6 chars. Remainder of message length is iBytesRead - 6
nNumAddresses = net_message.cursize - sizeof(int) - sizeof(unsigned char) - sizeof(byte);

// Each address is 6 bytes long
//
nNumAddresses /= 6;

while (nNumAddresses-- > 0)
{
memset(szAddress, 0, 128);

for (i = 0; i < 4; i++)
cIP[i] = MSG_ReadByte();

sprintf( szAddress, "%i.%i.%i.%i"
, (int)cIP[0]
, (int)cIP[1]
, (int)cIP[2]
, (int)cIP[3]
);

iIPPort = BigShort((unsigned short)MSG_ReadShort());

//if ( count <= 100 )
// Con_Printf( "%4i: %s:%i\n", count, szAddress, iIPPort );

count++;
}

Con_Printf( "%i servers\n", count );

if ( unique != 0 ) // More servers left, send another request to master
{
unsigned char c[10];
int i = 0;

// Request a server list from the master servers.
NET_Config( true ); // Initialize networking

c[0] = A2M_GET_SERVERS_BATCH;
i++;

*(int *)&c[1] = unique; // Request servers starting after this unique id.
i += sizeof( int );

// send to valve master
Con_DPrintf ("Requesting next batch ( %i ) server list from %s\n", unique, NET_AdrToString ( net_from ) );
NET_SendPacket (NS_CLIENT, i, c, net_from );
}
else
Con_Printf( "Done.\n" );
}

[jim mcnamara]

float32 is double in C++, so it would also be a double in VB.
It's a 32 bit floating point number. I cannot guarantee that it will correctly map to VB double variable.

[CornedBee]

double is 64-bit and maps to vb double. BUT float32 is a real float and therefore maps to single in VB