This chapter is for RPC programmers. It documents the XDR routines in the RPC Run-Time Library (RTL). These routines are the programming interface to RPC.
This section explains what XDR routines do and when you would call them. It also provides quick reference and detailed reference sections describing each XDR routine.
Most XDR routines share these characteristics:
• They convert data in two directions: from the host's local data format to XDR format (called encoding or marshalling), or the other way around (called decoding or unmarshalling).
• They use xdrs, a structure containing instructions for encoding, decoding, and deallocating memory.
• They return a boolean value to indicate success or failure.
Some XDR routines allocate memory while decoding an argument. To free this memory, call the xdr_free routine after the program is done with the decoded value.
XDR Actions shows the order in which XDR routines perform encoding and decoding.
|
Client |
Server |
|
1. Encodes arguments |
1. Decodes arguments |
Under most circumstances, you are not likely to call any XDR routines directly. The clnt_call and svc_sendreply routines call the XDR routines.
You would call the XDR routines directly only when you write your own routines to convert data to or from XDR format.
XDR Encoding and Decoding Routines lists the XDR routines that encode and decode data.
|
Table 12-2 XDR Encoding and Decoding Routines |
|
|
This routine... |
Encodes and decodes... |
|
xdr_array |
Variable-length array |
|
xdr_bool |
Boolean value |
|
xdr_bytes |
Bytes |
|
xdr_char |
Character |
|
xdr_double |
Double-precision floating point number |
|
xdr_enum |
Enumerated type |
|
xdr_float |
Floating point value |
|
xdr_hyper |
VAX quad word to an XDR hyper-integer, or the other way |
|
xdr_int |
Four-byte integer |
|
xdr_long |
Longword |
|
xdr_opaque |
Contents of a buffer (treats the data as a fixed length of bytes and does not attempt to interpret them) |
|
xdr_pointer |
Pointer to a data structure |
|
xdr_reference |
Pointer to a data structure (the address must be non-zero) |
|
xdr_short |
Two-byte unsigned integer |
|
xdr_string |
Null-terminated string |
|
xdr_u_char |
Unsigned character |
|
xdr_u_hyper |
VAX quad word to an XDR unsigned hyper-integer |
|
xdr_u_int |
Four-byte unsigned integer |
|
xdr_u_long |
Unsigned longword |
|
xdr_u_short |
Two-byte unsigned integer |
|
xdr_union |
Union |
|
xdr_vector |
Vector (fixed length array) |
|
xdr_void |
Nothing |
|
xdr_wrapstring |
Null-terminated string |
XDR Support Routines lists the XDR routines that perform various support functions.
|
Table 12-3 XDR Support Routines |
|
|
This routine... |
Does this... |
|
xdr_free |
Deallocates a data structure from memory |
|
xdrmem_create |
Creates a memory buffer XDR stream |
|
xdrrec_create |
Creates a record-oriented XDR stream |
|
xdrrec_endofrecord |
Marks the end of a record |
|
xdrrec_eof |
Goes to the end of the current record, then verifies whether any more data can be read |
|
xdrrec_skiprecord |
Goes to the end of the current record |
|
xdrstdio_create |
Initializes an stdio stream |
Upper Layer XDR Routines lists the upper layer XDR routines that support RPC.
|
Table 12-4 Upper Layer XDR Routines |
|
|
This routine... |
Encodes and decodes... |
|
xdr_accepted_reply |
Part of an RPC reply message after the reply is accepted |
|
xdr_authunix_parms |
UNIX-style authentication information |
|
xdr_callhdr |
Static part of an RPC request message header (encoding only) |
|
xdr_callmsg |
RPC request message |
|
xdr_netobj |
Data in the netobj structure |
|
xdr_opaque_auth |
Authentication information |
|
xdr_pmap |
Port Mapper parameters |
|
xdr_pmaplist |
List of Port Mapping data |
|
xdr_rejected_reply |
Part of an RPC reply message after the reply is rejected |
|
xdr_replymsg |
RPC reply header; it then calls the appropriate routine to convert the rest of the message |
The following sections describe each XDR routine in detail.
Converts an RPC reply message from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_accepted_reply (XDR *xdrs, struct accepted_reply *ar);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
ar
Address of the structure containing the RPC reply message.
Description
The xdr_replymsg routine calls the xdr_accepted_reply routine.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_replymsg
Converts a variable-length array from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_array (XDR *xdrs, u_char **addrp, u_long *sizep, u_long maxsize, u_long elsize, xdrproc_t elproc);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
addrp
Address of the address containing the array being converted. If addrp is zero, then xdr_array allocates ((*sizep) *elsize) number of bytes when it decodes.
sizep
Address of the number of elements in the array.
maxsize
Maximum number of elements the array can hold.
elsize
Size of each element, in bytes.
elproc
XDR routine that handles each array element.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts UNIX-style authentication information from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_authunix_parms (XDR *xdrs, struct authunix_parms *aupp);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
aupp
UNIX-style authentication information being converted.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts a boolean value from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_bool (XDR *xdrs, bool_t *bp);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
bp
Address of the boolean value.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts bytes from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_bytes (XDR *xdrs, u_char **cpp, u_long *sizep, u_long maxsize);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
cpp
Address of the address of the buffer containing the bytes being converted. If *cpp is zero, xdr_bytes allocates maxsize bytes when it decodes.
sizep
Address of the actual number of bytes being converted.
maxsize
Maximum number of bytes that can be used. The server protocol determines this number.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Encodes the static part of an RPC request message header.
Format
#include
bool_t xdr_callhdr (XDR *xdrs, struct rpc_msg *chdr);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
chdr
Address of the data being converted.
Description
The xdr_callhdr routine converts the following fields: transaction ID, direction, RPC version, server program number, and server version. It converts the last four fields once, when the client handle is created.
The clnttcp_create and clntudp_create / clntudp_bufcreate routines call the xdr_callhdr routine.
Diagnostics
This routine always returns TRUE.
See Also
clnt_call, clnttcp_create, clntudp_create / clntudp_bufcreate, xdr_callmsg
Converts an RPC request message from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_callmsg (XDR *xdrs, struct rpc_msg *cmsg);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
cmsg
Address of the message being converted.
Description
The xdr_callmsg routine converts the following fields: transaction ID, RPC direction, RPC version, program number, version number, procedure number, client authentication.
The pmap_rmtcall, svc_sendreply, and svc_sendreply_dq routines call xdr_callmsg.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_callhdr
Converts a character from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_char (XDR *xdrs, char *cp);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
cp
Address of the character being converted.
Description
This routine provides the same functionality as the xdr_u_char routine.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_u_char
Converts a double-precision floating point number between local and XDR format.
Format
#include
bool_t xdr_double (XDR *xdrs, double *dp);
Arguments
xdrs
Pointer to an XDR stream handle created by one of the XDR stream handle creation routines.
dp
Pointer to the double-precision floating point number.
Description
This routine provides a filter primitive that translates between double-precision numbers and their external representations. It is actually implemented by four XDR routines:
|
xdr_double_D |
Converts VAX D format floating point numbers |
|
xdr_double_G |
Converts VAX G format floating point numbers |
|
xdr_double_T |
Converts IEEE T format floating point numbers |
|
xdr_double_X |
Converts IEEE X format floating point numbers |
You can reference these routines explicitly or you can use compiler settings to control which routine is used when you reference the xdr_double routine.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts an enumerated type from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_enum (XDR *xdrs, enum_t *ep);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
ep
Address containing the enumerated type.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts a floating point value from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_float (XDR *xdrs, float *fp);
Arguments
xdrs
Pointer to an XDR stream handle created by one of the XDR stream handle creation routines.
fp
Pointer to a single-precision floating point number.
Description
This routine provides a filter primitive that translates between double-precision numbers and their external representations. It is actually implemented by four XDR routines:
|
xdr_float_F |
Converts VAX F format floating point numbers |
|
xdr_float_S |
Converts IEEE T format floating point numbers |
You can reference these routines explicitly or you can use compiler settings to control which routine is used when you reference the xdr_float routine.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Deallocates a data structure from memory.
Format
#include
void xdr_free (xdrproc_t proc, u_char *objp);
Arguments
proc
XDR routine that describes the data structure.
objp
Address of the data structure.
Description
Call this routine after decoded data is no longer needed. Do not call it for encoded data.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts a VAX quad word to an XDR hyper-integer, or the other way around.
Format
bool_t xdr_hyper (XDR *xdrs, quad *ptr);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
ptr
Address of the structure containing the quad word. The quad word is stored in standard VAX quad word format, with the low-order longword first in memory.
Description
This routine provided the same functionality as the xdr_u_hyper routine.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_u_hyper
Converts one four-byte integer from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_int (XDR *xdrs, int *ip);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
ip
Address containing the integer.
Description
This routine provides the same functionality as the xdr_u_int, xdr_long, and xdr_u_long routines.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_u_int,
xdr_long, xdr_u_long
Converts one longword from local format to XDR format, or the other way around.
Format
#include
bool_t xdr_long (XDR *xdrs, u_long *lp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
lp
Address containing the longword.
Description
This routine provides the same functionality as the xdr_u_long, xdr_int, and xdr_u_int routines.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_u_long,
xdr_int, xdr_u_int
Converts data in the netobj structure from the local data format to XDR format, or the other way around.
Format
bool_t xdr_netobj (XDR *xdrs, netobj *ptr);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
ptr
Address of the following structure:
typedef struct
{
u_long n_len;
byte *n_bytes;
} netobj;
This structure defines the data being converted.
Description
The netobj structure is an aggregate data structure that is opaque and contains a counted array of 1024 bytes.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts the contents of a buffer from the local data format to XDR format, or the other way around. This routine treats the data as a fixed length of bytes and does not attempt to interpret them.
Format
#include
bool_t xdr_opaque (XDR *xdrs, char *cp, u_long cnt);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
cp
Address of the buffer containing opaque data.
cnt
Byte length.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts authentication information from the local data format to XDR format, or the other way around.
Format
#include
bool_t xdr_opaque_auth (XDR *xdrs, struct opaque_auth *ap);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
ap
Address of the authentication information. This data was created by the authnone_create, authunix_create, or authunix_create_default routine.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts Port Mapper parameters from the local data format to XDR format, or the other way around.
Format
#include "MULTINET_INCLUDE:PMAP_PROT.H"
bool_t xdr_pmap (XDR *xdrs, struct pmap *regs);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
regs
Address of a structure containing the program number, version number, protocol number, and port number. This is the data being converted.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts a list of Port Mapping data from the local data format to XDR format, or the other way around.
Format
#include "TCPIP$RPC:PMAP_PROT.H"
bool_t xdr_pmaplist (XDR *xdrs, struct pmaplist **rpp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
rpp
Address of the address of the structure containing Port Mapper data. If this routine is used to decode a Port Mapper listing, rpp is set to the address of the newly allocated linked list of structures.
Diagnostics
This routine returns TRUE if it succeeds and
FALSE if it fails.
Converts a recursive data structure from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_pointer (XDR *xdrs,
u_char **objpp, u_long obj_size,
xdrproc_t xdr_obj);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
objpp
Address of the address containing the data being converted. May be zero.
obj_size
Size of the data structure in bytes.
xdr_obj
XDR routine that describes the object being pointed to. This routine can describe complex data structures, and these structures may contain pointers.
Description
An XDR routine for a data structure that contains pointers to other structures, such as a linked list, would call the xdr_pointer routine. The xdr_pointer routine encodes a pointer from an address into a boolean. If the boolean is TRUE, the data follows the boolean.
Example
bool_t xdr_pointer( xdrs, objpp,
obj_size, xdr_obj)
XDR *xdrs;
char **objpp;
longw obj_size;
xdrproc_t xdr_obj;
{
bool_t more_data;
/*
** determine if the pointer is a valid address (0 is invalid)
*/
if( *objpp != NULL)
more_data = TRUE;
else
more_data = FALSE;
/*
** XDR the flag
** If we are decoding, then
more_data is overwritten.
*/
if( !xdr_bool( xdrs, &more_data))
return( FALSE);
/*
** If there is no more data, set the pointer to 0 (No effect if we
** were encoding) and return TRUE
*/
if( !more_data)
{
*objpp = NULL;
return( TRUE);
}
/*
** Otherwise, call xdr_reference. The result is that xdr_pointer is
** the same as xdr_reference, except that xdr_pointer adds a Boolean
** to the encoded data and will properly handle NULL pointers.
*/
return( xdr_reference( xdrs, objpp, obj_size, xdr_obj));
} /* end function xdr_pointer() */
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
This routine recursively converts a structure that is referenced by a pointer inside the structure.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_reference (XDR *xdrs, u_char **objpp, u_long obj_size, xdrproc_t xdr_obj);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
objpp
Address of the address of a structure containing the data being converted. If objpp is zero, the xdr_reference routine allocates the necessary storage when decoding. This argument must be non-zero when encoding.
When xdr_reference encodes data, it passes *objpp to xdr_obj. When decoding, xdr_reference allocates memory if *objpp equals zero.
obj_size
Size of the referenced structure.
xdr_obj
XDR routine that describes the object being pointed to. This routine can describe complex data structures, and these structures may contain pointers.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts the remainder of an RPC reply message after the header indicates that the reply is rejected.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_rejected_reply (XDR *xdrs, struct rejected_reply *rr);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
rr
Address of the structure containing the reply message.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts the RPC reply header, then calls the appropriate routine to convert the rest of the message.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_replymsg (XDR *xdrs, struct rpc_msg *rmsg);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
rmsg
Address of the structure containing the reply message.
Description
The xdr_replymsg routine calls the xdr_rejected_reply or xdr_accepted_reply routine to convert the body of the RPC reply message from the local data format to XDR format, or the other way around.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_accepted_reply,
xdr_rejected_reply
Converts a two-byte integer from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_short (XDR *xdrs, short *sp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
sp
Address of the integer being converted.
Description
This routine provides the same functionality as xdr_u_short.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_u_short
Converts a null-terminated string from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_string (XDR *xdrs, char **cpp, u_long maxsize);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
cpp
Address of the address of the first byte in the string.
maxsize
Maximum length of the string. The service protocol determines this value.
Description
The xdr_string routine is the same as the xdr_wrapstring routine, except xdr_string allows you to specify the maxsize.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_wrapstring
Converts an unsigned character from local format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_u_char (XDR *xdrs, u_char bp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
bp
Address of the character being converted.
Description
This routine provides the same functionality as xdr_char.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_char
Converts a VAX quad word to an XDR unsigned hyper-integer, or the other way around.
Format
bool_t xdr_u_hyper (XDR *xdrs, quad *ptr);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
ptr
Address of the structure containing the quad word. The quad word is stored in standard VAX format, with the low-order longword first in memory.
Description
This routine provides the same functionality as the xdr_hyper routine.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_hyper
Converts a four-byte unsigned integer from local format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_u_int (XDR *xdrs, int *ip);
Arguments
xdrs
Address of a structure containing XDR encoding and decoding information.
ip
Address of the integer.
Description
This routine provides the same functionality as xdr_int, xdr_long, and xdr_u_long.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_int
Converts an unsigned longword from local format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_u_long (XDR *xdrs, u_long *lp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
lp
Address of the longword.
Description
This routine provides the same functionality as xdr_long, xdr_int, and xdr_u_int.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_long,
xdr_int, xdr_u_int
Converts a two-byte unsigned integer from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_u_short (XDR *xdrs, u_short *sp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
sp
Address of the integer being converted.
Description
This routine provides the same functionality as xdr_short.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_short
Converts a union from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_union (XDR *xdrs, enum_t *dscmp, u_char *unp, xdr_discrim *choices, xdrproc_t dfault);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
dscmp
Integer from the choices array.
unp
Address of the union.
choices
Address of an array. This array maps integers to XDR routines.
dfault
XDR routine that is called if the dscmp integer is not in the choices array.
Description
The xdr_union routine searches the array choices for the value of dscmp. If it finds the value, it calls the corresponding XDR routine to process the remaining data. If xdr_union does not find the value, it calls the dfault routine.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts a vector (fixed length array) from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_vector (XDR *xdrs, u_char *basep, u_long nelem, u_long elmsize, xdrproc_t xdr_elem);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
basep
Address of the array.
nelem
Number of elements in the array.
elmsize
Size of each element.
xdr_elem
Converts each element from the local data format to XDR format, or the other way around.
Diagnostics
This routine returns TRUE if it succeeds
and FALSE if it fails.
Converts nothing.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_void (XDR *xdrs, u_char *ptr);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
ptr
Ignored.
Description
Use this routine as a place-holder for a program that passes no data. The server and client expect an XDR routine to be called, even when there is no data to pass.
Diagnostics
This routine always returns TRUE.
Converts a null-terminated string from the local data format to XDR format, or the other way around.
Format
#include tcpip$rpc:xdr.h
bool_t xdr_wrapstring (XDR *xdrs, char **cpp);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
cpp
Address of the address of the first byte in the string.
Description
The xdr_wrapstring routine calls the xdr_string routine. The xdr_wrapstring routine hides the maxsize argument from the programmer. Instead, the maximum size of the string is assumed to be 232 - 1.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdr_string
Creates a memory buffer XDR stream.
Format
#include tcpip$rpc:xdr.h
void xdrmem_create (XDR *xdrs, u_char *addr, u_long size, enum xdr_op op);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
addr
Address of the buffer containing the encoded data.
size
Size of the addr buffer.
op
Operations you will perform on the buffer. Valid values are XDR_ENCODE, XDR_DECODE, and XDR_FREE. You may change this value.
Description
The xdrmem_creater outine initializes a
structure so that other XDR routines can write to a buffer.
Creates a record-oriented XDR stream.
Format
#include tcpip$rpc:xdr.h
void xdrrec_create (XDR *xdrs,
u_long sendsize, u_long recvsize,
u_char *tcp_handle, int (*readit)(), int (*writeit)();
Arguments
xdrs
Address of the structure being created. The xdrrec_create routine will write XDR encoding and decoding information to this structure.
sendsize
Size of the send buffer in bytes. The minimum size is 100 bytes. If you specify fewer than 100 bytes, 4000 bytes is used as the default.
recvsize
Size of the receive buffer in bytes. The minimum size is 100 bytes. If you specify fewer than 100 bytes, 4000 bytes is used as the default.
tcp_handle
Address of the client or server handle.
readit
Address of a user-written routine that reads data from the stream transport. This routine must use the following format:
int readit(tcp_handle, buffer, len)
u_char *tcp_handle;
u_char *buffer;
u_long len;
*tcp_handle is the client or server handle
*buffer is the
buffer to fill
len is the
number of bytes to read
The readit routine returns either the number of bytes read, or -1 if an error occurs.
writeit
Address of a user-written routine that writes data to the stream transport. This routine must use the following format:
int writeit(tcp_handle,
buffer, len)
u_char *tcp_handle;
u_char *buffer;
u_long len;
*tcp_handle is the client or server handle.
*buffer is the address of the buffer being written.
len is the number of bytes to write.
The writeit routine returns either the number of bytes written, or -1 if an error occurs.
Description
The xdrrec_create routine requires one of the following:
• The TCP transport
• A stream-oriented interface (such as file I/O) not supported by MultiNet. The stream consists of data organized into records. Each record is either an RPC request or reply.
The clnttcp_create and svcfd_create routines call the xdrrec_create routine.
See Also
clnttcp_create, svcfd_create, xdrrec_endofrecord, xdrrec_eof, xdrrec_skiprecord
Marks the end of a record.
Format
#include tcpip$rpc:xdr.h
bool_t xdrrec_endofrecord (XDR *xdrs, bool_t sendnow);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
sendnow
Indicates when the calling program will send the record to the writeit routine (see xdrrec_create).
If sendnow is TRUE, xdrrec_endofrecord sends the record now. If sendnow is FALSE, xdrrec_endofrecord writes the record to a buffer and sends the buffer when it runs out of buffer space.
Description
A client or server program calls the xdrrec_endofrecord routine when it reaches the end of a record it is writing. The program must call the xdrrec_create routine before calling xdrrec_endofrecord.
Diagnostics
This routine returns TRUE if it succeeds and FALSE if it fails.
See Also
xdrrec_create,
xdrrec_eof, xdrrec_skiprecord
Goes to the end of the current record, then verifies whether any more data can be read.
Format
#include tcpip$rpc:xdr.h
bool_t xdrrec_eof (XDR *xdrs);
Argument
xdrs
Address of the structure containing XDR encoding and decoding information.
Description
The client or server program must call the xdrrec_create routine before calling xdrrec_eof.
Diagnostics
This routine returns TRUE if it reaches the end of the data stream, and FALSE if it finds more data to read.
See Also
xdrrec_create,
xdrrec_endofrecord, xdrrec_skiprecord
Goes to the end of the current record.
Format
#include tcpip$rpc:xdr.h
bool_t xdrrec_skiprecord (XDR *xdrs);
Argument
xdrs
Address of the structure containing XDR encoding and decoding information.
Description
A client or server program calls the xdrrec_skiprecord routine before it reads data from a stream. This routine ensures that the program starts reading a record from the beginning.
The xdrrec_skiprecord routine is similar to the xdrrec_eof routine, except that xdrrec_skiprecord does not verify whether any more data can be read.
The client or server program must call the xdrrec_create routine before calling xdrrec_skiprecord.
Diagnostics
This routine returns TRUE if it has skipped to the start of a record. Otherwise, it returns FALSE.
See Also
xdrrec_create,
xdrrec_endofrecord, xdrrec_eof
Initializes a stdio XDR stream.
Format
#include tcpip$rpc:xdr.h
void xdrstdio_create (XDR *xdrs, FILE *file, enum xdr_op op);
Arguments
xdrs
Address of the structure containing XDR encoding and decoding information.
file
File pointer FILE *, which is to be associated with the stream.
op
An XDR operation, one of: XDR_ENCODE, XDR_DECODE, or XDR_FREE.
Description
The xdrstdio_create routine initializes an stdio stream for the specified file.