op_pk_fields_convert_from_vvec()
Abstract
Iterates through a formatted packet’s fields, calling each field’s conversion method.
Syntax
op_pk_fields_convert_from_vvec (pk_ptr, vvec, vvec_index)
| Argument | Type | Description |
|---|---|---|
pk_ptr |
Packet* |
pointer to a formatted packet [Typically this formatted packet has just been converted to a given format from a value vector packet.] |
vvec |
Vvec_vector |
value vector to use when setting the packet’s fields [Typically this is a value vector contained in a value vector packet that is being converted to a formatted packet.] |
vvec_index |
int |
index of the position in the value vector containing the packet’s first bit |
Return Type
Compcode
Completion code indicating whether the conversion completed successfully.
[This value can be compared to the two symbolic constants OPC_COMPCODE_SUCCESS and OPC_COMPCODE_FAILURE.]
Example
/* A value vector representing an entire TCP packet has */
/* been compeletly received on an esys interface. */
tcp_pkptr = op_pk_create_fmt ("tcp_seg");
/* Have each field's conversion-from-vvec method called */
/* to set values in the opnet TCP packet. */
op_pk_convert_fields_from_vvec (tcp_pkptr, vvec, 0);
/* The opnet TCP packet models the checksum field as */
/* "information", but the value vector for this */
/* packet contains a real checksum value. */
/* Test the checksum value... */
Details
This function, typically called by a packet conversion method, uses a value vector to systematically set each field in a formatted packet based upon the contents of a value vector. It iterates through each field using the fields’ conversion order attributes to determine the order in which the fields are converted. When setting a field, the KP calls the conversion specified by the field’s conversion method attribute.
WARNING: The field conversion routines of a newly created packet format are set to “NONE”, which does nothing. Before calling this function, the required conversion routines must be set to either a user-supplied routine or one of the default routines provided with Riverbed Modeler (these are applicable only to a small subset of packet formats; see Chapter 5 “Packet Conversion” on page 29.
op_pk_fields_convert_from_vvec() passes four arguments to the field conversion method:
- a pointer to a packet
- a field index
- a value vector of type bit (Vvec_Vector)
- an index for the position of the field in the value vector. The Simulation Kernel calculates this position using the bit sizes of the fields defined in the packet format.
Note: The Simulation Kernel may not always be able to calculate the exact position. In such cases, it passes OPC_FIELD_BIT_POSE_UNKNOWN to the conversion method.)
The conversion method prototype is:
typedef int
default_convert_field_from_vvec (Packet * pkptr, int fd_index,
Vvec_Vector vvec, int pos)
This function must be in a .pff.c file named to match the base part of the function name. For example, a function named my_convert_field_from_vvec() should be defined in the file my_convert.pff.c.
Purpose
Provides a mechanism to iterate through a packet’s fields, calling their conversion methods.
Errors
- Program Abort: Packet pointer references unowned packet (
<pk_id>). - Program Abort: Packet format (
<packet_format>) does not support converison. - Program Abort: Packet ID (
<pk_id>) can not be converted. (packet is unformatted) - Recoverable Error: Value vector is not owned by client.
- Recoverable Error: Value vector is invalid.
- Program Abort: Conversion of packet ID (
<pk_id>) from value vector form to field form failed.
Reference
- Use the KP op_pk_type() to determine the type of a packet
- Use the KP op_pk_vvec_is_set() to see if a value vector packet contains a value vector
- Use the KP op_pk_convert_from_vvec() to convert a value vector packet to a formatted packet
- Use the KP op_pk_fields_convert_to_vvec() to use the field conversion orders and methods specified in the packet format when converting a formatted packet to a value vector packet