Jack Christensen
743b98b298
Though this doesn't follow Go naming conventions exactly it makes names more consistent with PostgreSQL and it is easier to read. For example, TIDOID becomes TidOid. In addition this is one less breaking change in the move to V3.
51 lines
1.4 KiB
Go
51 lines
1.4 KiB
Go
package pgtype
|
|
|
|
import (
|
|
"io"
|
|
)
|
|
|
|
// Xid is PostgreSQL's Transaction ID type.
|
|
//
|
|
// In later versions of PostgreSQL, it is the type used for the backend_xid
|
|
// and backend_xmin columns of the pg_stat_activity system view.
|
|
//
|
|
// Also, when one does
|
|
//
|
|
// select xmin, xmax, * from some_table;
|
|
//
|
|
// it is the data type of the xmin and xmax hidden system columns.
|
|
//
|
|
// It is currently implemented as an unsigned four byte integer.
|
|
// Its definition can be found in src/include/postgres_ext.h as TransactionId
|
|
// in the PostgreSQL sources.
|
|
type Xid pguint32
|
|
|
|
// ConvertFrom converts from src to dst. Note that as Xid is not a general
|
|
// number type ConvertFrom does not do automatic type conversion as other number
|
|
// types do.
|
|
func (dst *Xid) ConvertFrom(src interface{}) error {
|
|
return (*pguint32)(dst).ConvertFrom(src)
|
|
}
|
|
|
|
// AssignTo assigns from src to dst. Note that as Xid is not a general number
|
|
// type AssignTo does not do automatic type conversion as other number types do.
|
|
func (src *Xid) AssignTo(dst interface{}) error {
|
|
return (*pguint32)(src).AssignTo(dst)
|
|
}
|
|
|
|
func (dst *Xid) DecodeText(src []byte) error {
|
|
return (*pguint32)(dst).DecodeText(src)
|
|
}
|
|
|
|
func (dst *Xid) DecodeBinary(src []byte) error {
|
|
return (*pguint32)(dst).DecodeBinary(src)
|
|
}
|
|
|
|
func (src Xid) EncodeText(w io.Writer) (bool, error) {
|
|
return (pguint32)(src).EncodeText(w)
|
|
}
|
|
|
|
func (src Xid) EncodeBinary(w io.Writer) (bool, error) {
|
|
return (pguint32)(src).EncodeBinary(w)
|
|
}
|