Data in D records for n columns is placed side by side:
For each column, the data consists of a null indicator followed by the data itself. If nulls are allowed (byte 22 of C record = Y), then bytes 44-49 of each C record points to the null indicator that precedes the data for that column. If byte 22 = N (nulls not allowed), then bytes 44-49 points to the data itself. However, in the latter case, space for the null indicator is left in the data record. The first position in bytes 44-49 is represented by a value of 1, which points to byte 6 of a D record (bytes 1 through 5 are ignored).
The representation of the null indicator depends on what is specified for OUTPUTMODE: character or binary. OUTPUTMODE is reflected in byte 43 of the T record: C for character or M for machine (binary). When data format is character, one byte is used for the null indicator:
See Figure 71 for an illustration of two D records showing data that is null in one case and not null in the other.
When the data format is binary, two bytes are used for the null indicator:
Figure 72 which shows two D records, illustrating both null and non-null data indicators.
[ Previous Page | Next Page | Contents | Index ]