Hitachi MK-96RD647-01 manual FXotm with Variable-Length Record Format

FXotm with Variable-Length Record Format

FXotm operations can be performed on variable-length source files only if delimiters have already been added to the source file (e.g., from a previous FXmto operation). If a variable-length source file without delimiters is processed, FCU will use the maximum record length to construct the target data entities, thereby corrupting the data and rendering the dataset unusable. FCU extracts but does not add delimiters to FXotm source files.

With delimiters. The figure below shows an FXotm operation for a variable- length source file with delimiters. FCU extracts and transfers the data entities to the target dataset, and automatically adds the four-byte RL field. The delimiters are not transferred. The target dataset must have variable-length record format.

Open-System Source File

Record length – 2 or 1 byte
Data entity 1	D
Data entity 2	D	→
Data entity 3	D
Data entity + 1 or 2 bytes

Target Dataset

	Record length
RL	Record 1: Data entity 1
RL	Record 2: Data entity 2
RL	Record 3: Data entity 3

Figure 2-19 FXotm with Variable-Length Records: Delimiters

If the length of any data entity in a UNIX source file is greater than the maximum record length minus one byte (CR or LF delimiter), FCU aborts the operation and reports an error. If the length of any data entity in a Windows 2000/2003/Windows NT source file is greater than the maximum record length minus two bytes (CR+LF delimiter), FCU aborts the operation and reports an error.

2-28 About Cross-OS File Exchange Operations

Hitachi USP V Cross-OS File Exchange User’s Guide