4.2.7 Execution environment option (-option)

This option specifies each execution environment option.

Format

-option [ colseq=col-def [ ,col-def ...] [ colseq= ...] ]
        [ fifo ]
        [ icode=input-code ]
        [ iconv=input-conversion ]
        [ memsize=memory-size ]
        [ msgfile=message-file ]
        [ msglevel=message-level ]
        [ supfile=startup-file ]
        [ tmpdir=temp-directory [ ,temp-directory ...] [ tmpdir= ...] ]

4.2.7.1 colseq operand

Specifies information required to modify the collation sequence.
If col is specified in the key field data format, selection field (compared field or comparison field) data format, with the modify collation sequence function, comparison is made in the collation sequence specified in the colseq operand. It is possible to specify information to modify multiple byte collating sequence values by separating the col-defs with commas (,).

Format

colseq=col-def [ ,col-def ...] [ colseq= ...]

col-def

This specifies information required to modify the collation sequence.
When describing in the format cctt and comparing records, if a field contains a byte equivalent to the hexadecimal number cc, that byte is evaluated as the hexadecimal number tt and compared. A hexadecimal number that is not specified in cc is not replaced and is compared as is.

Example

To compare the hexadecimal number 0x20 as the hexadecimal number 0x23, specify as follows:
```
colseq=2023
```
If a col-def is specified more than once for the same hexadecimal number, the last specification is valid. In the example below, the hexadecimal number 0x20 is compared as the hexadecimal number 0x3A.
```
colseq=2023,2040,203A
```
Specification when figures of ASCII code (0x30-0x39) are compared as a figure of the EBCDIC code (0xF0-0xF9) is as follows:
```
colseq=30F0,31F1,32F2,33F3,34F4,35F5,36F6,37F7,38F8,39F9
```

4.2.7.2 fifo operand

This specifies the first-in first-out (FIFO) option.
When the vales of the specified key fields are equal, the firstly input records are output first. This option is effective in the Sort option (-sort). When omitted, the output order is not necessarily first-in first-out. When specified at the same time as the Merge option (-merge), the Copy option (-copy), the Summation option (field operand), and the Suppression option (suppress operand), the specification of the first-in first-out (FIFO) option is disregarded.

Format

fifo

Note

A priority level specified of the FIFO option is as follows.

Environment variable BSORT_FIFO
fifo operand of -option option
BSORT_FIFO of startup file

4.2.7.3 icode operand

This specifies input code system options.

Format

icode=input-code

input-code

Specify input code system.
Select and specify one of the following.

input-code	Meaning
au	ASCII code system (NetCOBOL form) (Default value)
ac	ASCII code system (Micro Focus COBOL form)
eb	EBCDIC code system (Note)
u2	Unicode system (UCS-2 form)
u3	Unicode system (UTF-32 form)
u8	Unicode system (UTF-8 form)

NOTE

The EBCDIC code system can be specified for binary file.

4.2.7.4 iconv operand

This specifies conversion methods between ASCII code and EBCDIC code.

Information

In the following cases, the code conversion between ASCII code and the EBCDIC code is done according to the iconv operand.

Comparison of key field when data form of key field is EBCDIC code and when input file code system (icode operand) is ASCII code system.
Comparison of key field when data form of key field is ASCII code and when input file code system (icode operand) is EBCDIC code system.
Conversion of character strings of a literal value, when data form of selection field is EBCDIC code.
Conversion of character strings of a literal value, when data form of reconstruction field is EBCDIC code.

Format

iconv=input-conversion

input-conversion

Specify conversion methods between ASCII code and EBCDIC code.
Select and specify one of the following.

input-conversion	Meaning
0	It executes the code conversion between the EBCDIC code and US ASCII code. (Default value)
1	It executes the code conversion between the EBCDIC (Japanese kana) code and ASCII (JIS8) code.
2	It executes the code conversion between the EBCDIC (Lowercase letters) code and ASCII (JIS7) code.

4.2.7.5 memsize operand

Specifies the size of the work area (memory size) that PowerBSORT uses.
When the memsize operand is omitted or 0 is specified, if the BSORT_MEMSIZE of startup file is specified, the work area of the size is used and if it has not been specified, the work area is automatically set.

Format

memsize=memory-size

memory-size

This specifies the size of the work area that PowerBSORT uses in kilobytes.
Specify within the range from 64 to 2097151 kilobytes. When the value of 2097152 or more is specified, it is considered that 2097151 were specified.

Note

The memsize operand takes priority over BSORT_MEMSIZE of startup file.

4.2.7.6 msgfile operand

This specifies file path name to output the messages.
By default, if the BSORT_MSGFILE of startup file is specified, it is output to the file. If it is not specified, output the message to the standard error output.

Format

msgfile=message-file

message-file

This specifies file path name to output the messages.
To specify a file name that contains a space, enclose the whole name with double quotation marks (").

Note

The msgfile operand takes priority over BSORT_MSGFILE of startup file.
When the message file is specified, the information message is output to the message file. The warning message and the error message are output to the message file and the standard error output.

4.2.7.7 msglevel operand

Specify messages to be output.
When msglevel operand is omitted, BSORT_MSGLEVEL of startup file is used. When msglevel operand and the BSORT_MSGLEVEL of startup file specification are omitted, W (error message and warning message are output) is assumed for message-level. The output destination of the message is decided by specifying msgfile operand or the BSORT_MSGFILE of startup file.

Format

msglevel=message-level

message-level

The type of the output message is specified. The message level that can be specified are shown below.

message-level	Meaning
N	Nothing is output.
E	The error message is output.
W	The error message and the warning message are output. (Default value)
I	The error message, the warning message, and the information message are output.

The type of the output message used in previous versions can also be specified.

message-level	Meaning
0	Nothing is output.
1	The error message and the warning message are output. (Default value)
2	The error message, the warning message, and the information message are output.

Note

The errors of the command syntax and command help are output to the standard error output even when specifying that nothing is output.

The msglevel operand takes priority over BSORT_MSGLEVEL of startup file.

4.2.7.8 supfile operand

This specifies file path name of startup file.
By default, use "$HOME/.bsortrc" as a startup file. "$HOME" means PowerBSORT user's home directory.

Format

supfile=startup-file

startup-file

This specifies file path name of startup file.
To specify a file name that contains any blanks, enclose the whole name with double quotation marks (").

4.2.7.9 tmpdir operand

This specifies directory names of the temporary files.
Two or more directory names can be specified by delimiting with comma (,). Moreover, two or more tmpdir operands can be specified. When two or more directories are specified, the temporary file is distributed to the specified directory.

Format

tmpdir=temp-directory [ ,temp-directory ...] [ tmpdir= ...]

temp-directory

This specifies directory names of the temporary file.
When the directory names contain any blanks, enclose it with double quotation marks (").

Note

Refer to "Priority level of the directory that creates the temporary file" for the priority level of the directory specification that creates the temporary file.
When a wrong directory name is specified, the file not intended might be deleted.