ProSort Control Statements

This chapter describes control statements available in ProSort scripts.

1. Overview

ProSort supports the following control statements.

Sorting

Statement Description

SORT

Performs a sort or copy operation.

MERGE

Performs a merge operation.

OPTION

Overrides and sets some parameter values.
Record Filtering

Statement Description

INCLUDE

Includes specific records.

OMIT

Omits specific records.

OUTFIL

Creates new records by applying a condition to records resulting from an operation.

Statement	Description
SORT	Performs a sort or copy operation.
MERGE	Performs a merge operation.
OPTION	Overrides and sets some parameter values.

Statement	Description
INCLUDE	Includes specific records.
OMIT	Omits specific records.
OUTFIL	Creates new records by applying a condition to records resulting from an operation.

Record Reformatting

Statement	Description
INREC	Reformats input records into new records with a specified format before an operation.
OUTREC	Reformats input records into new records with a specified format after an operation.
OUTFIL	Reformats records resulting from an operation.

Statement

Description

INREC

Reformats input records into new records with a specified format before an operation.

OUTREC

Reformats input records into new records with a specified format after an operation.

OUTFIL

Reformats records resulting from an operation.

Others

Statement	Description
ALTSEQ	Changes the sorting order of particular characters by using an ALTSEQ conversion table. SORT, MERGE, INCLUDE, OMIT: AQ format is used. INREC, OUTREC, OUTFIL OUTREC: TRAN=ALTSEQ is used.
COLSEQ	Allows users to change a sorting order.
DATASIZE	Sets an expected input data size.
DEFREC	Sets a record format.
END	Ends a ProSort script, and ignores statements written after the END statement.
INFILE	Sets an input file’s path and expected size.
INSERT	Inserts a specified file into a script.
MEMORY	Sets the memory size used when performing operations.
MODS	Registers a function as User Exit Function by using a ProSort script.
SUM	Outputs one of sorted or merged records that have the same key field.
WORKSPACE	Specifies directories to save temporary files used for external sorting.

Statement

Description

ALTSEQ

Changes the sorting order of particular characters by using an ALTSEQ conversion table.

SORT, MERGE, INCLUDE, OMIT: AQ format is used.
INREC, OUTREC, OUTFIL OUTREC: TRAN=ALTSEQ is used.

COLSEQ

Allows users to change a sorting order.

DATASIZE

Sets an expected input data size.

DEFREC

Sets a record format.

END

Ends a ProSort script, and ignores statements written after the END statement.

INFILE

Sets an input file’s path and expected size.

INSERT

Inserts a specified file into a script.

MEMORY

Sets the memory size used when performing operations.

MODS

Registers a function as User Exit Function by using a ProSort script.

SUM

Outputs one of sorted or merged records that have the same key field.

WORKSPACE

Specifies directories to save temporary files used for external sorting.

2. ALTSEQ

Changes the sorting order of particular characters by using an ALTSEQ conversion table. This is used only to adjust a sorting order for sort and merge operations, and has no influence on actual output characters.

The ALTSEQ conversion table can be used as follows:

For SORT, MERGE, INCLUDE, and OMIT

The AQ format is used. Existing data is converted through ALTSEQ and then applied to an operation. Key orders can be changed, and different filtering conditions can be used.
For INREC, OUTREC, and OUTFIL OUTREC

Characters are converted with TRAN=ALTSEQ. Existing data is converted through ALTSEQ, and the converted data is applied to the next process.

The details of ALTSEQ are as follows:

Syntax

ALTSEQ
Component

Component Description

ff

Hexadecimal character to change.

tt

New hexadecimal character.
Example

The following performs a SORT operation after changing the character X'5B' to X’EA'.
```
SORT FILEDS=(1,8,AQ,A)
ALTSEQ CODE=(5BEA)
```
The following performs a MERGE operation after changing the character X'5B' to X’EA'.
```
MERGE FIELDS=(1,8,AQ,A)
ALTSEQ=(5BEA)
```
The following performs a COPY operation and outputs bytes 1 through 80 of each record. The character X'00' is changed to X'40'.
```
SORT FIELDS=(COPY)
ALTSEQ CODE=(0040)
OUTREC FIELDS=(1,80,TRAN=ALTSEQ)
```

Component	Description
ff	Hexadecimal character to change.
tt	New hexadecimal character.

3. COLSEQ

Allows users to change a sorting order. Character strings are compared to each other while a SORT or MERGE operation is performed.

Unlike comparing integers (for example, 1 < 2), comparing character strings is not easy and requires a sorting order rule such as ASCII and EBCDIC. For example, '9' < 'A' in ASCII code, but 'A' < '9' in EBCDIC code. Users can define such a sorting order rule by using COLSEQ.

The details of COLSEQ are as follows:

Syntax

COLSEQ

Component

Component	Description
DEFAULT	Default sorting order for CH fields.
CQ1	Sorting order for CQ1 fields.
CQ2	Sorting order for CQ2 fields.
CQ3	Sorting order for CQ3 fields.
CQ4	Sorting order for CQ4 fields.
ASCII	Uses ASCII code for sorting.
EBCDIC	Uses EBCDIC code for sorting.
HIGH	Characters in which a sorting order is not specified have greater collation values than characters in which a sorting order is specified, which means that the former characters have lower priority.
LOW	Characters in which a sorting order is not specified have less collation values than characters in which a sorting order is specified, which means that the former characters have higher priority.
c_string	Either C’x₁' or C’x₁x₂', where x₁ and x₂ are arbitrary characters (except for some special characters such as a single quotation mark).
x_string	Either X’x₁x₂' or X’x₁x₂x₃x₄', where x₁, …, and x₄ are hexadecimal characters.
MODIFICATION	Specifies that the collation value of a left-side character is equal to, less than, or greater than that of a right-side character.
NULLCHAR	Characters to be ignored during a sort operation. A sorting key is created without the specified characters.
NONPRINTABLE	Non-displayed control characters.
NONDICTIONARY	All characters except for alphabets, digits, blank spaces, and tabs.
FOLDED	All lowercase alphabetic characters have the same collation value as corresponding uppercase alphabetic characters.

Description

DEFAULT

Default sorting order for CH fields.

CQ1

Sorting order for CQ1 fields.

CQ2

Sorting order for CQ2 fields.

CQ3

Sorting order for CQ3 fields.

CQ4

Sorting order for CQ4 fields.

ASCII

Uses ASCII code for sorting.

EBCDIC

Uses EBCDIC code for sorting.

HIGH

Characters in which a sorting order is not specified have greater collation values than characters in which a sorting order is specified, which means that the former characters have lower priority.

LOW

Characters in which a sorting order is not specified have less collation values than characters in which a sorting order is specified, which means that the former characters have higher priority.

c_string

Either C’x₁' or C’x₁x₂', where x₁ and x₂ are arbitrary characters (except for some special characters such as a single quotation mark).

x_string

Either X’x₁x₂' or X’x₁x₂x₃x₄', where x₁, …, and x₄ are hexadecimal characters.

MODIFICATION

Specifies that the collation value of a left-side character is equal to, less than, or greater than that of a right-side character.

NULLCHAR

Characters to be ignored during a sort operation. A sorting key is created without the specified characters.

NONPRINTABLE

Non-displayed control characters.

NONDICTIONARY

All characters except for alphabets, digits, blank spaces, and tabs.

FOLDED

All lowercase alphabetic characters have the same collation value as corresponding uppercase alphabetic characters.

Example
- Order specification
  
  Users can list characters, or specify a character range with the same type and length.
  
  The followings are available.
  C'a'-C'z' C'AA'-C'AB' X'010f'-X'010f' X'10'-X'20'
  The followings are unavailable.
  C'a'-X'0f' C'A'-C'각'
- MODIFICATION
  
  In the following example, the character 'a' has the same collation value with the character 'z'.
  COLSEQ CQ1 ASCII MODIFICATION C'a' = C'z'
  In the following example, the collation value of 'a' is greater than that of 'd' but less than that of 'e'.
  COLSEQ CQ1 ASCII MODIFICATION C'a' > C'd'
  For example, a, b, c, d, e, and f in a record are sorted to b, c, d, a, e, and f in ascending order.
  
  The following example has the same result as the previous example. The collation value of 'a' is less than that of 'e' but greater than that of 'd'.
  COLSEQ CQ1 ASCII MODIFICATION C'a' < C'e'
- NULLCHAR
  
  In the following example, the character 'a' is excluded from sorting.
  COLSEQ CQ1 ASCII NULLCHAR C'a'
  For example, aE, Da, aCa, aaBa, and aaAaa in a record are sorted to aaAaa, aaBa, aCa, Da, and aE in ascending order.
  
  If nothing of NULLCHAR, HIGH, and LOW is specified in COLSEQ, all characters that are not specified in COLSEQ will be automatically specified as NULLCHAR.
- FOLDED
  
  The following sorts data in the order of a - z and 0 - 9. Characters in which a sorting order is not specified have greater collation values than characters in which a sorting order is specified (lowercase alphanumeric characters).
  SORT FILEDS=(1,8,CQ1,A) COLSEQ CQ1 C'a'-C'z', C'0'-C'9' HIGH
  The following sorts data in the order of X'010F' - X'01FF' (double-byte characters). Lowercase alphabetic characters are ignored. Characters in which a sorting order is not specified have less collation values than characters in which a sorting order is specified (X'010F' - X'01FF' and lowercase alphabetic characters).
  SORT FILEDS=(1,4,CQ2,A) COLSEQ CQ2 X'010F'-X'01FF' NULLCHAR C'a'-C'z' LOW

4. DATASIZE

Sets an expected input data size. The size is specified only when calling a script by using ProSort API.

The details of DATASIZE are as follows:

Syntax

DATASIZE
Component

Component Description

n

Expected input data size in n bytes.

nK

Expected input data size in n KB.

nM

Expected input data size in n MB.

nG

Expected input data size in n GB.
Example

The following sets the expected input data size to 100 MB.
```
DATASIZE 100M
```

Component	Description
n	Expected input data size in n bytes.
nK	Expected input data size in n KB.
nM	Expected input data size in n MB.
nG	Expected input data size in n GB.

5. DEFREC

Sets a record format.

The details of DEFREC are as follows:

Syntax

DEFREC
Component

Component Description

SIZE

Input record length.

DELIMITER

Record delimiter. Only CR (carriage return) is supported currently.
Example

The following sets a fixed-length record with a size of 10.
```
DEFREC FIXED, SIZE = 10
```
The following sets a variable-length record, and CR is used as a delimiter.
```
DEFREC VARIABLE, DELIMITER = CR
```

Component	Description
SIZE	Input record length.
DELIMITER	Record delimiter. Only CR (carriage return) is supported currently.

6. END

Ends a ProSort script execution.

The details of END are as follows:

Syntax

END
Component

Component Description

END

Ends a ProSort script execution. Statements written after the END statement in the script are not executed.

Component	Description
END	Ends a ProSort script execution. Statements written after the END statement in the script are not executed.

7. INCLUDE, OMIT

Includes or omits specific records.

The details of INCLUDE and OMIT are as follows:

Syntax

INCLUDE

OMIT

INCLUDE, OMIT - condition

Component

Component	Description
condition	Condition used to include records in the INCLUDE statement or used to exclude records in the OMIT statement.
f	Default format used when f₁ or f₂ is omitted.
ALL	Includes all records in the INCLUDE statement and omits all records in the OMIT statement.
NONE	Includes no records in the INCLUDE statement and omits no records in the OMIT statement.
logical_expression	Logical expression that specifies a condition used to include or omit records. A logical expression consists of one or more relational expressions combined with AND (or &), OR (or \|), and parentheses. The priorities of the operators are as follows: parentheses > AND (or &) > OR (or \|). The leftmost relational expression is evaluated first. In a logical expression with relational expressions combined with AND (or &), the result is true when all the relational expressions are true. In a logical expression with relational expressions combined with OR (or \|), the result is true when one or more relational expressions are true.
comparison	Relational expression that compares two fields (field 1 and field 2). Field 1 is the input record’s p₁th to (p₁ + m₁ - 1)th bytes in the f₁ format. If f₁ is not specified, f is used. Field 2 is the input record’s p₂th to (p₂ + m₂ - 1)th bytes in the f₂ format or a constant. If f₂ is not specified, f is used. The following operators are available. EQ: true if field 1 is equal to field 2. NE: true if field 1 is not equal to field 2. GT: true if field 1 is greater than field 2. GE: true if field 1 is greater than or equal to field 2. LT: true if field 1 is less than field 2. LE: true if field 1 is less than or equal to field 2.
constant	Constant used for comparison. For more information, refer to Constants.
bit_logic_test	Relational expression for bit logical evaluation with input record’s p-th to (p + m - 1)th bytes in the BI format. The evaluation is performed as follows: Uses a bit mask (bit_mask) or bit pattern (bit_pattern). The bit mask is a hexadecimal (X’x₁…x_n') or binary (B’x₁…x_n') string. Each bit that is on (1) in the bit mask is compared with each bit in a corresponding position in an input field. The following operators are available. BO or ALL: true if all target bits are on. BM or SOME: true if some (not all) target bits are on. BZ or NONE: true if no target bits are on. BNO or NOTALL: true if some or no target bits are on. BNM or NOTSOME: true if all or no target bits are on. BNZ or NOTNONE: true if all or some target bits are on. The bit pattern is a string in the format of B’x₁…x_n', where x₁, …, and x_n is 1, 0, or a period (.). Each bit that is on (1) and off (0) in a bit pattern is compared with each bit in a corresponding position in an input field. The following operators are available. EQ: true if each bit state (on or off) in the bit pattern is the same as that in an input field. NE: true if each bit state (on or off) in the bit pattern is not the same as that in an input field.
substring_ comparison_test	Relational expression for comparing a string constant and a CH field value at p-th to (p + m - 1)th bytes of an input record. The comparison is performed as follows: If m is greater than the length of the string constant, the input field value will be searched for the constant. If not, the string constant will be searched for the input field value. The following operators are available. EQ: true if the string constant and the input field value match is found. NE: true if the string constant and the input field value match is not found.
numeric_test	Relational expression for deciding whether a field (input record’s p-th to (p + m - 1)th bytes) is a numeric type. x1y1...xmym If the field is expressed as a previous hexadecimal value and the operator is EQ, the expression result is as follows: When f is CSF (or FS) or FORMAT=CSF (or FS) is specified, the result is true if x₁=…=x_m=3 and 0≤y₁,…,y_m≤9 or false if not. When f is ZD or FORMAT=ZD is specified, the result is true if x₁=…=x_m-1=3, x_m∈{3,4,5,7}, and 0≤y₁,…,y_m≤9 or false if not. When f is PD or FORMAT=PD is specified, the result is true if 0≤x₁,…,x_m≤9, 0≤y₁,…,y_m-1≤9, and y_m∈{C,D,F} or false if not. If the operator is NE, the result is contrary to that when the operator is EQ.

Description

condition

Condition used to include records in the INCLUDE statement or used to exclude records in the OMIT statement.

Default format used when f₁ or f₂ is omitted.

ALL

Includes all records in the INCLUDE statement and omits all records in the OMIT statement.

NONE

Includes no records in the INCLUDE statement and omits no records in the OMIT statement.

logical_expression

Logical expression that specifies a condition used to include or omit records.

A logical expression consists of one or more relational expressions combined with AND (or &), OR (or |), and parentheses.

The priorities of the operators are as follows: parentheses > AND (or &) > OR (or |). The leftmost relational expression is evaluated first. In a logical expression with relational expressions combined with AND (or &), the result is true when all the relational expressions are true. In a logical expression with relational expressions combined with OR (or |), the result is true when one or more relational expressions are true.

comparison

Relational expression that compares two fields (field 1 and field 2).

Field 1 is the input record’s p₁th to (p₁ + m₁ - 1)th bytes in the f₁ format. If f₁ is not specified, f is used.
Field 2 is the input record’s p₂th to (p₂ + m₂ - 1)th bytes in the f₂ format or a constant. If f₂ is not specified, f is used.

The following operators are available.
- EQ: true if field 1 is equal to field 2.
- NE: true if field 1 is not equal to field 2.
- GT: true if field 1 is greater than field 2.
- GE: true if field 1 is greater than or equal to field 2.
- LT: true if field 1 is less than field 2.
- LE: true if field 1 is less than or equal to field 2.

constant

Constant used for comparison. For more information, refer to Constants.

bit_logic_test

Relational expression for bit logical evaluation with input record’s p-th to (p + m - 1)th bytes in the BI format.

The evaluation is performed as follows:

Uses a bit mask (bit_mask) or bit pattern (bit_pattern).
The bit mask is a hexadecimal (X’x₁…x_n') or binary (B’x₁…x_n') string. Each bit that is on (1) in the bit mask is compared with each bit in a corresponding position in an input field.

The following operators are available.
- BO or ALL: true if all target bits are on.
- BM or SOME: true if some (not all) target bits are on.
- BZ or NONE: true if no target bits are on.
- BNO or NOTALL: true if some or no target bits are on.
- BNM or NOTSOME: true if all or no target bits are on.
- BNZ or NOTNONE: true if all or some target bits are on.
The bit pattern is a string in the format of B’x₁…x_n', where x₁, …, and x_n is 1, 0, or a period (.). Each bit that is on (1) and off (0) in a bit pattern is compared with each bit in a corresponding position in an input field.

The following operators are available.
- EQ: true if each bit state (on or off) in the bit pattern is the same as that in an input field.
- NE: true if each bit state (on or off) in the bit pattern is not the same as that in an input field.

substring_ comparison_test

Relational expression for comparing a string constant and a CH field value at p-th to (p + m - 1)th bytes of an input record.

The comparison is performed as follows:

If m is greater than the length of the string constant, the input field value will be searched for the constant. If not, the string constant will be searched for the input field value.

The following operators are available.
- EQ: true if the string constant and the input field value match is found.
- NE: true if the string constant and the input field value match is not found.

numeric_test

Relational expression for deciding whether a field (input record’s p-th to (p + m - 1)th bytes) is a numeric type.

x1y1...xmym

If the field is expressed as a previous hexadecimal value and the operator is EQ, the expression result is as follows:

When f is CSF (or FS) or FORMAT=CSF (or FS) is specified, the result is true if x₁=…=x_m=3 and 0≤y₁,…,y_m≤9 or false if not.
When f is ZD or FORMAT=ZD is specified, the result is true if x₁=…=x_m-1=3, x_m∈{3,4,5,7}, and 0≤y₁,…,y_m≤9 or false if not.
When f is PD or FORMAT=PD is specified, the result is true if 0≤x₁,…,x_m≤9, 0≤y₁,…,y_m-1≤9, and y_m∈{C,D,F} or false if not.

If the operator is NE, the result is contrary to that when the operator is EQ.

Example

The following includes records in which bytes 5 through 12 are greater than bytes 13 through 20 or bytes 105 through 108 are less than 1000.
```
INCLUDE COND=(5,8,GT,13,8,|,105,4,LE,1000),FORMAT=CSF
```
The following includes records in which bytes 11 through 16 contain the string OK or bytes 21 through 23 contain the string J69.
```
INCLUDE FORMAT=SS,COND(11,6,EQ,C'OK',OR,21,3,EQ,C'J69')
```

8. INFILE

Sets an input file’s path and expected size.

The details of INFILE are as follows:

Syntax

INFILE
Component

Component Description

file_path

Input file path.

file_size

Expected input file size.
Example

The following sets the input file’s path and expected size to IN and 1 MB respectively.
```
INFILE=(IN,1M)
```

Component	Description
file_path	Input file path.
file_size	Expected input file size.

9. INREC, OUTREC

Reformats input records into new records with a specified format. INREC is performed before SORT, MERGE, and COPY operations. OUTREC is performed after the operations.

The details of INREC and OUTREC are as follows:

Syntax

INREC

OUTREC

INREC, OUTREC - build_item

INREC, OUTREC - overlay_item

INREC, OUTREC - findrep_clause

INREC, OUTREC - ifthen_clause

Component

Component	Description
PARSE	Refer to parse_item.
CONVERT	Converts a variable-length input record to a fixed-length output record.
c	Field position. Unused spaces followed by the separation field are padded with blanks. For example, specifying 30:C’Tmax' pads the columns 1 through 29 with blanks and inserts 'Tmax' into the columns 30 through 33. Specifying 20:5,1 pads the columns 1 through 19 with blanks and inserts the input field’s byte 5 into the column 20.
s	Outputs one of the following separation fields. nX: n blanks (X'20'). If n is not specified, the default value is 1. nZ: n nulls (X'00'). If n is not specified, the default value is 1. nC'…': n strings enclosed in single quotation marks. If n is not specified, the default value is 1. nX'…': n hexadecimal strings enclosed in single quotation marks. If n is not specified, the default value is 1. DATE=(abcd): current date in the format of C’adbdc', where a,b,c∈{M,D,Y,4} and d is a delimiter. M: month (01-12) D: day (01-31) Y: last two digits of a year 4: four-digit year (0001-9999) &DATE=(abcd): same as DATE=(abcd). DATE or &DATE: same as DATE=(MDY/). DATENS=(abc): current date in the format of C’abc', where a,b,c∈{M,D,Y,4}. M: month (01-12) D: day (01-31) Y: last two digits of a year 4: four-digit year (0001-9999) &DATENS=(abc): same as DATENS=(abc). n/: n record delimiters. If n is not specified, the default value is 1.
%nn	Specifies the parsed field of the %nn label.
p,m	Specifies field position (p) and length (m). The last field’s length can be omitted. If the length is omitted, the last field ends at the end of the record.
a	Specifies field alignment. The following values are available. H: Halfword aligned F: Fullword aligned D: Doubleword aligned
HEX	Converts a value into a hexadecimal value. For example, a string AB is converted into C1C2.
TRAN= [ LTOU \| UTOL \| ALTSEQ ]	If TRAN=LTOU, lowercase letters are converted to uppercase letters. If TRAN=UTOL, uppercase letters are converted to lowercase letters. If TRAN=ALTSEQ, a defined ALTSEQ table is converted.
lookup	Finds a specific string in a field, and then converts the field value according to the result. v: length of the changed field value. find: string to find in the format of C’x₁…x_n', X’x₁…x_n' or B’x₁…x_n'. set₁: new value of the field if the string have been found. The new value is a constant in the format of C’x₁…x_n' or X’x₁…x_n', or a field in the format of p, m. set₂: new value of the field if it failed to find the string. The new value is a constant in the format of C’x₁…x_n' or X’x₁…x_n', or a field in the format of p, m.
arexp	Operation result.
arexp_term	Operand. p,m,f: field position, length, and format. n: integer value.
arexp_operator	Operator. The result of each operator is as follows: MIN: value of an operand with the smallest value. MAX: value of an operand with the greatest value. MUL: product of the operands. DIV: value of dividing the left operand by the right operand. MOD: remainder of dividing the left operand by the right operand. ADD: sum of the operands. SUB: value of subtracting the right operand from the left operand. Operators' priority is as follows: MIN, MAX MUL, DIV, MOD ADD, SUB
edit	Changes a field with a desired format and length. For more information, refer to [app_mask].
to	Changes a field with a desired format (fo) and length (n).
seqnum	Outputs sequence numbers according to defined rules. n: Specifies the number of digits for the sequence number. fs: Specifies the string format used to represent the sequence number. START=j: Sets the starting value of the sequence number to j. If omitted, START defaults to 1. INCR=i: Sets the increment value of the sequence number to i. If omitted, INCR defaults to 1. RESTART=(p,m): Resets SEQNUM to 1 when the specified field does not match the corresponding field of the previous record.
findrep_clause	Finds incon in a record and changes it to outcon. incon: string to find in the format of C’x₁…x_n', nC’x₁…x_n', X’x₁…x_n', or nX’x₁…x_n'. outcon: string to output in the format of C'', C’x₁…x_n', nC’x₁…x_n', X’x₁…x_n', or nX’x₁…x_n'. p: search start position. q: search end position.
WHEN=INIT, [ BUILD \| OVERLAY \| FINDREP ]	Performs BUILD, OVERLAY, or FINDREP operations on all input records. An IFTHEN clause that includes WHEN=INIT is processed before other IFTHEN clauses.
WHEN=(logical_expression), [ BUILD \| OVERLAY \| FINDREP ]	Performs BUILD, OVERLAY, or FINDREP operations on input records that meet the specific condition (logical_expression).
WHEN=GROUP [,BEGIN=(logical_expression) \| ,END=(logical_expression) \| ,RECORDS=n] ,PUSH=(items)	The BEGIN, END, and RECORDS operations are applied to all input records. When the specified conditions are met, records are assigned to or released from a GROUP. BEGIN=(logical_expression): Assigns a GROUP starting from the record that satisfies the logical_expression. END=(logical_expression): Releases the GROUP after the record that satisfies the logical_expression. RECORDS=n: Ends the GROUP when the number of records in a single GROUP reaches n. PUSH=(items): Overlays the values specified in items when the record belongs to a GROUP. The available item types are as follows: c:p,m: Overlays the value corresponding to p,m from the first record of the GROUP at position c. c:ID=n: Represents a GROUP ID of length n and overlays the current GROUP ID at position c. c:SEQ=n: Represents the record sequence within a GROUP with length n and overlays the current record sequence at position c.

Description

PARSE

Refer to parse_item.

CONVERT

Converts a variable-length input record to a fixed-length output record.

Field position.

Unused spaces followed by the separation field are padded with blanks. For example, specifying 30:C’Tmax' pads the columns 1 through 29 with blanks and inserts 'Tmax' into the columns 30 through 33. Specifying 20:5,1 pads the columns 1 through 19 with blanks and inserts the input field’s byte 5 into the column 20.

Outputs one of the following separation fields.

nX: n blanks (X'20'). If n is not specified, the default value is 1.
nZ: n nulls (X'00'). If n is not specified, the default value is 1.
nC'…': n strings enclosed in single quotation marks. If n is not specified, the default value is 1.
nX'…': n hexadecimal strings enclosed in single quotation marks. If n is not specified, the default value is 1.
DATE=(abcd): current date in the format of C’adbdc', where a,b,c∈{M,D,Y,4} and d is a delimiter.
- M: month (01-12)
- D: day (01-31)
- Y: last two digits of a year
- 4: four-digit year (0001-9999)
&DATE=(abcd): same as DATE=(abcd).
DATE or &DATE: same as DATE=(MDY/).
DATENS=(abc): current date in the format of C’abc', where a,b,c∈{M,D,Y,4}.
- M: month (01-12)
- D: day (01-31)
- Y: last two digits of a year
- 4: four-digit year (0001-9999)
&DATENS=(abc): same as DATENS=(abc).
n/: n record delimiters. If n is not specified, the default value is 1.

%nn

Specifies the parsed field of the %nn label.

p,m

Specifies field position (p) and length (m).

The last field’s length can be omitted. If the length is omitted, the last field ends at the end of the record.

Specifies field alignment.

The following values are available.

H: Halfword aligned
F: Fullword aligned
D: Doubleword aligned

HEX

Converts a value into a hexadecimal value. For example, a string AB is converted into C1C2.

TRAN= [ LTOU | UTOL | ALTSEQ ]

If TRAN=LTOU, lowercase letters are converted to uppercase letters. If TRAN=UTOL, uppercase letters are converted to lowercase letters.
If TRAN=ALTSEQ, a defined ALTSEQ table is converted.

lookup

Finds a specific string in a field, and then converts the field value according to the result.

v: length of the changed field value.
find: string to find in the format of C’x₁…x_n', X’x₁…x_n' or B’x₁…x_n'.
set₁: new value of the field if the string have been found. The new value is a constant in the format of C’x₁…x_n' or X’x₁…x_n', or a field in the format of p, m.
set₂: new value of the field if it failed to find the string. The new value is a constant in the format of C’x₁…x_n' or X’x₁…x_n', or a field in the format of p, m.

arexp

Operation result.

arexp_term

Operand.

p,m,f: field position, length, and format.
n: integer value.

arexp_operator

Operator.

The result of each operator is as follows:

MIN: value of an operand with the smallest value.
MAX: value of an operand with the greatest value.
MUL: product of the operands.
DIV: value of dividing the left operand by the right operand.
MOD: remainder of dividing the left operand by the right operand.
ADD: sum of the operands.
SUB: value of subtracting the right operand from the left operand.

Operators' priority is as follows:

MIN, MAX
MUL, DIV, MOD
ADD, SUB

edit

Changes a field with a desired format and length. For more information, refer to [app_mask].

Changes a field with a desired format (fo) and length (n).

seqnum

Outputs sequence numbers according to defined rules.

n: Specifies the number of digits for the sequence number.
fs: Specifies the string format used to represent the sequence number.
START=j: Sets the starting value of the sequence number to j. If omitted, START defaults to 1.
INCR=i: Sets the increment value of the sequence number to i. If omitted, INCR defaults to 1.
RESTART=(p,m): Resets SEQNUM to 1 when the specified field does not match the corresponding field of the previous record.

findrep_clause

Finds incon in a record and changes it to outcon.

incon: string to find in the format of C’x₁…x_n', nC’x₁…x_n', X’x₁…x_n', or nX’x₁…x_n'.
outcon: string to output in the format of C'', C’x₁…x_n', nC’x₁…x_n', X’x₁…x_n', or nX’x₁…x_n'.
p: search start position.
q: search end position.

WHEN=INIT, [ BUILD | OVERLAY | FINDREP ]

Performs BUILD, OVERLAY, or FINDREP operations on all input records. An IFTHEN clause that includes WHEN=INIT is processed before other IFTHEN clauses.

WHEN=(logical_expression), [ BUILD | OVERLAY | FINDREP ]

Performs BUILD, OVERLAY, or FINDREP operations on input records that meet the specific condition (logical_expression).

WHEN=GROUP

[,BEGIN=(logical_expression) |

,END=(logical_expression) |

,RECORDS=n]

,PUSH=(items)

The BEGIN, END, and RECORDS operations are applied to all input records. When the specified conditions are met, records are assigned to or released from a GROUP.

BEGIN=(logical_expression): Assigns a GROUP starting from the record that satisfies the logical_expression.
END=(logical_expression): Releases the GROUP after the record that satisfies the logical_expression.
RECORDS=n: Ends the GROUP when the number of records in a single GROUP reaches n.
PUSH=(items): Overlays the values specified in items when the record belongs to a GROUP. The available item types are as follows:
- c:p,m: Overlays the value corresponding to p,m from the first record of the GROUP at position c.
- c:ID=n: Represents a GROUP ID of length n and overlays the current GROUP ID at position c.
- c:SEQ=n: Represents the record sequence within a GROUP with length n and overlays the current record sequence at position c.

Example

The following two examples have the same result.

INREC FIELDS=(10,3,20,8)
SORT FIELDS=(5,6,CH,A)

OUTREC FIELDS=(10,3,20,8)
SORT FIELDS=(21,6,CH,A)

The first example reformats bytes 10 through 12 and bytes 20 through 27 of each input record into a new record with the length of 11 bytes, and then sorts the new records based on their bytes 5 through 10 in ascending order. The second example sorts input records based on bytes 21 through 26, and then reformats the result records in the same way as in the first example.

The following performs INREC and OUTREC.

INREC FIELDS=(14,4, 6,3)
SORT FIELDS=(1,4,A, 5,3,A), FORMAT=AC
OUTREC FIELDS=(3X, 1,4,H, 13:1,3, 5,2, 4X'61', X'0A')

Input record: ABCDE567LMNOP1234

The results are as follows:

The result of performing INREC

Classification Description

Bytes 1 through 4

Bytes 14 through 17 of the input record ('1234')

Bytes 5 through 7

Bytes 6 through 8 of the input record ('567')

Classification	Description
Bytes 1 through 4	Bytes 14 through 17 of the input record ('1234')
Bytes 5 through 7	Bytes 6 through 8 of the input record ('567')

Result of performing OUTREC

Classification	Description
Bytes 1 through 3	Space characters (20 20 20 in HEX)
Byte 4	NULL padded for halfword alignment (00 in HEX)
Bytes 5 through 8	Bytes 1 through 4 of the INREC result ('1234')
Bytes 9 through 12	Space characters (20 20 20 20 in HEX)
Bytes 13 through 15	Bytes 1 through 3 of the INREC result ('123')
Bytes 16 through 17	Bytes 5 through 6 of the INREC result ('56')
Bytes 18 through 21	0x61 ('aaaa')
Byte 22	Carriage return (0A in HEX)

Classification

Description

Bytes 1 through 3

Space characters (20 20 20 in HEX)

Byte 4

NULL padded for halfword alignment (00 in HEX)

Bytes 5 through 8

Bytes 1 through 4 of the INREC result ('1234')

Bytes 9 through 12

Space characters (20 20 20 20 in HEX)

Bytes 13 through 15

Bytes 1 through 3 of the INREC result ('123')

Bytes 16 through 17

Bytes 5 through 6 of the INREC result ('56')

Bytes 18 through 21

0x61 ('aaaa')

Byte 22

Carriage return (0A in HEX)

9.1. parse_item

The details of parse_item are as follows.

Syntax

INREC, OUTREC - parse_item

Component

Component	Description
%nn	Converts variable-length input records into fixed-length output records. Parsed fields can be used wherever (p,m) is specified in BUILD or OVERLAY.
FIXLEN=m	Specifies the length of the parsed field. This attribute must be specified for all parsed fields.
ABSPOS=p	Specifies the absolute starting position p for parsing in the input record.
ADDPOS=x	Increases the parsing start position in the input record by x.
SUBPOS=y	Decreases the parsing start position in the input record by y.
STARTAFT	STARTAFT=string: The parsed field starts after the first occurrence of string in the input record. STARTAFT=BLANKS: The parsed field starts after the first sequence of spaces from the beginning of the input record.
STARTAT	STARTAT=string: The parsed field starts at the first occurrence of string in the input record, including the string. STARTAT=BLANKS: The parsed field starts at the first sequence of spaces from the beginning of the input record. STARTAT=NONBLANK: The parsed field starts at the first non-space character from the beginning of the input record.
ENDBEFR	ENDBEFR=string: The parsed field ends before the first occurrence of string in the input record. ENDBEFR=BLANKS: The parsed field ends before the first sequence of spaces from the beginning of the input record.
ENDAT	ENDAT=string: The parsed field ends at the first occurrence of string in the input record, including the string. ENDAT=BLANKS: The parsed field ends at the first sequence of spaces from the beginning of the input record, including the spaces.
PAIR	PAIR=APOST: Text enclosed in apostrophes (') is not treated as BLANKS or strings. PAIR=QUOTE: Text enclosed in quotation marks (") is not treated as BLANKS or strings.

Description

%nn

Converts variable-length input records into fixed-length output records.

Parsed fields can be used wherever (p,m) is specified in BUILD or OVERLAY.

FIXLEN=m

Specifies the length of the parsed field. This attribute must be specified for all parsed fields.

ABSPOS=p

Specifies the absolute starting position p for parsing in the input record.

ADDPOS=x

Increases the parsing start position in the input record by x.

SUBPOS=y

Decreases the parsing start position in the input record by y.

STARTAFT

STARTAFT=string: The parsed field starts after the first occurrence of string in the input record.
STARTAFT=BLANKS: The parsed field starts after the first sequence of spaces from the beginning of the input record.

STARTAT

STARTAT=string: The parsed field starts at the first occurrence of string in the input record, including the string.
STARTAT=BLANKS: The parsed field starts at the first sequence of spaces from the beginning of the input record.
STARTAT=NONBLANK: The parsed field starts at the first non-space character from the beginning of the input record.

ENDBEFR

ENDBEFR=string: The parsed field ends before the first occurrence of string in the input record.
ENDBEFR=BLANKS: The parsed field ends before the first sequence of spaces from the beginning of the input record.

ENDAT

ENDAT=string: The parsed field ends at the first occurrence of string in the input record, including the string.
ENDAT=BLANKS: The parsed field ends at the first sequence of spaces from the beginning of the input record, including the spaces.

PAIR

PAIR=APOST: Text enclosed in apostrophes (') is not treated as BLANKS or strings.
PAIR=QUOTE: Text enclosed in quotation marks (") is not treated as BLANKS or strings.

10. INSERT

Inserts a specified file into a script.

The details of INSERT are as follows:

Syntax

INSERT
Component

Component Description

file_path

Path of a file to be inserted into a script.
Example

The following inserts the ins.scr file into a script.
```
INSERT ins.scr
```

Component	Description
file_path	Path of a file to be inserted into a script.

11. MEMORY

Sets the memory size used when performing operations.

The details of MEMORY are as follows:

Syntax

MEMORY
Component

Component Description

n

Memory size in n bytes.

nK

Memory size in n KB.

nM

Memory size in n MB.

nG

Memory size in n GB.
Example

The following sets 64 MB memory.
```
MEMORY 64M
```
The following set 1 GB memory.
```
MEMORY 1G
```

Component	Description
n	Memory size in n bytes.
nK	Memory size in n KB.
nM	Memory size in n MB.
nG	Memory size in n GB.

12. MODS

Registers a function as User Exit Function by using a ProSort script.

For more information about User Exit Function, refer to User Exit Function.

The details of MODS are as follows:

Syntax

MODS
Component

Component Description

func_name

Function name to register.

lib_path

Path of library file to dynamically refer to.

lib_type

Either C for COBOL functions or N for non-COBOL function.
Example

The following registers the exit_none and exit_insert functions as User Exit Function E15 and E32 respectively. The lib_exit.so library file is referred to dynamically.
```
MODS E15 = (exit_none, /path/lib_exit.so, N),
     E32 = (exit_insert, /path/lib_exit.so, N)
```

Component	Description
func_name	Function name to register.
lib_path	Path of library file to dynamically refer to.
lib_type	Either C for COBOL functions or N for non-COBOL function.

13. OPTION

Sets values of options such as USE_ALTERNATIVE_TP_TYPE.

The details of OPTION are as follows:

Syntax

OPTION

Component

Component	Description
EQUALS	Guarantees to preserve the original sequence of records that have the same key field.
NOEQUALS	Does not guarantee to preserve the original sequence of records that have the same key field.
SKIPREC=n	Skips the first n input records and passes the remaining records to E15 user exit or a routine that processes an INCLUDE or OMIT statement. Refer to SORT and COPY Operation Procedure (with User Exit Function).
STOPAFT=n	Passes up to the first n input records to a routine that processes an INREC statement if the INREC statement exists or a routine that processes a SORT or COPY statement if an INREC statement does not exist. Refer to SORT and COPY Operation Procedure (with User Exit Function).
[ CENTURY \| CENTWIN \| Y2PAST ] = [ s \| f ]	Supplements the first two-digit year value for Y2x fields because the fields have only the last two-digit year value. CENTURY and CENTWIN are the same as Y2PAST. Y2PAST can have a value with one of the following types. s: sliding century window. An integer between 0 and 100. A Y2x field value is interpreted as a value between (the current year - s) and (the current year - s + 99). f: fixed century window. An integer between 1000 and 3000. A Y2x field value is interpreted as a value between f and (f + 99).
COPY	Performs the COPY operation.
MSGPRT	Messages to output. Only syntax-checked. (Alias: PRINT)
VLSHRT	Action to take when a field specified in a SORT, MERGE, FILTER (INCLUDE, OMIT), or SUM operation does not exist or only a part of it exists in an input record. For more information, refer to Example.
NOVLSHRT	An error message is displayed and ProSort terminates when a field specified in a SORT, MERGE, FILTER (INCLUDE, OMIT), or SUM operation does not exist or only a part of it exists in an input record. If both VLSHRT and NOVLSHRT are not specified, NOVLSHRT is used.
VLTESTI=(0 \| 1 \| 2)	Action to take when a field specified in a reformat (INCLUDE, OMIT) operation does not exist or only a part of it exists in an input record. The following describes each option: 0: In INCLUDE or OMIT operations, the full range of all specified fields must exist in the input record. Otherwise, an error occurs. If the VLTESTI option is not specified, the operation behaves as if VLTESTI=0 were specified. 1: For INCLUDE operations, if the specified field does not exist in the input record or exists only partially, the record is not output. For OMIT operations, if the specified field does not exist in the input record or exists only partially, the record is output. 2: In INCLUDE or OMIT operations, if the specified field does not exist in the input record or exists only partially, the relational expression is evaluated as false.
USE_ALTERNATIVE_ TP_TYPE	How to express the TP positive numbers. For more information, refer to Miscellaneous.

Description

EQUALS

Guarantees to preserve the original sequence of records that have the same key field.

NOEQUALS

Does not guarantee to preserve the original sequence of records that have the same key field.

SKIPREC=n

Skips the first n input records and passes the remaining records to E15 user exit or a routine that processes an INCLUDE or OMIT statement. Refer to SORT and COPY Operation Procedure (with User Exit Function).

STOPAFT=n

Passes up to the first n input records to a routine that processes an INREC statement if the INREC statement exists or a routine that processes a SORT or COPY statement if an INREC statement does not exist. Refer to SORT and COPY Operation Procedure (with User Exit Function).

[ CENTURY | CENTWIN | Y2PAST ] = [ s | f ]

Supplements the first two-digit year value for Y2x fields because the fields have only the last two-digit year value. CENTURY and CENTWIN are the same as Y2PAST.

Y2PAST can have a value with one of the following types.

s: sliding century window. An integer between 0 and 100. A Y2x field value is interpreted as a value between (the current year - s) and (the current year - s + 99).
f: fixed century window. An integer between 1000 and 3000. A Y2x field value is interpreted as a value between f and (f + 99).

COPY

Performs the COPY operation.

MSGPRT

Messages to output. Only syntax-checked. (Alias: PRINT)

VLSHRT

Action to take when a field specified in a SORT, MERGE, FILTER (INCLUDE, OMIT), or SUM operation does not exist or only a part of it exists in an input record.

For more information, refer to Example.

NOVLSHRT

An error message is displayed and ProSort terminates when a field specified in a SORT, MERGE, FILTER (INCLUDE, OMIT), or SUM operation does not exist or only a part of it exists in an input record.

If both VLSHRT and NOVLSHRT are not specified, NOVLSHRT is used.

VLTESTI=(0 | 1 | 2)

Action to take when a field specified in a reformat (INCLUDE, OMIT) operation does not exist or only a part of it exists in an input record.

The following describes each option:

0: In INCLUDE or OMIT operations, the full range of all specified fields must exist in the input record. Otherwise, an error occurs. If the VLTESTI option is not specified, the operation behaves as if VLTESTI=0 were specified.
1: For INCLUDE operations, if the specified field does not exist in the input record or exists only partially, the record is not output. For OMIT operations, if the specified field does not exist in the input record or exists only partially, the record is output.
2: In INCLUDE or OMIT operations, if the specified field does not exist in the input record or exists only partially, the relational expression is evaluated as false.

USE_ALTERNATIVE_ TP_TYPE

How to express the TP positive numbers. For more information, refer to Miscellaneous.

Example

When VLSHRT is specified, each operation behaves as follows.
- SORT, MERGE
  
  If a specified field does not exist or only a part of it exists in an input record, binary zeros are padded.
  SORT FIELDS=(1,2,CH,A) Input: A B AA BB Output: A AA B BB
- INCLUDE, OMIT
  
  If a specified field does not exist or only a part of it exists in an input record, the relational expression is regarded as false.
  INCLUDE COND=(1,2,CH,EQ,C'AA') Input: A AA B BB Output: AA
- SUM
  
  If a specified field does not exist or only a part of it exists in an input record, the record is excluded from a SUM operation.
  SORT FIELDS=(1,1,CH,A) SUM FIELDS=(2,2,ZD) Input: A03 A04 B B05 C Output: A07 B05

14. OUTFIL

Outputs new records created by applying a condition to records resulting from a SORT, MERGE, or COPY operation. Its alias is OUTFILE.

The details of OUTFIL are as follows:

Syntax

OUTFIL

Component

Component	Description
filename	File name to output.
STARTREC=n	Outputs result records from the n-th record.
ENDREC=n	Outputs result records to the n-th record.
INCLUDE=condition	Outputs result records that meet the condition. For more information, refer to INCLUDE, OMIT.
OMIT=condition	Does not output result records that meet the condition. For more information, refer to INCLUDE, OMIT.
SAVE	Outputs records that are not outputted for any other OUTFIL with INCLUDE or OMIT.
OUTREC or BUILD	Refer to INREC, OUTREC.
OVERLAY	Refer to INREC, OUTREC.
FINDREP	Refer to INREC, OUTREC.
IFTHEN	Refer to INREC, OUTREC.
VTOF or CONVERT	Converts a variable-length input record to a fixed-length output record.
VLFILL=c	Character (c) with which an input record is right-padded if a specific field does not exist or only a part of it exists in the input record.
FTOV	Converts a variable-length input record to a fixed-length output record.
VLTRIM=c	Trailing character (c) to remove from a result record. The number of the characters can be one or more.
NULLOFL= [ RC0 \| RC4 \| RC16 ]	Action to take when there is no operation result record. RC0: set the return value to 1, and then continue with the next operation. RC4: set the return value to 4, and then continue with the next operation. RC16: stop the operation and return 16.
LINES=n	Number of lines (n) displayed in one page of a report.
HEADER1	Report header. For more information, refer to header_item.
HEADER2	Page header. For more information, refer to header_item.
SECTIONS	Adds a header or trailer whenever the data defined in a section changes. Multiple sections can be defined with sections_item. For more information, refer to sections_item.
TRAILER2	Page trailer. For more information, refer to trailer_item.
TRAILER1	Report trailer. For more information, refer to trailer_item.
NODETAIL	Outputs a report that includes result records without data records.
REMOVECC	Outputs result records without ANSI control characters.

Description

filename

File name to output.

STARTREC=n

Outputs result records from the n-th record.

ENDREC=n

Outputs result records to the n-th record.

INCLUDE=condition

Outputs result records that meet the condition. For more information, refer to INCLUDE, OMIT.

OMIT=condition

Does not output result records that meet the condition. For more information, refer to INCLUDE, OMIT.

SAVE

Outputs records that are not outputted for any other OUTFIL with INCLUDE or OMIT.

OUTREC or BUILD

Refer to INREC, OUTREC.

OVERLAY

Refer to INREC, OUTREC.

FINDREP

Refer to INREC, OUTREC.

IFTHEN

Refer to INREC, OUTREC.

VTOF or CONVERT

Converts a variable-length input record to a fixed-length output record.

VLFILL=c

Character (c) with which an input record is right-padded if a specific field does not exist or only a part of it exists in the input record.

FTOV

Converts a variable-length input record to a fixed-length output record.

VLTRIM=c

Trailing character (c) to remove from a result record. The number of the characters can be one or more.

NULLOFL= [ RC0 | RC4 | RC16 ]

Action to take when there is no operation result record.

RC0: set the return value to 1, and then continue with the next operation.
RC4: set the return value to 4, and then continue with the next operation.
RC16: stop the operation and return 16.

LINES=n

Number of lines (n) displayed in one page of a report.

HEADER1

Report header. For more information, refer to header_item.

HEADER2

Page header. For more information, refer to header_item.

SECTIONS

Adds a header or trailer whenever the data defined in a section changes. Multiple sections can be defined with sections_item. For more information, refer to sections_item.

TRAILER2

Page trailer. For more information, refer to trailer_item.

TRAILER1

Report trailer. For more information, refer to trailer_item.

NODETAIL

Outputs a report that includes result records without data records.

REMOVECC

Outputs result records without ANSI control characters.

Example

The following is an example that records in which bytes 15 through 20 are AAA are outputted to the file AAA, the records in which bytes 15 through 200 are BBB are outputted to the file BBB, and the remaining records are outputted to the file UNKNOWN.
```
OUTFIL INCLUDE=(15,6,AC,EQ,C'AAA'), FNAMES=AAA
OUTFIL INCLUDE=(15,6,AC,EQ,C'BBB'), FNAMES=BBB
OUTFIL SAVE, FNAME=UNKNOWN
```

14.1. header_item

The details of header_item are as follows:

Syntax

OUTFIL - header_item

Component

Component	Description
s	Count to repeat a string in the format of n’xx…x'.
p,m	Input field position (p) and length (m).
DATENS=(abc)	Date in the format of abc. a, b, and c can be one of M, D, Y, and 4. Each value cannot be defined multiple times. M: month (01-12) D: day (01-31) Y: last two digits of a year 4: four-digit year (0001-9999) For example, DATENS=(DMY) expresses June 3, 2009 as 030609.
DATE=(abcd)	Date in the format of abcd. abc: same as abc for DATENS. d: delimiter added between each token (year, month, and day). For example, DATE=(4MD-) expresses June 3, 2009 as 2009-06-03.
TIMENS=(ab)	Time in the format of ab (either 12 or 24). If ab is 24, time is expressed as hhmmss (24-hour clock). hh: hour (00-23) mm: minute (00-59) ss: second (00-59) For example, 4:15:37 p.m. is expressed as 161537. If ab is 12, time is expressed as hhmmss xx (12-hour clock). hh: hour (00-23) mm: minute (00-59) ss: second (00-59) xx: either am or pm. For example, 4:15:37 p.m. is expressed as 041537 pm.
TIME=(abc)	Time in the format of abc. ab: same as ab for TIMENS. c: delimiter added between each token (hour, minute, and second). For example, TIME=(12.) expresses 4:15:37 p.m. as 04.15.37 pm.
PAGE	Page number. For information about 'edit' and 'to', refer to INREC, OUTREC.

Description

Count to repeat a string in the format of n’xx…x'.

p,m

Input field position (p) and length (m).

DATENS=(abc)

Date in the format of abc. a, b, and c can be one of M, D, Y, and 4. Each value cannot be defined multiple times.

M: month (01-12)
D: day (01-31)
Y: last two digits of a year
4: four-digit year (0001-9999)

For example, DATENS=(DMY) expresses June 3, 2009 as 030609.

DATE=(abcd)

Date in the format of abcd.

abc: same as abc for DATENS.
d: delimiter added between each token (year, month, and day).

For example, DATE=(4MD-) expresses June 3, 2009 as 2009-06-03.

TIMENS=(ab)

Time in the format of ab (either 12 or 24).

If ab is 24, time is expressed as hhmmss (24-hour clock).

hh: hour (00-23)
mm: minute (00-59)
ss: second (00-59)

For example, 4:15:37 p.m. is expressed as 161537.

If ab is 12, time is expressed as hhmmss xx (12-hour clock).

hh: hour (00-23)
mm: minute (00-59)
ss: second (00-59)
xx: either am or pm.

For example, 4:15:37 p.m. is expressed as 041537 pm.

TIME=(abc)

Time in the format of abc.

ab: same as ab for TIMENS.
c: delimiter added between each token (hour, minute, and second).

For example, TIME=(12.) expresses 4:15:37 p.m. as 04.15.37 pm.

PAGE

Page number. For information about 'edit' and 'to', refer to INREC, OUTREC.

14.2. sections_item

The details of sections_item are as follows:

Syntax

OUTFIL - sections_item

Component

Component	Description
p,m	Position (p) and length (m) of a section field in an output record. Before outputting the first record, the first section header is outputted. If the section field value of a current record is different from that of the previous record, the previous section trailer and the current section header are outputted. After outputting the last record, the last section trailer is outputted.
SKIP=L	Outputs one empty line between sections.
SKIP=nL	Outputs n empty lines between sections.
HEADER3	Section header. For more information, refer to header_item.
TRAILER3	Section trailer. For more information, refer to trailer_item.

Description

p,m

Position (p) and length (m) of a section field in an output record.

Before outputting the first record, the first section header is outputted. If the section field value of a current record is different from that of the previous record, the previous section trailer and the current section header are outputted. After outputting the last record, the last section trailer is outputted.

SKIP=L

Outputs one empty line between sections.

SKIP=nL

Outputs n empty lines between sections.

HEADER3

Section header. For more information, refer to header_item.

TRAILER3

Section trailer. For more information, refer to trailer_item.

14.3. trailer_item

The details of trailer_item are as follows:

Syntax

OUTFIL - trailer_item

Component

Component	Description
s	Refer to header_item.
p,m	Refer to header_item.
DATENS=(abc)	Refer to header_item.
DATE=(abcd)	Refer to header_item.
TIMENS=(ab)	Refer to header_item.
TIME=(abc)	Refer to header_item.
PAGE	Refer to header_item.
COUNT	For report trailers Total number of output records. For page/section trailers Number of records displayed in a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC. If both 'edit' and 'to' are not specified, an 8-digit integer is outputted.
COUNT15	Same as COUNT except that the count is a 15-digit integer.
SUBCOUNT	For report trailers Total number of output records. For page/section trailers Accumulated number of records displayed up to a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC. If both 'edit' and 'to' are not specified, an 8-digit integer is outputted.
SUBCOUNT15	Same as SUBCOUNT except that a 15-digit integer is output instead of an 8-digit integer.
TOTAL or TOT	A field at bytes p through (p + m - 1) of each output recor is used as an aggregate field. For report trailers Total of aggregate fields in the specified format (f) of all output records. For page/section trailers Total of aggregate fields in the specified format (f) of records displayed in a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.
MIN	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Minimum value among aggregate fields in the specified format (f) of all output records. For page/section trailers Minimum value among aggregate fields in the specified format (f) of records displayed in a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.
MAX	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Maximum value among aggregate fields in the specified format (f) of all output records. For page/section trailers Maximum value among aggregate fields in the specified format (f) of records displayed in a current page/section. For information about 'edi’t and 'to', refer to INREC, OUTREC.
AVG	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Average value of aggregate fields in the specified format (f) of all output records. For page/section trailers Average value of aggregate fields in the specified format (f) of records displayed in a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.
[ SUBTOTAL \| SUBTOT \| SUB ]	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Total value of aggregate fields in the specified format (f) of all output records. For page/section trailers Total value of aggregate fields in the specified format (f) of records displayed up to a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.
SUBMIN	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Minimum value among aggregate fields in the specified format (f) of all output records. For page/section trailers Minimum value among aggregate fields in the specified format (f) of records displayed up to a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.
SUBMAX	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Maximum value among aggregate fields in the specified format (f) of all output records. For page/section trailers Maximum value among aggregate fields in the specified format (f) of records displayed up to a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.
SUBAVG	A field at bytes p through (p + m - 1) of each output record is used as an aggregate field. For report trailers Average value of aggregate fields in the specified format (f) of all output records. For page/section trailers Average value of aggregate fields in the specified format (f) of records displayed up to a current page/section. For information about 'edit' and 'to', refer to INREC, OUTREC.

Description

Refer to header_item.

p,m

Refer to header_item.

DATENS=(abc)

Refer to header_item.

DATE=(abcd)

Refer to header_item.

TIMENS=(ab)

Refer to header_item.

TIME=(abc)

Refer to header_item.

PAGE

Refer to header_item.

COUNT

For report trailers

Total number of output records.
For page/section trailers

Number of records displayed in a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC. If both 'edit' and 'to' are not specified, an 8-digit integer is outputted.

COUNT15

Same as COUNT except that the count is a 15-digit integer.

SUBCOUNT

For report trailers

Total number of output records.
For page/section trailers

Accumulated number of records displayed up to a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC. If both 'edit' and 'to' are not specified, an 8-digit integer is outputted.

SUBCOUNT15

Same as SUBCOUNT except that a 15-digit integer is output instead of an 8-digit integer.

TOTAL or TOT

A field at bytes p through (p + m - 1) of each output recor is used as an aggregate field.

For report trailers

Total of aggregate fields in the specified format (f) of all output records.
For page/section trailers

Total of aggregate fields in the specified format (f) of records displayed in a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

MIN

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Minimum value among aggregate fields in the specified format (f) of all output records.
For page/section trailers

Minimum value among aggregate fields in the specified format (f) of records displayed in a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

MAX

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Maximum value among aggregate fields in the specified format (f) of all output records.
For page/section trailers

Maximum value among aggregate fields in the specified format (f) of records displayed in a current page/section.

For information about 'edi’t and 'to', refer to INREC, OUTREC.

AVG

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Average value of aggregate fields in the specified format (f) of all output records.
For page/section trailers

Average value of aggregate fields in the specified format (f) of records displayed in a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

[ SUBTOTAL | SUBTOT | SUB ]

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Total value of aggregate fields in the specified format (f) of all output records.
For page/section trailers

Total value of aggregate fields in the specified format (f) of records displayed up to a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

SUBMIN

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Minimum value among aggregate fields in the specified format (f) of all output records.
For page/section trailers

Minimum value among aggregate fields in the specified format (f) of records displayed up to a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

SUBMAX

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Maximum value among aggregate fields in the specified format (f) of all output records.
For page/section trailers

Maximum value among aggregate fields in the specified format (f) of records displayed up to a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

SUBAVG

A field at bytes p through (p + m - 1) of each output record is used as an aggregate field.

For report trailers

Average value of aggregate fields in the specified format (f) of all output records.
For page/section trailers

Average value of aggregate fields in the specified format (f) of records displayed up to a current page/section.

For information about 'edit' and 'to', refer to INREC, OUTREC.

15. SORT, MERGE

Performs a sort, copy, or merge operation. SORT and MERGE statements consist of a field list and options. If all the fields have the same format, it needs to set the format only once after the field list.

The details of SORT and MERGE are as follows:

Syntax

SORT

MERGE

SORT, MERGE - smj_option

Component

Component	Description
COPY	Performs a COPY operation instead of a SORT operation.
p,m,f,s	Position (p), length (m), and format (f) of a field in each input record, and a sorting order (s). If f is omitted, the format set in the FORMAT option is used. s can be set to either A (ascending order) or D (descending order).
smj_option	Refer to OPTION.

Description

COPY

Performs a COPY operation instead of a SORT operation.

p,m,f,s

Position (p), length (m), and format (f) of a field in each input record, and a sorting order (s).

If f is omitted, the format set in the FORMAT option is used. s can be set to either A (ascending order) or D (descending order).

smj_option

Refer to OPTION.

Example

The following sorts records based on 8-byte ZD fields at byte 3 in ascending order and 6-byte AC fields at byte 40 in descending order.
```
SORT FIELDS=(3,8,ZD,A,40,6,AC,D)
```
The following sorts records based on 8-byte ZD fields at byte 3 in ascending order and 6-byte AC fields at byte 40 in descending order. Since EQUALS is used, records with the same key field are sorted in the original order (stable sort).
```
SORT FIELDS=(3,8,ZD,A,40,6,AC,D), EQUALS
```
The following sorts records based on 8-byte ZD fields at byte 3 in ascending order and 6-byte AC fields at byte 40 in descending order. Since SKIPREC and STOPAFT are used, only records 101 through 300 are sorted.
```
SORT FIELDS=(3,8,ZD,A,40,6,AC,D), SKIPREC = 100, STOPAFT = 200
```
The following merges records based on 4-byte ZD fields at byte 25 and 8-byte ZD fields at byte 48 in ascending order.
```
MERGE FIELDS=(25,4,A,48,8,A),FORMAT=ZD
```
The following examples only copy records to output.
```
MERGE FIELDS=COPY
```
```
SORT FIELDS=COPY
```

16. SUM

Outputs one of sorted or merged records that have the same key field. Specified fields (summary fields) of the records are added and the sum is outputted.

The details of SUM are as follows:

Syntax

SUM

Component

Component	Description
NONE	Performs a SORT or MERGE operation and then outputs one of the records that have the same key field.
p,m,f	Position (p), length (m), and format (f) of a summary field in each input record. Only one of the records that have the same key field is outputted. Summary fields in the format of f of the records are added, and the sum in the format of f is outputted in bytes p through (p + m -1) of the output record. If f is omitted, the format set in the FORMAT option is used.

Description

NONE

Performs a SORT or MERGE operation and then outputs one of the records that have the same key field.

p,m,f

Position (p), length (m), and format (f) of a summary field in each input record.

Only one of the records that have the same key field is outputted. Summary fields in the format of f of the records are added, and the sum in the format of f is outputted in bytes p through (p + m -1) of the output record. If f is omitted, the format set in the FORMAT option is used.

Example

The following specifies an 8-byte packed decimal field at byte 21 and a 4-byte fixed-integer field at byte 11 as summary fields.
```
SUM FIELDS=(21,8,PD,11,4,FI)
```
The following only eliminates duplicate records.
```
SUM FIELDS=NONE
```
The following specifies an 8-byte ZD field at byte 41 and a 4-byte ZD field at byte 49 as summary fields.
```
SUM FIELDS=(41,8,49,4),FORMAT=ZD
```

17. WORKSPACE

Specifies the directories to save temporary files used for external sorting.

The details of WORKSPACE are as follows:

Syntax

WORKSPACE
Component

Component Description

directory_path

Path of a directory to save temporary files.

Component	Description
directory_path	Path of a directory to save temporary files.

Example

WORKSPACE = (/tmp, /home/tmp0, /home/tmp1)