GCSORT 1.03.

06
[15 GEN 2015 Version]
User’s Guide
1nd Edition, 15 Janury 2016
Sauro Menna
mennasauro@gmail.com

GCSORT Copyright © 2016-2023 Sauro Menna

GCSORT Copyright © 2009 Cedric Issaly

Under the terms of the GNU General Public License

Document Copyright © 2016 Sauro Menna

Permission is granted to copy, distribute and/or modify this document

under the terms of the GNU Free Documentation License [FDL], Version 1.3
or any later version published by the Free Software Foundation.

15 Jan 2016 Version Pag. 1

 This work is dedicated to the memory of my niece Federica,
a strong young woman, sweet and resourceful.
You will always be in my heart and mind.

15 Jan 2016 Version Pag. 2

Summary of Changes
Edition v Date Change Description
1st 15 Jan 2016 INITIAL RELEASE OF DOCUMENT
09 Nov 2016 UPGRADE version with integration of LIBCOB
New Data Types
Search Substring search Conditional
1.0.1 15 Oct 2020 New option in command line -fsign=EBCDIC/ASCII for NUMERIC field.
1.0.1 09 Jan 2021 INREC OVERLAY – OUTREC OVERLAY
1.03.02 18 Jan 2022 RECORD CONTROL STATEMENT / DATE - Currente Date : DATE1, DATE2, DATE3, DATE4 / INREC
CHANGE / OUTREC CHANGE / MODS E15 – E35
1.03.03 27 Mar 2022 JOIN Statement
1.03.04 4 Agu 2022 FINDREP in INREC/OUREC Control statement
1.03.05 13 Mar 2023 OUTFIL changes
1.03.06 29 Mar 2023 SubString new format type

15 Jan 2016 Version Pag. 3

Table of Contents
1. Introduction ............................................................................................................................................... 6
1.1. What is GCSort? ................................................................................................................................. 6
2. Features ..................................................................................................................................................... 6
3. Environment and first use........................................................................................................................ 10
3.1. Following the steps for the first use ................................................................................................ 10
3.2. Modify first environment variables ................................................................................................. 10
3.3. Use TAKE command ......................................................................................................................... 10
1. Process Schema ....................................................................................................................................... 11
2. Sort .......................................................................................................................................................... 12
3. Merge ...................................................................................................................................................... 12
4. File Organization and Record Type .......................................................................................................... 12
5. Field Type ................................................................................................................................................. 13
Date Format ................................................................................................................................................. 14
6. Commands ............................................................................................................................................... 15
6.1. SORT................................................................................................................................................. 15
6.2. MERGE ............................................................................................................................................. 15
6.3. COPY ................................................................................................................................................ 15
6.4. FIELDS .............................................................................................................................................. 15
6.5. USE ................................................................................................................................................... 16
6.6. GIVE ................................................................................................................................................. 17
6.7. INCLUDE/OMIT ................................................................................................................................ 17
6.8. INREC/OUTREC ................................................................................................................................ 20
6.9. SUM FIELDS...................................................................................................................................... 23
6.10. RECORD ........................................................................................................................................ 23
6.11. OUTFIL ......................................................................................................................................... 24
6.12. OPTION ........................................................................................................................................ 25
Exit Routines .................................................................................................................................................... 25
7. JOIN Statement ........................................................................................................................................ 27
Join Process Schema .................................................................................................................................... 28
8. Environment Variables............................................................................................................................. 32
8.1. Byte Order ....................................................................................................................................... 32
8.2. Temporary Files................................................................................................................................ 32
8.3. Memory Allocation .......................................................................................................................... 32
8.4. Statistics ........................................................................................................................................... 33

15 Jan 2016 Version Pag. 4

9. Command Line ......................................................................................................................................... 36
10. Padding and Truncating ....................................................................................................................... 36
11. Retun Code .......................................................................................................................................... 36
12. File Conversion..................................................................................................................................... 37
13. Performance and Tuning...................................................................................................................... 37
14. Limits ................................................................................................................................................... 38
15. Errors and Warnings ............................................................................................................................ 38
16. GCSort by examples ............................................................................................................................. 39
16.1. SORT............................................................................................................................................. 39
16.2. MERGE ......................................................................................................................................... 39
16.3. COPY ............................................................................................................................................ 40
16.4. SUMFIELDS................................................................................................................................... 40
16.5. OUTREC ........................................................................................................................................ 41
16.6. OUTFIL ......................................................................................................................................... 42
16.7. INREC/OUREC CHANGE................................................................................................................ 42
16.8. DATE ............................................................................................................................................. 44
16.9. RECORD CONTROL STATEMENT ................................................................................................... 44
16.10. DATE – Option Y2PAST ................................................................................................................. 44

15 Jan 2016 Version Pag. 5

1. Introduction
1.1.What is GCSort?

This document describes the features of the GCSORT utility.

GCSORT is an open-source tool for operations of sort/merge/copy files (Line Sequential, Sequential,
Indexed and Relative) produced by GNUCobol compiler.
The principal developers of GCSORT are Cedric Issaly and Sauro Menna.
This document was intended to serve as a full-function reference and user’s guide for GCSORT utility.

2. Features

Version 1.3.2 of GCSort contains a follow constructs:

_______________________________________________________________________________________

gcsort help
gcsort is a program to sort, merge and copy records in a file into a specified order
________________________________________________________________________________________
Syntax case insensitive
Return code : 0 (ok) - 4 (warning) - 16 (error)
________________________________________________________________________________________
Usage with file parameters : gcsort <options> take filename
Usage from command line : gcsort <options> <control statements>
________________________________________________________________________________________
gcsort options
-fsign=[ASCII|EBCDIC] define display sign representation
-fcolseq=[NATIVE|ASCII|EBCDIC] collating sequence to use
-febcdic-table=<cconv-table>/<file> EBCDIC/ASCII translation table
________________________________________________________________________________________
gcsort control statements
Notations: '{name}' = parameters , '|' = Alternative format of control statement
________________________________________________________________________________________
SORT | MERGE | COPY FIELDS Control statement for Sort or Merge file(s)
USE Declare input file(s)
GIVE Declare output file
[ SUM FIELDS ] Sum fields for same record key, or eliminate duplicate keys)
[ RECORD ] Record control statement
[ INCLUDE ] Select input records that respect include condition(s)
[ OMIT ] Omit input records that respect include condition(s)
[ INREC ] Reformat input record Before sort, merge or copy operation
[ OUTREC ] Reformat input record After sort, merge or copy operation
[ OUTFIL ] Create one or more output files for sort,merge or copy operation
[ OPTION ] Specifies option for control statements
________________________________________________________________________________________
gcsort
SORT | MERGE | COPY

15 Jan 2016 Version Pag. 6

 FIELDS({Pos},{Len},{FormatType},{Order}, ...) |
FIELDS({Pos},{Len},{Order}, ...),FORMAT={FormatType} |
FIELDS=COPY

USE {Filename}
ORG {Org}
RECORD [F,{RecordLen}] | [V,{MinLen},{MaxLen}]
[KEY ({Pos},{Len},{KeyType})]

GIVE same parameters of USE

SUM FIELDS = [({Pos},{Len},{FormatType2}, ...)] |

[({Pos},{Len}, ...)],FORMAT={FormatType2} |
[NONE] | [(NONE)]

RECORD [TYPE=[{V}(Variable-length)/{F}(Fixed-length)]],[LENGTH=[{len}(L1-Input record

length)]
','[{len}(L2-Record
length)]
','[{len}(L3-Output
record length)]
INCLUDE | OMIT
COND=({Condition})[,FORMAT={FormatType}]

INREC FIELDS | INREC BUILD =({FieldSpec})

INREC OVERLAY =({FieldSpec})
OUTREC FIELDS | OUTREC BUILD =({FieldSpec})
OUTREC OVERLAY =({FieldSpec})

OUTFIL
INCLUDE | OMIT ({Condition})[,FORMAT={FormatType}]
OUTREC = ({FieldSpec})
FILES/FNAMES= {Filename} | (file1, file2, file3,...)
STARTREC={nn} Start from record nn
ENDREC={nn} Skip record after nn
SAVE
SPLIT Split 1 record output for file group (file1, file2, file3,...)
SPLITBY={nn} Split n records output for file group (file1, file2, file3,...)

OPTION
SKIPREC={nn} Skip nn records from input
STOPAFT={nn} Stop read after nn records
VLSCMP 0 disabled , 1 = enabled -- temporarily replace any
missing compare field bytes with binary zeros
VLSHRT 0 disabled , 1 = enabled -- treat any comparison
involving a short field as false
Y2PAST (YY) - Sliding, (YYYY) century
MODS E15=(<name>) [,] <name>= Name E15 Cobol Program for input
E35=(<name>) <name>= Name E35 Cobol Program for ouput
________________________________________________________________________________________
___{Parameters}____________________________|___{Relational}_____________________________
{FileName} = Filename or Env. Variable | EQ = Equal
{Pos} = Field Position | GT = GreaterThan
{Len} = Field Length | GE = GreaterEqual
{RecordLen}= Record Length | LT = LesserThan
{MinLen} = Min size of record | LE = LesserEqual

15 Jan 2016 Version Pag. 7

 {MaxLen} = Max size of record | NE = NotEqual
{Order} = A(ascending) | D(descending)| SS = SubString (only for Field Type 'CH')
___________________________________________|____________________________________________
___{Condition}__________________________________________________________________________
Format 1 - (Pos,Len,{FormatType},{Relational},[AND|OR],Pos,Len,{FormatType})
Format 2 - (Pos,Len,{FormatType},{Relational},[X|C'[value]'] | numeric value)]
Format 3 - ( {Condition} ,[AND|OR],{Condition} )
Format 4 - ( Pos,Len,{FormatType},{Relational}, [DATE1][(+/-)num] | [DATE2][(+/-)num]
[DATE3][(+/-)num] | [DATE4][(+/-)num]
DATE - Currente Date : DATE1 (C'yyyymmdd'), DATE2 (C'yyyymm'),
DATE3 (C'yyyyddd'), DATE4 (C'yyyy-mm-dd') (no Timestamp)
[(+/-)num] [+num] future date, [-num] past date)
________________________________________________________________________________________
___{Org}___File Organization_______________|___{KeyType}____Mandatory for ORG = IX______
LS = Line Sequential | P = Primary Key
SQ = Sequential Fixed or Variable | A = Alternative Key
IX = Indexed Fixed or Variable | D = Alternative Key with Duplicates
RL = Relative Fixed or Variable | C = Continue definition
___________________________________________|____________________________________________
__{FormatType}____Field Format Type________|___{FormatType2}____Format Type SumField____
CH = Char | BI = Binary unsigned
BI = Binary unsigned | FI = Binary signed
FI = Binary signed | FL = Floating Point
FL = Floating Point | PD = Packed
PD = Packed | ZD = Zoned
ZD = Zoned | CLO = Numeric sign leading
CLO = Numeric sign leading | CSL = Numeric sign leading separate
CSL = Numeric sign leading separate | CST = Numeric sign trailing separate
CST = Numeric sign trailing separate | SS = Search Substring
___________________________________________|____________________________________________
_Format_Len_Type__Date field_______________|_Format_Len_Type__Date field________________
Y2T = 8 ZD CCYYMMDD | Y2D = 1 PD YY
Y2T = 4 ZD YYXX | Y2P = 2 PD YY
Y2T = 2 ZD YYX | Y2U = 3 PD YYDDD
Y2T = 3 ZD YY | Y2S = 2 ZD YY
Y2T = 5 ZD YYDDD | Y2V = 4 PD YYMMDD
Y2T = 6 ZD YYMMDD | Y2X = 3 PD DDDYY
Y2B = 1 BI YY | Y2Y = 4 PD MMDDYY
Y2C = 2 ZD YY | Y2Z = 2 ZD YY
___________________________________________|____________________________________________
____{FieldSpec}___Field Specification___________________________________________________
pos, len pos = position input record, len = length of field
posOut:pos,len posOut = position output, pos = position input , len = length
n:X Filling with Blank character from last position to n
(absolute position of output record).
n:Z Filling with zero Binary character from last position to n
(absoluteposition of output record).
C'constant' constant character value.
nC'constant' repeat n times constant character value.
nX repeat n times Blank character.
nZ repeat n times Binary (0x00) character.
X'hh....hh' hexdecimal characters.
nX'hh...hh' repeat n times hexdecimal characters.
CHANGE=(vlen,[C | X]'<valueFind>',[C | X]'<valueSet>',.....),NOMATCH=([C |
X]'<valueSet>)
CHANGE=(vlen,[C | X]'<valueFind>', posIn, lenIn), NOMATCH = (posIn, posLen)

15 Jan 2016 Version Pag. 8

________________________________________________________________________________________
Environment Variables
________________________________________________________________________________________
COB_VARSEQ_FORMAT Used by GnuCOBOL
GCSORT_DEBUG 0 no print info, 1 info DEBUG, 2 for info Parser
GCSORT_MEMSIZE Memory Allocation in byte (Default 512000000 byte)
GCSORT_PATHTMP Pathname for temporary files (Default TMP / TEMP / TMPDIR)
GCSORT_STATISTICS 0 minimal informations, 1 for Summary, 2 for Details
GCSORT_TESTCMD 0 for normal operations , 1 for ONLY test command line (NO SORT)
________________________________________________________________________________________

15 Jan 2016 Version Pag. 9

3. Environment and first use
GCSort is a executable program written in ‘C’.
Dependencies of executable GCSort are:
- libcob - GNUCobol
- libm - Math library

3.1. Following the steps for the first use

- Make executable gcsort
- Set environment variable to find library at runtime
- Run gcsort <option> <command line>
o <option> -fsign=[EBCDIC | ASCII]
The -fsign=EBCDIC option can be used for files with ZONED fields and EBCDIC sign.

3.2. Modify first environment variables

- Set Memory Allocation (GCSORT_MEMSIZE)
- Set Statistics (GCSORT_STATISTICS) to view details of execution

3.3. Use TAKE command

- Create file text

- Insert command. Single row o one row for command.

- In the file TAKE the ‘*’ character indicates that the rest of the line is treated as a comment

- Run : gcsort TAKE filename

Example to create TAKE file with script sh.

export LD_LIBRARY_PATH=/usr/local/lib
export GCSORT_MEMSIZE=1024000000
export GCSORT_BYTEORDER=0
export GCSORT_STATISTICS=2
echo " * This is comment " >TAKEFILE.PRM
echo "SORT FIELDS(4,1,CH,A) " >TAKEFILE.PRM
echo "SUM FIELDS=(1,2,ZD,4,2,ZD,7,4,ZD,12,4,ZD) " >>TAKEFILE.PRM
echo "USE ../files/SQZD03 RECORD F,396 ORG SQ " >>TAKEFILE.PRM
echo "GIVE ../files/SQZD03.SRT RECORD F,396 ORG SQ " >>TAKEFILE.PRM
../bin/gcsort TAKE TAKEFILE.PRM

15 Jan 2016 Version Pag. 10

1. Process Schema

This picture show logical schema of utility GCSort for SORT operations.

This picture show logical schema of utility GCSort for MERGE operations.

15 Jan 2016 Version Pag. 11

2. Sort
The purpose of SORT is read one or more files and create a output file with data ordered as indicated by
the sort key fields.

3. Merge
The purpose of MERGE is read one or more files and create a output file with data ordered as indicated by
the merge key fields.
It is mandatory that the input data is already sorted.

4. File Organization and Record Type

File organization identifies the type of file.
The types of file organization utility managed GCSORT are:

LS = Line Sequential
LSF = Line Sequential Fixed
SQ = Sequential
IX = Indexed
RL = Relative

15 Jan 2016 Version Pag. 12

Use LSF file organization when the record to be sorted contains trailing spaces and you need fixed-length
records (GCSort does not delete trailing spaces). Record type identifies the record structure

Record type are

F = Fixed
V = Variable (first n byte record len, verify COB_VARSEQ_FORMAT in GNUCobol )

5. Field Type
Field type detects typology of field, Field type used are:

Type Description
CH Char
BI Binary unsigned
FI Binary signed
FL Floating Point
PD Packed
ZD Zoned
CLO Numeric sign leading
CSL Numeric sign leading separate
CST Numeric sign trailing separate
SS Search Substring

15 Jan 2016 Version Pag. 13

Date Format

Field Formats and Lengths for date.

_________________________________________________________________________________
Format_Len_Type__Date field_______________|_Format_Len_Type__Date field__________
Y2T = 8 ZD CCYYMMDD | Y2D = 1 PD YY
Y2T = 4 ZD YYXX | Y2P = 2 PD YY
Y2T = 2 ZD YYX | Y2U = 3 PD YYDDD
Y2T = 3 ZD YY | Y2S = 2 ZD YY
Y2T = 5 ZD YYDDD | Y2V = 4 PD YYMMDD
Y2T = 6 ZD YYMMDD | Y2X = 3 PD DDDYY
Y2B = 1 BI YY | Y2Y = 4 PD MMDDYY
Y2C = 2 ZD YY | Y2Z = 2 ZD YY

[ DATE - Currente Date : DATE4 ]

INCLUDE COND=(1,13,CH,GT,DATE4)
USE ../files/inp5000.txt ORG LS RECORD F,5000
GIVE ../files/inp5000.txt.srt ORG LS RECORD F,5000
SORT FIELDS=(35,5,ZD,A)

[ DATE + / - day - month ]

COND=(1,13,CH,GT,DATE1+5)
COND=(1,13,CH,GT,DATE1-5)
COND=(1,13,CH,GT,DATE2+3)
COND=(1,13,CH,GT,DATE2-8)
COND=(1,13,CH,GT,DATE3+150)
COND=(1,13,CH,GT,DATE3-15)

[ DATE4 ]
OMIT COND=(1,13,CH,GT,DATE4)

15 Jan 2016 Version Pag. 14

6. Commands

6.1. SORT
SORT is command for ordering data.

Format 1 SORT

6.2.MERGE
MERGE is command for merging data.

Format 1 MERGE

6.3.COPY
In SORT or MERGE command FIELDS=COPY copy data from input to output file.

Format 1 FIELDS=COPY

6.4.FIELDS
This command specify fields for sort/merge operations. The fields are the key for order or merging data
from files.

Format 1 FIELDS (pos,len,type,order, ...) |

Format 2 FIELDS ((pos,len, order, ...),FORMAT=TYPE |
Format 3 FIELDS=COPY

FIELDS (pos, len, type, order,….)

pos specifies the first byte of a control field relative to the beginning of the input record.
The first data byte of a fixed-length record has relative position 1.
The first data byte of a variable-length record has relative position 1.
len specifies the length of the field. Values for all fields must be expressed in integer numbers
of bytes.
type specifies the format of the data of field.

Type Description
CH Char
BI Binary unsigned
FI Binary signed
FL Floating Point
PD Packed
ZD Zoned
15 Jan 2016 Version Pag. 15
 CLO Numeric sign leading
CSL Numeric sign leading separate
CST Numeric sign trailing separate
SS Search Substring

order specifies how the field is to be ordered. The valid codes are:
A ascending order
D descending order

FIELDS ((pos,len,order, ...),FORMAT=type

FORMAT=type can be used to specify a particular format for one or more control fields. f from FORMAT=f is
used for p,m,s fields.

FIELDS=COPY or FIELDS=(COPY)

Causes GCSORT to copy a file input to the output data sets. Records can be edited INCLUDE/OMIT, INREC,
OUTREC, and OUTFIL statements; and SKIPREC and STOPAFT parameters.

6.5.USE
USE command declare input file for SORT and MERGE operation.

Format for USE:

USE <filename > ORG <organization> RECORD [<record format>,< length>]
[KEY ({Pos},{Len},{KeyType})

USE <filename > ORG <organization> RECORD [<record format>, <lenght min>,< length max>]
[KEY ({Pos},{Len},{KeyType})

filename Input file name, with or without pathname

organization LS = Line Sequential
SQ = Sequential
RL = Relative
IX = Indexed

record format F = Fixed

V = Variable
length Length of record

length min Minimun length of record

length max Maximum length of record

15 Jan 2016 Version Pag. 16

Structure of key (Mandatory for ORG = IX)
Pos Position of key
Len Length of key
KeyType P = Primary Key
A = Alternative Key
D = Alternative Key with Duplicates
C = Continue definition

6.6.GIVE
GIVE command declare output file for SORT and MERGE operation.

Same rules of USE control statement.

Format for GIVE:
GIVE <filename > ORG <organization> RECORD [<record format>,< length>]
[KEY ({Pos},{Len},{KeyType})

GIVE <filename > ORG <organization> RECORD [<record format>, <lenght min>,< length max>]
[KEY ({Pos},{Len},{KeyType})

6.7.INCLUDE/OMIT

INCLUDE condition statement is used for select records to insert in the file output.
OMIT condition statement is used for exclude certain records from the file input.

INCLUDE/OMIT COND=(condition) [FORMAT=type]

condition
Format 1 (pos , len , type , cond, pos , len , type)
Format 2 (pos , len , type , cond, [X|C|Z]'[value]')
Format 3 (condition , relcond , condition)

Format 1 (pos , len , type , cond, relcond , pos , len , type)

pos specifies the first byte of a control field relative to the beginning of the input record.
The first data byte of a fixed-length record has relative position 1.
The first data byte of a variable-length record has relative position 1.
len specifies the length of the field. Values for all fields must be expressed in integer numbers of
bytes.
type specifies the format of the data of field.

Type Description
CH Char
BI Binary unsigned
FI Binary signed

15 Jan 2016 Version Pag. 17

 FL Floating Point
PD Packed
ZD Zoned
CLO Numeric sign leading
CSL Numeric sign leading separate
CST Numeric sign trailing separate
SS Search Substring

cond Comparison operators are as follows:

EQ Equal to
NE Not equal to
GT Greater than
GE Greater than or equal to
LT Less than
LE Less than or equal to
SS Search Substring

With the SearchSubstring option, you can search for substrings within a field. The length can be greater
than the length of the substring. It is possible to search for multiple substrings within the field.

Examples:

INCLUDE COND=(1,100,SS,EQ,C'66666')

INCLUDE FORMAT=SS,COND=(18,2,EQ,C'00,88,99')

Format 2 (pos , len , type , cond, [X|C]'[value]')|[+/-nnnn]

pos specifies the first byte of a control field relative to the beginning of the input record.
The first data byte of a fixed-length record has relative position 1.
The first data byte of a variable-length record has relative position 1.
len specifies the length of the field. Values for all fields must be expressed in integer numbers of
bytes.
type specifies the format of the data of field.
Type Description
CH Char
BI Binary unsigned
FI Binary signed
FL Floating Point
PD Packed
ZD Zoned
CLO Numeric sign leading
CSL Numeric sign leading separate
CST Numeric sign trailing separate
SS Search Substring

cond Comparison operators are as follows:

15 Jan 2016 Version Pag. 18
 EQ Equal to
NE Not equal to
GT Greater than
GE Greater than or equal to
LT Less than
LE Less than or equal to

C’cc…c’ Character String Format . The value c is a ASCII character/string.

X’hh..hh’ Hexadecimal String Format. The value hh represents any pair of hexadecimal digits.

+/- nnnn.. Decimal Number Format

Format 3 (condition , relcond , condition)

condition Format 1 or Format 2

relcond Relational conditions can be logically combined, with AND or OR.

The relational condition specifies that a comparison test be performed.
Relational conditions can be logically combined, with AND or OR.

15 Jan 2016 Version Pag. 19

 Format 4 (pos, len , CHANGE=(vlen, [X|C]'[value Find]', [X|C]'[value Set]' ….
NOMATCH=([X|C]'[value]')

CHANGE Specifies how the input field or parsed input field is to be changed to the output field, using a
lookup table.

NOMATCH if an input field value does not match any of the find constants, NOMATCH values is used for
output field.

Format 5 (pos, len , CHANGE=(vlen, [X|C]'[value Find]', posFind,

lenFind …. NOMATCH=(posNoMatch, lenNomatch)

CHANGE Specifies how the input field or parsed input field is to be changed to the output field, using
position(posFind) and length(lenFind) of input record.

NOMATCH if an input field value does not match any of the find constants, NOMATCH input record position
and length are used for output field.

6.8.INREC/OUTREC

INREC redefines the structure of record input. This operation is executed after read file input e before all
operations.
The INREC control statement reformat the input records before they are sorted, merged, or copied.
All fields specifications presents in OUTREC, Sort Key, … must be referred to a new structure defined by
INREC.

Format 1 INREC FIELDS=(FIELD-SPEC...)

Format 2 INREC BUILD=(FIELD-SPEC...)
Format 3 INREC OVERLAY=(FIELD-SPEC...)
Format 4 INREC FINDREP=(FIELD-FINDEREP-SPEC

OUTREC defines structure record output for output file.

Format 1 OUTREC FIELDS=(FIELD-SPEC...)
Format 2 OUTREC BUILD=(FIELD-SPEC...)
Format 3 OUTREC OVERLAY=(FIELD-SPEC...)
Format 4 INREC FINDREP=(FIELD-FINDEREP-SPEC

Use OVERALY only to overwrite existing columns or to add fields at end of every record.

Field specification is the same for INREC and OUTREC.

15 Jan 2016 Version Pag. 20

BUILD or FIELDS are synonymous.

15 Jan 2016 Version Pag. 21

FIELD-SPEC ( pos, len | posOut:pos,len | n:X | n:Z |nC’constant’ | nX | nZ, |X’hh’ )
One or more occurrence of follow elements, separated by comma.

pos, len pos = position input record, len = length of field

posOut:pos,len posOut = position output, pos = position input record, len = length of field
n:X Filling with Blank character (0x20) from last position to n (absolute
position of output record).
n:Z Filling with zero Binary (0x00) character from last position to n (absolute
position of output record).
C’constant’ constant character value.
nC’constant’ repeat n times constant character value.
nX repeat n times Blank character.
nZ repeat n times Binary (0x00) character.
X’hh…hh’ hexdecimal string .
nX’hh…hh’ repeat n times hexdecimal string .

_______________________________________________________________________________________

FIELD-FINDREP-SPEC__Field Find/Replace Specification

IN=C'constant' , OUT=C'constant' constant character value.
IN=(C'constant', C'constant' ....) , OUT=C'constant' constant character value.
INOUT=(C'constantIn', C'constantOut' , C'constantIn', C'constantOut', ....)
STARTPOS=pos pos = Start Position to find/replace
ENDPOS=pos pos = End Position to find/replace
DO=n n=Maximum number of times find and replace
MAXLEN=n n=Maximum len of record n
OVERRUN=TRUNC|ERROR Truncate or Error(Default) for overrun
SHIFT=YES|NO Shift data or no (default) when different length between find replace
_______________________________________________________________________________________

15 Jan 2016 Version Pag. 22

 6.9.SUM FIELDS

SUM FIELDS is command for aggregate record and summarize value for numeric fields.
All fields present in SUM FIELDS are aggregate when more records has same key.

Format 1 SUM FIELDS = (pos,len,type, ...)

Format 2 SUM FIELDS = (NONE) or SUM FIELDS = NONE

There are two formats for SUM FIELD, the first summarize numeric fields, the send NOT summarize, but
eliminate duplicate key.

Format 1 SUM FIELDS = (pos,len,type, ...)

pos specifies the first byte of a control field relative to the beginning of the input record.
The first data byte of a fixed-length record has relative position 1.
The first data byte of a variable-length record has relative position 1.
len specifies the length of the field. Values for all fields must be expressed in integer numbers of bytes.
type specifies the format of the data of field.
Type Description
BI Binary unsigned
FI Binary signed
FL Floating Point
PD Packed
ZD Zoned
CLO Numeric sign leading
CSL Numeric sign leading separate
CST Numeric sign trailing separate

Format 2 SUM FIELDS = (NONE) or SUM FIELDS = NONE

In this case Format2 insert into output file one occurrence of same key specified by SORT KEY.
The record output contains the first record in order of reading.
For identify a first occurrence of data, GCSORT verified the value of pointer of record into file input,
selecting the lowest value.

6.10. RECORD
RECORD control statement is option to specify the type and lengths of the records.

RECORD [TYPE=[{V}/{F}(Fixed-length)]] , [LENGTH=[{len}(L1-Input record length)]

','[{len}(L2-Record length)]
','[{len}(L3-Output record length)]

TYPE = V (Variable-length) / F (Fixed-length)

LENGTH = (L1, L2, L3)

15 Jan 2016 Version Pag. 23

 L1 = Input length
L2 = Record length after E15
L3 = Output record length
L1 is ignored if the input record length is available from USE command.
L2 is ignored if E15 is not used.
L3 is ignored if the input record length is available from GIVE command.

Example:
[ RECORD CONTROL STATEMENT ]

SORT FIELDS=(8,5,CH,A) USE ../files/sqbig01.dat ORG SQ GIVE ../files/sqbig01_gcs.srt ORG SQ RECORD TYPE=F, LENGTH=500

RECORD TYPE=F, LENGTH=(500)

RECORD TYPE=F, LENGTH=(500, ,500)

RECORD TYPE=F LENGTH=(,,500)

RECORD TYPE=F,LENGTH=(,,500)

6.11. OUTFIL
OUTFIL is command to create one or more output file for a sort, copy, or merge operation.
Each file output is defined from OUTFIL command

FORMAT

OUTFIL
FILES/FNAMES= (environment variable)
STARTREC=nn
ENDREC=nn
[SAVE|[INCLUDE|OMIT] (CONDITION) [FORMAT=TYPE]]
SPLIT
OUTREC = (FIELD-SPEC...)

OUTFIL
FILES/FNAMES=filename filename = Identify a environment variable the contain the file
name
STARTREC=nn Start write after nn records
ENDREC=nn Stop write after nn records

SAVE Save records that not used by command INCLUDE/OMIT.

15 Jan 2016 Version Pag. 24

 INCLUDE/OMIT (CONDITION) [FORMAT=TYPE]] Same definition for COND-FIELD (INCLUDE/OMIT)
SPLIT Split 1 record for each File in Group definition (FILE=file1,file,file2)
SPLITBY=n Split n records for each File in Group definition (FILE=file1,file,file2)
OUTREC = (FIELD-SPEC...) Define structure output data. Same definition for
(FIELD-SPEC...).

If the environment variable filename for FILES/FNAMES is not defined, GCSort writes output file in local
folder assuming the name equal at value of identifier filename (FILES/FNAMES=filename).

If OUTFIL does not include the definition of FNAMES/FILES the input data will be written to the GIVE file.

6.12. OPTION

This command allows you to change the behavior of the utility.

Format1 OPTION [SKIPREC=nn]|[ STOPAFT=nn]|[ VLSCMP]|[ VLSHRT] | [Y2PAST=[YY] | [YYYY]]
SKIPREC=nn Skip nn records from input
STOPAFT=nn Stop read after nn records
VLSCMP 0 disabled , 1 = enabled -- temporarily replace any
missing compare field bytes with binary zeros
VLSHRT 0 disabled , 1 = enabled -- treat any comparison
involving a short field as false
Y2PAST=YY (YY) – Sliding = Numbers of years to subtract from the current
year.
(YYYY) – Century= Specifies the beginning of the fixed century
window.
MODS [E15 =(<name>)] [E35=(<name>)] Routine name E15 and/or E35 Cobol Program.

Exit Routines

E15 – Routine called after file read

E15 routine is a COBOL program.

Linkage :
LINKAGE for fixed records
01 RECORD - FLAGS PIC 9(8) BINARY.
88 FIRST - REC VALUE 00.
88 MIDDLE - REC VALUE 04.
88 END - REC VALUE 08.
01 NEW-REC PIC X(nn).
01 RETURN-REC PIC X(nn).
01 UNUSED1 PIC 9(8) BINARY.
01 UNUSED2 PIC 9(8) BINARY.
01 NEW-REC-LEN PIC 9(8) BINARY
(Only for Variable Length)
01 RETURN-REC-LEN PIC 9(8) BINARY (Only

15 Jan 2016 Version Pag. 25

for Variable Length)
01 UNUSED5 PIC 9(8) BINARY.
01 EXITAREA-LEN PIC 9(4) BINARY.
01 EXITAREA.
05 EAREA OCCURS 1 TO 256 TIMES
DEPENDING ON EXITAREA-LEN PIC X.

E35 – Routine called before write output

E35 routine is a COBOL program.

LINKAGE for fixed records

01 RECORD-FLAGS PIC 9(8) BINARY.
88 FIRST-REC VALUE 00.
88 MIDDLE-REC VALUE 04.
88 END-REC VALUE 08.
01 LEAVING-REC.
05 LREC OCCURS 1 TO 200 TIMES
DEPENDING ON LEAVING-REC-LEN PIC X.
01 RETURN-REC.
05 RREC OCCURS 1 TO 200 TIMES
DEPENDING ON RETURN-REC-LEN PIC X.
01 OUTPUT-REC.
05 OREC OCCURS 1 TO 200 TIMES
DEPENDING ON OUTPUT-REC-LEN PIC X.
01 UNUSED1 PIC 9(8) BINARY.
01 LEAVING-REC-LEN PIC 9(8) BINARY.
01 RETURN-REC-LEN PIC 9(8) BINARY.
01 OUTPUT-REC-LEN PIC 9(8) BINARY.
01 EXITAREA-LEN PIC 9(4) BINARY.
01 EXITAREA.
05 EAREA OCCURS 1 TO 256 TIMES
DEPENDING ON EXITAREA-LEN PIC X.

E15 - Return code

00 - No Action
04 - Record deleted
08 - Do Not Return
12 - Record inserted
16 - Terminate DFSORT
20 – Record Altered or Replaced

E35 – Return code

00 - No Action
04 – Record deleted
08 - Do Not Return
12 - Insert record
16 - End of GCSort

15 Jan 2016 Version Pag. 26

7. JOIN Statement

The purpose of the JOIN statement is to perform JOIN between two files (F1 and F2).
You can perform different types of join on two files (F1 and F2) by one or more keys with
GCSort using the following statements:

JOINKEYS

JOINKEYS specifies the definition of the JOIN key.

It is necessary to specify a JOINKEYS statement for each file, one for F1 and one for F2.
Each JOINKEYS statement must specify the starting position, the length and the sequence of the keys that
file. You can also optionally specify if the file is already sorted by the keys and if sequence checking of the
keys is not needed, or stop reading the file after n records.

JOIN

JOIN tells gcsort how to match records in the JOIN command.

Inner join – Default, only paired records from F1 and F2 are processed.
Left outer join - Unpaired F1 records as well as paired records.
Right outer join - Unpaired F2 records as well as paired records.
Full outer join - unpaired F1 and F2 records as well as paired records.
Unpaired F1,ONLY - Only unpaired F1 records
Unpaired F2,ONLY - Only unpaired F2 records
Unpaired F1,F2,ONLY / Unpaired,ONLY- Only unpaired F1 and F2 records

REFORMAT

REFORMAT statement specified the fields of F1 and/or F2 in the joined records.

15 Jan 2016 Version Pag. 27

Join Process Schema

15 Jan 2016 Version Pag. 28

For details gcsort -- help JOIN

gcsort --help JOIN

________________________________________________________________________________________
gcsort help
gcsort is a utility to sort, merge, copy and join records in a file into a
specified order in GnuCOBOL environment.
________________________________________________________________________________________
Syntax case insensitive
Return code : 0 (ok) - 4 (warning) - 16 (error)
________________________________________________________________________________________
Usage with file parameters : gcsort <options> take filename
Usage from command line : gcsort <options> <control statements>
________________________________________________________________________________________
gcsort options
-fsign=[ASCII|EBCDIC] define display sign representation
-fcolseq=[NATIVE|ASCII|EBCDIC] collating sequence to use
-febcdic-table=<cconv-table>/<file> EBCDIC/ASCII translation table
________________________________________________________________________________________
========================================================================================
Section for JOIN control statement
========================================================================================
JOIN file(s)
USE Declare input file F1
USE Declare input file F2
GIVE Declare output file
JOINKEYS FILES=F1.. Declare keys file F1
[ INCLUDE] Input file F1 - Select input records that respect include
condition(s)
[ OMIT ] Input file F1 - Omit input records that respect omit condition(s)
JOINKEYS FILES=F2.. Declare keys file F2
[ INCLUDE] Input file F2 - Select input records that respect include
condition(s)
[ OMIT ] Input file F2 - Omit input records that respect omit condition(s)
UNPAIRED Declare join type
REFORMAT FIELDS Declare output format
[ INCLUDE] Output file - Select input records that respect include condition(s)
[ OMIT ] Output file - Omit input records that respect omit condition(s)
[ INREC ] Output file - Reformat input record before join operation
[ OUTFIL ] Output file - Create one or more output files from join operation
________________________________________________________________________________________
JOIN
USE {Filename} [File F1]
ORG {Org}
RECORD [F,{RecordLen}] | [V,{MinLen},{MaxLen}]
[KEY ({Pos},{Len},{KeyType})]

USE {Filename} [File F2]

ORG {Org}
RECORD [F,{RecordLen}] | [V,{MinLen},{MaxLen}]
[KEY ({Pos},{Len},{KeyType})]

GIVE same parameters of USE

JOINKEYS FILES=F1,FIELDS=[({Pos},{Len},{Order}, ...)]

[,SORTED] [,STOPAFT={nn]]
[, INCLUDE ] | [, OMIT]
[ COND=({Condition})[,FORMAT={FormatType}] ]

JOINKEYS FILES=F2,FIELDS=[({Pos},{Len},{Order}, ...)]

[,SORTED] [,STOPAFT={nn]]

15 Jan 2016 Version Pag. 29

 [, INCLUDE ] | [, OMIT]
[ COND=({Condition})[,FORMAT={FormatType}] ]

JOIN UNPAIRED [,F1][,F2][,ONLY]

UNPAIRED,F1,F2 or UNPAIRED
Unpaired records from F1 and F2 as well as paired records (Full outer join).
UNPAIRED,F1
Unpaired records from F1 as well as paired records (Left outer join).
UNPAIRED,F2
Unpaired records from F2 as well as paired records (Right outer join).
UNPAIRED,F1,F2,ONLY or UNPAIRED,ONLY
Unpaired records from F1 and F2.
UNPAIRED,F1,ONLY
Unpaired records from F1.
UNPAIRED,F2,ONLY
Unpaired records from F2.

REFORMAT FIELDS=({File}:{Pos},{Len},{?},{File}:{Pos},{Len}.....) [,FILL=[C'constant']

| [X'hh']
______Commands for output file__________________________________________________________
INCLUDE | OMIT
COND=({Condition})[,FORMAT={FormatType}]

INREC FIELDS | INREC BUILD =({FieldSpec})

INREC OVERLAY =({FieldSpec})
OUTREC FIELDS | OUTREC BUILD =({FieldSpec})
OUTREC OVERLAY =({FieldSpec})

OUTFIL
INCLUDE | OMIT ({Condition})[,FORMAT={FormatType}]
OUTREC BUILD | BUILD = ({FieldSpec})
FILES/FNAMES= {Filename}

________________________________________________________________________________________
___{Parameters}____________________________|___{Parameters}_____________________________
{File} = F1 or F2 | ? = 1-byte indicator joined record
{Pos} = Field Position | 'B' = 'Both' - Key found in F1 and F2
{Len} = Field Length | '1' = Key found in F1, but not in F2
{Order} = A(ascending) | D(descending)| '2' = Key found in F1, but not in F1
C'Constant'= Character fill byte | nn = Numbers of records from input file
X'hh' = Hexadecimal fill byte (00-FF). |
________________________________________________________________________________________
___{Parameters}____________________________|___{Relational}_____________________________
{FileName} = Filename or Env. Variable | EQ = Equal
{Pos} = Field Position | GT = GreaterThan
{Len} = Field Length | GE = GreaterEqual
{RecordLen}= Record Length | LT = LesserThan
{MinLen} = Min size of record | LE = LesserEqual
{MaxLen} = Max size of record | NE = NotEqual
{Order} = A(ascending) | D(descending)| SS = Substring (only for Field Type 'CH')
___________________________________________|____________________________________________
___{Condition}__________________________________________________________________________
Format 1 - (Pos,Len,{FormatType},{Relational},[AND|OR],Pos,Len,{FormatType})
Format 2 - (Pos,Len,{FormatType},{Relational},[X|C'[value]'] | numeric value)]
Format 3 - ( {Condition} ,[AND|OR],{Condition} )
Format 4 - ( Pos,Len,{FormatType},{Relational}, [DATE1][(+/-)num] | [DATE2][(+/-)num]
[DATE3][(+/-)num] | [DATE4][(+/-)num]
DATE - Currente Date : DATE1 (C'yyyymmdd'), DATE2 (C'yyyymm'),
DATE3 (C'yyyyddd'), DATE4 (C'yyyy-mm-dd') (no Timestamp)
[(+/-)num] [+num] future date, [-num] past date) only for DATE1,DATE2,DATE3
________________________________________________________________________________________
___{Org}___File Organization_______________|___{KeyType}____Mandatory for ORG = IX______
LS = Line Sequential | P = Primary Key
SQ = Sequential Fixed or Variable | A = Alternative Key

15 Jan 2016 Version Pag. 30

 IX = Indexed Fixed or Variable | D = Alternative Key with Duplicates
RL = Relative Fixed or Variable | C = Continue definition
___________________________________________|____________________________________________
__{FormatType}____Field Format Type________|___{FormatType2}____Format Type SumField____
CH = Char | BI = Binary unsigned
BI = Binary unsigned | FI = Binary signed
FI = Binary signed | FL = Floating Point
FL = Floating Point | PD = Packed
PD = Packed | ZD = Zoned
ZD = Zoned | CLO = Numeric sign leading
CLO = Numeric sign leading | CSL = Numeric sign leading separate
CSL = Numeric sign leading separate | CST = Numeric sign trailing separate
CST = Numeric sign trailing separate | SS = Substring
___________________________________________|____________________________________________
_Format_Len_Type__Date field_______________|_Format_Len_Type__Date field________________
Y2T = 8 ZD CCYYMMDD | Y2D = 1 PD YY
Y2T = 4 ZD YYXX | Y2P = 2 PD YY
Y2T = 2 ZD YYX | Y2U = 3 PD YYDDD
Y2T = 3 ZD YY | Y2S = 2 ZD YY
Y2T = 5 ZD YYDDD | Y2V = 4 PD YYMMDD
Y2T = 6 ZD YYMMDD | Y2X = 3 PD DDDYY
Y2B = 1 BI YY | Y2Y = 4 PD MMDDYY
Y2C = 2 ZD YY | Y2Z = 2 ZD YY
___________________________________________|____________________________________________
____{FieldSpec}___Field Specification___________________________________________________
pos, len pos = position input record, len = length of field
posOut:pos,len posOut = position output, pos = position input , len = length
n:X Filling with Blank character from last position to n
(absolute position of output record).
n:Z Filling with zero Binary character from last position to n
(absoluteposition of output record).
C'constant' constant character value.
nC'constant' repeat n times constant character value.
nX repeat n times Blank character.
nZ repeat n times Binary (0x00) character.
X'hh....hh' hexdecimal characters.
nX'hh...hh' repeat n times hexdecimal characters.
CHANGE=(vlen,[C|X]'<valueFind>',[C|X]'<valueSet>',.....),NOMATCH=([C|X] '<valueSet>')
CHANGE=(vlen,[C|X]'<valueFind>', posIn, lenIn), NOMATCH = (posIn, posLen)
________________________________________________________________________________________
Environment Variables
________________________________________________________________________________________
COB_VARSEQ_FORMAT Used by GnuCOBOL
GCSORT_DEBUG 0 no print info, 1 info DEBUG, 2 for info Parser
GCSORT_MEMSIZE Memory Allocation in byte (Default 512000000 byte)
GCSORT_PATHTMP Pathname for temporary files (Default TMP / TEMP / TMPDIR)
GCSORT_STATISTICS 0 minimal informations, 1 for Summary, 2 for Details
GCSORT_TESTCMD 0 for normal operations , 1 for ONLY test command line (NO SORT)
________________________________________________________________________________________

15 Jan 2016 Version Pag. 31

8. Environment Variables
8.1. Byte Order
GCSort can treat numeric fields in both binary format BigEndian or Native. To indicate a byte order is used
environment variable GCSORT_BYTEORDER that assume 0 for Native or 1 for BigEndian. This value affects
the treatment of SORT and SUM KEY FIELDS.

8.2.Temporary Files
When dimension of files input is greater of memory available, GCSort creates temporary files for sort
operation. Temporary files is created in pathname specified from GCSORT_TMPFILE environment variable, if
this value is not available, GCSort use TMP/TEMP environment variable or use current directory. For
Windows the filename is composed from:
- Prefix = Srt
- Name = name ( created from GetTempFileName())
- Extension = .tmp
-
For Linux file name is composed from:
- Prefix = Srt
- Name = PID of process GCSort
- Num = Progressive of file
- Extension = .tmp
Temporary files are destroyed after sort operation.

8.3. Memory Allocation

The environment variable GCSORT_MEMSIZE specify amount of memory that GCSORT will use for sort
operation.

GCSort analyze the value and made two area for sort operation:

(1) Key Area : this area is used for sort in memory

(2) Data Area : this area contains data record

The optimization for use of memory GCSort check dimension of key and record.

Key Area = [GCSORT_MEMSIZE] * ((Key Length + 8 + 4 + 8) / Record Length)

Data Area = [GCSORT_MEMSIZE] - Key Area

(8 + 4 + 8) 8 is pointer of record into file, 4 record length, 8 pointer to record area in memory.

15 Jan 2016 Version Pag. 32

If value of ((Key Length + 8 + 4 + 8)/ Record Length) is minor of 15% or major of 50%, GCSORT force this
value to 15%.

8.4.Statistics

GCSort produce in output a lot of information about execution.

You can setting GCSORT_STATISTICS environment variable to three values:

0 = minimal information

Example:
========================================================
GCSort Version 01.00.00
========================================================
TAKE file name
D:\GNU_COBOL\GCSort_1_0_0\gcsort_testcase\take\par_SORT_debug.par
========================================================
File : D:\GCSORTTEST\OCFILES\TEST9\INP000.txt
Size : 1194
========================================================
Record Number Total : 15
Record Write Sort Total : 0
Record Write Output Total : 15
========================================================
Start : Mon Jan 25 11:17:55 2016
End : Mon Jan 25 11:17:55 2016
Elapsed Time 00hh 00mm 00ss 000ms

Sort OK

1 = medium information

Example

========================================================
GCSORT
File TAKE : D:\GNU_COBOL\GCSort_1_0_0\gcsort_testcase\take\par_SORT_debug.par
========================================================
SORT FIELDS(3,1,CH,A)
USE D:\GCSORTTEST\OCFILES\TEST9\INP000.txt ORG LS RECORD V,1,27990
GIVE D:\GCSORTTEST\OCFILES\TEST9\OUT000.SRT ORG LS RECORD V,1,27990

========================================================
GCSort Version 01.00.00
========================================================
TAKE file name
D:\GNU_COBOL\GCSort_1_0_0\gcsort_testcase\take\par_SORT_debug.par
========================================================
Operation : SORT

15 Jan 2016 Version Pag. 33

INPUT FILE :
D:\GCSORTTEST\OCFILES\TEST9\INP000.txt VARIABLE (1,27990) LS
OUTPUT FILE :
D:\GCSORTTEST\OCFILES\TEST9\OUT000.SRT VARIABLE (1,27990) LS
SORT FIELDS : (3,1,CH,A)
========================================================
File : D:\GCSORTTEST\OCFILES\TEST9\INP000.txt
Size : 1194
========================================================
Record Number Total : 15
Record Write Sort Total : 0
Record Write Output Total : 15
========================================================
Start : Mon Jan 25 11:20:01 2016
End : Mon Jan 25 11:20:01 2016
Elapsed Time 00hh 00mm 00ss 000ms

Sort OK

2 = details information
========================================================
GCSORT
File TAKE : D:\GNU_COBOL\GCSort_1_0_0\gcsort_testcase\take\par_SORT_debug.par
========================================================
SORT FIELDS(3,1,CH,A)
USE D:\GCSORTTEST\OCFILES\TEST9\INP000.txt ORG LS RECORD V,1,27990
GIVE D:\GCSORTTEST\OCFILES\TEST9\OUT000.SRT ORG LS RECORD V,1,27990

========================================================
GCSort Version 01.00.00
========================================================
TAKE file name
D:\GNU_COBOL\GCSort_1_0_0\gcsort_testcase\take\par_SORT_debug.par
========================================================
Operation : SORT

INPUT FILE :
D:\GCSORTTEST\OCFILES\TEST9\INP000.txt VARIABLE (1,27990) LS
OUTPUT FILE :
D:\GCSORTTEST\OCFILES\TEST9\OUT000.SRT VARIABLE (1,27990) LS
SORT FIELDS : (3,1,CH,A)
========================================================
File : D:\GCSORTTEST\OCFILES\TEST9\INP000.txt
Size : 1194
After job_loadFiles - Mon Jan 25 11:21:44 2016
After job_sort - Mon Jan 25 11:21:44 2016
After job_save - Mon Jan 25 11:21:44 2016
========================================================
Record Number Total : 15
Record Write Sort Total : 0
Record Write Output Total : 15
========================================================

Memory size for GCSort data : 133875000

Memory size for GCSort key : 23625000
BufferedReader MAX_BUFFER : 4063232
MAX_SIZE_CACHE_WRITE : 4063232
MAX_SIZE_CACHE_WRITE_FINAL : 4063232
MAX_MLTP_BYTE : 63
BYTEORDER : 0

15 Jan 2016 Version Pag. 34

===============================================

Start : Mon Jan 25 11:21:44 2016

End : Mon Jan 25 11:21:44 2016
Elapsed Time 00hh 00mm 00ss 000ms

Sort OK

15 Jan 2016 Version Pag. 35

9. Command Line
GCSort command line accepts the following parameters:

gcsort print version and options.

gcsort --help print help.

gcsort --help SORT | MERGE | COPY | JOIN print help for specific control statement.

gcsort --version print version.

gcsort --config print the value of environment variables.

gcsort command line execute command line.

gcsort TAKE filename read filename where are present commands for Sort/Merge.

The file used in the TAKE command is free format.

10. Padding and Truncating

GCSort uses LIBCOB that defines how made record in write output operation.

Use LSF file organization when the record to be sorted contains trailing spaces and you need fixed-length
records (GCSort does not delete trailing spaces).

Otherwise, you can set the environment variable COB_LS_FIXED=1 before running the GCSort command to
NOT delete trailing spaces.

11. Retun Code

GCSort has two values for return code:

0 for Success

4 for Warning

16 for Failure

15 Jan 2016 Version Pag. 36

12. File Conversion
GCSort permit to specify ‘ORGANIZATION’ and ‘RECORD TYPE’ for output data different structure from input
data, to permit the conversion of file format.

In this case GCSort convert data from a structure to another structure, for example, from Sequential to Line
Sequential or vice versa.

If you want sort a text file (LS) and you don’t know the record length, you can specify RECORD V with max
len very large, example:

SORT KEY (1,20,CH,A)

USE F1.TXT ORG LS RECORD V,1,3000
GIVE F1.TXT.OUT ORG LS RECORD V,1,3000

13. Performance and Tuning

For tuning performance of GCSort is good practices modify the settings of value for memory allocation and
modify dimension of area for Memory Mapped File.

GCSORT_MEMSIZE Indicate amount of memory for sort.

GCSORT_MLT Indicate the number of views for MMF in temporary files. This number is multiplied
by Page Size of system (example 65536). Increasing this value the view for read file
in memory is more greater and can reduce the elapsed time.(Temporary files).

By default GCSORT_MLT assume 63 ( Example: 63 * 65536 = 4Mbyte dimension of view for MMF).

15 Jan 2016 Version Pag. 37

14. Limits
The max numbers of input files for Merge is 16.

The max numbers of temporary files is 16. The temporary files is reused when the size of files input is more
of size of (Memory GCSORT_MEMSIZE * 16 files).

15. Errors and Warnings

GCSORT produces two types of messages:

- Error format ‘*GCSort*Snnn’

- Warning format ‘*GCSort*Wnnn’

For Error message GCSort break execution and terminate operation with message and return code.

For Warning message GCSort continue execution and continue operation with message.

The message string identify a specific condition of error or warning, in the of warning print a specific action.

15 Jan 2016 Version Pag. 38

16. GCSort by examples

16.1. SORT
SORT single file

====================================================================
SORT FIELDS(3,1,CH,A)
USE ../PJTestCaseSort/SQBI01 RECORD F,51 ORG SQ
GIVE ../PJTestCaseSort/SQBI01.SRT.TST RECORD F,51 ORG SQ
====================================================================

SORT single file with INCLUDE condition

Order KEY
1) Position 37, Len 1, Character, Descending
2) Position 18, Len 17, Character, Ascending
Filter only records with character in position 37 Equal ‘C’.
=====================================================
SORT FIELDS=(37,1,CH,D,18,17,CH,A)
INCLUDE COND=(37,1,EQ,C'C') FORMAT=CH
USE FIL_100.TXT RECORD F,3000 ORG LS
GIVE FIL_100.TXT.SRT RECORD F,3000 ORG LS
=====================================================

16.2. MERGE
MERGE

Merge files with KEY Position 1, Len 50, Char, Ascending

Input files sorted
Input Record Variable from 1 to 27990 ORGanization Sequential
Output Record Variable from 1 to 27990 ORGanization Sequential
========================================================================
MERGE FIELDS(1,50,CH,A)
USE D:\GCSORTTEST\OCFILES\RGX10.DAT RECORD V,1,27990 ORG SQ
USE D:\GCSORTTEST\OCFILES\RGX10.DAT RECORD V,1,27990 ORG SQ
USE D:\GCSORTTEST\OCFILES\RGX10.DAT RECORD V,1,27990 ORG SQ
GIVE D:\GCSORTTEST\OCFILES\RGX10.DAT.MRG RECORD V,1,27990 ORG SQ
========================================================================
MERGE

FIELDS=COPY
Copy records from input to output.
Include condition check binary value (low-value)
Pos Len Condition Value
from 305 04 Not Equal Hex '00000000'

15 Jan 2016 Version Pag. 39

========================================================================
USE D:\GCSORTTEST\FilesT\FIL_OUTFIL_500.TXT ORG LS RECORD F,3000
GIVE D:\GCSORTTEST\FilesT\FIL_OUTFIL_500_023.TXT.SRT ORG LS RECORD F,3000
OPTION VLSHRT,VLSCMP,EQUALS
MERGE FIELDS=COPY
INCLUDE COND=(305,4,NE,X'00000000'),FORMAT=CH
========================================================================

16.3. COPY

COPY

Copy data from input to output with record filter.

Input FIXED Line Sequential, Output FIXED Line Sequential
Omitted (not insert in output file) records with condition:
a) Position 1, Len 12, EQual , Character '000000006060'
OR
b) Position 1, Len 12, EQual , Character '000000000030'
OR
c) Position 1, Len 12, EQual , Character '000000000051'
=====================================================
USE F1IN.DAT RECORD F,3000 ORG LS
GIVE F1IN.DAT_002.SRT RECORD F,3000 ORG LS
MERGE FIELDS=COPY
OMIT COND=(01,12,EQ,C'000000006060',OR,
01,12,EQ,C'000000000030',OR,
01,12,EQ,C'000000000051'),FORMAT=CH
=====================================================

SORT without duplicates

Sort Key Pos 5, len 6, Ascending

SUM FIELDS = (NONE) delete duplicates
=====================================================
USE FIL_OUTFIL_100.TXT ORG LS RECORD F,3000
GIVE FIL_OUTFIL_100_020.TXT.SRT ORG LS RECORD F,3000
SORT FIELDS=(5,6,A),FORMAT=CH,EQUALS
SUM FIELDS=(NONE)
=====================================================

16.4. SUMFIELDS

SUMFIELDS

Sort Key Pos 1, len 1, Ascending

SUM FIELDS Binary fields
========================================================================
SORT FIELDS(3,1,CH,A)
SUM FIELDS=(1,2,BI,7,3,BI,15,4,BI,20,3,BI,29,4,BI,34,8,BI,43,8,BI)

15 Jan 2016 Version Pag. 40

USE ../PJTestCaseSort/SQBI01 RECORD F,51 ORG SQ
GIVE ../PJTestCaseSort/SQBI01.SRT.TST RECORD F,51 ORG SQ

16.5. OUTREC
OUTREC FIELDS/BUILD

SORT FIELDS = COPY (copy record NO Sort)

Format output : OUTREC
Output structure
Pos Len Value
01 16 Record input Pos:1,Len 16
17 2 Blank ('X' = blank)
19 2 Record input Pos:18,Len 2
21 1 Character '-'
23 2 Record input Pos:20,Len 2
25 1 Character '-'
26 2 Record input Pos:22,Len 2
28 2 2 blank
========================================================================
USE ../Files/FIL_OUTFIL_200.TXT ORG LS RECORD F,3000
GIVE ../Files/FIL_OUTFIL_200_007.TXT.SRT ORG LS RECORD F,3000
SORT FIELDS=COPY
OUTREC=(01,16,2X,18,2,C'-',20,2,C'-',22,2,2X)
END

OUTREC FIELDS=(8,2, 20:5,10,3C'ABC’,80:X)

Position Len Position Len output Value

Input Input output
8 2 1 2

5 10 20 10 Characters from pos 5, len10 from

input
30 9 (3 times x 3 ‘ABCABCABC’
char)
80 Padding from 39 to 80

OUTREC FIELDS=(5C'LITERAL -',10X'414243',3X'525558',120,18)

Position Len Position Len output Value

Input Input output
1 45 (5 time x 9 ‘LITERAL -LITERAL -LITERAL LITERAL-
char) LITERAL-’

46 30 (10 times 1 ‘ABCABCABCABCABCABCABCABCABCABC’

char hex)
76 9 (3 times x 3 ‘RUXRUXRUX’
char hex)
80 18 85 18 Input record from 80 for 18
characters

15 Jan 2016 Version Pag. 41

OUTREC FIELDS=(1,40,60:Z,81:X)

Position Len Position Len output Value

Input Input output
1 40 1 40 Input record from 1 for 40 characters
41 20 (60 abs 20 characters with ‘00’ binary
position – 40
current
position)
61 20 21 characters with ‘20’ space

16.6. OUTFIL
OUTFIL INCLUDE
Example with more files for OUTFIL
Each file output with Include condition
The purpose is merge files and write four output.
FNAMES=FOUT201_1
FOUT201_1 Environment Variable
FOUT201_2 Environment Variable
FOUT201_3 Environment Variable
FOUT201_SAVE Environment Variable
========================================================================
USE ../FIL_OUTFIL_001.TXT ORG LS RECORD F,3000
GIVE ../FIL_OUTFIL_001.TXT.OUT ORG LS RECORD F,3000
MERGE FIELDS=COPY
OUTFIL INCLUDE=(01,03,CH,EQ,C'201',AND,24,03,CH,LE,C'999'),FNAMES=FOUT201_1
OUTFIL INCLUDE=(01,03,CH,EQ,C'210',AND,24,04,CH,GT,C'0000',AND,24,04,CH,LE,C'9999'),FNAMES=FOUT201_2
OUTFIL INCLUDE=(01,03,CH,EQ,C'230',AND,36,04,CH,GT,C'0000',AND,36,04,CH,LE,C'9999'),FNAMES=FOUT201_3
OUTFIL SAVE,FNAMES=FOUT201_SAVE
========================================================================

OUTFIL OMIT
Format output record
OMIT Condition for input.
FOUTKEY_YES Environment Variable
FOUTKEY_NO Environment Variable
========================================================================
USE D:\GCSORTTEST\FilesT\FIL_OUTFIL_050.txt ORG LS RECORD F,3000
GIVE D:\GCSORTTEST\FilesT\FIL_OUTFIL_050.txt.OUT ORG LS RECORD F,3000
SORT FIELDS=COPY
OUTFIL OMIT=(156,15,CH,LT,141,15,CH,AND,005,10,CH,EQ,C'KEYMAX800E'),FNAMES=FOUTKEY_YES
OUTFIL SAVE,FNAMES=FOUTKEY_NO
END
========================================================================

16.7. INREC/OUREC CHANGE

[ INREC CHANGE ]
INREC FIELDS=(15,6,25,3,CHANGE=(1,C'K12',X'41',C'M22',X'42',C'P32',X'43'),NOMATCH=(X'49'))
INREC
FIELDS=(1,15,16,2,CHANGE=(1,C'22',X'41',C'88',X'48',C'44',X'42',C'66',X'43'),NOMATCH=(X'49'),17,83)

15 Jan 2016 Version Pag. 42

[ OUTREC CHANGE ]
OUTREC FIELDS=(15,6,25,3,CHANGE=(1,C'K12',X'41',C'M22',X'42',C'P32',X'43'),NOMATCH=(X'49'),26,4974)

[ CHANGE - Position ]
OUTREC FIELDS=(1,1,CHANGE=(6,C'2',28,6),NOMATCH=(2,6),X,8,19,35,15,51,59)

15 Jan 2016 Version Pag. 43

 16.8. DATE
[ DATE - Currente Date : DATE4 ]
INCLUDE COND=(1,13,CH,GT,DATE4)
USE ../files/inp5000.txt ORG LS RECORD F,5000
GIVE ../files/inp5000.txt.srt ORG LS RECORD F,5000
SORT FIELDS=(35,5,ZD,A)

[ DATE + / - day - month ]

COND=(1,13,CH,GT,DATE1+5)
COND=(1,13,CH,GT,DATE1-5)

COND=(1,13,CH,GT,DATE2+3)
COND=(1,13,CH,GT,DATE2-8)

COND=(1,13,CH,GT,DATE3+150)
COND=(1,13,CH,GT,DATE3-15)

[ DATE4 ]
OMIT COND=(1,13,CH,GT,DATE4)

16.9. RECORD CONTROL STATEMENT

[ RECORD CONTROL STATEMENT ]

SORT FIELDS=(8,5,CH,A) USE ../files/sqbig01.dat ORG SQ GIVE ../files/sqbig01_gcs.srt ORG SQ

RECORD TYPE=F, LENGTH=500

RECORD TYPE=F, LENGTH=(500)

RECORD TYPE=F, LENGTH=(500, ,500)

RECORD TYPE=F LENGTH=(,,500)

RECORD TYPE=F,LENGTH=(,,500)

16.10. DATE – Option Y2PAST

[ DATE – Y2PAST ]
SORT FIELDS=(10,8,Y2T,A)
USE FDate.dat RECORD F,85 ORG SQ
GIVE FDate.dat.Y2T8.srt RECORD F,85 ORG SQ
OPTION Y2PAST=80

15 Jan 2016 Version Pag. 44