VSAM to DB2

Online And Batch Program Conversion Strategy

1
Version 1.0

VSAM to DB2 Online and Batch Program Conversion Strategy

Preface
Scope and audience
This document is intended to assist in converting online programs using COBOL CICS
and batch programs using COBOL which access VSAM files to access DB2 tables.
.
Objectives
VSAM to DB2 conversion
Structure of this document
Document discusses schema translation strategy initially. Translation of batch COBOL
access calls for VSAM to DB2 is discussed. Translation of CICS access calls to DB2
access calls is detailed out next for each of the CICS commands for KSDS files. The
document ends with the performance tuning issues to be considered.
Related documents
VSAM to DB2 error handling document

VSAM to DB2 Online and Batch Program Conversion Strategy

Table of contents
1

OVERVIEW............................................................................................................................................6

SCHEMA TRANSLATION...................................................................................................................6

CONVERSION STRATEGY................................................................................................................7
3.1
BATCH PROGRAM TRANSLATION.......................................................................................................11
3.1.1
Specific Translation Rules......................................................................................................11
3.1.2
Sequential Organization Sequential Access...........................................................................12
3.1.2.1
3.1.2.2
3.1.2.3

3.1.3
3.1.3.1
3.1.3.2
3.1.3.3

3.1.4
3.1.4.1
3.1.4.2
3.1.4.3

3.1.5
3.1.5.1
3.1.5.2
3.1.5.3

INPUT.............................................................................................................................................12
I-O...................................................................................................................................................13
OUTPUT/EXTEND.......................................................................................................................13

Indexed Organization Sequential Access...............................................................................13

INPUT.............................................................................................................................................13
I-O...................................................................................................................................................15
OUTPUT / EXTEND.....................................................................................................................17

Indexed Organization Random Access...................................................................................17

INPUT.............................................................................................................................................17
I-O...................................................................................................................................................17
OUTPUT........................................................................................................................................17

Indexed Organization Dynamic Access..................................................................................17

INPUT.............................................................................................................................................18
I-O...................................................................................................................................................22
OUTPUT........................................................................................................................................26

3.2
ONLINE PROGRAM TRANSLATION.....................................................................................................27
3.2.1
READ......................................................................................................................................29
3.2.1.1
3.2.1.2
3.2.1.3

3.2.2
3.2.2.1
3.2.2.2
3.2.2.3

3.2.3
3.2.3.1
3.2.3.2
3.2.3.3

3.2.4
3.2.4.1
3.2.4.2
3.2.4.3

3.2.5
3.2.5.1
3.2.5.2
3.2.5.3

3.2.6
3.2.6.1
3.2.6.2
3.2.6.3

3.2.7
3.2.7.1
3.2.7.2
3.2.7.3

3.2.8
3.2.8.1
3.2.8.2
3.2.8.3

3.2.9

Function..........................................................................................................................................29
CICS Syntax...................................................................................................................................29
Translation of Syntax....................................................................................................................29

STARTBR................................................................................................................................31
Function..........................................................................................................................................31
CICS Syntax...................................................................................................................................31
Translation of Syntax......................................................................................................................31

READNEXT / READPREV.....................................................................................................34
Function..........................................................................................................................................34
CICS Syntax...................................................................................................................................34
Translation of Syntax.....................................................................................................................35

WRITE....................................................................................................................................38
Function..........................................................................................................................................38
CICS Syntax...................................................................................................................................38
Translation of Syntax.....................................................................................................................38

REWRITE...............................................................................................................................39
Function..........................................................................................................................................39
CICS Syntax...................................................................................................................................39
Translation of Syntax.....................................................................................................................39

DELETE.................................................................................................................................40
Function...........................................................................................................................................40
CICS Syntax...................................................................................................................................40
Translation Of Syntax....................................................................................................................40

UNLOCK................................................................................................................................42
Function..........................................................................................................................................42
CICS Syntax...................................................................................................................................42
Translation of Syntax....................................................................................................................42

ENDBR...................................................................................................................................43
Function..........................................................................................................................................43
CICS Syntax...................................................................................................................................43
Translation Of Syntax....................................................................................................................43

RESETBR...............................................................................................................................44

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.9.1
3.2.9.2
3.2.9.3

OPTIMIZING FOR PERFORMANCE.............................................................................................48

4.1
4.2
4.3

Function..........................................................................................................................................44
CICS Syntax...................................................................................................................................44
Translation Of Syntax....................................................................................................................44

SELECTING ONLY THE REQUIRED FIELDS..........................................................................................48

ADDING COMMIT LOGIC..................................................................................................................48
UPDATE ONLY THE REQUIRED COLUMNS..........................................................................................48

APPENDIX...........................................................................................................................................49
5.1.1
5.1.2
5.1.3
5.1.4
5.1.5
5.1.6
5.1.7
5.1.8
5.1.9
5.1.10
5.1.11
5.1.12
5.1.13
5.1.14
5.1.15
5.1.16
5.1.17
5.1.18
5.1.19
5.1.20

Sequential Read OPEN-CURSOR..........................................................................................49

Sequential Read START-CURSOR.........................................................................................50
Sequential Read READ-CURSOR..........................................................................................51
Random Read.........................................................................................................................52
Insert......................................................................................................................................53
Random Rewrite.....................................................................................................................54
Delete.....................................................................................................................................56
OPEN-CURSOR.....................................................................................................................57
START-CURSOR....................................................................................................................58
READ-CURSOR.....................................................................................................................59
Dependant-Cursor-1..............................................................................................................60
Dependant-Cursor-n..............................................................................................................60
Dependant-Updatable-Cursor-1............................................................................................60
Dependant-Updatable-Cursor-n............................................................................................60
FETCH-PARA-1.....................................................................................................................61
FETCH-PARA-n.....................................................................................................................61
INSERT-PARA-1.....................................................................................................................62
INSERT-PARA-n.....................................................................................................................62
UPDATE-PARA-1...................................................................................................................63
UPDATE-PARA-n...................................................................................................................63

VSAM to DB2 Online and Batch Program Conversion Strategy

1 Overview
This document discusses translation strategy for VSAM access calls appearing in online COBOL CICS and
batch COBOL programs by SQLs to access DB2 table.
Emphasis is on one to one translation of each VSAM access call to DB2 access SQLs.

2 Schema Translation
During this phase of the migration, VSAM file schema needs to be translated to relational schema.
For this a complete mapping document needs to be prepared for the VSAM files being converted to DB2
tables. This document has the VSAM file name, the record name, copybook name for this file, the field
names of the record, with the corresponding DB2 table names, column names, data description for the
columns, indicators to identify whether nulls are allowed, the column is an index and mapping exceptions
of any kind. It is understood that in such conversion there are cases of multiple mapping, like one VSAM
file mapped to many tables, many VSAM files mapped to one table, different record types of a VSAM file
mapped to different tables, one field of a VSAM record mapped to multiple columns of a table, many
fields of a VSAM record mapped to one table column.

VSAM to DB2 Online and Batch Program Conversion Strategy

3 Conversion Strategy
The conversion is based on the following translation rules :
1.

COBOL source code will remain same except for the introduction of new code and commenting of
existing code as part of translation. New code will be either embedded SQL statements or COBOL
constructs or INCLUDE copybooks, to support translation. Comments will be added to indicate the
program changes done.

2.

DCLGEN variables will be used as host variables. These values will be moved into the VSAM file
structure for the program processing.

3.

VSAM sequential operations will be translated using cursors on the associated tables.

4.

I/O operations on array fields of the record will be translated using cursors for normalised cases. In
denormalised cases the array fields will be translated to the corresponding columns on the DB2 table.

5.

No cursor will be defined if file is opened in OUTPUT or EXTEND mode.

6.

Any supported COBOL construct related to VSAM file, which is translated to relational equivalent or
not needed any more will be commented.

7.

In case of commenting OPEN statements, if the same statement opens some other non-VSAM files or
VSAM files which will not be converted, an OPEN statement to open those files will be added in the
program and the original OPEN statement will be commented.

8.

In case of commenting CLOSE statements, if the same statement closes some other non-VSAM files or
VSAM files which will not be converted, an CLOSE statement to close those files will be added in the
program and the original CLOSE statement will be commented.

9.

At any point of time, there will be only one cursor opened per table. Cursors will be defined and
opened with various conditions based on organization, access mode and file operation.

10. Cursor will be defined

With ORDER BY condition for indexed files.
With WHERE condition for START and READ
With FOR FETCH for read only operation
With FOR UPDATE for update operations
11. START operations will be translated to cursor declaration with WHERE condition. The condition used
in the WHERE condition will be same as the condition stated in the START. If START uses EQUAL
TO, then the = will be used in the WHERE condition. If START uses GREATER THAN, then > will
be used in the WHERE condition.
12. START can use a full-key or partial key. The translation will depend on the type of key used as well as
the key length specified.If START uses full key then the columns mapped to this will be used in the
WHERE clause. If START statement uses partial key there are 2 cases :
a.

if the keylength is a constant known at conversion time, the START will be translated to
DECLARE CURSOR using the table columns corresponding to this key length specified.

b.

If the key length is calculated at run time, it needs to be analysed and suitable WHERE clause
should be used in the DECLARE CURSOR clause.

13. Any column corresponding to key fields of the VSAM file needs to be used as an index on the table.
Alternate record keys can be considered as index on tables. This needs table to be indexed on columns
corresponding to alternate record keys.

VSAM to DB2 Online and Batch Program Conversion Strategy

14. CLOSE will be translated to closing the cursor(s) defined on the corresponding table. If same CLOSE
statement closes non-VSAM files or VSAM files not being converted , separate close statement to
close the these files will be generated and included in the source code.
15. Standard error Paragraphs will be performed to take care of any SQLCODE checking.
16. All the FILE STATUS logic used by the program will be retained for conditional processing. The FILE
STATUS for the converted VSAM file will be set by the standard error processing logic, depending on
the value of SQLCODE returned
17. One-to-one mapping will be handled by translating the VSAM access to the corresponding SQL
statement.
Sequential Operation for VSAM file
a.

Sequential READ operations will be translated to cursor FETCH.

b.

Sequential REWRITE will be translated to UPDATE with CURRENT OF condition.

c.

Sequential DELETE will be translated to DELETE with CURRENT OF condition.

Random Operation for VSAM file

a.

Random READ operations will be translated to SELECT with WHERE condition.

b.

Random REWRITE will be translated to UPDATE with WHERE condition.

c.

Random DELETE will be translated to DELETE with WHERE condition.

d.

WRITE will be translated to INSERT with VALUES clause.

18. In the case of one-to-many mapping, the following scenarios are possible :
a.

One VSAM file having different record types mapped to different tables
Sequential Operation on the VSAM file
Example of this is header and detail records in a VSAM file. In DB2 there can be table
defined for header records and another table defined for detail records. Any sequential read of
this VSAM file will be translated to opening a cursor on the header table, then fetch cursor
on the header table. Using the key retrieved from the header table open a cursor on the
detail table and continue fetch using this cursor until end-of-cursor, then close the cursor on
the detail table. Again fetch the cursor on the header table and continue the process of open
cursor and fetch cursor on the detail table.
For any sequential update or delete operations on this VSAM file, the same translation
strategy will be used but with the cursor option FOR UPDATE.
Random Operation on the VSAM file
If the VSAM file is being read randomly using a key value moved within the program, then
the program flow needs to be analysed to determine which table needs to be accessed.

b.

Fields of one VSAM file mapped to different tables

In such cases any input-output operation on this VSAM file will be translated to an equivalent
SQL statements to access data from all the associated tables.
Sequential Operation on the VSAM file
For a sequential read on this file, a cursor will be defined which will join required data from
all the associated tables.
For any sequential update or delete on the file, individual cursors will be defined on the
associated tables with the FOR UPDATE clause. In the case of a VSAM file mapped to two

VSAM to DB2 Online and Batch Program Conversion Strategy

DB2 tables, the cursor on the first table will be fetched, then the cursor on the second table
will be fetched. If both the rows fetched have the same key, then the DCLGEN values are
formatted as required and moved to the COBOL layout of the file. Values for any updates to
be made are moved in, and the tables are updated. In case the keys fetched by the two cursors
are not same, then this may be either because of some invalid data in one of the tables, or
because some rows are being added into either of the tables from some other VSAM file.
Since this may be a valid case as per application design, additional analysis will be required.
Random Operation on the VSAM file
For a random read, translate to a SELECT clause using a join from the associated tables.
For write, rewrite or delete, translate to a INSERT, UPDATE or DELETE with conditional
execution. If the operation on the first table was successful, then only attempt a similar
operation on the other table.
c.

Many VSAM files mapped to one table

Sequential Operation on the VSAM file
Translate the sequential operation to cursor operation on the table using only the columns
from table which has been mapped from the fields of this VSAM file. While inserting use
default values for the columns coming from other files. While updating, update only the table
columns mapped from the file being rewritten.
Random Operation on the VSAM file
For read, select only the table columns which have been mapped from the file being read.
For delete, verify if the row is present in the table before attempting a delete from the table,
If present then do a delete, else do not attempt to delete. This is because the delete operation
which might have been performed on the other VSAM file which also maps to this table,
might have already deleted the entire row. For update, update only the table columns which
have been mapped from this file. For insert, verify if the key being inserted is already present
in the table, if present then do an update instead of an insert, else do an insert. This is because
this row might already have been inserted when a write was performed on the other file which
also maps to this table. While inserting use default values for the columns coming from other
files.

d.

One field mapped to multiple columns

One field split into different columns
Situations were a field from the VSAM file is split into table columns will be taken care of as
per the splitting specified in the mapping input. This will be valid for any fields in the file
which is type DISPLAY. But any attempt to split a field defined as COMP or COMP-3 will
need additional analysis.
One field mapped to different columns
In situations were a whole field can be mapped to different columns depending on some
conditional processing, the exception logic associated with the mapping for that field will
have to be used. This exception logic will also be used to assign the proper DCLGEN
field to the COBOL field.

19. While populating default values for any NOT NULL fields in the DB2 table, assume 0 for numeric and
SPACES for CHAR fields as default.
Any NULL fetched from the table will be interpreted as a 0 if the receiving field is numeric or as
SPACES if the receiving field is alphanumeric.
20. Before updating or inserting into any column which has a numeric data type in DB2, perform a
numeric-check. If the value is non-numeric, assign the null indicator for nullable fields and
0 for non-nullable fields.

10

VSAM to DB2 Online and Batch Program Conversion Strategy

21. For groups containing numeric items mapped to a single numeric column, the group will be treated as
an alphanumeric and moved into the numeric column. Such translation will hold good only if the
numeric values in the sub-fields can be treated as an alphanumeric value from the business point of
view. For example it is ok to map a group item having sub-items as account and sub-account, both
numeric, into a single numeric table column as account. But if the same translation is performed for
items like amounts, then the result will be unpredictable.
22. In many situations a field defined in the VSAM layout and the corresponding DB2 table column will
have different lengths, this is because the datatypes in VSAM and DB2 are different and have their
own field length restrictions.
23. For variable length VSAM files being used in the program, the table will have columns corresponding
to the OCCURS fields. The number of columns will be the maximum possible value of the OCCURS
clause. After conversion to tables there will be a column for each occurance of the variable occurance
field, although all the columns may not be used or not have values populated in them. Within the
program the logic will be to use all the occurances of the field, but this may not be in tune with the
existing program logic.
24. New fields defined in the tables for any future use will be populated with the default values.
25. In case of CICS programs a record can be read into an area within the program, or into the CICS
buffer. In the later situation a pointer is used for accessing the buffer area where the record is read into.
After translation the use of this pointer will be avoided for data read from tables. Define a 01 level
item with length equal to the record length for the file for which the pointer was used. After accessing
data from the tables, format and move this data to the record layout as well as to this newly defined
data item. Any use of the data stored in the buffer identified by the pointer will be replaced by using
the newly defined data item.
26. Translate a START browse, READNEXT operation into OPEN cursor and FETCH cursor. After
each fetch, the key values retrieved will be saved and will be compared with the key value before the
next fetch. If the key values match continue with the fetch, else close the cursor and then open a new
cursor using the current key values and continue the fetch operation. This feature will allow the skip
sequential processing to be converted correctly.
27. Any VSAM files used in internal program sorting will also be translated to SQL statements.
28. After conversion, many file OPEN statements will have been commented because they are not required
for an indexed file which is converted to DB2 table. In such cases the file status for these commented
files have to be set up properly so that program processing continues normally. This can be done by
setting the file status to 00 for the file OPEN which was commented out.

11

VSAM to DB2 Online and Batch Program Conversion Strategy

3.1

Batch Program Translation

3.1.1

Specific Translation Rules

Following sections specify the translation strategy for each of the valid file I/O operation for VSAM files.
The rules are organized as follows:

For every organization (sequential/indexed/relative)

For every access mode (sequential/random/dynamic)
For every IO mode (input/output/input-output/extend)
All the IO statements applicable

Following <placeholder> convention is used in defining cursors and queries.

Name

Description

Dependant-cursor-1

Cursor defined on the tables containing columns corresponding to the

fields of the first array

Dependant-cursor-n

Cursor defined on the tables containing columns corresponding to the

fields of the nth array

Array-1

First array in the record

Db2-array-1

Newly defined structure similar to the first array but scalar, to be used
in moving data between tables and the array

Array-1-table-1

First table corresponding to the first array

array-1-table-1-col-1

First column in the first table corresponding to the first array

db2-array-1-table-1-col-1-fld-1

Field in the newly defined record corresponding to the first field of the
first array, which has mapping column in the first table

array-1-table-1-col-n

Nth column in the first table corresponding to the first array

db2-array-1-table-1-col-n-fld-n

Field in the newly defined record corresponding to the nth field of the
first array, which has mapping column in the first table

array-1-table-n

Nth table corresponding to the first array

array-1-table-n-col-1

First column in the nth table corresponding to the first array

db2-array-1-table-n-col-1-fld-1

Field in the newly defined record corresponding to the first field of the
first array, which has mapping column in the nth table

array-1-table-n-col-n

Nth column in the nth table corresponding to the first array

db2-array-1-table-n-col-n-fld-n

Field in the newly defined record corresponding to the nth field of the
first array, which has mapping column in the nth table

Db2-array-1-size

Newly generated data item to store the size of the first array. This data
item will be used in determining the number of times
fetch/select/insert/update to be done on the tables corresponding to the
array

Array-n

Nth array in the record

Db2-array-n

Newly defined structure similar to the nth array, to be used in moving

data between tables and the array

array-n-table-1

First table corresponding to the nth array

12

VSAM to DB2 Online and Batch Program Conversion Strategy

array-n-table-1-col-1

First column in the first table corresponding to the nth array

db2-array-n-table-1-col-1-fld-1

Field in the newly defined record corresponding to the first field of the
nth array, which has mapping column in the first table

array-n-table-1-col-n

Nth column in the first table corresponding to the nth array

db2-array-n-table-1-col-n-fld-n

Field in the newly defined record corresponding to the nth field of the
nth array, which has mapping column in the first table

array-n-table-n

Nth table corresponding to the nth array

array-n-table-n-col-1

First column in the nth table corresponding to the nth array

db2-array-n-table-n-col-1-fld-1

Field in the newly defined record corresponding to the first field of the
nth array, which has mapping column in the nth table

array-n-table-n-col-n

Nth column in the nth table corresponding to the nth array

db2-array-n-table-n-col-n-fld-n

Field in the newly defined record corresponding to the nth field of the
nth array, which has mapping column in the nth table

Db2-array-n-size

Newly generated data item to store the size of the nth array. This data
item will be used in determining the number of times
fetch/select/insert/update to be done on the tables corresponding to the
array

table-1-key-col-1

First key column in the first table containing columns corresponding

to the scalar fields of the record

table-1-key-col-n

Nth key column in the first table containing columns corresponding to

the scalar fields of the record

at-end-statements

Imperative statements to be executed on end of file

not-at-end-statements

Imperative statements to be executed on not end of file

Invalid-key-statements

Imperative statements to be executed on invalid key

Not-invalid-key-statements

Imperative statements to be executed on valid key

Db2-count

Newly generated data item to keep the loop count.

3.1.2

Sequential Organization Sequential Access

All the I/O operations in files with sequential organization and sequential access mode are sequential.

3.1.2.1 INPUT
Cursor Definition :
OPEN-CURSOR
Operation
OPEN
READ
CLOSE

Description
Open the defined cursor.

If the CLOSE statement closes any VSAM

file, then close the cursor opened for that
file.

Translated Code
EXEC SQL
OPEN <file1>-OPEN-CURSOR
END-EXEC.
Sequential-Read-OPEN-CURSOR
EXEC SQL
CLOSE <file1>-OPEN-CURSOR
END-EXEC

13

VSAM to DB2 Online and Batch Program Conversion Strategy

3.1.2.2 I-O
Cursor Definition :
OPEN-CURSOR
Operation
OPEN

Description
Open the defined cursor.

READ
REWRITE
CLOSE

Close the opened cursor.

Translated Code
EXEC SQL
OPEN <file1>-OPEN-CURSOR
END-EXEC.
Sequential-Read-OPEN-CURSOR
Random-Rewrite
EXEC SQL
CLOSE <file1>-OPEN-CURSOR
END-EXEC

3.1.2.3 OUTPUT/EXTEND
Operation
OPEN
WRITE
CLOSE

3.1.3

Description
Comment CLOSE statement
Insert new rows into respective tables.
Comment the CLOSE statement.

Translated Code
None
Insert
None

Indexed Organization Sequential Access

All the I/O operations in a file with index organization and sequential access are in the ascending order of
the value of the record key. File positioning is achieved using either OPEN or START. For files opened in
INPUT or I-O mode, query translation varies based on the existence of START on the file. In each of the
above two open modes, there are two sections, START EXISTS and START DOES NOT EXIST,
specifying the corresponding translation strategies.

3.1.3.1 INPUT
3.1.3.1.1 START EXISTS
This section specifies the translation strategy for VSAM files with indexed organization, sequential access
mode opened in INPUT mode where START is used.
Define a cursor for each VSAM file, to select all the rows in the corresponding tables.
If START is used and if it uses full record key in key comparison, define a cursor to select all the rows in
the corresponding tables where key columns have values greater than or equal to the record key values.
Define the above cursors to be ORDER BY primary key columns of one of the tables.
Define a condition name in the working storage section, to be set and reset by START and OPEN
statements when they open and close the cursor. This condition name will indicate the current opened
cursor to be used by other operations.
Definition of the condition name will be
01 <file1>-CURSOR PIC 9 VALUE 0.
88 <file1>-OPEN
88 <file1>-START-RK

value 1.
value 2.

Where <file1>-OPEN will be set when the cursor corresponding to OPEN is opened
<file1>-START-RK will be set when the cursor corresponding to START is opened.
Cursor definition for OPEN :

14

VSAM to DB2 Online and Batch Program Conversion Strategy

OPEN-CURSOR
Cursor definition for START :
START-CURSOR
Operation
OPEN

START

READ or
READ
NEXT

CLOSE

Description
Open the cursor <file1>-OPEN-CURSOR. This
will select all the rows in the table. Set this cursor
to be the current valid cursor, which can then be
used by successive I/O operations till START
repositions the file position indicator or CLOSE
closes the file.
Start repositions the file position indicator by
populating record key fields with appropriate
values. If the corresponding cursor is already
opened, close the cursor and open it again.
Reopening the cursor will select the rows based on
the current value of the host variables used in the
WHERE clause. Set <file1>-START-RKCURSOR to be the current valid cursor which can
be used by successive I/O operations till another
START repositions the file position indicator or
CLOSE closes the file.

Translated Code
EXEC SQL
OPEN <file1>-OPENCURSOR
END-EXEC.
MOVE 1 TO <file1>-CURSOR.

In sequential access files, READ NEXT is same as

READ. Both of them are sequential read.
READ / READ NEXT can follow either any one
of OPEN, START or another READ/READ
NEXT.
If START exists, then cursor to fetch will be
decided based on the flag which states the current
valid cursor. COBOL EVALUATE statement will
be used to make the decision.
If START exists, there will be two opened cursors.
Close both the cursors at that point. Set the flag to
indicate no cursors are available.

EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
Sequential-Read-OPENCURSOR
WHEN <file1>-START-RK
Sequential-Read-STARTCURSOR
END-EVALUATE

IF <file1>-START-RK
THEN
EXEC SQL
CLOSE <file1>-STARTRK-CURSOR
END-EXEC
END-IF.
EXEC SQL
OPEN <file1>-START-RKCURSOR
END-EXEC.
MOVE 2 TO <file1>-CURSOR.

EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
EXEC SQL
CLOSE <file1>-STARTRK-CURSOR
END-EXEC
MOVE 0 TO <file1>-CURSOR.

3.1.3.1.2 START DOES NOT EXIST

This section specifies the translation strategy for VSAM files with indexed organization, sequential access
mode opened in INPUT mode where START is not used.
Define a cursor for each VSAM file, to select all the rows in the corresponding table, ORDER BY primary
key columns.
Cursor definition for OPEN :
OPEN-CURSOR

15

VSAM to DB2 Online and Batch Program Conversion Strategy

Operation
OPEN

Description
Open the cursor <file1>-OPEN-CURSOR.

READ or
READ
NEXT
CLOSE

In sequential access files, READ NEXT is same as

READ. Both of them are sequential read. FETCH
the cursor defined on the corresponding table.
Close the opened cursor.

Translated Code
EXEC SQL
OPEN <file1>-OPENCURSOR
END-EXEC.
Sequential-Read-OPEN-CURSOR
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC

3.1.3.2 I-O
3.1.3.2.1 START-EXISTS
This section specifies the translation strategy for VSAM files with indexed organization, sequential access
mode opened in I-O mode where START is used.
Define a cursor for each VSAM file, to select all the rows in the corresponding table.
If start is used and if start uses full record key in key comparison, define a cursor to select all the rows in
the corresponding tables where key columns have values greater than or equal to the record key values.
Define the above cursors to be ORDER BY primary key columns.
Define a condition name in the working storage section, to be set and reset by START and OPEN
statements when they open and close the cursor. This condition name will indicate the current opened
cursor to be used by other operations.
Definition of the condition name will be
01 <file1>-CURSOR PIC 9 VALUE 0.
88 <file1>-OPEN
88 <file1>-START-RK

value 1.
value 2.

Where <file1>-OPEN will be set when the cursor corresponding to OPEN is opened
<file1>-START-RK will be set when the cursor corresponding to START is opened.
Cursor definition for OPEN :
OPEN-CURSOR
Cursor definition for START :
START-CURSOR
Operation
OPEN

START

Description
Open the cursor <file1>-OPEN-CURSOR.
This will select all the rows in the table. Set
<file1>-OPEN-CURSOR to be the current
valid cursor which can then be used by
successive I/O operations till START
repositions the file position indicator or
CLOSE closes the file.
Start repositions the file position indicator
by populating record key fields with
appropriate values. If the corresponding
cursor is already opened, close the cursor
and open it again. Reopening the cursor will

Translated Code
EXEC SQL
OPEN <file1>-OPEN-CURSOR
END-EXEC.
MOVE 1 TO <file1>-CURSOR.

IF <file1>-START-RK
THEN
EXEC SQL
CLOSE <file1>-START-RKCURSOR

16

VSAM to DB2 Online and Batch Program Conversion Strategy

READ

REWRITE
WRITE
DELETE
CLOSE

select the rows based on the current value of

the host variables used in the WHERE
clause. Set <file1>-START-RK-CURSOR to
be the current valid cursor which can be
used by successive I/O operations till
another START repositions the file position
indicator or CLOSE closes the file.

END-EXEC
END-IF.
EXEC SQL
OPEN <file1>-START-RKCURSOR
END-EXEC.
MOVE 2 TO <file1>-CURSOR.

In sequential access files, READ NEXT is

same as READ. Both of them are sequential
read.
READ / READ NEXT can follow either any
one of OPEN, START or another
READ/READ NEXT.
If START exists, then cursor to fetch will be
decided based on the flag which states the
current valid cursor. COBOL EVALUATE
statement will be used to make the decision.

EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
Sequential-Read-OPEN-CURSOR
WHEN <file1>-START-RK
Sequential-Read-START-CURSOR
END-EVALUATE

Insert the new rows into respective tables.

As there will be two opened cursors, one by
the OPEN statement and another by START
statement, both the cursors will be closed at
this point.

Random-Rewrite
Insert
Delete
EXEC SQL
CLOSE <file1>-OPEN-CURSOR
END-EXEC
EXEC SQL
CLOSE <file1>-START-RKCURSOR
END-EXEC
MOVE 0 TO <file1>-CURSOR

3.1.3.2.2 STARTDOES NOT EXIST

This section specifies the translation strategy for VSAM files with indexed organization, sequential access
mode opened in I-O mode where START is not used.
Define a cursor for each VSAM file, to select all the rows in the corresponding table, ORDER BY primary
key columns.
Define the above cursors to be ORDER BY primary key columns.
Cursor definition for OPEN :
OPEN-CURSOR
Operation
OPEN

Description
Open the cursor defined for the
corresponding file.

READ

In sequential access files, READ NEXT is

same as READ. Both of them are sequential
read.
READ / READ NEXT can follow either any
one of OPEN, START or another
READ/READ NEXT. Fetch the <file1>OPEN-CURSOR.

REWRITE
WRITE

Insert new rows into respective tables.

Translated Code
EXEC SQL
OPEN <file1>-OPEN-CURSOR
END-EXEC.
Sequential-Read-OPEN-CURSOR

Random-Rewrite
Insert

17

VSAM to DB2 Online and Batch Program Conversion Strategy

DELETE
CLOSE

Close the cursor opened corresponding to

the OPEN statement.

Delete
EXEC SQL
CLOSE <file1>-OPEN-CURSOR
END-EXEC

3.1.3.3 OUTPUT / EXTEND

In this mode, there will be no READ / READ NEXT / START statements. Only allowed operations are
OPEN, WRITE and CLOSE.
Operation
OPEN
WRITE
CLOSE

3.1.4

Description
Comment OPEN statement.
Insert the new row into the corresponding
table.
Multiple INSERT
Comment CLOSE statement

Translated Code
None
Insert
None

Indexed Organization Random Access

All the I/O operations in a file with index organization and random access mode, are random i.e., will use
key values to insert or retrieve records. All the queries will be translated using WHERE condition.

3.1.4.1 INPUT
Operation
OPEN
READ
CLOSE

Description
Comment the OPEN statement.
Select the record with primary key
columns having values equal to record key
values.
Comment the CLOSE statement

Translated Code
None
Random-Read
None

3.1.4.2 I-O
Operation
OPEN
READ
REWRITE
INSERT
DELETE
CLOSE

Description
Comment the OPEN statement
Select the row with primary key columns
having values equal to record key values.
Update the row with primary key columns
having values equal to record key values.
Multiple UPDATE
Insert the new row into the table
corresponding to the file.
Delete the row with primary key columns
having values equal to record key values.
Comment CLOSE statement.

Translated Code
None
Random-Read
Random-Rewrite
Insert
Delete
None

3.1.4.3 OUTPUT
Operation
OPEN
WRITE
CLOSE

3.1.5

Description
Comment the OPEN statement
Insert the new row into the table
corresponding to the file.
Comment the CLOSE statement

Translated Code
None
Insert
None

Indexed Organization Dynamic Access

I/O operations in files with indexed organization and dynamic access, can be either sequential or random or
both. OPEN opens the file with file position indicator pointing to first record. File position indicator can be

18

VSAM to DB2 Online and Batch Program Conversion Strategy

repositioned using START or random READ. Records can be read sequentially using READ NEXT and
randomly using RANDOM READ.
Translation strategy varies based on the existence of START, READ and READ NEXT. File position
indicator can be repositioned using either random READ or START.
For files opened in INPUT or I-O mode, file I/O operations can be sequential or random or both based on
the existence of READ NEXT, RANDOM READ and START.
Following are six possible cases classifying the file operation to be sequential or random or both, based on
the existence of READ NEXT, RANDOM READ and START.
Sl.No.
1
2
3
4
5
6

READ
NEXT
Yes
Yes
Yes
Yes
No
No

RANDOM READ

START

Operation

No
Yes
No
Yes
Yes
Yes

No
No
Yes
Yes
Yes
No

Sequential
Sequential and Random
Sequential and Random
Sequential and Random
Random
Random

Query translation for

Case 1 is same as for files with indexed organization and sequential access without start
Case 3 is same as for files with indexed organization and sequential access with start
Case 5 and 6 is same as for files with indexed organization and random access. Refer to the
corresponding sections for their translation strategy. Following sections specify the translation
strategies for cases 2 and 4.

3.1.5.1 INPUT
3.1.5.1.1 Case 2
This section specifies the translation strategy for VSAM files with indexed organization, dynamic access
mode opened in INPUT mode where READ exists, RANDOM READ exists and START does not exist.
Define a cursor for each VSAM file, to select all the rows in the corresponding table.
If RANDOM READ is used and if it uses record key in key comparison, define a cursor to select all the
rows in the corresponding table where key columns have values greater than record key values.
Define the above cursors to be ORDER BY primary key columns.
Define a condition name in the working storage section, to be set and reset by RANDOM READ and
OPEN statements when they open and close the cursor. This condition name will indicate the current
opened cursor to be used by other operations.
Definition of the condition name will be
01 <file1>-CURSOR PIC 9 VALUE 0.
88 <file1>-OPEN
88 <file1>-READ-RK

value 1.
value 2.

Where <file1>-OPEN will be set when the cursor corresponding to OPEN is opened
<file2>-READ-RK will be set when the cursor corresponding to READ is opened.
Cursor definition for OPEN :
OPEN-CURSOR
Cursor definition for READ :
READ-CURSOR
Operation

Description

Translated Code

19

VSAM to DB2 Online and Batch Program Conversion Strategy

OPEN

Open the cursor <file1>-OPEN-CURSOR.

RANDOM
READ

RANDOM READ translates to SELECT with

WHERE clause. As RANDOM READ
repositions the file position indicator, any
previously opened cursor should be closed to
prevent operations following RANDOM READ,
from using the cursor.

READ
NEXT

READ NEXT is a sequential operation and

should follow successful OPEN or RANDOM
READ or READ NEXT.
If the previous operation is OPEN, then that
would have opened the appropriate cursor and
hence FETCH on the cursor corresponding to
OPEN will get the correct row.
If the previous operation is RANDOM READ,
then that would have selected the correct record.
Record key fields( host variables ) would have
got primary key column values of the selected
record. Opening the cursor, <file1>-READ-RKCURSOR , will select all the rows in the
corresponding table with primary key columns
have values greater than record key values and
FETCH on that cursor will get the correct
record.
If the previous operation is READ NEXT, it
would have opened the appropriate cursor and
hence fetch on that cursor will get the correct
record.
Close the opened cursors.

CLOSE

EXEC SQL
OPEN <file1>-OPENCURSOR
END-EXEC.
MOVE 1 TO <file1>-CURSOR.
Random-Read
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
WHEN <file1>-READ-RK
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
END-EVALUATE
MOVE 0 TO <file1>-CURSOR
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
Sequential-Read-OPENCURSOR
WHEN <file1>-READ-RK
Sequential-Read-READCURSOR
WHEN 0
EXEC SQL
OPEN <file1>-READ-RKCURSOR.
END-EXEC
Sequential-Read-READ-CURSOR
MOVE 2 TO <file1>-CURSOR
END-EVALUATE

IF <file1>-OPEN-CURSOR
THEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
END-IF.
IF <file1>-READ-RK-CURSOR
THEN
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
END-IF.

20

VSAM to DB2 Online and Batch Program Conversion Strategy

3.1.5.1.2 Case 4
This section specifies the translation strategy for VSAM files with indexed organization, dynamic access
mode opened in INPUT mode where READ exists, RANDOM READ exists and START exists.
Define a cursor for each VSAM file, to select all the rows in the corresponding table.
If RANDOM READ is used and if it uses record key in key comparison, define a cursor to select all the
rows in the corresponding table where key columns have values greater than the record key values.
If START is used and if it uses record key in key comparison, define a cursor to select all the rows in the
corresponding table where key columns have values greater than or equal to the record key values.
Define the above cursors to be ORDER BY primary key columns.
Define a condition name in the working storage section, to be set and reset by RANDOM READ, START
and OPEN statements when they open and close the cursor. This condition name will indicate the current
opened cursor to be used by other operations.
Definition of the condition name will be
01 <file1>-CURSOR PIC 9 VALUE 0.
88 <file1>-OPEN
88 <file1>-READ-RK
88 <file1>-START-RK

value 1.
value 2.
value 3.

Where <file1>-OPEN will be set when the cursor corresponding to OPEN is opened,
<file1>-READ-RK will be set when the cursor corresponding to READ is opened,
<file1>-START-RK will be set when the cursor corresponding to START is opened.
Cursor definition for OPEN :
OPEN-CURSOR
Cursor definition for START :
START-CURSOR
Cursor definition for READ :
READ-CURSOR
Operation
OPEN

Description
Open the cursor defined corresponding to OPEN
statement.

START

Start repositions the file position indicator by

populating record key fields with appropriate
values.
If the corresponding cursor is already opened,
close the cursor and open it again. Reopening
the cursor will select the rows based on the
current value of the host variables used in the
WHERE clause. Set <file1>-START-RKCURSOR to be the current valid cursor which
can be used by successive I/O operations till
another START/RANDOM READ repositions
the file position indicator or CLOSE closes the
file.

RANDOM
READ

RANDOM READ translates to SELECT with

WHERE clause. As RANDOM READ

Translated Code
EXEC SQL
OPEN <file1>-OPENCURSOR
END-EXEC.
MOVE 1 TO <file1>-CURSOR.
IF <file1>-START-RK
THEN
EXEC SQL
CLOSE <file1>-START-RKCURSOR
END-EXEC
END-IF.
EXEC SQL
OPEN <file1>-START-RKCURSOR
END-EXEC.
MOVE 3 TO <file1>-CURSOR.
Random-Read
EVALUATE <file1>-CURSOR

21

VSAM to DB2 Online and Batch Program Conversion Strategy

repositions the file position indicator, any

previously opened cursor should be closed to
prevent operations following RANDOM READ,
from using the cursor.

READ
NEXT

CLOSE

READ NEXT is a sequential operation and

should follow either successful OPEN or
RANDOM READ or READ NEXT or START.
If the previous operation is OPEN or START,
then that would have opened the appropriate
cursor and hence FETCH on the cursor
corresponding to OPEN or START will get the
correct row.
If the previous operation is RANDOM READ,
then that would have selected the correct record.
Record key fields( host variables ) would have
got primary key column values of the selected
record. Opening the cursor, <file1>-READ-RKCURSOR , will select all the rows in the
corresponding table with primary key columns
have values greater than or equal to record key
values and FETCH on that cursor will get the
correct record.
If the previous operation is READ NEXT, it
would have opened the appropriate cursor and
hence on that cursor will get the correct record.
Close the opened cursors. Set the flag to indicate
that there are no cursors available.

WHEN <file1>-OPEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
WHEN <file1>-READ-RK
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
WHEN <file1>-START-RK
EXEC SQL
CLOSE <file1>-START-RKCURSOR
END-EXEC
END-EVALUATE
MOVE 0 TO <file1>-CURSOR
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
Sequential-Read-OPENCURSOR
WHEN <file1>-READ-RK
Sequential-Read-READCURSOR
WHEN <file1>-START-RK
Sequential-Read-STARTCURSOR
WHEN 0
EXEC SQL
OPEN <file1>-READ-RKCURSOR.
END-EXEC
Sequential-Read-READ-CURSOR
MOVE 2 TO <file1>-CURSOR
END-EVALUATE

IF <file1>-OPEN-CURSOR
THEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
END-IF.
IF <file1>-READ-RK-CURSOR
THEN
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
END-IF.
IF <file1>-START-RK-CURSOR
THEN
EXEC SQL
CLOSE <file1>-START-RK-

22

VSAM to DB2 Online and Batch Program Conversion Strategy

CURSOR
END-EXEC
END-IF.
MOVE 0 TO <file1>-CURSOR

3.1.5.2 I-O
3.1.5.2.1

Case 2

This section specifies the translation strategy for VSAM files with indexed organization, dynamic access
mode opened in I-O mode where READ exists, RANDOM READ exists and START does not exist.
Define a cursor for each VSAM file, to select all the rows in the corresponding table.
If RANDOM READ is used and if it uses record key in key comparison, define a cursor to select all the
rows in the corresponding table where key columns have values greater than record key values.
Define the above cursors to be ORDER BY primary key columns.
Define a condition name in the working storage section, to be set and reset by RANDOM READ and
OPEN statements when they open and close the cursor. This condition name will indicate the current
opened cursor to be used by other operations.
Definition of the condition name will be
01 <file1>-CURSOR PIC 9 VALUE 0.
88 <file1>-OPEN
88 <file1>-READ-RK

value 1.
value 2.

Where <file1>-OPEN will be set when the cursor corresponding to OPEN is opened
<file2>-READ-RK will be set when the cursor corresponding to READ is opened.
Cursor definition for OPEN :
OPEN-CURSOR
Cursor definition for READ :
READ-CURSOR
Operation
OPEN

Description
Open the cursor <file1>-OPEN-CURSOR
defined corresponding to OPEN statement.

RANDOM
READ

RANDOM READ translates to SELECT with

WHERE clause. As RANDOM READ
repositions the file position indicator, any
previously opened cursor should be closed to
prevent operations following RANDOM READ,
from using the cursor.

Translated Code
EXEC SQL
OPEN <file1>-OPENCURSOR
END-EXEC.
MOVE 1 TO <file1>-CURSOR.
Random-Read
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
WHEN <file1>-READ-RK
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
END-EVALUATE

23

VSAM to DB2 Online and Batch Program Conversion Strategy

READ
NEXT

REWRITE

DELETE

WRITE
CLOSE

READ NEXT is a sequential operation and

should follow successful OPEN or RANDOM
READ or READ NEXT.
If the previous operation is OPEN, then that
would have opened the appropriate cursor and
hence FETCH on the cursor corresponding to
OPEN will get the correct row.
If the previous operation is RANDOM READ,
then that would have selected the correct record.
Record key fields( host variables ) would have
got primary key column values of the selected
record. Opening the cursor, <file1>-READ-RKCURSOR , will select all the rows in the
corresponding table with primary key columns
have values greater than record key values and
FETCH on that cursor will get the correct
record.
If the previous operation is READ NEXT, it
would have opened the appropriate cursor and
hence fetch on that cursor will get the correct
record.
Record to be rewritten, is specified by the value
contained in the prime RECORD KEY.
REWRITE will translate to UPDATE with
WHERE clause.
Record to be deleted is specified by the value
contained in the prime RECORD KEY.
DELETE will translate to DELETE with
WHERE clause.
Record to be written is specified by the value
contained in the prime RECORD KEY. WRITE
will translate to INSERT with WHERE clause.
Close the opened cursors. Set the flag to indicate
that there are no cursors available.

MOVE 0 TO <file1>-CURSOR
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
Sequential-Read-OPENCURSOR
WHEN <file1>-READ-RK
Sequential-Read-READCURSOR
WHEN 0
EXEC SQL
OPEN <file1>-READ-RKCURSOR.
END-EXEC
Sequential-Read-READ-CURSOR
MOVE 2 TO <file1>-CURSOR
END-EVALUATE

Random-Rewrite

Delete

Insert
IF <file1>-OPEN-CURSOR
THEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
END-IF.
IF <file1>-READ-RK-CURSOR
THEN
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
END-IF.
MOVE 0 TO <file1>-CURSOR

3.1.5.2.1 Case 4
Define a cursor for each VSAM file, to select all the rows in the corresponding table.
If RANDOM READ is used and if it uses record key in key comparison, define a cursor to select all the
rows in the corresponding table where key columns have values greater than record key values.

24

VSAM to DB2 Online and Batch Program Conversion Strategy

If START is used and if it uses record key in key comparison, define a cursor to select all the rows in the
corresponding table where key columns have values greater than or equal to the record key values.
Define the above cursors to be ORDER BY primary key columns.
Define a condition name in the working storage section, to be set and reset by RANDOM READ, START
and OPEN statements when they open and close the cursor. This condition name will indicate the current
opened cursor to be used by other operations.
Definition of the condition name will be
01 <file1>-CURSOR PIC 9 VALUE 0.
88 <file1>-OPEN
88 <file1>-READ-RK
88 <file1>-START-RK

value 1.
value 2.
value 3.

Where <file1>-OPEN will be set when the cursor corresponding to OPEN is opened,
<file1>-READ-RK will be set when the cursor corresponding to READ is opened,
<file1>-START-RK will be set when the cursor corresponding to START is opened.
Cursor definition for OPEN :
OPEN-CURSOR
Cursor definition for READ :
READ-CURSOR
Cursor definition for START :
START-CURSOR
Operation
OPEN

Description
Open the cursor <file1>-OPEN-CURSOR,
defined corresponding to OPEN statement.

START

Start repositions the file position indicator by

populating relative key fields with appropriate
values.
If the corresponding cursor is already opened,
close the cursor and open it again. Reopening the
cursor will select the rows based on the current
value of the host variables used in the WHERE
clause. Set <file1>-START-RK-CURSOR to be
the current valid cursor which can be used by
successive I/O operations till another
START/RANDOM READ repositions the file
position indicator or CLOSE closes the file.

RANDO
M READ

RANDOM READ translates to SELECT with

WHERE clause. As RANDOM READ
repositions the file position indicator, any
previously opened cursor should be closed to
prevent operations following RANDOM READ,
from using the cursor.

Translated Code
EXEC SQL
OPEN <file1>-OPENCURSOR
END-EXEC.
MOVE 1 TO <file1>-CURSOR.
IF <file1>-START-RK
THEN
EXEC SQL
CLOSE <file1>-START-RKCURSOR
END-EXEC
END-IF.
EXEC SQL
OPEN <file1>-START-RKCURSOR
END-EXEC.
MOVE 3 TO <file1>-CURSOR.
Random-Read
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
WHEN <file1>-READ-RK
EXEC SQL
CLOSE <file1>-READ-RKCURSOR

25

VSAM to DB2 Online and Batch Program Conversion Strategy

READ
NEXT

REWRIT
E
DELETE
WRITE
CLOSE

READ NEXT is a sequential operation and

should follow either successful OPEN or
RANDOM READ or READ NEXT or START.
If the previous operation is OPEN or START,
then that would have opened the appropriate
cursor and hence FETCH on the cursor
corresponding to OPEN or START will get the
correct row.
If the previous operation is RANDOM READ,
then that would have selected the correct record.
Record key fields( host variables ) would have
got primary key column values of the selected
record. Opening the cursor, <file1>-READ-RKCURSOR , will select all the rows in the
corresponding table with primary key columns
have values greater than record key values and
FETCH on that cursor will get the correct record.
If the previous operation is READ NEXT, it
would have opened the appropriate cursor and
hence fetch on that cursor will get the correct
record.
Record to be rewritten is specified by the value
contained in the prime RECORD KEY.
REWRITE will translate to UPDATE with
WHERE clause.
Record to be deleted is specified by the value
contained in the prime RECORD KEY. DELETE
will translate to DELETE with WHERE clause.
Record to be written is specified by the value
contained in the prime RECORD KEY. WRITE
will translate to INSERT with WHERE clause.
Close the opened cursors.

END-EXEC
WHEN <file1>-START-RK
EXEC SQL
CLOSE <file1>-START-RKCURSOR
END-EXEC
END-EVALUATE
MOVE 0 TO <file1>-CURSOR
EVALUATE <file1>-CURSOR
WHEN <file1>-OPEN
Sequential-Read-OPENCURSOR
WHEN <file1>-READ-RK
Sequential-Read-READCURSOR
WHEN <file1>-START-RK
Sequential-Read-STARTCURSOR
WHEN 0
EXEC SQL
OPEN <file1>-READ-RKCURSOR.
END-EXEC
Sequential-Read-READ-CURSOR
MOVE 2 TO <file1>-CURSOR
END-EVALUATE

Random-Rewrite

Delete
Insert
IF <file1>-OPEN-CURSOR
THEN
EXEC SQL
CLOSE <file1>-OPENCURSOR
END-EXEC
END-IF.
IF <file1>-READ-RK-CURSOR
THEN
EXEC SQL
CLOSE <file1>-READ-RKCURSOR
END-EXEC
END-IF.
IF <file1>-START-RK-CURSOR
THEN

26

VSAM to DB2 Online and Batch Program Conversion Strategy

EXEC SQL
CLOSE <file1>-START-RKCURSOR
END-EXEC
END-IF.
MOVE 0 TO <file1>-CURSOR

3.1.6.3 OUTPUT
In this mode, there will be no READ / READ NEXT / START statements. Only allowed operations are
OPEN, WRITE and CLOSE.
Operation
OPEN
WRITE
CLOSE

Description
Comment OPEN statement.
Insert the record into the corresponding
table.
Comment CLOSE statement

Translated Code
None
Insert
None

27

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2

Online Program Translation

Each original program with embedded VSAM access call is assumed to have following template structure.
Template also identifies some record structures that will be added to the working storage section of the
program.
IDENTIFICATION DIVISION.
PROGRAM-ID. Cicsprgm.
ENVIRONMENT DIVISION.
DATA DIVISION.
WORKING-STORAGE SECTION.
01

DATA-RECORD.
05 DATA-FIELD-1 .
05 DATA-FIELD-2 .
.
.

01

KEY-RECORD.
05 KEY-FIELD-1 .
05 KEY-FIELD-2 .

01
1
1
1
1
01

DATA-REC-LENGTH.
KEY-REC-LENGTH.
TOKEN-FIELD.
REQID-FILED
COUNT-FIELD
NUMERIC-COUNT

**** record structures below needs to be added in the new code ****
1

DCLGEN-RECORD.
5
DCLGEN-KEY.
10
DCLGEN-KEY-FIELD1
10
DCLGEN-KEY-FIELD2
.
.
5
DCLGEN-DATA.
10
DCLGEN-DATA-FIELD1
10
DCLGEN-DATA-FIELD2
.
.

/* this definition needs to be repeated appropriately */

/* for each TOKEN used in program
*/
01
TOKEN-DCLGEN-KEY.
5
TOKEN-DCLGEN-KEY-FIELD1
5
TOKEN-DCLGEN-KEY-FIELD2
.
.

28

VSAM to DB2 Online and Batch Program Conversion Strategy

/*Indicator to identify if the cursor is defined for a READNEXT or READPREV operation */

01
WS-FUNCTION-CURSOR
PIC X(01)
VALUE .
88
WS-NEXT-CURSOR
VALUE N.
88
WS-PREV-CURSOR
VALUE P.
88
WS-NO-CURSOR
VALUE .
/* this definition needs to be repeated appropriately for each RESETBR on a VSAM file using different
browse criteria */
01

****

WS-FUNCTION-CURSOR-1
88
WS-NEXT-CURSOR-1
88
WS-PREV-CURSOR-1
88
WS-NO-CURSOR-1

PIC X(01)

End of the new addition

VALUE .
VALUE N.
VALUE P.
VALUE .

****

PROCEDURE DIVISION.
.
.
EXEC CICS
Vsam access call
END-EXEC.
.
.
Discussion below will assume this program template and will identify necessary changes to the code
corresponding to translation of each command.

29

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.1

READ

3.2.1.1 Function
Read a record from a file.

3.2.1.2 CICS Syntax

EXEC CICS
READ
FILE ( vsam-filename )

file to be read

INTO ( DATA-RECORD )

data area where to read the data

LENGTH ( DATA-REC-LENGTH)

CICS will populate with actual

record length

RIDFLD ( KEY-RECORD)

key record, This will identify

record to be read
no of position of key to be matched
for partial/generic search (
mandatory when generic option is
specified)
partial key match

KEYLENGTH ( KEY-REC-LENGTH)

GENERIC

GTEQ / EQUAL

match key for Equal or Greater

than or Equal condition

RBA

when this option is specified

RIDFLD contains actual relative
byte address

RRN

when this option is specified

RIDFLD contains actual Relative
Record Number

UPDATE

TOKEN ( TOKEN-FIELD )

read record will be used for

UPDATE/REWRITE
used to identify multiple records
read for UPDATE/REWRITE

END-EXEC

3.2.1.1 Translation of Syntax

If Relational operator in the READ is =
.
MOVE KEY-RECORD TO

DCLGEN-KEY. {need to move appropriate fields depending on sizes }

30

VSAM to DB2 Online and Batch Program Conversion Strategy

EXEC SQL
SELECT
INTO
FROM
WHERE

*
{ :DCLGEN-RECORD fields. }
{ db2-tablename corresponding to vsam-filename }
{
If ( KEYLENGTH is specified )
{
identify exact fields from KEY-RECORD;
corresponding to these fields identify DCLGEN-KEY fields
}
else
{
all fields of DCLGEN-KEY will participate in WHERE clause
}
using DB2 column names and corresponding DCLGEN-KEY fields as
identified before and by using relational operator form the WHERE
clause
}

If Relational operator in the READ is >=

Declare a cursor to select from the corresponding DB2 table using the key fields and key values in the
WHERE clause. Since this cursor will be used only to read the first record matching the WHERE clause,
The OPTIMIZE FOR 1 ROW option will be used to improve performance.
Cursor declaration will be as follows :
EXEC SQL DECLARE cursor-name CURSOR FOR
SELECT
*
FROM
db2-table-name
WHERE
db2-column-name >= : dclgen-record fields
OPTIMIZE FOR 1 ROW
FOR FETCH ONLY
END-EXEC.
Translation for the READ will be as follows :
MOVE KEY-RECORD TO

DCLGEN-KEY. {need to move appropriate fields depending on sizes }

EXEC SQL
OPEN cursor-name
END-EXEC.
EXEC SQL
FETCH cursor-name
INTO : dclgen-record fields
END-EXEC.
EXEC SQL
OPEN cursor-name
END-EXEC.
For all the above cases:

31

VSAM to DB2 Online and Batch Program Conversion Strategy

IF ( READ is for UPDATE )

{
/* if token is specified explicitly in program find appropriate TOKEN-DCLGEN-KEY */
/* correspondingly.
*/
MOVE DCLGEN-KEY to
TOKEN-DCLGEN-KEY
}

3.2.2

STARTBR

3.2.2.1 Function
Start browse of a file. This is used to position the file position indicator at the desired record so that
VSAM read can be done from this record onwards.

3.2.2.2 CICS Syntax

EXEC CICS
STARTBR
FILE ( VSAM-FILENAME )

file to be read

RIDFLD ( KEY-RECORD)

KEYLENGTH ( KEY-REC-LENGTH)

GENERIC

key record, need to be populated

before call
no of position of key to be matched
for partial/generic search (
mandatory when generic option is
specified)
partial key match

GTEQ / EQUAL

match key for Equal or Greater

than or Equal condition

RBA

when this option is specified

RIDFLD contains actual relative
byte address

RRN

when this option is specified

RIDFLD contains actual Relative
Record Number

REQID ( REQID-FIELD )

used to identify is there are more

than one browse on same file

END-EXEC

3.2.2.3 Translation of Syntax

.
.
IF ( REQID-FIELD is specified )
{

32

VSAM to DB2 Online and Batch Program Conversion Strategy

use it in determining NEXT-CURSORNAME and PREV-CURSORNAME
}
If ( KEYLENGTH is specified )
{
identify exact fields from KEY-RECORD;
corresponding to these fields identify DCLGEN-KEY fields
}
else
{
all fields of DCLGEN-KEY will participate in WHERE clause
}
CURSOR DECLARATION
If only READNEXT operation is present in the program for a VSAM file, then declare only
the cursor for readnext (NEXT-CURSORNAME). If only READPREV operation is present in the program
for a VSAM file then declare only the cursor for readprev (PREV-CURSORNAME). Both the cursors
will be declared if READPREV and READNEXT is being done on a file in the program.
EXEC SQL
DECLARE CURSOR NEXT-CURSORNAME FOR
SELECT
*
FROM
{ db2-tablename corresponding to vsam-filename }
WHERE
{ appropriate where clause with tablefields and >= operator }
END-EXEC
EXEC SQL
DECLARE CURSOR PREV-CURSORNAME FOR
SELECT
*
FROM
{ db2-tablename corresponding to vsam-filename }
WHERE
{ appropriate where clause with tablefields and <= operator }
END-EXEC
REPLACING THE STARTBR COMMAND
The WHERE EXISTS clause is used first to do a SELECT on the table. If any rows are found with
matching condition then only the cursor is opened. For this purpose the Standard Dummy table
SYSIBM.SYSDUMMY1 is used.
The translated code will look as given below :
EXEC SQL SELECT IBMREQD FROM SYSIBM.SYSDUMMY1
INTO :WS-IBMREQD
WHERE EXISTS
(SELECT * FROM db2-table-name
WHERE db2-column-name = (or >= ) : dclgen-record fields)
END-EXEC.

33

VSAM to DB2 Online and Batch Program Conversion Strategy

The STARTBR is translated to a OPEN CURSOR assuming that the next function will be READNEXT.
This is due to the fact that the same STARTBR command can be used either for a READNEXT or
READPREV function, and this will not be known at the time of translating the STARTBR to the OPEN
CURSOR.
IF SQLCODE = +0
{
Move the VSAM key values to the dclgen-record fields to use in the WHERE clause
SET WS-NEXT-CURSOR TO TRUE
EXEC SQL
OPEN NEXT-CURSORNAME
END-EXEC
}

34

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.3

READNEXT / READPREV

3.2.3.1 Function
Read next record or previous record during a browse of a file.

3.2.3.2 CICS Syntax

EXEC CICS
READNEXT / READPREV
FILE ( vsam-filename )

file to be read

INTO ( DATA-RECORD )

data area where to read the data

LENGTH ( DATA-REC-LENGTH)

CICS will populate with actual

record length

RIDFLD ( KEY-RECORD)

key record size should be

sufficient to hold complete key

KEYLENGTH ( KEY-REC-LENGTH)

no of position of key to be matched

for partial/generic search (
mandatory when generic option is
specified)

RBA

when this option is specified

RIDFLD contains actual relative
byte address

RRN

when this option is specified

RIDFLD contains actual Relative
record Number

REQID ( REQID-FIELD )

used to identify is there are more

than one browse on same file

END-EXEC

35

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.3.3 Translation of Syntax

FOR READNEXT COMMAND
IF WS-NEXT-CURSOR
(
EXEC SQL
FETCH NEXT-CURSORNAME
INTO :DCLGEN-RECORD
END-EXEC
)
ELSE
IF WS-PREV-CURSOR
(
EXEC SQL
CLOSE PREV-CURSORNAME
END-EXEC
Move the VSAM key values to the dclgen-record fields to use in the WHERE clause
SET WS-NEXT-CURSOR TO TRUE
EXEC SQL
OPEN NEXT-CURSORNAME
END-EXEC
EXEC SQL
FETCH NEXT-CURSORNAME
INTO :DCLGEN-RECORD
END-EXEC
)
END-IF
* this logic is to take care of the RESETBR done on the file, with different criteria than the STARTBR
IF WS-NEXT-CURSOR-1
(
EXEC SQL
FETCH NEXT-CURSORNAME -1
INTO :DCLGEN-RECORD
END-EXEC
)
ELSE
IF WS-PREV-CURSOR-1
(
EXEC SQL
CLOSE PREV-CURSORNAME-1
END-EXEC
Move the VSAM key values to the dclgen-record fields to use in the WHERE clause
SET WS-NEXT-CURSOR TO TRUE
EXEC SQL
OPEN NEXT-CURSORNAME-1
END-EXEC
EXEC SQL
FETCH NEXT-CURSORNAME-1

36

VSAM to DB2 Online and Batch Program Conversion Strategy

INTO
END-EXEC

:DCLGEN-RECORD

)
END-IF
FOR READPREV COMMAND
IF WS-PREV-CURSOR
(
EXEC SQL
FETCH PREV-CURSORNAME
INTO :DCLGEN-RECORD
END-EXEC
)
ELSE
IF WS-NEXT-CURSOR
(
EXEC SQL
CLOSE NEXT-CURSORNAME
END-EXEC
Move the VSAM key values to the dclgen-record fields to use in the WHERE clause
SET WS-PREV-CURSOR-1 TO TRUE
EXEC SQL
OPEN PREV-CURSORNAME
END-EXEC
EXEC SQL
FETCH PREV-CURSORNAME
INTO :DCLGEN-RECORD
END-EXEC
)
END-IF
* this logic is to take care of the RESETBR done on the file, with different criteria than the STARTBR
IF WS-PREV-CURSOR-1
(
EXEC SQL
FETCH PREV-CURSORNAME-1
INTO :DCLGEN-RECORD
END-EXEC
)
ELSE
IF WS-NEXT-CURSOR-1
(
EXEC SQL
CLOSE NEXT-CURSORNAME-1
END-EXEC
Move the VSAM key values to the dclgen-record fields to use in the WHERE clause
SET WS-PREV-CURSOR-1 TO TRUE
EXEC SQL
OPEN PREV-CURSORNAME-1
END-EXEC
EXEC SQL

37

VSAM to DB2 Online and Batch Program Conversion Strategy

FETCH PREV-CURSORNAME-1
INTO :DCLGEN-RECORD
END-EXEC
)
END-IF

MOVE DCLGEN-RECORD
.
.

TO

DATA-RECORD.

38

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.4

WRITE

3.2.4.1 Function
Write a record using the key values specified.

3.2.4.2 CICS Syntax

EXEC CICS
WRITE
FILE ( vsam-filename )

FROM ( DATA-RECORD )

LENGTH ( DATA-REC-LENGTH)

file to be read

data area from where to record

needs to be written
length of data record to be written

RIDFLD ( KEY-RECORD)

KEYLENGTH ( KEY-REC-LENGTH)

RBA

RRN

key record, need to be populated

before call
generally this is specified as length
of complete key field
when this option is specified
RIDFLD contains actual relative
byte address
when this option is specified
RIDFLD contains actual Relative
Record Number

END-EXEC

3.2.4.3 Translation of Syntax

.
.
MOVE DATA-RECORD
EXEC SQL
INSERT
VALUES
END-EXEC

TO
INTO
ARE

DCLGEN-RECORD.

{ db2-tablename corresponding to vsam-filename }

{
:DCLGEN-RECORD fields }

39

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.5

REWRITE

3.2.5.1 Function
Update a record in a file for the key values specified. REWRITE can be done after executing a READ FOR
UPDATE or it can be a random REWRITE also.

3.2.5.2 CICS Syntax

EXEC CICS
REWRITE
FILE ( vsam-filename )

file to be read

FROM ( DATA-RECORD )

data area from where to record

needs to be written

LENGTH ( DATA-REC-LENGTH)
TOKEN ( TOKEN-FIELD )

length of data record to be rewritten

O

used to identify multiple records

read for UPDATE/REWRITE

END-EXEC

3.2.5.3 Translation of Syntax

.
.
MOVE DATA-RECORD

TO

DCLGEN-RECORD

If ( token is explicitly specified in command}

{
use it to identify corresponding TOKEN-DCLGEN-KEY
which can be used in where clause
}
EXEC SQL
UPDATE
SET
WHERE
END-EXEC
.

{ db2-tablename corresponding to vsam-filename }

db2-column-names = :DCLGEN-RECORD fields
db2-column-names = :TOKEN-DCLGEN-KEY or DCLGEN-RECORD-KEY

40

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.6

DELETE

3.2.6.1 Function
Delete a record from a file

3.2.6.2 CICS Syntax

EXEC CICS
DELETE
FILE ( vsam-filename )

file for delete

RIDFLD ( KEY-RECORD)

KEYLENGTH ( KEY-REC-LENGTH)

GENERIC

key record, need to be populated

before call
no of position of key to be matched
for partial/generic match (
mandatory when generic option is
specified)
partial key match

RBA

when this option is specified

RIDFLD contains actual relative
byte address

RRN

when this option is specified

RIDFLD contains actual Relative
Record Number

TOKEN ( TOKEN-FIELD )

NUMERIC ( NUMERIC-COUNT )

used to identify record to be

deleted which was identified by
read for UPDATE/REWRITE
This variable is populated with
number of records deleted

END-EXEC

3.2.6.3 Translation Of Syntax

IF ( Token is Specified )
{
MOVE TOKEN-DCLGEN-KEY TO
}

DCLGEN-KEY.

IF ( RIDFLD is specified )
{
MOVE KEY-RECORD to DCLGEN-KEY ;

41

VSAM to DB2 Online and Batch Program Conversion Strategy

If ( KEYLENGTH is specified )
{
identify exact fields from KEY-RECORD;
corresponding to these fields identify DCLGEN-KEY fields
}
else
{
all fields of DCLGEN-KEY will participate in WHERE clause
}
}
Relational Operator is =
using DB2 columnnames and corresponding DCLGEN-KEY fields as identified before and by using
relational operator , form the WHERE clause ;
EXEC SQL
DELETE (db2-tablename corresponding to vsam-filename )
WHERE ( where clause as formated before )
END-EXEC
MOVE SQLERRD(3)

TO

NUMERIC-COUNT

42

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.7

UNLOCK

3.2.7.1 Function
Release exclusive control of the file. UNLOCK file is usually done after a READ FOR UPDATE is
executed on the file and it is decided that no update needs to be done to the record. UNLOCK also needs to
be performed after VSAM MASSINSERT.

3.2.7.2 CICS Syntax

EXEC CICS
UNLOCK
FILE ( vsam-filename )

file to be unlocked

TOKEN ( TOKEN-FIELD )

Explicitly specified READ for

update to be unlocked

END-EXEC

3.2.7.3 Translation of Syntax

No SQL statement will be added for this because DB2 takes care of data locking depending on the
bind options used. Any resources locked will be released after a COMMIT.
Since the READ with UPDATE option is being translated to a SQL SELECT, and the corresponding
REWRITE is being translated to the an UPDATE on the table, the scenario of the FILE UNLOCK does
not appear in the table operation.
Replace this command with initialization of Temporary Storage Records for READ with UPDATE.
If specific TOKEN is specified then initialize the specific record corresponding to that TOKEN.

43

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.8

ENDBR

3.2.8.1 Function
End browse of a file.

3.2.8.2 CICS Syntax

EXEC CICS
ENDBR
FILE ( vsam-filename )

file to be read

REQID ( REQID-FIELD )

used to identify if there are more

than one browse on same file

END-EXEC

3.2.8.3 Translation Of Syntax

IF ( REQID-FIELD is specified )
{
use it in determining NEXT-CURSORNAME and PREV-CURSORNAME
}
If ( READNEXT operation exists)
{
EXEC SQL
CLOSE NEXT-CURSORNAME
END-EXEC
}
else
If ( READPREV operations exists )
{
EXEC SQL
CLOSE PREV-CURSORNAME
END-EXEC
}

44

VSAM to DB2 Online and Batch Program Conversion Strategy

3.2.9

RESETBR

3.2.9.1 Function
Reset start of browse. The file position indicator is set to the position at the record with the new key values
specified.

3.2.9.2 CICS Syntax

EXEC CICS
RESETBR
FILE ( VSAM-FILENAME )

file to be read

RIDFLD ( KEY-RECORD)

KEYLENGTH ( KEY-REC-LENGTH)

GENERIC

key record, need to be populated

before call
no of position of key to be matched
for partial/generic search (
mandatory when generic option is
specified)
partial key match

GTEQ / EQUAL

match key for Equal or Greater

than or Equal condition

RBA

when this option is specified

RIDFLD contains actual relative
byte address

RRN

when this option is specified

RIDFLD contains actual Relative
Record Number

REQID ( REQID-FIELD )

used to identify is there are more

than one browse on same file

END-EXEC

3.2.9.3 Translation Of Syntax

IF ( REQID-FIELD is specified )
{
use it in determining NEXT-CURSORNAME and PREV-CURSORNAME
}
Move the key values specified in the RESETBR to the DCLGEN fields.
CURSOR DECLARATION
The RESETBR can use either the same criteria that the STARTBR used or it can reset the file position
indicator using some other criteria, example, the STARTBR may be having a EQUAL condition, but the

45

VSAM to DB2 Online and Batch Program Conversion Strategy

RESETBR can use a GTEQ condition. Verify if the conditions specified in the RESETBR is different than
in the STARTBR. If the conditions are different then new cursors will have to be defined for RESETBR
using the new criteria.
In case new cursors have to be defined, if only READNEXT operation is present in the program for the
VSAM file, then declare only the cursor for readnext (NEXT-CURSORNAME). If only READPREV
operation is present in the program for a VSAM file then declare only the cursor for readprev (PREVCURSORNAME). Both the cursors will be declared if READPREV and READNEXT is being done on a
file in the program.
EXEC SQL
DECLARE CURSOR NEXT-CURSORNAME-1 FOR
SELECT
*
FROM
{ db2-tablename corresponding to vsam-filename }
WHERE
{ appropriate where clause with tablefields and >= operator }
END-EXEC
EXEC SQL
DECLARE CURSOR PREV-CURSORNAME-1 FOR
SELECT
*
FROM
{ db2-tablename corresponding to vsam-filename }
WHERE
{ appropriate where clause with tablefields and <= operator }
END-EXEC
REPLACING THE RESETBR COMMAND
If WS-NEXT-CURSOR
{
SET WS-NO-CURSOR TO TRUE
EXEC SQL
CLOSE NEXT-CURSORNAME
END-EXEC
}
else
If WS-PREV-CURSOR
{
SET WS-NO-CURSOR TO TRUE
EXEC SQL
CLOSE PREV-CURSORNAME
END-EXEC
}.
If WS-NEXT-CURSOR-1
{
SET WS-NO-CURSOR-1 TO TRUE
EXEC SQL
CLOSE NEXT-CURSORNAME-1
END-EXEC
}
else
If WS-PREV-CURSOR-1
{
SET WS-NO-CURSOR-1 TO TRUE
EXEC SQL
CLOSE PREV-CURSORNAME-1

46

VSAM to DB2 Online and Batch Program Conversion Strategy

END-EXEC
}.

The WHERE EXISTS clause is used first to do a SELECT on the table. If any rows are found with
matching condition then only the cursor is opened. For this purpose the Standard Dummy table
SYSIBM.SYSDUMMY1 is used.
The translated code will look as given below :
EXEC SQL SELECT IBMREQD FROM SYSIBM.SYSDUMMY1
INTO :WS-IBMREQD
WHERE EXISTS
(SELECT * FROM db2-table-name
WHERE db2-column-name = (or >= ) : dclgen-record fields)
END-EXEC.

47

VSAM to DB2 Online and Batch Program Conversion Strategy

The RESETBR is translated to a OPEN CURSOR assuming that the next function will be READNEXT.
This is due to the fact that the same RESETBR command can be used either for a READNEXT or
READPREV function, and this will not be known at the time of translating the RESETBR to the OPEN
CURSOR.
IF SQLCODE = +0
{
Move the VSAM key values to the dclgen-record fields to use in the WHERE clause
SET WS-NEXT-CURSOR-1 TO TRUE
EXEC SQL
OPEN NEXT-CURSORNAME
END-EXEC
}

48

VSAM to DB2 Online and Batch Program Conversion Strategy

4 Optimizing For Performance

4.1

Selecting only the required fields

It is preferred that the only the required columns are selected in the query.

4.2

Adding Commit logic

Batch programs which handle large volume of data and do updates, inserts or deletes on the DB2 tables will
have better performance if commit logic is added to commit the changes after the number of rows
processed reaches a predetermined value. However for any cursor operations for which the commit logic is
used should have the cursor defined with option WITH HOLD, otherwise the cursor position is lost after
the commit.

4.3

Update only the required columns

Only columns required to be updated should be retained in the update statements. This is especially true if
the column is indexed and is being updated unnecessarily.

49

VSAM to DB2 Online and Batch Program Conversion Strategy

5 Appendix
5.1.1

Sequential Read OPEN-CURSOR

EXEC SQL
FETCH <file1>-OPEN-CURSOR
INTO <respective-fields>
END-EXEC
IF SQLCODE = 100
THEN
<at-end-imperative-stmts> | NEXT SENTENCE
ELSE
IF SQLCODE = 0
THEN
EXEC SQL
OPEN CURSOR dependant-cursor-1
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-1 UNTIL <db2-count> > <db2-array-1-size> AND
SQLCODE = 100
EXEC SQL
CLOSE CURSOR dependant-cursor-1
END-EXEC
...
EXEC SQL
OPEN CURSOR dependant-cursor-n
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-n UNTIL <db2-count> > <db2-array-n-size> AND
SQLCODE =100
EXEC SQL
CLOSE CURSOR dependant-cursor-n
END-EXEC
<not-at-end-imperative-stmts>
ELSE
DISPLAY ERROR IN FETCH
END-IF
END-IF.

50

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.2

Sequential Read START-CURSOR

EXEC SQL
FETCH <file1>-START-CURSOR
INTO <respective-fields>
END-EXEC
IF SQLCODE = 100
THEN
<at-end-imperative-stmts> | NEXT SENTENCE
ELSE
IF SQLCODE = 0
THEN
EXEC SQL
OPEN CURSOR dependant-cursor-1
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-1 UNTIL <db2-count> > <db2-array-1-size> AND
SQLCODE = 100
EXEC SQL
CLOSE CURSOR dependant-cursor-1
END-EXEC
...
EXEC SQL
OPEN CURSOR dependant-cursor-n
END-EXEC
MOVE 1 TO <db2-count>
MOVE <array-n-size> TO <DB2-ARRAY-n-SIZE>
PERFORM FETCH-PARA-n UNTIL <db2-count> <= <db2-array-n-size> AND
SQLCODE = 100
EXEC SQL
CLOSE CURSOR dependant-cursor-n
END-EXEC
<not-at-end-imperative-stmts>
ELSE
DISPLAY ERROR IN FETCH
END-IF
END-IF.

51

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.3

Sequential Read READ-CURSOR

EXEC SQL
FETCH <file1>-READ-CURSOR
INTO <respective-fields>
END-EXEC
IF SQLCODE = 100
THEN
<at-end-imperative-stmts> | NEXT SENTENCE
ELSE
IF SQLCODE = 0
THEN
EXEC SQL
OPEN CURSOR dependant-cursor-1
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-1 UNTIL <db2-count> > <db2-array-1-size> AND
SQLCODE = 100
EXEC SQL
CLOSE CURSOR dependant-cursor-1
END-EXEC
...
EXEC SQL
OPEN CURSOR dependant-cursor-n
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-n UNTIL <db2-count> > <db2-array-n-size> AND
SQLCODE = 100
EXEC SQL
CLOSE CURSOR dependant-cursor-n
END-EXEC
<not-at-end-imperative-stmts>
ELSE
DISPLAY ERROR IN FETCH
END-IF
END-IF.

52

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.4

Random Read

EXEC SQL
SELECT <table-1-col-1> ... <table-1-col-n>,
...,
<table-n-col-1>...<table-n-col-n>
FROM <table-1> ...<table-n>
INTO <respective-fields>
END-EXEC
IF SQLCODE = 0
THEN
EXEC SQL
OPEN CURSOR dependant-cursor-1
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-1 UNTIL <db2-count> > <db2-array-1-size> AND SQLCODE =
100
EXEC SQL
CLOSE CURSOR dependant-cursor-1
END-EXEC
...
EXEC SQL
OPEN CURSOR dependant-cursor-n
END-EXEC
MOVE 1 TO <db2-count>
PERFORM FETCH-PARA-n UNTIL <db2-count> > <db2-array-n-size> AND SQLCODE = 100
EXEC SQL
CLOSE CURSOR dependant-cursor-n
END-EXEC
<not-invalid-key-imperative-stmts>
ELSE
<invalid-key-imperative-statements> | NEXT SENTENCE
END-IF.

53

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.5

Insert

EXEC SQL
INSERT INTO <table-1>(<table-1-col-1> ... <table-1-col-n>)
VALUES (<respective-fields>)
END-EXEC.
IF SQLCODE = 0
THEN
EXEC SQL
INSERT INTO <table-2>(<table-2-col-1> ... <table-2-col-n>)
VALUES (<respective-fields>)
END-EXEC
END-IF
...
IF SQLCODE = 0
THEN
EXEC SQL
INSERT INTO <table-n>(<table-n-col-1> ... <table-n-col-n>)
VALUES (<respective-fields>)
END-EXEC
END-IF
IF SQLCODE = 0
THEN
MOVE 1 TO <db2-count>
PERFORM INSERT-PARA-1 UNTIL <db2-count> > <db2-array-1-size> AND SQLCODE = 0
...
MOVE 1 TO <db2-count>
PERFORM INSERT-PARA-n UNTIL <db2-count> > <db2-array-n-size> AND SQLCODE = 0
END-IF
IF SQLCODE = 0
THEN
<not-invalid-key-statements> | NEXT SENTENCE
ELSE
<invalid-key-statements> | DISPLAY Error In Insert
END-IF

54

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.6

Random Rewrite

EXEC SQL
UPDATE <table-1>
SET
<table-1-col-1> = <table-1-col-1-fld-1>,
...
<table-1-col-n> = <table-1-col-n-fld-n>
WHERE <table-1-key-col-1> = <table-1-key-col-1-key-fld-1>,
...
<table-1-key-col-n> = <table-1-key-col-n-key-fld-n>
END-EXEC
IF SQLCODE = 0
THEN
EXEC SQL
UPDATE <table-2>
SET
<table-2-col-1> = <table-2-col-1-fld-1>,
...
<table-2-col-n> = <table-2-col-n-fld-n>
WHERE<table-2-key-col-1> = <table-2-key-col-1-key-fld-1>,
...
<table-2-key-col-n> = <table-2-key-col-n-key-fld-n>
END-EXEC
END-IF
...
IF SQLCODE = 0
THEN
EXEC SQL
UPDATE <table-n>
SET
<table-n-col-1> = <table-n-col-1-fld-1>,
...
<table-n-col-n> = <table-n-col-n-fld-n>
WHERE<table-n-key-col-1> = <table-n-key-col-1-key-fld-1>,
...
<table-n-key-col-n> = <table-n-key-col-n-key-fld-n>
END-EXEC
END-IF
IF SQLCODE = 0
THEN
EXEC SQL
OPEN CURSOR dependant-updatable-cursor-1
END-EXEC
MOVE 1 TO <db2-count>
PERFORM UPDATE-PARA-1 UNTIL <db2-count> > <db2-array-1-size> AND SQLCODE = 0
EXEC SQL
CLOSE CURSOR dependant-updatable-cursor-1
END-EXEC
...
EXEC SQL
OPEN CURSOR dependant-updatable-cursor-n
END-EXEC
MOVE 1 TO <db2-count>.
PERFORM UPDATE-PARA-n UNTIL <db2-count> > <db2-array-n-size> AND SQLCODE = 0
EXEC SQL
CLOSE CURSOR dependant-updatable-cursor-n
END-EXEC

55

VSAM to DB2 Online and Batch Program Conversion Strategy

END-IF
IF SQLCODE = 0
THEN
<not-invalid-key-statements> | NEXT SENTENCE
ELSE
<invalid-key-statements> | DISPLAY Error In Update
END-IF

56

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.7

Delete

EXEC SQL
DELETE <table-1>
WHERE<table-1-key-col-1> = <table-1-key-col-1-key-fld-1>,
...
<table-1-key-col-n> = <table-1-key-col-n-key-fld-n>
END-EXEC.
IF SQLCODE = 0
THEN
EXEC SQL
DELETE <table-2>
WHERE<table-2-key-col-1> = <table-2-key-col-1-key-fld-1>,
...
<table-2-key-col-n> = <table-2-key-col-n-key-fld-n>
END-EXEC.
END-IF
...
IF SQLCODE = 0
THEN
EXEC SQL
DELETE <table-n>
WHERE<table-n-key-col-1> = <table-n-key-col-1-key-fld-1>,
...
<table-n-key-col-n> = <table-n-key-col-n-key-fld-n>
END-EXEC.
END-IF
IF SQLCODE = 0
THEN
<not-invalid-key-statements> | NEXT SENTENCE
ELSE
<invalid-key-statements> | DISPLAY Error In Delete
END-IF

57

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.8

OPEN-CURSOR

If file organization is sequential then

EXEC SQL
DECLARE <file1>-OPEN-CURSOR CURSOR FOR
SELECT <table-1-col-1> ... <table-1-col-n>,
...,
<table-n-col-1>...<table-n-col-n>
FROM <table-1> ...<table-n>
END-EXEC
If file organization is indexed or relative then
EXEC SQL
DECLARE <file1>-OPEN-CURSOR CURSOR FOR
SELECT <table-1-col-1> ... <table-1-col-n>,
...,
<table-n-col-1>...<table-n-col-n>
FROM <table-1> ...<table-n>
ORDER BY <table-1-key-col-1> ... <table-1-key-col-n>
END-EXEC

58

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.9

START-CURSOR

EXEC SQL
DECLARE <file1>-START-RK-CURSOR CURSOR FOR
SELECT <table-1-col-1> ... <table-1-col-n>,
...,
<table-n-col-1>...<table-n-col-n>
FROM <table-1> ...<table-n>
WHERE <table-1-key-col-1> >= <rec-key-fld-1>,
<table-1-key-col-n> >= <rec-key-fld-n>,
<table-2-key-col-n> = <table-1-key-col-1>
<table-2-key-col-n> = <table-1-key-col-n>
... ,
<table-n-key-col-n> = <table-1-key-col-1>
<table-n-key-col-n> = <table-1-key-col-n>
ORDER BY <table-1-key-col-1> ... <table-1-key-col-n>
END-EXEC

59

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.10 READ-CURSOR
EXEC SQL
DECLARE <file1>-READ-RK-CURSOR CURSOR FOR
SELECT <table-1-col-1> ... <table-n-col-n>,
...,
<table-n-col-1>...<table-n-col-n>
FROM <table-1> ...<table-n>
WHERE <table-1-key-col-1> >= <rec-key-fld-1>,
<table-1-key-col-n> >= <rec-key-fld-n>,
<table-2-key-col-n> = <table-1-key-col-1>
<table-2-key-col-n> = <table-1-key-col-n>
... ,
<table-n-key-col-n> = <table-1-key-col-1>
<table-n-key-col-n> = <table-1-key-col-n>
ORDER BY <table-1-key-col-1> ... <table-1-key-col-n>
END-EXEC

60

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.11 Dependant-Cursor-1
EXEC SQL
DECLARE <dependant-cursor-1> CURSOR FOR
SELECT <array-1-table-1-col-1> ... <array-1-table-n-col-n>,
...,
<array-1-table-n-col-1>...<array-1-table-n-col-n>
FROM <array-1-table-1> ...<array-1-table-n>
END-EXEC

5.1.12 Dependant-Cursor-n
EXEC SQL
DECLARE <dependant-cursor-n> CURSOR FOR
SELECT <array-n-table-1-col-1> ... <array-n-table-n-col-n>,
...,
<array-n-table-n-col-1>...<array-n-table-n-col-n>
FROM <array-n-table-1> ...<array-n-table-n>
END-EXEC

5.1.13 Dependant-Updatable-Cursor-1
EXEC SQL
DECLARE <dependant-updatable-cursor-1> CURSOR FOR
SELECT <array-1-table-1-col-1> ... <array-1-table-1-col-n>
FROM <array-1-table-1> FOR UPDATE OF <columns>
END-EXEC

5.1.14 Dependant-Updatable-Cursor-n
EXEC SQL
DECLARE <dependant-updatable-cursor-n> CURSOR FOR
SELECT <array-n-table-1-col-1> ... <array-n-table-1-col-n>
FROM <array-n-table-1> FOR UPDATE OF <columns>
END-EXEC

61

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.15 FETCH-PARA-1
EXEC-SQL
FETCH dependant-cursor-1
INTO <db2-array-1-table-1-col-1-fld-1>, .. <db2-array-1-table-1-col-n-fld-n>
...
<db2-array-1-table-n-col-1-fld-1>, .. <db2-array-1-table-n-col-n-fld-n>
END-EXEC
MOVE <db2-array-1> TO <array-1>(<db2-count>)
COMPUTE <db2-count> = <db2-count> + 1

5.1.16 FETCH-PARA-n
EXEC-SQL
FETCH dependant-cursor-n
INTO <db2-array-1-table-1-col-1-fld-1>, .. <db2-array-1-table-1-col-n-fld-n>
...
<db2-array-1-table-n-col-1-fld-1>, .. <db2-array-1-table-n-col-n-fld-n>
END-EXEC.
MOVE <db2-array-1> TO <array-n>(<db2-count>)
COMPUTE <db2-count> = <db2-count> + 1

62

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.17 INSERT-PARA-1
MOVE <ARRAY-1>(<db2-count>) TO <DB2-ARRAY-1>.
EXEC SQL
INSERT INTO <array-1-table-1>(<array-1-table-1-col-1>... <array-1-table-1-col-n>)
VALUES (<db2-array-1-table-1-col-1-fld-1>, .. <db2-array-1-table-1-col-n-fld-n>)
END-EXEC
IF SQLCODE = 0
THEN
EXEC SQL
INSERT INTO <array-1-table-2>(<array-1-table-2-col-1> ... <array-1-table2-col-n>)
VALUES (<db2-array-1-table-2-col-1-fld-1>, .. <db2-array-1-table-2-col-nfld-n>)
END-EXEC
END-IF
...
IF SQLCODE = 0
THEN
EXEC SQL
INSERT INTO <array-1-table-n>(<array-1-table-n-col-1> ... <array-1-tablen-col-n>)
VALUES (<db2-array-1-table-n-col-1-fld-1>, .. <db2-array-1-table-n-col-nfld-n>)
END-EXEC
END-IF
COMPUTE <db2-count> = <db2-count> + 1

5.1.18 INSERT-PARA-n.
MOVE <array-n>(<db2-count>) TO <db2-array-n>.
EXEC SQL
INSERT INTO <array-n-table-1>(<array-n-table-1-col-1>... <array-n-table-1-col-n>)
VALUES (<db2-array-1-table-1-col-1-fld-1>, .. <db2-array-1-table-1-col-n-fld-n>)
END-EXEC.
IF SQLCODE = 0
THEN
EXEC SQL
INSERT INTO <array-n-table-2>(<array-n-table-2-col-1> ... <array-n-table2-col-n>)
VALUES (<db2-array-1-table-2-col-1-fld-1>, .. <db2-array-1-table-2-col-nfld-n>)
END-EXEC
END-IF
...
IF SQLCODE = 0
THEN
EXEC SQL
INSERT INTO <array-n-table-n>(<array-n-table-n-col-1> ... <array-n-tablen-col-n>)
VALUES (<db2-array-1-table-n-col-1-fld-1>, .. <db2-array-1-table-n-col-nfld-n>)
END-EXEC
END-IF
COMPUTE <db2-count> = <db2-count> + 1

63

VSAM to DB2 Online and Batch Program Conversion Strategy

5.1.19 UPDATE-PARA-1
EXEC-SQL
FETCH
INTO <db2-array-1-table-1-col-1-fld-1>, .. <db2-array-1-table-1-col-n-fld-n>
...
<db2-array-1-table-n-col-1-fld-1>, .. <db2-array-1-table-n-col-n-fld-n>
END-EXEC
MOVE <array-1>(<db2-count>) TO <db2-array-1>
EXEC SQL
UPDATE <array-1-table-1>
SET
<array-1-table-1-col-1> = <db2-array-1-table-1-col-1-fld-1>,
...
<array-1-table-1-col-n> = <db2-array-1-table-1-col-n-fld-n>
WHERE CURRENT OF
END-EXEC
COMPUTE <db2-count> = <db2-count> + 1

5.1.20 UPDATE-PARA-n
EXEC-SQL
FETCH
INTO <db2-array-1-table-1-col-1-fld-1>, .. <db2-array-1-table-1-col-n-fld-n>
...
<db2-array-1-table-n-col-1-fld-1>, .. <db2-array-1-table-n-col-n-fld-n>
END-EXEC
MOVE <array-1>(<db2-count>) TO <db2-array-1>
EXEC SQL
UPDATE <array-1-table-1>
SET
<array-1-table-1-col-1> = <db2-array-1-table-1-col-1-fld-1>,
...
<array-1-table-1-col-n> = <db2-array-1-table-1-col-n-fld-n>
WHERE CURRENT OF
END-EXEC
COMPUTE <db2-count> = <db2-count> + 1

64

VSAM to DB2 Online and Batch Program Conversion Strategy

65