Data Type Conversion
A COBOL data item must be compatible with the Ingres value it represents. Numeric Ingres values can be set by and retrieved into COBOL numeric and numeric edited items, and Ingres character values can be set by and retrieved into COBOL character data items, that is, alphabetic, alphanumeric, and alphanumeric edited items.
Data type conversion occurs automatically for different numeric types such as from floating-point Ingres database column values into integer (COMP) COBOL data items, and for different length character strings, such as from varying‑length Ingres character fields into COBOL alphabetic and alphanumeric data items.
Ingres does not automatically convert between numeric and character types, such as from Ingres integer fields into COBOL alphanumeric data items. You must use the Ingres type conversion functions, the Ingres ascii function, or the COBOL STRING statement to effect such conversions.
The following table shows the default type compatibility for each Ingres data type in UNIX and VMS. Note that some COBOL types are omitted from the table because they do not exactly match an Ingres type. Use of those types necessitates some runtime conversion, which may possibly result in some loss of precision.
Windows and UNIX:
There is no exact match for float, so use COMP‑3.
Ingres types and their corresponding COBOL data types are listed in the following table:
*This type may not map to 8-byte integers with some COBOL compilers.
Note that Ingres stores decimal as signed. Thus, use a signed decimal variable if it interacts with an Ingres decimal type. Also, Ingres allows a maximum precision of 39 while COBOL allows only 18.
Decimal Type Conversion
An Ingres decimal value that will not fit into a COBOL variable will either be truncated if there is loss of scale or cause a runtime error if loss of significant digits.
Runtime Numeric Type Conversion
The Ingres runtime system provides automatic data type conversion between numeric‑type values in the database and the forms system and numeric COBOL data items. It follows the standard COBOL type conversion rules. For example, if you assign the value in a scaled COMP‑3 data item (UNIX and Windows) or COMP‑1 data item (VMS) to an integer‑valued field in a form, the digits after the decimal point of the data item's value are truncated. Runtime errors are generated for overflow on conversion.
The preprocessor generates COBOL MOVE statements or calls Ingres convert routines that convert various COBOL data types. These can again be converted at runtime by Ingres based on the final value being set or retrieved. The standard COBOL data conversion rules hold for all these generated MOVE statements, with a potential loss of precision.
Floats are coerced to decimal types by Ingres at runtime.
The preprocessor uses temporary data items when moving values between numeric DISPLAY data items and Ingres objects. Depending on the PICTURE clause of the DISPLAY item shown below, these temporary data items are either:
• COMP-3 or 4‑byte COMP-5 (UNIX or
• COMP-2 or 4‑byte COMP (VMS)
The following table lists numeric DISPLAY items and temporary data items:
COMP-3 items used to set or receive Ingres values also require some runtime conversion. This is not true if you are setting or receiving decimal data. This is true for Micro Focus COBOL when float values are received into COMP‑3.
The preprocessor also generates code to use a temporary data item when Ingres data is to interact with a COBOL unscaled COMP data item whose picture string is exactly 10. Because a COBOL non‑scaled numeric item whose picture contains 10 or fewer digits is regarded as compatible with the Ingres integer type, ESQL/COBOL assigns such data to a temporary COBOL 4‑byte COMP‑5 data item to allow it to interact with Ingres integer data. Note that the range of the Ingres i4 type does not include all 10‑digit numbers. If you have 10‑digit numeric data outside the Ingres range, you should use a COMP‑3 (UNIX) or for VMS use COMP‑1 or COMP‑2 data item and choose the Ingres float type. Or with decimal you can use COMP‑3 and choose a decimal Ingres type.
You can use only COMP data items or items that get assigned to temporary 4‑byte COMP‑5 (UNIX) or COMP (VMS) data items to set the values of Ingres integer objects, such as table field row numbers. You can, however, use any numeric data items to set and retrieve numeric values in Ingres database tables or forms.
Windows and UNIX:
The Ingres money type is represented as a COMP‑3 data item.
VMS:
The Ingres money type is represented as an 8‑byte floating-point value, COMP‑2.
Recall that a COBOL non-scaled numeric item with a picture that contains 10 or fewer digits is regarded as compatible with the Ingres integer type. (For details, see
Variable and Type Declarations in this chapter.) However, the VAX standard data type for an unscaled 10‑digit COMP item is a quadword (8 bytes). Therefore, ESQL/ COBOL assigns such data to a temporary COBOL 4‑byte COMP data item to allow it to interact with Ingres integer data. Note that the range of the Ingres integer4 type does not include all 10‑digit numbers.
Runtime Character and Varchar Type Conversion
Automatic conversion occurs between Ingres character string values and COBOL character variables (alphabetic, alphanumeric, and alphanumeric edited data items). The string-valued Ingres objects that can interact with character string variables are:
• Ingres names, such as form and column names
• Database columns of type character
• Database columns of type varchar
• Form fields of type character
• Database columns of type long varchar
Several considerations apply when dealing with character string conversions, both to and from Ingres.
The conversion of COBOL character variables used to represent Ingres names is simple: trailing blanks are truncated from the variables because the blanks make no sense in that context. For example, the string constants empform and empform refer to the same form.
The conversion of other Ingres objects is a bit more complicated. First, the storage of character data in Ingres differs according to whether the medium of storage is a database column of type character, a database column of type varchar, or a character form field. Ingres pads columns of type character with blanks to their declared length. Conversely, it does not add blanks to the data in columns of type varchar or long varchar, or in form fields.
Second, the COBOL convention is to blank‑pad fixed‑length character strings. For example, the character string abc may be stored in a COBOL PIC X(5) data item as the string abc followed by two blanks.
When character data is retrieved from a database column or form field into a COBOL character variable and the variable is longer than the value being retrieved, the variable is padded with blanks. If the variable is shorter than the value being retrieved, the value is truncated. You must always ensure that the variable is at least as long as the column or field, in order to avoid truncation of data. You should note that, when a value is transferred into a data item from an Ingres object, it is copied directly into the variable storage area without regard to the COBOL special insertion rules.
When inserting character data into an Ingres database column or form field from a COBOL variable, note the following conventions:
• When data is inserted from a COBOL variable into a database column of type character and the column is longer than the variable, the column is padded with blanks. If the column is shorter than the variable, the data is truncated to the length of the column.
• When data is inserted from a COBOL variable into a database column of type varchar or long varchar and the column is longer than the variable, no padding of the column takes place. Furthermore, by default, all trailing blanks in the data are truncated before the data is inserted into the varchar column. For example, when a string abc stored in a COBOL PIC X(5) data item as abc (see above) is inserted into the varchar column, the two trailing blanks are removed and only the string abc is stored in the database column. To retain such trailing blanks, you can use the Ingres notrim function. It has the following syntax:
notrim(:charvar)
where charvar is a character variable. The following example demonstrates this feature. If the varchar column is shorter than the variable, the data is truncated to the length of the column.
• When data is inserted from a COBOL variable into a character form field and the field is longer than the variable, no padding of the field takes place. In addition, all trailing blanks in the data are truncated before the data is inserted into the field. If the field is shorter than the data (even after all trailing blanks have been truncated), the data is truncated to the length of the field.
When comparing character data in an Ingres database column with character data in a COBOL variable, note the following convention:
• When comparing data in character or varchar database columns with data in a character variable, all trailing blanks are ignored. Initial and embedded blanks are significant.
Caution! As described above, the conversion of character string data between Ingres objects and COBOL variables often involves the trimming or padding of trailing blanks, with resultant change to the data. If trailing blanks have significance in your application, give careful consideration to the effect of any data conversion. For a more complete description of the significance of blanks in string comparisons, see the SQL Reference Guide.
The Ingres date data type is represented as a 25‑byte character string: PIC X(25).
The program fragment in the following example demonstrates the notrim function and the truncation rules explained above.
Example: notrim function usage
DATA DIVISION.
WORKING-STORAGE SECTION.
EXEC SQL INCLUDE SQLCA END-EXEC.
EXEC SQL BEGIN DECLARE SECTION END-EXEC.
EXEC SQL DECLARE varychar TABLE
(row integer,
data varchar(10))
END-EXEC.
01 ROW PIC S9(4) USAGE COMP.
01 DATA PIC X(7).
EXEC SQL END DECLARE SECTION END-EXEC.
PROCEDURE DIVISION.
BEGIN.
* DATA will hold "abc " followed by 4 blanks.
MOVE "abc " TO DATA.
* The following INSERT adds the string "abc"
* (blanks truncated).
EXEC SQL INSERT INTO varychar (row, data)
VALUES (1, :DATA)
END-EXEC.
* This statement adds the string "abc ", with 4 trailing
* blanks left intact by using the NOTRIM function
EXEC SQL INSERT INTO varychar (row, data)
VALUES (2, NOTRIM(:DATA))
END-EXEC.
* This SELECT will retrieve the second row,
* because the NOTRIM
* function of the previous INSERT statement
* left trailing blanks in the "data" variable.
EXEC SQL SELECT row
INTO :ROW
FROM varychar
WHERE length(data) = 7
END-EXEC.
DISPLAY "Row found = " ROW.