CCAPRINT: A Newsletter for Model 204® and System 1032® Users	October 10, 2004

	Model 204: Becoming a V6R1 Beta Site
	System 1032: Back to Basics, Part 3 – Attribute Qualities	Printer-friendly version
	Model 204: MONITOR SERV - Tracking Servers in a Model 204 Online	Printer-friendly version

Model 204

USE OF AND ACCESS TO PRODUCTS AND FEATURES ARE IN ACCORDANCE WITH THE TERMS AND CONDITIONS OF THE USER’S SOFTWARE LICENSE. THE PRESENTATION OF MATERIAL HEREIN DOES NOT, IN ANY MANNER, MODIFY SUCH TERMS AND CONDITIONS.

Becoming a V6R1Beta Site

Computer Corporation of America is pleased to announce our plans to release Model 204 in the Spring of 2005. Between now and then, we need you and your data to give V6R1 a tryout shake down. We cannot promise your name in lights, but we do promise prompt, friendly, experienced help as you test Version 6.1. This version might just include features and enhancements you have longed for:

		Large Object data support
		The TCP/IP Socket Interface for Model 204
		Automatic logon password expiration and renewal control
		Dynamic support to refresh APSY procedure compilations and subsystem definition attributes
		A dynamic data compactor
		Plus, an assortment of command and parameter upgrades and performance improvements that you have come to expect from Model 204.

If you are interested, please contact your account representative or call 508-270-6666 and ask to be directed to your representative.

Back to Basics, Part 3 – Attribute Qualities
By Tym Stegner

In the previous Back to Basics article in this series, we reviewed the definition of attributes that are the database fields of the System 1032 world. In this article we continue our examination of attribute definitions, focusing upon attribute qualities that refine attribute capabilities.

Overview of Attribute Qualities
Attribute qualities are used to augment and extend the basic definition of an attribute. Furthermore, some qualities govern appropriate data entry. As you would think, not all qualities are applicable to all data types and some of them are mutually exclusive, as will be noted as we describe them.

Attribute qualities fall into three broad categories: Structural, Usage, and Index. The first two categories are discussed in this article, because they influence how data is stored or whether data values are accepted. We are saving the Index qualities for another article, because they influence how queries are accomplished.

Introducing Structural Qualities
Structural qualities affect how attribute data is stored. You must declare these qualities in the original attribute definition, because you cannot modify them later. Structural qualities include: RANGE, INITIALLY, PRECISION, and MISSING.

RANGE lb : ub
In System 1032, Numeric and Date-Time attributes can be automatically constrained to a contiguous set of values. When you define a RANGE quality, it has two boundary parameters, lower and upper, separated by a colon, that denote the inclusive endpoints of the valid values. Text and Binary Varying attributes are excluded from RANGE operations. Other data types have a default range comprised of all allowed values for that data type. Specification of the RANGE option allows a restriction of the default range.

In a study of this topic, I found the break-even point to be around the 80-character mark. If your text values are 80 characters or fewer, just use fixed-length Text for your data, instead of incurring the overhead of Text Varying.

PRECISON units
The PRECISION quality controls the granularity of the data values stored. For Decimal and Real attributes, you can specify an integer for the number of significant digits for the value. For date-time attributes, the PRECISION setting controls the least significant element of the date and/or time to be stored--in descending order: years, months, days, hours, minutes, seconds, milliseconds, microseconds.

Usage Qualities
The Usage qualities control how attributes behave within certain PL1032 commands. You can modify these qualities after you define and use attribute definition, if you wish. Usage qualities include: OR, COLUMN, LENGTH, COMMENT, DEMAND, PROMPT, Summaries, TITLE, FORMAT, and UPDATE.

OR name
The OR quality provides one or more alternate names for the attribute—up to 126 other names. Often, the primary name of an attribute is very descriptive and long. Repeatedly entering a long name can become tiring, so defining a short alternate name, or names, can be helpful. In addition, there are efficiency considerations in using a short name as the primary attribute name, and then using the long descriptive name as one of the alternate names.

COLUMN col1 col2
The COLUMN quality predefines the location of the attribute data location in the dataset Record Descriptor (RD). The COLUMN quality has two integer parameters to denote the starting and ending column position of the data in an external file. Specify COLUMN when the order of the data in the external file does not match the attribute order. COLUMN is mutually exclusive with LENGTH.

LENGTH n
The LENGTH quality denotes the number of characters of the attribute data in an external file. Specify LENGTH when the attribute is arranged sequentially within the external file. The LENGTH quality has an integer parameter. Although you cannot use LENGTH and COLUMN on the same attribute, individual attributes in the same dataset can use LENGTH or COLUMN, depending on the order of the data values within the external record.

COMMENT text
The COMMENT quality is often used by the dataset designer to describe the use or purpose of the attribute. Some sites have completely documented the value distribution of an attribute in an extended comment. The COMMENT quality takes a single text parameter. The delimited text type code (#TD/…/) is often used with the comment option to enable multi-line comments to be specified.

DEMAND
The DEMAND quality is used by the ADD command to insist that a data entry user specify a value for this attribute during interactive data entry.

PROMPT text
The PROMPT quality specifies a text string that is used as prompting text during interactive data entry. The PROMPT quality has a text string parameter.

Summaries
Summaries can include MINIMUM, MAXIMUM, AVERAGE, TOTAL. When you specify an attribute with a Summary quality within a PRINT command item list, the attribute quality displays one or more automatic summary conditions for the attribute. If summary types are specified, they are automatically calculated and displayed whenever the attribute is printed.

TITLE text [position]
The TITLE quality is a text string for the column title that is displayed during PRINT operations. The TITLE quality also provides an optional position keyword for title position: LEFT, CENTERED, RIGHT. For a multiple line title you can use a backslash character (\) within the title string.

FORMAT fmtspec
The FORMAT quality specifies how an attribute value is displayed. Numeric formats control the number of digits, the placement of a decimal point or sign, exponentation and thousands quoting. Date-Time formats control the physical output representation of the date and/or time. Text formats control number of characters, line/character wrapping and non-printing character disposition. The FORMAT quality requires a text parameter. (Because the formatting possibilities are extensive we will cover them in depth in an upcoming article.)

Examining Attribute Qualities and System 1032 Defaults

In Figure A a simple dataset with three attributes is created specifying only the minimum required syntax for the attributes.

Figure A. Creating dataset TMP1 with three attributes

1032> create dataset tmp1 1032 Data Definition Language DDL> attribute one text 1 keyed DDL_ attribute two integer keyed DDL_ attribute third_one date DDL_ end

Figure B is a SHOW ATTRIBUTE FULL command that displays the complete attribute definitions that includes the System 1032 defaults for attributes. Note the default title and prompt are both based upon the primary name of the attribute.

Figure B. Examining the system generated defaults for attribute qualities

1032> show attribute full Attribute ONE Text 1 Title Centered "ONE" Format A1 – Prompt "ONE" Missing System_Default Initially Missing – Column 1 1 Keyed Ignore_Case No_Multinational; Attribute TWO Integer Range -2147483647 : 2147483647 Title – Centered "TWO" Format N10 Prompt "TWO" Missing 2147483648 – Initially Missing Column 2 12 Keyed; Attribute THIRD_ONE Date Range '01/02/0000' : '01/01/9999' – Precision Days Title Centered "THIRD_ONE" Format D3 – Prompt "THIRD_ONE" Missing '01/01/0000' Initially Missing – Column 13 22;

In Figure C a series of attribute qualities are defined for the same attributes in TMP1, including COLUMN qualities to pick up data from an external file not in defined attribute order.

Figure C. Examining an attribute with more than two names

1032> create dataset tmp3 1032 Data Definition Language DDL> attribute one text 1 column 10 10 title "First/Attribute" – DDL_ Missing "Z" DDL> attribute two integer column 1 4 title "Only/An/Integer" – DDL_ Missing -1 Range 0:100 DDL> attribute nxt or third_one date range '1/1/1900':'1/1/2033' – DDL_ title "A Date" precision months column 11 20 DDL_ end_dataset

Compare Figures C and D. Note that NXT has become the primary name for the original attribute THIRD_ONE. Because NXT is first name defined, it is now the primary name. NXT is now the Prompt value for that attribute and will be the column heading in PRINT command processes.

Figure D. Examining the RANGE quality and System 1032 default handling

1032> show attribute full Attribute ONE Text 1 Title Centered "First/Attribute" – Format A1 Prompt "ONE" Missing "Z" Initially Missing Column 10 10; Attribute TWO Integer Range -1 : 100 Title Centered – "Only/An/Integer" Format N3 Prompt "TWO" Missing -1 Initially Missing Column 1 4; Attribute NXT or THIRD_ONE Date_Time – Range '01/01/1900' : '01/01/2033' Precision Months – Title Centered "A Date" Format D3 Prompt "NXT" – Missing '12/01/1899' Initially Missing Column 11 20;

Again compare Figures C and D, this time looking at both RANGE and MISSING qualities. For Attribute TWO we defined a RANGE of 0 to100 and MISSING as –1, which is one less unit (an integer) outside the range. Because we defined the MISSING value, System 1032 increased the RANGE value to include the MISSING value.

However, for attribute NXT we defined a RANGE of 01/01/1900 to 01/01/2033, but we did not define a MISSING value. So System 1032 set the MISSING value to one less unit (a month) than the RANGE value, but did not change the RANGE values themselves.

In Summary
As in the last article, the ability to store and represent data is the purpose of a DBMS. System 1032 attributes and their qualities lets you accurately represent and present your data values.

USER SVR   BUF FLS  PCPU  SMPLS    RUNG    REDY    BLKI    WTSV    BLKO   SWPG
 152   1 59988 130 0.665     30   0.766   0.466  95.566   0.000   0.000   0.000

USER SVR USERID     SERVSIZE  LSERVER FUNC  CNCT        CPU  SEQIO QUE  WT FLGS
  44   1 ARWEN        800000   290816 EVAL   512      0.116     31 BLKI  3  67
       2              800000   290816                              WTUS
  29   3 HARPO        650000   352256 EVAL    16      0.085     23 REDY
  33   4 LEGOLAS      650000   143360 EVAL   771      0.268     67 RUNG
  48   5 DEWEY        650000   290816 EVAL   102      0.059     21 BLKI  3  67
  17   6 PARIS        650000   282624 EVAL  3877      1.064    227 BLKI  3  67
       7              425000   221184                              WTUS
  69   8 MUSTAFA      425000   290816 EVAL  1202      0.615    165 BLKI  3  67
 101   9 SAMWISE      425000   290816 EVAL  6853     28.980    962 BLKI  3  67
  31  10 CHER         425000   290816 EVAL  2359      3.116    145 BLKI  3  67

Copyright © 2008 Computer Corporation of America.
All right reserved. Published in the United States of America.