This documentation is not maintained. Please refer to doc.castsoftware.com/technologies to find the latest updates.

Summary: This document provides information about the extension that provides data column access support for the SQL Analyzer extension (from version ≥ 2.4). See also Data Sensitivity for more information about other technologies that support Data Sensitivity checks.

Extension ID

com.castsoftware.datacolumnaccess

What's new?

Please see Data Column Access - 2.2 - Release Notes for more information.

Description

The Data Column Access extension provides the following features:

  • Data sensitivity checks on table columns (custom and predefined)
  • Creation of column access links

In what situation should you install this extension?

  • If you need to check that the application is data sensitive compliant using custom and predefined indicators for table columns
  • If you need to see Access Read and Access Write links for columns on SELECT, INSERT, UPDATE, DELETE and MERGE statements
  • If you need to see Access Write links between columns and Cobol data, from INSERT, UPDATE and SELECT .. INTO statements
  • If you need to see Access Write links for columns selected in the DECLARE cursor statement and the Cobol data fetched in FETCH cursor statement

AIP Core compatibility

AIP Core releaseSupported
8.3.x(tick)

Supported DBMS servers used for CAST AIP schemas

AIP Core releaseCSS/PostgreSQL
All supported releases(tick)

Supported client languages

All supported client languages are listed here : SQL Analyzer - 3.6#clientsidesupportClientSideSupportSQL Analyzer - 3.6#SpecialnotesaboutQualityRulesonclientside

Prerequisites

(tick)An installation of any compatible release of CAST AIP (see table above)
(tick)An installation of SQL Analyzer extension (from version ≥ 2.4)

Download and installation instructions

This extension will be automatically downloaded and installed when SQL source code is delivered for analysis. It can be managed using the Application - Extensions interface:


Note that if you are using the legacy workflow without Fast Scan in CAST Console (not required when using the Fast Scan workflow where this option is automatically enabled), you must also manually enable the Data Sensitivity option as part of the source code delivery process - see Application onboarding without Fast Scan - Standard onboarding - perform all actions - choose objectives:

Data sensitivity configuration

See Data Sensitivity for more information about setting a property on specific objects for the following:

  • custom sensitivity
  • GDPR
  • PCI-DSS

What results can you expect?

Table/column properties for data sensitivity (custom and predefined)

  • Data Sensitivity Indicator: displays the most sensitive value (Sensitive, Very Sensitive, Highly Sensitive) for any of the columns in the selected table.
  • List of the specific Sensitive / Very sensitive / Highly sensitive columns

Example:

  • Links are created for transaction and function point needs.
  • You can expect the following links on the DDL side within the same sql file:
    • accessRead from View / Procedure / Function / Trigger / Event  to Column
    • accessWrite from Procedure / Function / Trigger / Event  to Column
  • You can expect the same links for the following client side, only if the server-side code has been analyzed with the SQL Analyzer extension and only if they have dependencies with the SQL Analyzer analysis results:

COBOL PB VB .NET JAVAC/C++/OBJC IOS PYTHON

  • When Cobol Data are created during Cobol Analysis and the production option Save data and links to other data is activated, you can expect to the following links:
    • accessWrite from Cobol Data  to Column, for the case of insert into and update. E.g. : Insert into table1 (col1, col2, ...) values (:data1, data2) the data1, data2 will access write the col1, col2. Update table1 set col1 = :data1, col2 = :data2 the data1, data2 will access write col1, col2
      • Sometimes those links have bookmarks but sometimes the bookmarks are missing, for details refer: Limitations. When the bookmark exists, it will pointing on the Cobol Data definition and not on the SQL statement.
    • accessWrite from Column to Cobol Data, for the case of select into. E.g. : Select col1, col2, ... into :data1, :data2 the col1, col2 will access write the data1, data2. 
      • Those links have no bookmarks, for details refer: Limitations.
    • accessWrite from Column selected in DECLARE cursor to Cobol Data fetched in FECTH cursor. E.g. : declare toto_curs cursor for select col_toto from toto, followed by fetch toto_curs into :var1. The col_toto will access write the var1. 
      • Those links have no bookmarks, for details refer: Limitations.

Why does CAST not create accessRead links between COBOL data variables and SQL columns ?

CAST does not create links between the COBOL data variables and selected columns in the WHERE clauses because those links are not significant for "how data circulates in an application". To resume, CAST only creates accessWrite links as follows:

  • SELECT: Only the INTO clause is considered because the SQL Column will write into the Cobol Data variables
  • UPDATE: Only the SET clause is be considered because he Cobol Data variables will write into the SQL Column
  • INSERT: Only the Cobol Data variables that will write into the SQL Columns is considered
  • FETCH: The SQL columns selected in the declare cursor will write into the fetched Cobol Data variables

Examples

Select Into, Move, Update

The value we have in the column COL_TOTO will be written in the Cobol data VAR1. What we have in VAR1 will move into the Cobol Data VAR2. At the end, VAR2 will update the column COL_TOTO, here is the source code :

       IDENTIFICATION DIVISION.                                         00010000
       PROGRAM-ID.   TESTMOVE.                                          00020000
       DATA DIVISION.                                                   00340000
       WORKING-STORAGE SECTION.                                         00010000
       01 VAR1 PIC X.                                                   00010000
       02 VAR2 PIC X.                                                   00010000
                                                                        00010000
       PROCEDURE DIVISION.                                              00010000
                                                                        00010000
       EXEC SQL                                                         00010000
        SELECT COL_TOTO                                                 00010000
            INTO :VAR1                                                  00010000
        FROM TOTO                                                       00010000
       END-EXEC                                                         00010000
                                                                        00010000
       MOVE VAR1 TO VAR2                                                00010000
                                                                        00010000
       EXEC SQL                                                         00010000
       UPDATE TITI                                                      00010000
        SET COL_TITI = :VAR2                                            00010000
       END-EXEC                                                         00010000


Insert Into

Update

Select into

 

Declare, Fetch, Move and Insert

The value we have in the column COL_TOTO is selected in DECLARE TOTO_CURS, than it will be written during the FETCH TOTO_CURS in the Cobol data VAR1. What we have in VAR1 will move into the Cobol Data VAR2. At the end, VAR2 will be inserted in the table TITI, column COL_TITI, here is the source code:

       IDENTIFICATION DIVISION.                                         00010000
       PROGRAM-ID.   TESTCURSOR.                                        00020000
       DATA DIVISION.                                                   00340000
       WORKING-STORAGE SECTION.                                         00010000
       01 VAR1 PIC X.                                                   00010000
       02 VAR2 PIC X.                                                   00010000
                                                                        00010000
       PROCEDURE DIVISION.                                              00010000
                                                                        00010000
       EXEC SQL
        DECLARE TOTO_CURS CURSOR FOR                                    00010000
        SELECT COL_TOTO                                                 00010000
        FROM TOTO                                                       00010000
       END-EXEC                                                         00010000
                                                                        00010000
       EXEC SQL                                                         00010000
       FETCH TOTO_CURS                                                  00010000
        INTO :VAR1                                                      00010000
       END-EXEC                                                         00010000
                                                                        00010000
       MOVE VAR1 TO VAR2                                                00010000
                                                                        00010000
       EXEC SQL                                                         00010000
       INSERT INTO TITI (COL_TITI)                                      00010000
        VALUES (:VAR2)                                                  00010000
       END-EXEC                                                         00010000
                                                                        00010000
                                                                        00010000



Performance

When using the Data Column Access extension 1.0.0, analysis execution time is slightly different (usually slightly longer). Here are some details of the performance we have experienced using AIP Core 8.3.24 and SQL Analyzer 3.4.4-funcrel:

Use caseTechnical detailsExecution details

100% SQL (Teradata)

1 single SQL file, file size 14.5MB

Detected variant: Ansisql

129,529 LOC

Analysis duration for 1 analysis unit:

  • without Data Column Access extension = 20 min 20 sec
  • with Data Column Access extension = 21 min 26 sec  

No data sensitivity

988 access links are added

Cobol + SQL (Db2)

SQL: 1 single file, file size 289KB

Detected variant: Db2

Cobol: 1,996 files, 11 folders, 52MB

443.8K LOC

Analysis duration for 2 analysis units:

  • without Data Column Access extension = 19 min 19 sec
  • with Data Column Access extension = 19 min 27 sec

No data sensitivity

15,245 access links are added, all of them on the Cobol side

.NET + SQL (Microsoft SQL Server)

SQL: 6 files (3 files of 6 are .sqltablesize), 1 folder, 4.08 MB

Detected variant: SQL Server

.NET: 14,062 files, 2,747 folders, 1.04GB

Assemblies: 218 files, 214 folders, 68.2MB

698K LOC

Analysis duration for 185 analysis units:

  • without Data Column Access extension = 2 hours 9 min 34 sec
  • with Data Column Access extension = 2 hours 5min 27 sec

No data sensitivity

From a total of 2,562 added access links, 6 are detected on the .NET side