A Framework for Dimensional Benchmarking of SQL Query Performance

A few years ago I wanted to investigate the performance of different SQL queries for the same problem, and wanted to include analysis of how the queries’ performance varied with problem size. In order to do this efficiently I wrote an Oracle framework consisting of tables, packages, types etc., and which I have now published here, dim_bench_sql_oracle on GitHub. As well as the obvious cpu and elapsed times, I included statistics contained in the execution plan tables, and also the differences in v$ view statistics that are gathered in the well known Runstats scripts, (originally developed by Tom Kyte, and for which there now seem to be lots of variations around, such as
Runstats utility
). My approach is to collect these statistics in tables keyed by both query and dimensions to allow for more elaborate reporting, and to easily detect unscaleable queries, for example that use resources at a rate that grows quadratically, or worse, with problem size, as in one of the demo queries. Output goes to both a text log file, and to summary csv files for importing to Excel.

This article has design information on the framework, and is best read in conjunction with the first article on its use for a specific problem, Dimensional Benchmarking of Oracle v10-v12 Queries for SQL Bursting Problems, which includes the output log produced. This problem is included in the GitHub project, so anyone interested can run it themselves fairly easily.

Update, 26 November 2016: A notes section has been added discussing design issues and features.

Here is one of the Scribd articles for which I originally developed the framework:

Loading...

Bench Data Model – ERD

bench-1-0-erd

Code Structure Diagram

bench-1-1-csd

Test_Queries Package Call Structure Table

Level 1 Level 2 Level 3 Package
Add_Query Test_Queries
Init_Statistics Test_Queries
Plan_Lines Test_Queries
Get_SQL_Id Utils
Display_Cursor DBMS_XPlan
Write_Plan_Statistics Test_Queries
Get_Queries Test_Queries
Write_Log Utils
Flush_Buf Test_Queries
Init_Time Timer_Set
Put_Line UTL_File
Increment_Time Timer_Set
Write_Line Test_Queries
Flush_Buf Test_Queries
Open_File Test_Queries
Construct Timer_Set
Fopen UTL_File
Close_File Test_Queries
Flush_Buf Test_Queries
FClose UTL_File
Write_Times Timer_Set
Outbound_Interface Test_Queries
Construct Timer_Set
Open_File Test_Queries
Write_Line Test_Queries
Process_Cursor Test_Queries
Write_Log Utils
Init_Statistics Test_Queries
Init_Time Timer_Set
Increment_Time Timer_Set
Write_Line Test_Queries
Flush_Buf Test_Queries
Write_Plan_Statistics Test_Queries
Write_Plan Utils
Write_Times Timer_Set
Get_Timer_Stats Timer_Set
Close_File Test_Queries
Write_Log Utils
Write_Size_list Test_Queries
Write_Log Utils
Write_Twice Test_Queries
Write_Line Test_Queries
Write_CSV_Fields Utils
Write_Data_Points Test_Queries
Heading Utils
Write_Line Test_Queries
Write_Twice Test_Queries
Write_Distinct_Plans Test_Queries
Heading Utils
Write_Log Utils
Write_Rows Test_Queries
Heading Utils
Write_Line Test_Queries
Write_Twice Test_Queries
Write_Rows Test_Queries
Write_Stat Test_Queries
Write_Log Utils
Write_Rows Test_Queries
Write_All_Facts Test_Queries
Write_Distinct_Plans Test_Queries
Write_Data_Points Test_Queries
Write_Rows Test_Queries
Write_Stat Test_Queries
Heading Utils
Close_File Test_Queries
Write_Stats Test_Queries
Open_File Test_Queries
Write_All_Facts Test_Queries
Write_Log Utils
Term_Run Test_Queries
Write_Times Timer_Set
Get_Run_Details Test_Queries
Construct Timer_Set
Create_Log Utils
Write_Log Utils
Heading Utils
Write_Size_list Test_Queries
Get_Queries Test_Queries
Set_Data_Point Test_Queries
Init_Time Timer_Set
Get_CPU_Time DBMS_Utility
Setup_Data Query_Test_Set
Increment_Time Timer_Set
Run_One Test_Queries
Construct Timer_Set
Outbound_Interface Test_Queries
Write_Log Utils
Write_Times Timer_Set
Create_Run Test_Queries
Write_Log Utils
Main Test_Queries
Main Test_Queries
Main Test_Queries
Get_Run_Details Test_Queries
Set_Data_Point Test_Queries
Run_One Test_Queries
Increment_Time Timer_Set
Write_Stats Test_Queries
Write_Log Utils
Write_Other_Error Utils

 
Outputs

out folder

The demo script Test_Bur.sql writes the log data to ..\out\Test_Bur.LST.

The program loops over each (W, D) data point included in the driving script lists, and outputs for each query in the group data including:

  • Full execution plan
  • CPU and elapsed timings of query execution steps, file writing and data setup

On completion of the loops, summary reports are written to both the main log and to summary csv files, mentioned below, with information including

  • Distinct execution plans (main log only)
  • Data point statistics including setup timings and records created
  • Numbers of output records from the queries
  • cpu and elapsed times for the queries
  • Execution plan statistics
  • Numerous v$ After-Before statistic differences (following the Run_Stats model)

Oracle directory: output_dir

For a query group and query with data points W-D, the results of running the query are written to:

<query_group>_<query>_<W>-<D>.csv

For example:

BURST_MTH_QRY_30-30

Two summary files are written, with the bench run id as part of the name:

<bench run id>_W.csv
<bench run id>_D.csv

These files contain all the detailed statistics in csv format, so that they can be imported into Excel and used to obtain graphs. _W has the width parameter as the row and depth as the column, and _D the other way round.

Notes

Query Timing

Obtaining reliable comparative timings of queries in a test environment is not as straightforward as it may seem. Some of the issues are considered in this article, for example, Timing an ALL_ROWS query. For ad hoc tests, running the query in SQL*Plus after ‘SET AUTOTRACE TRACEONLY’ is one of the better approaches. However, in this framework a different approach is taken in order to simulate the performance that might be obtained if the query records were fetched in batches to be processed, say as an outbound interface, where they may be written to a file on the server. To do this, the query select list is converted into CSV format and the records are written to a file, with care taken to separate the timings of the query operations from those of the file processing.

Hard Parsing

This framework is not intended for testing OLTP SQL but relatively long-running batch-type SQL, where the cost of parsing is generally negligible. As the dataset sizes vary it is possible that the execution plan may vary, so it is important that the SQL engine performs a hard-parse on each execution of a query to ensure plan re-calculation. A hard parse is ensured by appending a placeholder field into the select list CSV string of the transformed queries, which is then replaced before each execution by a random number: The SQL engine considers the queries then to be distinct and therefore re-parses them.

Code Timing

The processing within the framework is heavily instrumented using the author’s own code timing utility package Code Timing and Object Orientation and Zombies. This is very low footprint in terms both of code and of performance, operating entirely in memory with individual timers keyed by name, and (logically) object oriented so that multiple timer sets can be running at once. Timings are printed to log, and the cpu and elapsed times for the query executed are summed from the individual components for the query, together with the times for any pre-query step:

  • Pre SQL
  • Open cursor
  • First fetch
  • Remaining fetches

Benchmarking Non-Query SQL

The framework is centred around the concept of a group of queries that are run in turn against the same dataset for each dataset point. However, non-query SQL can also be benchmarked in two ways: First, the query can include database PL/SQL functions; and secondly, the query metadata record includes a clob field for pre-query SQL that can be a PL/SQL block, while the actual query could just be ‘select 1 from dual’.

Query Transformation by Regular Expression Processing

The query output is written to file in csv format, includes the hint GATHER_PLAN_STATISTICS, and has a placeholder for a random number. Rather than cluttering up the input queries with this formatting, it seemed better to have the framework do the formatting. To this end the input queries instead have a select list with individual expressions and (mandatory) aliases, which can be simple or can be in double-quotes. The aliases form the header line of the csv file. To facilitate formatting the main query select list has to be of the form:

SELECT 
/* SEL */
        expr_1          alias_1,
        expr_2          alias_2
/* SEL */

Each expression must be on a separate line, and the list must be delimited by comment lines /* SEL */ as shown. The query formatting is performed in a procedure Get_Queries using some fairly complex regular expression processing.

Statistic Output Formatting

The various kinds of statistic (basic timing, execution plan aggregates, v$ statistics) are generally output in matrix format, both WxD and DxW. First the base numbers are printed for the whole grid for each query; then the last row for each query is printed, the deep or wide ‘slice’; then the same two sets of output are printed for the ratios of each number compared to the smallest number at the same data point across all queries.

Execution Plan Aggregation

After executing a query the execution plan statistics are copied from the system view v$sql_plan_statistics_all into bench_run_v$sql_plan_stats_all, and the formatted plan is written to a nested varray in bench_run_statistics using DBMS_XPlan.Display_Cursor.

At the end useful statistics in the plans are printed in aggregate by query execution, including maximum values of memory used and disk reads and writes, etc.

Estimated vs Actual Cardinalities

Oracle’s Cost Based Optimizer (CBO) uses estimated cardinalities at each step to evaluate candidate execution plans, and using the hint GATHER_PLAN_STATISTICS causes the actual cardinalities to be collected. Differences between estimated and actuals are generally recognised as being an important factor in whether or not a ‘good’ plan is chosen, so the maximum difference is included in the aggregates printed.

V$ Statistics

The statistics in the system views v$mystat, v$latch, v$sess_time_model are written to bench_run_v$stats before query execution (value_before, wait_before) and after execution (value_after, wait_after).

At the end a selection of the (after – before) differences of these statistics is written to log and csv file in the same format as the other statistics, based on the variance across the queries at the highest data point. A simple heuristic is included in the reporting query to restrict the statistics written to those deemed of most interest in comparing the queries, but all of the statistics remain available in bench_run_v$stats for ad hoc querying if desired.

See Also…






Design Patterns for Database API Testing 4: REF Cursor Getter

Last October I gave a presentation on database unit testing with utPLSQL, Oracle Unit Testing with utPLSQL. I mentioned design patterns as a way of reducing the effort of building unit tests and outlined some strategies for coding them effectively.

In the current set of articles, I develop the ideas further, starting from the idea that all database APIs can be considered in terms of the axes:

  • direction (i.e. getter or setter, noting that setters can also ‘get’)
  • mode (i.e. real time or batch)

For each cell in the implied matrix, I construct an example API (or view) with specified requirements against Oracle’s HR demo schema, and use this example to construct a testing program with appropriate scenarios as a design pattern. Concepts and common patterns and anti-patterns in automated API testing are discussed throughout, and these are largely independent of testing framework used. However, the examples use my own lightweight independent framework that is designed to help avoid many API testing anti-patterns. The code is available on GitHub here, BrenPatF/trapit_oracle_tester, and includes both framework and design pattern examples.

Behind the four examples, there is an underlying design pattern that involves wrapping the API call in a ‘pure’ procedure, called once per scenario, with the output ‘actuals’ array including everything affected by the API, whether as output parameters, or on database tables, etc. The inputs are also extended from the API parameters to include any other effective inputs. Assertion takes place after all scenarios and is against the extended outputs, with extended inputs also listed. This concept of the ‘pure’ function, central to Functional Programming, has important advantages in automated testing. I explained the concepts involved in a presentation at the Oracle User Group Ireland Conference in March 2018:

The Database API Viewed As A Mathematical Function: Insights into Testing

The previous example is here: Design Patterns for Database API Testing 3: Batch Loading of Flat Files. This fourth example covers getting data in real time via a REF cursor, such as would often be used as the source for a web service.

Requirement Summary

Modern Oracle SQL is very powerful and can apply complex logic within a single statement, reducing the need for more complex procedural code. In order to show how to API test SQL that might be used in a batch getter module, we previously devised a test view, HR_Test_V, having a range of features that we might want to test in general. We can use similar SQL to demonstrate API testing of a real time getter procedure that might be used by a web service, where reference cursors are often used as output parameters. The following list of features to test is taken, in slightly modified form, from Design Patterns for Database API Testing 2: Views 1 – Design

  • Inner joins suppress driving records where there is no joining record
  • Outer joins return driving records where there is no joining record
  • Analytic functions that partition by some key, and return aggregates on the returned record set
  • Functions based on aggregates over records that include those not in the returned record set
  • Constraints based on aggregates over records that include those not in the returned record set
  • Constraints on column values

The SQL functionality can be described in words as:

  • Selected values
    • Employee name, department name, and salary
    • Manager’s name
    • Ratio of employee’s salary to the department average (returned employees only)
    • Ratio of employee’s salary to the average salary of all employees
  • Constraints
    • Exclude employees in job ‘AD_ASST’
    • Return employees for a department passed as a bind parameter
    • Do not return any records if the total salary of all employees is below 1600
  • Outer join
    • Include employees both with and without a manager

The REF cursor SQL is:

  WITH all_emps AS (
        SELECT Avg (salary) avg_sal, SUM (salary) sal_tot_g
          FROM employees e
)
SELECT e.last_name, d.department_name, m.last_name manager, e.salary,
       Round (e.salary / Avg (e.salary) OVER (PARTITION BY e.department_id), 2) sal_rat,
       Round (e.salary / a.avg_sal, 2) sal_rat_g
  FROM all_emps a
 CROSS JOIN employees e
  JOIN departments d
    ON d.department_id = e.department_id
  LEFT JOIN employees m
    ON m.employee_id = e.manager_id
 WHERE e.job_id != 'AD_ASST'
   AND a.sal_tot_g >= 1600
   AND d.department_id = :1

Notes on API Testing REF Cursor Procedures

  • A new utility function has been added to Utils_TT, Cursor_to_Array, that converts an open reference cursor to a delimited list of strings (created from an initial stand-alone procedure: A Utility for Reading REF Cursors into a List of Delimited Strings)
  • Using this utility, very little code needs to be written once the test data has been set up: One call to return the reference cursor, and a second to return the actual values in a list, to be passed at the end in a single call to the library results checker

ERD

unit-testing-three-erd_rc

Design Pattern Groups

The API testing framework is centred around the concept of input and output groups, representing the data sets that respectively form the inputs to, and outputs from, the program. The records in each group are printed by the framework with column headers for each scenario. These groups are identified by the developer, and in this case they are as noted below.

Input Groups

  • Employees Table
  • Department Parameter

Output Groups

  • Select results
  • Timing of average call

Test Scenarios

The scenario descriptions start with a data set code, followed by a verbal description.

  1. DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep (10) – pass dep 10
  2. DS-2, testing same as 1 but with extra emp in another dep (20) – pass dep 10
  3. DS-2, as second scenario, but – pass dep 20
  4. DS-2, as second scenario, but – pass null dep
  5. DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) – pass dep 10

Package Structure Diagram

Call Structure Table – TT_Emp_WS.tt_AIP_Get_Dept_Emps

TT_Emp_WS.tt_AIP_Get_Dept_Emps – Emp REF Cursor Test Procedure

Declare section

PROCEDURE tt_AIP_Get_Dept_Emps IS

  c_proc_name           CONSTANT VARCHAR2(61) := 'TT_Emp_WS.tt_AIP_Get_Dept_Emps';
  c_dep_id_1            CONSTANT PLS_INTEGER := 10;
  c_dep_id_2            CONSTANT PLS_INTEGER := 20;
  c_dep_nm_1            CONSTANT VARCHAR2(100) := 'Administration';
  c_dep_nm_2            CONSTANT VARCHAR2(100) := 'Marketing';
  c_job_bad             CONSTANT VARCHAR2(100) := 'AD_ASST';
  c_job_good            CONSTANT VARCHAR2(100) := 'IT_PROG';
  c_base_sal            CONSTANT PLS_INTEGER := 1000;
  c_out_group_lis       CONSTANT L1_chr_arr := L1_chr_arr ('Select results');

  c_ln_pre              CONSTANT VARCHAR2(10) := DML_API_TT_HR.c_ln_pre;

  c_dep_lis             CONSTANT L1_chr_arr := L1_chr_arr (c_dep_id_1, c_dep_id_1, c_dep_id_2, NULL, c_dep_id_1);

  c_dataset_3lis        CONSTANT L3_chr_arr := L3_chr_arr (
                             L2_chr_arr (L1_chr_arr ('4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job'),
--                                         dep           job          salary
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '1000'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '2000'),
                               L1_chr_arr (NULL,         c_job_good,  '3000'),
                               L1_chr_arr (c_dep_id_1,   c_job_bad,   '4000')
                                             ),
                             L2_chr_arr (L1_chr_arr ('As dataset 1 but with extra emp-5, in second dep (20)'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '1000'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '2000'),
                               L1_chr_arr (NULL,         c_job_good,  '3000'),
                               L1_chr_arr (c_dep_id_1,   c_job_bad,   '4000'),
                               L1_chr_arr (c_dep_id_2,   c_job_good,  '5000')
                                             ),
                             L2_chr_arr (L1_chr_arr ('As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '100'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '200'),
                               L1_chr_arr (NULL,         c_job_good,  '300'),
                               L1_chr_arr (c_dep_id_1,   c_job_bad,   '400'),
                               L1_chr_arr (c_dep_id_2,   c_job_good,  '500')
                                             )
                        );

  c_exp_2lis            CONSTANT L2_chr_arr := L2_chr_arr (
                                               L1_chr_arr (
                                       Utils.List_Delim (c_ln_pre || '1',   c_dep_nm_1, NULL,            '1000', '.67',   '.4'),
                                       Utils.List_Delim (c_ln_pre || '2',   c_dep_nm_1, c_ln_pre || '1', '2000',  '1.33', '.8')
                                               ),
                                               L1_chr_arr (
                                       Utils.List_Delim (c_ln_pre || '1',   c_dep_nm_1, NULL,            '1000', '.67',  '.33'),
                                       Utils.List_Delim (c_ln_pre || '2',   c_dep_nm_1, c_ln_pre || '1', '2000',  '1.33', '.67')
                                               ),
                                               L1_chr_arr (
                                       Utils.List_Delim (c_ln_pre || '5',   c_dep_nm_2, c_ln_pre || '1', '5000',  '1',    '1.67')
                                               ),
                                               Utils_TT.c_empty_list,
                                               Utils_TT.c_empty_list
                        );

  c_scenario_ds_lis     CONSTANT L1_num_arr := L1_num_arr (1, 2, 2, 2, 3);
  c_scenario_lis        CONSTANT L1_chr_arr := L1_chr_arr (
                               'DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep (10) - pass dep 10',
                               'DS-2, testing same as 1 but with extra emp in another dep (20) - pass dep 10',
                               'DS-2, as second scenario, but - pass dep 20',
                               'DS-2, as second scenario, but - pass null dep',
                               'DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) - pass dep 10');

  c_inp_group_lis       CONSTANT L1_chr_arr := L1_chr_arr ('Employee', 'Department Parameter');
  c_inp_field_2lis      CONSTANT L2_chr_arr := L2_chr_arr (
                                                        L1_chr_arr (
                                                                '*Employee Id',
                                                                'Last Name',
                                                                'Email',
                                                                'Hire Date',
                                                                'Job',
                                                                '*Salary',
                                                                '*Manager Id',
                                                                '*Department Id',
                                                                'Updated'),
                                                        L1_chr_arr (
                                                                '*Department Id')
  );
  c_out_field_2lis      CONSTANT L2_chr_arr :=  L2_chr_arr ( L1_chr_arr (
                                'Name',
                                'Department',
                                'Manager',
                                '*Salary',
                                '*Salary Ratio (dep)',
                                '*Salary Ratio (overall)'));

  l_act_2lis                      L2_chr_arr := L2_chr_arr();
  c_ms_limit            CONSTANT PLS_INTEGER := 1;
  l_timer_set                    PLS_INTEGER;
  l_inp_3lis                     L3_chr_arr := L3_chr_arr();
  l_emp_csr                      SYS_REFCURSOR;

Notes on declare section

  • Data sets, scenarios, expected values etc. are stored in generic arrays, where:
    • L1_chr_arr is type of array of VARCHAR2(4000), same as standard type SYS.ODCIVarchar2List
    • L2_chr_arr is a type of array of L1_chr_arr
    • L3_chr_arr is a type of array of L2_chr_arr

Setup section

  PROCEDURE Setup_Array IS
  BEGIN

    l_act_2lis.EXTEND (c_exp_2lis.COUNT);
    l_inp_3lis.EXTEND (c_exp_2lis.COUNT);

    FOR i IN 1..c_exp_2lis.COUNT LOOP

      l_inp_3lis (i) := L2_chr_arr();
      l_inp_3lis (i).EXTEND(2);
      l_inp_3lis (i)(2) := L1_chr_arr (c_dep_lis(i));

    END LOOP;

  END Setup_Array;

  /***************************************************************************************************

  Setup_DB: Create test records for a given scenario for testing AIP_Get_Dept_Emps

  ***************************************************************************************************/
  PROCEDURE Setup_DB (p_call_ind           PLS_INTEGER,   -- index of input dataset
                      x_inp_lis        OUT L1_chr_arr) IS -- input list, employees

    l_emp_id            PLS_INTEGER;
    l_mgr_id            PLS_INTEGER;
    l_len_lis           L1_num_arr := L1_num_arr (1, -11, -13, -10, 10, -10);

  BEGIN

    Utils.Heading ('Employees created in setup: DS-' || p_call_ind || ' - ' || c_dataset_3lis (p_call_ind)(1)(1));
    Utils.Col_Headers (L1_chr_arr ('#', 'Employee id', 'Department id', 'Manager', 'Job id', 'Salary'), l_len_lis);
    x_inp_lis := L1_chr_arr();
    x_inp_lis.EXTEND (c_dataset_3lis (p_call_ind).COUNT - 1);
    FOR i IN 2..c_dataset_3lis (p_call_ind).COUNT LOOP

      l_emp_id := DML_API_TT_HR.Ins_Emp (
                            p_emp_ind  => i - 1,
                            p_dep_id   => c_dataset_3lis (p_call_ind)(i)(1),
                            p_mgr_id   => l_mgr_id,
                            p_job_id   => c_dataset_3lis (p_call_ind)(i)(2),
                            p_salary   => c_dataset_3lis (p_call_ind)(i)(3),
                            x_rec      => x_inp_lis(i - 1));
      Utils.Pr_List_As_Line (L1_chr_arr ((i-1), l_emp_id, Nvl (c_dataset_3lis (p_call_ind)(i)(1), ' '), Nvl (To_Char(l_mgr_id), ' '), c_dataset_3lis (p_call_ind)(i)(2), c_dataset_3lis (p_call_ind)(i)(3)), l_len_lis);
      IF i = 2 THEN
        l_mgr_id := l_emp_id;
      END IF;

    END LOOP;

  END Setup_DB;

Notes on setup section

  • c_dataset_3lis contains the data for all data sets indexed by (data set, record, field)
  • Setup_Array is called once to do some setup on the arrays
  • Setup_DB is called for a single data set at a time in each scenario
  • Description of the data set is contained in the array and printed out
  • Data set is printed out in tabular format. In the current version of the utility code, this is not strictly necessary, because all the input data is printed out before the outputs

'Pure' API wrapper procedure

  PROCEDURE Purely_Wrap_API (p_scenario_ds      PLS_INTEGER,   -- index of input dataset
                             p_dep_id           PLS_INTEGER,   -- input department id
                             x_inp_lis      OUT L1_chr_arr,    -- generic inputs list (for scenario)
                             x_act_lis      OUT L1_chr_arr) IS -- generic actual values list (for scenario)
  BEGIN

    Setup_DB (p_scenario_ds, x_inp_lis);
    Timer_Set.Increment_Time (l_timer_set, Utils_TT.c_setup_timer);

    Emp_WS.AIP_Get_Dept_Emps (p_dep_id  => p_dep_id,
                              x_emp_csr => l_emp_csr);
    x_act_lis := Utils_TT.List_or_Empty (Utils_TT.Cursor_to_Array (x_csr => l_emp_csr));
    Timer_Set.Increment_Time (l_timer_set, Utils_TT.c_call_timer);
    ROLLBACK;

  END Purely_Wrap_API;

Notes on 'Pure' API wrapper procedure

  • Setup_DB is called to create the data set for the scenario
  • Emp_WS.AIP_Get_Dept_Emps returns the ref cursor
  • Utils_TT.Cursor_to_Array gets the ref cursor result set into an array (which may be empty)

Main section

BEGIN
--
-- Every testing main section should be similar to this, with array setup, then loop over scenarios
-- making a 'pure'(-ish) call to specific, local Purely_Wrap_API, with single assertion call outside
-- the loop
--
  l_timer_set := Utils_TT.Init (c_proc_name);
  Setup_Array;

  FOR i IN 1..c_exp_2lis.COUNT LOOP

    Purely_Wrap_API (c_scenario_ds_lis(i), c_where_lis(i), l_inp_3lis(i)(1), l_act_2lis(i));

  END LOOP;

  Utils_TT.Is_Deeply (c_proc_name, c_scenario_lis, l_inp_3lis, l_act_2lis, c_exp_2lis, l_timer_set, c_ms_limit,
                      c_inp_group_lis, c_inp_field_2lis, c_out_group_lis, c_out_field_2lis);

EXCEPTION

  WHEN OTHERS THEN
    Utils.Write_Other_Error;
    RAISE;

END tt_AIP_Get_Dept_Emps;

Notes on main section

  • It's quite short isn't it 🙂
  • Setup is called to do array setup
  • Main section loops over the scenarios calling Purely_Wrap_API
  • Is_Deeply is called to do all the assertions within nested loops, then print the results

Test Output

TRAPIT TEST: TT_Emp_WS.tt_AIP_Get_Dept_Emps
===========================================

Employees created in setup: DS-1 - 4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job
==========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1658             10              IT_PROG           1000
2         1659             10        1658  IT_PROG           2000
3         1660                       1658  IT_PROG           3000
4         1661             10        1658  AD_ASST           4000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1662             10              IT_PROG           1000
2         1663             10        1662  IT_PROG           2000
3         1664                       1662  IT_PROG           3000
4         1665             10        1662  AD_ASST           4000
5         1666             20        1662  IT_PROG           5000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1667             10              IT_PROG           1000
2         1668             10        1667  IT_PROG           2000
3         1669                       1667  IT_PROG           3000
4         1670             10        1667  AD_ASST           4000
5         1671             20        1667  IT_PROG           5000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1672             10              IT_PROG           1000
2         1673             10        1672  IT_PROG           2000
3         1674                       1672  IT_PROG           3000
4         1675             10        1672  AD_ASST           4000
5         1676             20        1672  IT_PROG           5000

Employees created in setup: DS-3 - As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600
================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1677             10              IT_PROG            100
2         1678             10        1677  IT_PROG            200
3         1679                       1677  IT_PROG            300
4         1680             10        1677  AD_ASST            400
5         1681             20        1677  IT_PROG            500

SCENARIO 1: DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep (10) - pass dep 10 {
==================================================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1658  LN_1       EM_1   01-OCT-2016  IT_PROG    1000                         10  01-OCT-2016
                   1659  LN_2       EM_2   01-OCT-2016  IT_PROG    2000        1658             10  01-OCT-2016
                   1660  LN_3       EM_3   01-OCT-2016  IT_PROG    3000        1658                 01-OCT-2016
                   1661  LN_4       EM_4   01-OCT-2016  AD_ASST    4000        1658             10  01-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       10

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                      .4
                LN_2  Administration  LN_1       2000                1.33                      .8

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 2: DS-2, testing same as 1 but with extra emp in another dep (20) - pass dep 10 {
==========================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1662  LN_1       EM_1   01-OCT-2016  IT_PROG    1000                         10  01-OCT-2016
                   1663  LN_2       EM_2   01-OCT-2016  IT_PROG    2000        1662             10  01-OCT-2016
                   1664  LN_3       EM_3   01-OCT-2016  IT_PROG    3000        1662                 01-OCT-2016
                   1665  LN_4       EM_4   01-OCT-2016  AD_ASST    4000        1662             10  01-OCT-2016
                   1666  LN_5       EM_5   01-OCT-2016  IT_PROG    5000        1662             20  01-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       10

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                     .33
                LN_2  Administration  LN_1       2000                1.33                     .67

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 3: DS-2, as second scenario, but - pass dep 20 {
=========================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1667  LN_1       EM_1   01-OCT-2016  IT_PROG    1000                         10  01-OCT-2016
                   1668  LN_2       EM_2   01-OCT-2016  IT_PROG    2000        1667             10  01-OCT-2016
                   1669  LN_3       EM_3   01-OCT-2016  IT_PROG    3000        1667                 01-OCT-2016
                   1670  LN_4       EM_4   01-OCT-2016  AD_ASST    4000        1667             10  01-OCT-2016
                   1671  LN_5       EM_5   01-OCT-2016  IT_PROG    5000        1667             20  01-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       20

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 1, Expected = 1 {
        ================================================

            F?  Name  Department  Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  ----------  -------  ------  ------------------  ----------------------
                LN_5  Marketing   LN_1       5000                   1                    1.67

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 1: SUCCESS
=========================

SCENARIO 4: DS-2, as second scenario, but - pass null dep {
===========================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1672  LN_1       EM_1   01-OCT-2016  IT_PROG    1000                         10  01-OCT-2016
                   1673  LN_2       EM_2   01-OCT-2016  IT_PROG    2000        1672             10  01-OCT-2016
                   1674  LN_3       EM_3   01-OCT-2016  IT_PROG    3000        1672                 01-OCT-2016
                   1675  LN_4       EM_4   01-OCT-2016  AD_ASST    4000        1672             10  01-OCT-2016
                   1676  LN_5       EM_5   01-OCT-2016  IT_PROG    5000        1672             20  01-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 0, Expected = 0: SUCCESS
        =======================================================

} 0 failed, of 1: SUCCESS
=========================

SCENARIO 5: DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) - pass dep 10 {
==============================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1677  LN_1       EM_1   01-OCT-2016  IT_PROG     100                         10  01-OCT-2016
                   1678  LN_2       EM_2   01-OCT-2016  IT_PROG     200        1677             10  01-OCT-2016
                   1679  LN_3       EM_3   01-OCT-2016  IT_PROG     300        1677                 01-OCT-2016
                   1680  LN_4       EM_4   01-OCT-2016  AD_ASST     400        1677             10  01-OCT-2016
                   1681  LN_5       EM_5   01-OCT-2016  IT_PROG     500        1677             20  01-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       10

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 0, Expected = 0: SUCCESS
        =======================================================

} 0 failed, of 1: SUCCESS
=========================

TIMING: Actual = 7, Expected <= 1: FAILURE
==========================================

SUMMARY for TT_Emp_WS.tt_AIP_Get_Dept_Emps
==========================================

Scenario                                                                                              # Failed  # Tests  Status
----------------------------------------------------------------------------------------------------  --------  -------  -------
DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep (10) - pass dep 10         0        2  SUCCESS
DS-2, testing same as 1 but with extra emp in another dep (20) - pass dep 10                                 0        2  SUCCESS
DS-2, as second scenario, but - pass dep 20                                                                  0        1  SUCCESS
DS-2, as second scenario, but - pass null dep                                                                0        1  SUCCESS
DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) - pass dep 10                             0        1  SUCCESS
Timing                                                                                                       1        1  FAILURE
----------------------------------------------------------------------------------------------------  --------  -------  -------
Total                                                                                                        1        8  FAILURE
----------------------------------------------------------------------------------------------------  --------  -------  -------

Timer Set: TT_Emp_WS.tt_AIP_Get_Dept_Emps, Constructed at 01 Oct 2016 09:14:12, written at 09:14:13
===================================================================================================
[Timer timed: Elapsed (per call): 0.03 (0.000034), CPU (per call): 0.03 (0.000030), calls: 1000, '***' denotes corrected line below]

Timer       Elapsed         CPU         Calls       Ela/Call       CPU/Call
-------  ----------  ----------  ------------  -------------  -------------
Setup          0.08        0.05             5        0.01660        0.01000
Caller         0.03        0.01             5        0.00680        0.00200
(Other)        0.10        0.10             1        0.09500        0.10000
-------  ----------  ----------  ------------  -------------  -------------
Total          0.21        0.16            11        0.01927        0.01455
-------  ----------  ----------  ------------  -------------  -------------






 

Design Patterns for Database API Testing 3: Batch Loading of Flat Files

Last October I gave a presentation on database unit testing with utPLSQL, Oracle Unit Testing with utPLSQL. I mentioned design patterns as a way of reducing the effort of building unit tests and outlined some strategies for coding them effectively.

In the current set of articles, I develop the ideas further, starting from the idea that all database APIs can be considered in terms of the axes:

  • direction (i.e. getter or setter, noting that setters can also ‘get’)
  • mode (i.e. real time or batch)

For each cell in the implied matrix, I construct an example API (or view) with specified requirements against Oracle’s HR demo schema, and use this example to construct a testing program with appropriate scenarios as a design pattern. Concepts and common patterns and anti-patterns in automated API testing are discussed throughout, and these are largely independent of testing framework used. However, the examples use my own lightweight independent framework that is designed to help avoid many API testing anti-patterns. The code is available on GitHub here, BrenPatF/trapit_oracle_tester, and includes both framework and design pattern examples.

Behind the four examples, there is an underlying design pattern that involves wrapping the API call in a ‘pure’ procedure, called once per scenario, with the output ‘actuals’ array including everything affected by the API, whether as output parameters, or on database tables, etc. The inputs are also extended from the API parameters to include any other effective inputs. Assertion takes place after all scenarios and is against the extended outputs, with extended inputs also listed. This concept of the ‘pure’ function, central to Functional Programming, has important advantages in automated testing. I explained the concepts involved in a presentation at the Oracle User Group Ireland Conference in March 2018:

The Database API Viewed As A Mathematical Function: Insights into Testing

The previous example, on views, has its second part here: Design Patterns for Database API Testing 2: Views 2 – Code. This third example covers batch loading of data into a database table from file by means of external tables. There are many possible variants of this kind of interface, and I tried to combine the features from past projects that seemed to work best. In particular:

  • The load procedure is assumed to be executed from an operating system script (normally Unix) that manages the input files, copying from source files into a single file for reading by the external table, looping over multiple files where necessary, and archiving processed files
    • It is possible to dynamically map the external table to multiple files, but that approach involves executing DDL and seems to be generally more complex
    • The idea is to do the different types of processing at the level most appropriate, so avoiding excessive file processing within the database
  • SQL operations are performed at set level, rather than within a loop, and the set concerned comprises the entire contents of the external table
    • This assumes that the Oracle internal working space requirements (such as size of TEMP tablespace) are not exceeded
    • The alternative approach of fetching batches of records into arrays for loading tends to be more complex, and less efficient
    • It may be preferable to restrict the size of the input files where necessary instead, either at the source end, or from the controlling operating system script
    • Where loading occurs from staging tables (which is not in scope here) the use of the DBMS_Parallel_Execute package to control transaction size looks an attractive modern approach (from v11, DBMS_PARALLEL_EXECUTE)
  • Metadata tables are used for specifying job control parameters, and for recording run statistics
    • Run statistics include job status and numbers of records succeeding and failing at both external table and database level
    • A percentage threshold is included in the job control table, that causes the job to fail if a higher percentage of records fails
  • Oracle’s DML LOG ERRORS clause is used to divert failing records into an errors table, while processing the other records normally
    • While simpler than other approaches to error handling, prior to v12.1 this clause had a significant performance overhead, but from v12.1 this is no longer the case (see DML Error Logging in Oracle 10g Database Release 2 – it has a table of comparative timings up to v12.1 at the end)
    • The standard Oracle err$ table structure is used, with the addition of two columns, one to identify the associated job run, and the other the utid column used by the framework to identify test data
    • In a replication environment a unique identifier would be needed, but is not included here
    • The table is also used to capture records that fail custom validation (here when a record passed as an update does not match an existing record)

Requirement Summary

  • The procedure reads employee records from a single flat file
  • Employee id is an optional field, and if passed the record is treated as an update, otherwise it is an insert
  • A job statistics table is populated with a record on each run, and stores numbers of successful and failed records etc.
  • Records that fail at the database level are inserted into an errors table, extended from Oracle’s err$ table format
  • The external table has all fields defined as 4000 bytes character fields to minimise records failing to be read from the external table
  • An audit date column is included, and is set to the current date if any change is made in a record; if an unchanged record is detected no update should be made
  • The load program takes two parameters that would be passed in by an operating system calling script:
    • An identifier for the file, that may be a timestamped original file name (but the external table file has a fixed name)
    • A line count for the file
  • The passed file identifier is saved in the job statistics table, and a repeat identifier will only be accepted if previous records all have failed status
  • A job control table stores an error percentage threshold, above which the run is considered unsuccessfull, and an exception is raised

Notes on Testing Flat File Loads

  • It is considered best practice to keep testing code as self-contained as possible, and in order to avoid dependence on external data files, Oracle’s UTL_File package is used to delete and create the test files from arrays within the test code
    • The Utils package contains two wrapper procedures to facilitate this: Delete_File and Write_File

Extended ERD

unit-testing-two-ff-erd-h

Design Pattern Groups

The testing framework is centred around the concept of input and output groups, representing the data sets that respectively form the inputs to, and outputs from, the program. The records in each group are printed by the framework with column headers for each scenario. These groups are identified by the developer, and in this case they are as noted below.

Input Groups

  • Parameters
  • Input File
  • Batch Job Table
  • Job Statistics Table
  • Employees Table

Output Groups

  • Employees Table
  • Errors Table
  • Job Statistics Table
  • Exception
  • Timing of average call

Test Scenarios

Key
The scenario descriptions start with two sets of counts, followed by a verbal description. The first set is on the records coming in from file:

  • NV – new valid records
  • OV – old valid records
  • OU – old unchanged records
  • NI – new invalid records
  • OI – old valid records
  • EI – external table invalid records

The second set is on the existing records in job statistics and employees (other than are in the input file; these are counted as old records in the first set of counts):

  • J – job statistics records
  • E – employees records

Scenario List

  1. NV/OV/OU/NI/OI/EI: 1/0/0/0/0/0. Existing J/E: 0/0. [1 valid new record from scratch]
  2. NV/OV/OU/NI/OI/EI: 1/1/1/0/0/0. Existing J/E: 1/0. [3 valid records of each kind]
  3. NV/OV/OU/NI/OI/EI: 0/0/0/0/1/0. Existing J/E: 1/1. Uid not found [1 invalid old – exception]
  4. NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Email too long [1 invalid new – exception]
  5. NV/OV/OU/NI/OI/EI: 1/0/0/0/1/0. Existing J/E: 1/1. Name too long [1 valid new, 1 invalid old – no exception]
  6. NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Invalid job [1 invalid new – exception]
  7. NV/OV/OU/NI/OI/EI: 0/1/0/1/1/0. Existing J/E: 1/2. 2 invalid jobs [1 valid old, 2 invalid: old and new – no exception]
  8. NV/OV/OU/NI/OI/EI: 0/1/0/0/0/1. Existing J/E: 1/2. Name 4001ch [1 valid old, 1 invalid new for external table – no exception; also file had previously failed]
  9. NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. [File already processed – exception]

Package Structure Diagram

Call Structure Table – TT_Emp_Batch.tt_AIP_Load_Emps

Test Output

TRAPIT TEST for TT_Emp_Batch.tt_AIP_Load_Emps
=============================================
employees.dat was not present to delete!

SCENARIO 1: NV/OV/OU/NI/OI/EI: 1/0/0/0/0/0. Existing J/E: 0/0. [1 valid new record from scratch] {
==================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------
            ,LN 1,EM 1,01-JAN-2010,IT_PROG,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table (No records)
        ===================================

        GROUP Employees Table (No records)
        ==================================

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1627  LN 1  EM 1   01-JAN-2010  IT_PROG  10000   11-SEP-2016

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 1, Expected = 1 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                               3  LOAD_EMPS     employees_20160801.dat               1                  0                  0  11-SEP-2016  11-SEP-2016  S

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 4: SUCCESS
=========================

SCENARIO 2: NV/OV/OU/NI/OI/EI: 1/1/1/0/0/0. Existing J/E: 1/0. [3 valid records of each kind] {
===============================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      3

        }
        =

        GROUP File {
        ============

            Line
            -----------------------------------------
            ,LN 1,EM 1,01-JAN-2010,IT_PROG,10000
            1629,LN 2,EM 2,01-JAN-2010,IT_PROG,20000
            1630,LN 3U,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                           5  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1629  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010
                   1630  LN 3  EM 3   01-JAN-2010  IT_PROG   30000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 3, Expected = 3 {
        ==========================================

            F?  Employee Id  Name   Email  Hired        Job      Salary  Updated
            --  -----------  -----  -----  -----------  -------  ------  -----------
                       1629  LN 2   EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1630  LN 3U  EM 3   01-JAN-2010  IT_PROG  30000   11-SEP-2016
                       1631  LN 1   EM 1   01-JAN-2010  IT_PROG  10000   11-SEP-2016

        } 0 failed, of 3: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                               5  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                               6  LOAD_EMPS     employees_20160801.dat               2                  0                  0  11-SEP-2016  11-SEP-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 7: SUCCESS
=========================

SCENARIO 3: NV/OV/OU/NI/OI/EI: 0/0/0/0/1/0. Existing J/E: 1/1. Uid not found [1 invalid old - exception] {
==========================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            --------------------------------------
            99,LN 1,EM 1,01-JAN-2010,IT_PROG,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                           8  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1633  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1633  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$       ORA_ERR_OPTYP$  Employee Id  Name  Email  Hired        Job      Salary
            --  ----------------  ------------  ------------------  --------------  -----------  ----  -----  -----------  -------  ------
                               9                Employee not found  PK                       99  LN 1  EM 1   01-JAN-2010  IT_PROG  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                               8  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                               9  LOAD_EMPS     employees_20160801.dat               0                  0                  1  11-SEP-2016  11-SEP-2016  F

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  ------------------------------------------------------
                ORA-20001: Batch failed with too many invalid records!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 5: SUCCESS
=========================

SCENARIO 4: NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Email too long [1 invalid new - exception] {
===========================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------------------------------------
            ,LN 1,EM 1123456789012345678901234567890,01-JAN-2010,IT_PROG,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          11  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1635  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1635  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                             ORA_ERR_OPTYP$  Employee Id  Name  Email                               Hired        Job      Salary
            --  ----------------  ------------  ----------------------------------------------------------------------------------------  --------------  -----------  ----  ----------------------------------  -----------  -------  ------
                              12                ORA-12899: value too large for column "HR"."EMPLOYEES"."EMAIL" (actual: 34, maximum: 25)  I                      1636  LN 1  EM 1123456789012345678901234567890  01-JAN-2010  IT_PROG  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              11  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              12  LOAD_EMPS     employees_20160801.dat               0                  0                  1  11-SEP-2016  11-SEP-2016  F

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  ------------------------------------------------------
                ORA-20001: Batch failed with too many invalid records!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 5: SUCCESS
=========================

SCENARIO 5: NV/OV/OU/NI/OI/EI: 1/0/0/0/1/0. Existing J/E: 1/1. Name too long [1 valid new, 1 invalid old - no exception] {
==========================================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      2

        }
        =

        GROUP File {
        ============

            Line
            ----------------------------------------------------------------------
            1638,LN 1123456789012345678901234567890,EM 1,01-JAN-2010,IT_PROG,10000
            ,LN 3,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          14  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1638  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 2 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1638  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1639  LN 3  EM 3   01-JAN-2010  IT_PROG  30000   11-SEP-2016

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                                 ORA_ERR_OPTYP$  Employee Id  Name                                Email  Hired        Job      Salary
            --  ----------------  ------------  --------------------------------------------------------------------------------------------  --------------  -----------  ----------------------------------  -----  -----------  -------  ------
                              15                ORA-12899: value too large for column "HR"."EMPLOYEES"."LAST_NAME" (actual: 34, maximum: 25)  U                      1638  LN 1123456789012345678901234567890  EM 1   01-JAN-2010  IT_PROG  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              14  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              15  LOAD_EMPS     employees_20160801.dat               1                  0                  1  11-SEP-2016  11-SEP-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 6: SUCCESS
=========================

SCENARIO 6: NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Invalid job [1 invalid new - exception] {
========================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------
            ,LN 1,EM 1,01-JAN-2010,NON_JOB,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          17  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1641  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1641  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                    ORA_ERR_OPTYP$  Employee Id  Name  Email  Hired        Job      Salary
            --  ----------------  ------------  -------------------------------------------------------------------------------  --------------  -----------  ----  -----  -----------  -------  ------
                              18                ORA-02291: integrity constraint (HR.EMP_JOB_FK) violated - parent key not found  I                      1642  LN 1  EM 1   01-JAN-2010  NON_JOB  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              17  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              18  LOAD_EMPS     employees_20160801.dat               0                  0                  1  11-SEP-2016  11-SEP-2016  F

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  ------------------------------------------------------
                ORA-20001: Batch failed with too many invalid records!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 5: SUCCESS
=========================

SCENARIO 7: NV/OV/OU/NI/OI/EI: 0/1/0/1/1/0. Existing J/E: 1/2. 2 invalid jobs [1 valid old, 2 invalid: old and new - no exception] {
====================================================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      3

        }
        =

        GROUP File {
        ============

            Line
            -----------------------------------------
            ,LN 1,EM 1,01-JAN-2010,NON_JOB,10000
            1644,LN 2,EM 2,01-JAN-2010,NON_JOB,20000
            1645,LN 3U,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          20  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1644  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010
                   1645  LN 3  EM 3   01-JAN-2010  IT_PROG   30000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 2 {
        ==========================================

            F?  Employee Id  Name   Email  Hired        Job      Salary  Updated
            --  -----------  -----  -----  -----------  -------  ------  -----------
                       1644  LN 2   EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1645  LN 3U  EM 3   01-JAN-2010  IT_PROG  30000   11-SEP-2016

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Error: Actual = 2, Expected = 2 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                    ORA_ERR_OPTYP$  Employee Id  Name  Email  Hired        Job      Salary
            --  ----------------  ------------  -------------------------------------------------------------------------------  --------------  -----------  ----  -----  -----------  -------  ------
                              21                ORA-02291: integrity constraint (HR.EMP_JOB_FK) violated - parent key not found  U                      1644  LN 2  EM 2   01-JAN-2010  NON_JOB  20000
                              21                ORA-02291: integrity constraint (HR.EMP_JOB_FK) violated - parent key not found  I                      1646  LN 1  EM 1   01-JAN-2010  NON_JOB  10000

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              20  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              21  LOAD_EMPS     employees_20160801.dat               1                  0                  2  11-SEP-2016  11-SEP-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 7: SUCCESS
=========================

SCENARIO 8: NV/OV/OU/NI/OI/EI: 0/1/0/0/0/1. Existing J/E: 1/2. Name 4001ch [1 valid old, 1 invalid new for external table - no exception; also file had previously failed] {
============================================================================================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      2

        }
        =

        GROUP File {
        ============

            Line
            --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
            ,123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567
            1649,LN 3U,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          23  LOAD_EMPS     employees_20160801.dat               0                  0                  2  01-JAN-2010  01-JAN-2010  F

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1648  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010
                   1649  LN 3  EM 3   01-JAN-2010  IT_PROG   30000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 2 {
        ==========================================

            F?  Employee Id  Name   Email  Hired        Job      Salary  Updated
            --  -----------  -----  -----  -----------  -------  ------  -----------
                       1648  LN 2   EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1649  LN 3U  EM 3   01-JAN-2010  IT_PROG  30000   11-SEP-2016

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              23  LOAD_EMPS     employees_20160801.dat               0                  0                  2  01-JAN-2010  01-JAN-2010  F
                              24  LOAD_EMPS     employees_20160801.dat               1                  1                  0  11-SEP-2016  11-SEP-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 6: SUCCESS
=========================

SCENARIO 9: NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. [File already processed - exception] {
=====================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------
            ,LN 1,EM 1,01-JAN-2010,NON_JOB,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          26  LOAD_EMPS     employees_20160801.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1651  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1651  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 1, Expected = 1 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              26  LOAD_EMPS     employees_20160801.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  --------------------------------------------------------
                ORA-20002: File has already been processed successfully!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 4: SUCCESS
=========================

TIMING: Actual = 306, Expected <= 2: FAILURE
============================================

SUMMARY for TT_Emp_Batch.tt_AIP_Load_Emps
=========================================

Scenario                                                                                                                                                        # Failed  # Tests  Status
--------------------------------------------------------------------------------------------------------------------------------------------------------------  --------  -------  -------
NV/OV/OU/NI/OI/EI: 1/0/0/0/0/0. Existing J/E: 0/0. [1 valid new record from scratch]                                                                                   0        4  SUCCESS
NV/OV/OU/NI/OI/EI: 1/1/1/0/0/0. Existing J/E: 1/0. [3 valid records of each kind]                                                                                      0        7  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/0/1/0. Existing J/E: 1/1. Uid not found [1 invalid old - exception]                                                                           0        5  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Email too long [1 invalid new - exception]                                                                          0        5  SUCCESS
NV/OV/OU/NI/OI/EI: 1/0/0/0/1/0. Existing J/E: 1/1. Name too long [1 valid new, 1 invalid old - no exception]                                                           0        6  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Invalid job [1 invalid new - exception]                                                                             0        5  SUCCESS
NV/OV/OU/NI/OI/EI: 0/1/0/1/1/0. Existing J/E: 1/2. 2 invalid jobs [1 valid old, 2 invalid: old and new - no exception]                                                 0        7  SUCCESS
NV/OV/OU/NI/OI/EI: 0/1/0/0/0/1. Existing J/E: 1/2. Name 4001ch [1 valid old, 1 invalid new for external table - no exception; also file had previously failed]         0        6  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. [File already processed - exception]                                                                                0        4  SUCCESS
Timing                                                                                                                                                                 1        1  FAILURE
--------------------------------------------------------------------------------------------------------------------------------------------------------------  --------  -------  -------
Total                                                                                                                                                                  1       50  FAILURE
--------------------------------------------------------------------------------------------------------------------------------------------------------------  --------  -------  -------

Timer Set: TT_Emp_Batch.tt_AIP_Load_Emps, Constructed at 11 Sep 2016 16:10:46, written at 16:10:49
==================================================================================================
[Timer timed: Elapsed (per call): 0.04 (0.000036), CPU (per call): 0.03 (0.000030), calls: 1000, '***' denotes corrected line below]

Timer           Elapsed         CPU         Calls       Ela/Call       CPU/Call
-----------  ----------  ----------  ------------  -------------  -------------
Setup              1.10        0.30             9        0.12189        0.03333
Caller             1.53        1.16             5        0.30560        0.23200
Get_Tab_Lis        0.75        0.49             9        0.08378        0.05444
Get_Err_Lis        0.02        0.03             9        0.00244        0.00333
Get_Jbs_Lis        0.00        0.00             9        0.00011        0.00000
(Other)            0.34        0.35             1        0.33500        0.35000
-----------  ----------  ----------  ------------  -------------  -------------
Total              3.74        2.33            42        0.08898        0.05548
-----------  ----------  ----------  ------------  -------------  -------------

Suite Summary
=============

Package.Procedure                  Tests  Fails         ELA         CPU
---------------------------------  -----  -----  ----------  ----------
TT_Emp_WS.tt_AIP_Save_Emps            17      3        0.10        0.09
TT_Emp_WS.tt_AIP_Get_Dept_Emps         7      1        0.11        0.12
TT_View_Drivers.tt_HR_Test_View_V      9      0        0.12        0.12
TT_Emp_Batch.tt_AIP_Load_Emps         50      1        3.74        2.33
---------------------------------  -----  -----  ----------  ----------
Total                                 83      5        4.07        2.66
---------------------------------  -----  -----  ----------  ----------
Others error in (): ORA-20001: Suite BRENDAN returned error status: ORA-06512: at "DP_1.UTILS_TT", line 151
ORA-06512: at "DP_1.UTILS_TT", line 818
ORA-06512: at line 5

The fourth and final article in this series is here: Design Patterns for Database API Testing 4: REF Cursor Getter.






 

A Utility for Reading REF Cursors into a List of Delimited Strings

This is a generic utility function that I wrote for use in unit testing web services that return REF cursors. It takes as input an open REF cursor and an optional filter string and returns the output as an array of delimited strings matching the filter. This can then be used to assert against expected output records.

The initial technical difficulty I had with making this generic was in getting the structure of the cursor. This is resolved by translating the REF cursor into a cursor that can be managed by the package DBMS_SQL, using DBMS_SQL.To_Cursor_Number.

As I am using it for unit testing the function does not need to be scalable. For other uses involving large volumes it could be made scalable by, for example, converting to a pipelined function.

Data Types

The cursor needs to return a flat projection, and columns accounted for currently are:

  • NUMBER
  • VARCHAR2
  • DATE
  • TIMESTAMP

Custom Dependencies

  • L1_chr_arr – custom array type, equivalent to SYS.ODCIVarchar2List
  • Utils.List_Delim – function that turns an array of strings into one delimited string, included in my unit test framework, Brendan’s Database Unit Testing Framework

Code

The function wiil (shortly) be packaged within my unit testing package UT_Utils, but for demo purposes, here is a script that has a driving anonymous block with the function declared locally within that block.

DECLARE
  l_csr         SYS_REFCURSOR;
  l_res_lis     L1_chr_arr;
  c_query_1     CONSTANT VARCHAR2(4000) := 'SELECT * FROM employees ORDER BY employee_id';
 
  PROCEDURE Write_Log (p_line VARCHAR2) IS
  BEGIN
    DBMS_Output.Put_Line (p_line);
  END Write_Log;
 
FUNCTION Cursor_to_Array (p_csr IN OUT SYS_REFCURSOR, p_filter VARCHAR2 DEFAULT NULL) RETURN L1_chr_arr IS
 
  c_chr_type    CONSTANT PLS_INTEGER := 1; --DBMS_Types.TYPECODE_* do not seem to quite work
  c_num_type    CONSTANT PLS_INTEGER := 2;
  c_dat_type    CONSTANT PLS_INTEGER := 12;
  c_stp_type    CONSTANT PLS_INTEGER := 180;
  l_csr_id      PLS_INTEGER;
  l_n_cols      PLS_INTEGER;
  l_desctab     DBMS_SQL.DESC_TAB;
  l_chr_val     VARCHAR2(4000);
  l_num_val     NUMBER;
  l_dat_val     DATE;
  l_stp_val     TIMESTAMP;
  l_val_lis     L1_chr_arr;
  l_res_lis     L1_chr_arr := L1_chr_arr();
  l_rec         VARCHAR2(4000);
 
BEGIN
 
  l_csr_id := DBMS_SQL.To_Cursor_Number (p_csr);
  DBMS_SQL.Describe_Columns (l_csr_id, l_n_cols, l_desctab);
 
  FOR i IN 1..l_n_cols LOOP
 
    CASE l_desctab(i).col_type
 
      WHEN c_chr_type THEN
        DBMS_SQL.Define_Column (l_csr_id, i, l_chr_val, 4000);
      WHEN c_num_type THEN
        DBMS_SQL.Define_Column (l_csr_id, i, l_num_val);
      WHEN c_dat_type THEN
        DBMS_SQL.Define_Column (l_csr_id, i, l_dat_val);
      WHEN c_stp_type THEN
         DBMS_SQL.Define_Column (l_csr_id, i, l_stp_val);
     ELSE
        Write_Log ('Col type ' || l_desctab(i).col_type || ' not accounted for!');
 
    END CASE;
 
  END LOOP;
 
  WHILE DBMS_SQL.Fetch_Rows (l_csr_id) > 0 LOOP
 
    l_val_lis := L1_chr_arr();
    l_val_lis.EXTEND (l_n_cols);
    FOR i IN 1 .. l_n_cols LOOP
 
      CASE l_desctab(i).col_type
 
        WHEN c_chr_type THEN
          DBMS_SQL.Column_Value (l_csr_id, i, l_chr_val);
          l_val_lis(i) := l_chr_val;
        WHEN c_num_type THEN
          DBMS_SQL.Column_Value (l_csr_id, i, l_num_val);
          l_val_lis(i) := l_num_val;
        WHEN c_dat_type THEN
          DBMS_SQL.Column_Value (l_csr_id, i, l_dat_val);
          l_val_lis(i) := l_dat_val;
        WHEN c_stp_type THEN
          DBMS_SQL.Column_Value (l_csr_id, i, l_stp_val);
          l_val_lis(i) := l_stp_val;
 
      END CASE;
 
    END LOOP;
 
    l_rec := Utils.List_Delim (l_val_lis);
    IF l_rec LIKE '%' || p_filter || '%' THEN
      l_res_lis.EXTEND;
      l_res_lis (l_res_lis.COUNT) := l_rec;
    END IF;
 
  END LOOP;
 
  DBMS_SQL.Close_Cursor (l_csr_id);
  RETURN l_res_lis;
 
END Cursor_to_Array;
 
BEGIN
 
  OPEN l_csr FOR c_query_1;
 
  l_res_lis := Cursor_to_Array (l_csr, 'SA_REP');
 
  FOR i IN 1..l_res_lis.COUNT LOOP
    Write_Log (i || ': ' || l_res_lis(i));
  END LOOP;
 
END;
/

 
Output for Cursor against HR.employees

1: 150|Peter|Tucker|PTUCKER|011.44.1344.129268|30-JAN-05|SA_REP|10000|.3|145|80|
2: 151|David|Bernstein|DBERNSTE|011.44.1344.345268|24-MAR-05|SA_REP|9500|.25|145|80|
3: 152|Peter|Hall|PHALL|011.44.1344.478968|20-AUG-05|SA_REP|9000|.25|145|80|
4: 153|Christopher|Olsen|COLSEN|011.44.1344.498718|30-MAR-06|SA_REP|8000|.2|145|80|
5: 154|Nanette|Cambrault|NCAMBRAU|011.44.1344.987668|09-DEC-06|SA_REP|7500|.2|145|80|
6: 155|Oliver|Tuvault|OTUVAULT|011.44.1344.486508|23-NOV-07|SA_REP|7000|.15|145|80|
7: 156|Janette|King|JKING|011.44.1345.429268|30-JAN-04|SA_REP|10000|.35|146|80|
8: 157|Patrick|Sully|PSULLY|011.44.1345.929268|04-MAR-04|SA_REP|9500|.35|146|80|
9: 158|Allan|McEwen|AMCEWEN|011.44.1345.829268|01-AUG-04|SA_REP|9000|.35|146|80|
10: 159|Lindsey|Smith|LSMITH|011.44.1345.729268|10-MAR-05|SA_REP|8000|.3|146|80|
11: 160|Louise|Doran|LDORAN|011.44.1345.629268|15-DEC-05|SA_REP|7500|.3|146|80|
12: 161|Sarath|Sewall|SSEWALL|011.44.1345.529268|03-NOV-06|SA_REP|7000|.25|146|80|
13: 162|Clara|Vishney|CVISHNEY|011.44.1346.129268|11-NOV-05|SA_REP|10500|.25|147|80|
14: 163|Danielle|Greene|DGREENE|011.44.1346.229268|19-MAR-07|SA_REP|9500|.15|147|80|
15: 164|Mattea|Marvins|MMARVINS|011.44.1346.329268|24-JAN-08|SA_REP|7200|.1|147|80|
16: 165|David|Lee|DLEE|011.44.1346.529268|23-FEB-08|SA_REP|6800|.1|147|80|
17: 166|Sundar|Ande|SANDE|011.44.1346.629268|24-MAR-08|SA_REP|6400|.1|147|80|
18: 167|Amit|Banda|ABANDA|011.44.1346.729268|21-APR-08|SA_REP|6200|.1|147|80|
19: 168|Lisa|Ozer|LOZER|011.44.1343.929268|11-MAR-05|SA_REP|11500|.25|148|80|
20: 169|Harrison|Bloom|HBLOOM|011.44.1343.829268|23-MAR-06|SA_REP|10000|.2|148|80|
21: 170|Tayler|Fox|TFOX|011.44.1343.729268|24-JAN-06|SA_REP|9600|.2|148|80|
22: 171|William|Smith|WSMITH|011.44.1343.629268|23-FEB-07|SA_REP|7400|.15|148|80|
23: 172|Elizabeth|Bates|EBATES|011.44.1343.529268|24-MAR-07|SA_REP|7300|.15|148|80|
24: 173|Sundita|Kumar|SKUMAR|011.44.1343.329268|21-APR-08|SA_REP|6100|.1|148|80|
25: 174|Ellen|Abel|EABEL|011.44.1644.429267|11-MAY-04|SA_REP|11000|.3|149|80|
26: 175|Alyssa|Hutton|AHUTTON|011.44.1644.429266|19-MAR-05|SA_REP|8800|.25|149|80|
27: 176|Jonathon|Taylor|JTAYLOR|011.44.1644.429265|24-MAR-06|SA_REP|8600|.2|149|80|
28: 177|Jack|Livingston|JLIVINGS|011.44.1644.429264|23-APR-06|SA_REP|8400|.2|149|80|
29: 178|Kimberely|Grant|KGRANT|011.44.1644.429263|24-MAY-07|SA_REP|7000|.15|149||
30: 179|Charles|Johnson|CJOHNSON|011.44.1644.429262|04-JAN-08|SA_REP|6200|.1|149|80|
 
PL/SQL procedure successfully completed.






 

A Note on Dependencies and Database Unit Testing

Ideas on unit testing for the database are often heavily influenced by the world of object oriented programming (OOP), usually Java in practice. This is no doubt because much of modern thinking on development methodologies, including test driven development (TDD), originated in this world. Some of these ideas appear to translate very well into the database world, including that of TDD itself, with automated unit tests. However, some ideas may not translate so well, or even make sense, in database unit testing. For example, Roy Osherove (2011), Unit Test – Definition says:

A good unit test … runs in memory (no DB or File access, for example)

One concept that appears very important in the OOP world is that of dependencies, and of isolation of the code under test from its dependencies. This gives rise to complex mechanisms of ‘mocking’ and ‘dependency injection’ to bring about said isolation. Osherove mentions isolation in the same article as a requirement of good unit testing, and his view appears to be widespread. It’s worth mentioning though that not everyone in the OOP world shares his insistence. The influential Martin Fowler (2014) uses a nice terminology of ‘sociable’ tests (as opposed to ‘isolated’ tests) for tests that rely on other units to fulfill the behaviour under test, and he uses this approach himself when practicable, UnitTest.

In the case of database unit testing, it seems to me to make very little sense to think in terms of isolating code under test from its dependencies. The following two diagrams represent how I see the relationships between base code, dependencies and unit test code across two distinct phases.

Development Phase
UT Phases-dev

Regression Phase
UT Phases-reg

I would welcome any comments or opinions.






 

TRAPIT – TRansactional API Testing in Oracle

I recently wrote a number of articles on design patterns for database API testing, with code included for testing example procedures against Oracle’s HR demo schema. There are now four examples, intended together to represent the whole range of database API testing requirements, as follows:

ModeSetterGetter
Real TimeWeb Service SavingWeb Service Getting by REF Cursor
BatchBatch Loading of Flat FilesViews

The testing code used my own framework utility packages, and originally called procedures from the well-known utPLSQL framework. However, the use of utPLSQL was always minimal, and I later removed it to form my own lightweight independent framework that is designed to help avoid many API testing anti-patterns, and provide better output formatting than utPLSQL.

The framework is based on the idea that all API testing programs can follow a universal design pattern for testing APIs, using the concept of a ‘pure’ function as a wrapper to manage the ‘impurity’ inherent in database APIs. I explained the concepts involved in a presentation at the Oracle User Group Ireland Conference in March 2018:

The Database API Viewed As A Mathematical Function: Insights into Testing

The framework lists all inputs and outputs at scenario level (see box below), meaning that the API test logs become precise and accurate documents of what the program does. All the output is printed by a central library procedure so that work by individual test programs is minimised.

The code is available on GitHub here, and includes both framework and design pattern examples: BrenPatF/trapit_oracle_tester

See also the following articles describing the four design patterns. The log from my demo suite is included below for the first example to illustrate the general format of the log.

Design Patterns for Database API Testing 1: Web Service Saving 1 – Design
Design Patterns for Database API Testing 1: Web Service Saving 2 – Code
Design Patterns for Database API Testing 2: Views 1 – Design
Design Patterns for Database API Testing 2: Views 2 – Code
Design Patterns for Database API Testing 3: Batch Loading of Flat Files
Design Patterns for Database API Testing 4: REF Cursor Getter

SQL> DECLARE
  2  BEGIN
  3  
  4    Utils.Clear_Log;
  5    Utils_TT.Run_Suite (Utils_TT.c_tt_suite_bren);
  6  
  7  EXCEPTION
  8    WHEN OTHERS THEN
  9      Utils.Write_Other_Error;
 10  END;
 11  /

PL/SQL procedure successfully completed.

SQL> @L_Log_Default

TEXT
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

TRAPIT TEST: TT_Emp_WS.tt_AIP_Save_Emps
=======================================

SCENARIO 1: 1 valid record {
============================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 1  EM 1   IT_PROG    1000

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee id  Name  Email  Job      Salary
            --  -----------  ----  -----  -------  ------
                       1927  LN 1  EM 1   IT_PROG    1000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Output array: Actual = 1, Expected = 1 {
        ==============================================

            F?  Employee id  Description
            --  -----------  --------------------------------------
                       1927  ONE THOUSAND NINE HUNDRED TWENTY-SEVEN

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 2: 1 invalid job id {
==============================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 2  EM 2   NON_JOB    1500

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 0, Expected = 0: SUCCESS
        =================================================

        GROUP Output array: Actual = 1, Expected = 1 {
        ==============================================

            F?  Employee id  Description
            --  -----------  -------------------------------------------------------------------
                          0  ORA-02291: integrity constraint (.) violated - parent key not found

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 3: 1 invalid number {
==============================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 3  EM 3   IT_PROG   2000x

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 0, Expected = 0: SUCCESS
        =================================================

        GROUP Output array: Actual = 0, Expected = 0: SUCCESS
        =====================================================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Error message
            --  -------------------------------------------------------------------------------
                ORA-06502: PL/SQL: numeric or value error: character to number conversion error

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 4: 2 valid records, 1 invalid job id (2 deliberate errors) {
=====================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 4  EM 4   IT_PROG    3000
            LN 5  EM 5   NON_JOB    4000
            LN 6  EM 6   IT_PROG    5000

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 3 {
        ==========================================

            F?  Employee id  Name  Email  Job      Salary
            --  -----------  ----  -----  -------  ------
            F          1929  LN 4  EM 4   IT_PROG    3000
            >          1929  LN 4  EM 4   IT_PROG    1000
                       1931  LN 6  EM 6   IT_PROG    5000
            F
            >          1931  LN 6  EM 6   IT_PROG    5000

        } 2 failed, of 3: FAILURE
        =========================

        GROUP Output array: Actual = 3, Expected = 3 {
        ==============================================

            F?  Employee id  Description
            --  -----------  -------------------------------------------------------------------
                       1929  ONE THOUSAND NINE HUNDRED TWENTY-NINE
                          0  ORA-02291: integrity constraint (.) violated - parent key not found
                       1931  ONE THOUSAND NINE HUNDRED THIRTY-ONE

        } 0 failed, of 3: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 2 failed, of 7: FAILURE
=========================

TIMING: Actual = 1, Expected <= 2: SUCCESS
==========================================

SUMMARY for TT_Emp_WS.tt_AIP_Save_Emps
======================================

Scenario                                                 # Failed  # Tests  Status
-------------------------------------------------------  --------  -------  -------
1 valid record                                                  0        3  SUCCESS
1 invalid job id                                                0        3  SUCCESS
1 invalid number                                                0        3  SUCCESS
2 valid records, 1 invalid job id (2 deliberate errors)         2        7  FAILURE
Timing                                                          0        1  SUCCESS
-------------------------------------------------------  --------  -------  -------
Total                                                           2       17  FAILURE
-------------------------------------------------------  --------  -------  -------

Timer Set: TT_Emp_WS.tt_AIP_Save_Emps, Constructed at 22 Oct 2016 15:05:16, written at 15:05:16
===============================================================================================
[Timer timed: Elapsed (per call): 0.03 (0.000034), CPU (per call): 0.05 (0.000050), calls: 1000, '***' denotes corrected line below]

Timer       Elapsed         CPU         Calls       Ela/Call       CPU/Call
-------  ----------  ----------  ------------  -------------  -------------
Setup          0.00        0.00             1        0.00300        0.00000
Caller         0.00        0.02             3        0.00133        0.00667
SELECT         0.00        0.00             3        0.00033        0.00000
(Other)        0.08        0.06             1        0.07700        0.06000
-------  ----------  ----------  ------------  -------------  -------------
Total          0.09        0.08             8        0.01063        0.01000
-------  ----------  ----------  ------------  -------------  -------------

TRAPIT TEST: TT_Emp_WS.tt_AIP_Get_Dept_Emps
===========================================

Employees created in setup: DS-1 - 4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job
==========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1932             10              IT_PROG           1000
2         1933             10        1932  IT_PROG           2000
3         1934                       1932  IT_PROG           3000
4         1935             10        1932  AD_ASST           4000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1936             10              IT_PROG           1000
2         1937             10        1936  IT_PROG           2000
3         1938                       1936  IT_PROG           3000
4         1939             10        1936  AD_ASST           4000
5         1940             20        1936  IT_PROG           5000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1941             10              IT_PROG           1000
2         1942             10        1941  IT_PROG           2000
3         1943                       1941  IT_PROG           3000
4         1944             10        1941  AD_ASST           4000
5         1945             20        1941  IT_PROG           5000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1946             10              IT_PROG           1000
2         1947             10        1946  IT_PROG           2000
3         1948                       1946  IT_PROG           3000
4         1949             10        1946  AD_ASST           4000
5         1950             20        1946  IT_PROG           5000

Employees created in setup: DS-3 - As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600
================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1951             10              IT_PROG            100
2         1952             10        1951  IT_PROG            200
3         1953                       1951  IT_PROG            300
4         1954             10        1951  AD_ASST            400
5         1955             20        1951  IT_PROG            500

SCENARIO 1: DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep (10) - pass dep 10 {
==================================================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1932  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1933  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1932             10  22-OCT-2016
                   1934  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1932                 22-OCT-2016
                   1935  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1932             10  22-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       10

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                      .4
                LN_2  Administration  LN_1       2000                1.33                      .8

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 2: DS-2, testing same as 1 but with extra emp in another dep (20) - pass dep 10 {
==========================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1936  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1937  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1936             10  22-OCT-2016
                   1938  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1936                 22-OCT-2016
                   1939  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1936             10  22-OCT-2016
                   1940  LN_5       EM_5   22-OCT-2016  IT_PROG    5000        1936             20  22-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       10

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                     .33
                LN_2  Administration  LN_1       2000                1.33                     .67

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 3: DS-2, as second scenario, but - pass dep 20 {
=========================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1941  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1942  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1941             10  22-OCT-2016
                   1943  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1941                 22-OCT-2016
                   1944  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1941             10  22-OCT-2016
                   1945  LN_5       EM_5   22-OCT-2016  IT_PROG    5000        1941             20  22-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       20

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 1, Expected = 1 {
        ================================================

            F?  Name  Department  Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  ----------  -------  ------  ------------------  ----------------------
                LN_5  Marketing   LN_1       5000                   1                    1.67

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 1: SUCCESS
=========================

SCENARIO 4: DS-2, as second scenario, but - pass null dep {
===========================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1946  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1947  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1946             10  22-OCT-2016
                   1948  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1946                 22-OCT-2016
                   1949  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1946             10  22-OCT-2016
                   1950  LN_5       EM_5   22-OCT-2016  IT_PROG    5000        1946             20  22-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 0, Expected = 0: SUCCESS
        =======================================================

} 0 failed, of 1: SUCCESS
=========================

SCENARIO 5: DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) - pass dep 10 {
==============================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1951  LN_1       EM_1   22-OCT-2016  IT_PROG     100                         10  22-OCT-2016
                   1952  LN_2       EM_2   22-OCT-2016  IT_PROG     200        1951             10  22-OCT-2016
                   1953  LN_3       EM_3   22-OCT-2016  IT_PROG     300        1951                 22-OCT-2016
                   1954  LN_4       EM_4   22-OCT-2016  AD_ASST     400        1951             10  22-OCT-2016
                   1955  LN_5       EM_5   22-OCT-2016  IT_PROG     500        1951             20  22-OCT-2016

        }
        =

        GROUP Department Parameter {
        ============================

            Department Id
            -------------
                       10

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 0, Expected = 0: SUCCESS
        =======================================================

} 0 failed, of 1: SUCCESS
=========================

TIMING: Actual = 1, Expected <= 1: SUCCESS
==========================================

SUMMARY for TT_Emp_WS.tt_AIP_Get_Dept_Emps
==========================================

Scenario                                                                                              # Failed  # Tests  Status
----------------------------------------------------------------------------------------------------  --------  -------  -------
DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep (10) - pass dep 10         0        2  SUCCESS
DS-2, testing same as 1 but with extra emp in another dep (20) - pass dep 10                                 0        2  SUCCESS
DS-2, as second scenario, but - pass dep 20                                                                  0        1  SUCCESS
DS-2, as second scenario, but - pass null dep                                                                0        1  SUCCESS
DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) - pass dep 10                             0        1  SUCCESS
Timing                                                                                                       0        1  SUCCESS
----------------------------------------------------------------------------------------------------  --------  -------  -------
Total                                                                                                        0        8  SUCCESS
----------------------------------------------------------------------------------------------------  --------  -------  -------

Timer Set: TT_Emp_WS.tt_AIP_Get_Dept_Emps, Constructed at 22 Oct 2016 15:05:16, written at 15:05:16
===================================================================================================
[Timer timed: Elapsed (per call): 0.04 (0.000035), CPU (per call): 0.03 (0.000030), calls: 1000, '***' denotes corrected line below]

Timer       Elapsed         CPU         Calls       Ela/Call       CPU/Call
-------  ----------  ----------  ------------  -------------  -------------
Setup          0.04        0.02             5        0.00840        0.00400
Caller         0.01        0.02             5        0.00140        0.00400
(Other)        0.09        0.10             1        0.09200        0.10000
-------  ----------  ----------  ------------  -------------  -------------
Total          0.14        0.14            11        0.01282        0.01273
-------  ----------  ----------  ------------  -------------  -------------

TRAPIT TEST: TT_View_Drivers.tt_HR_Test_View_V
==============================================

Employees created in setup: DS-1 - 4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job
==========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1956             10              IT_PROG           1000
2         1957             10        1956  IT_PROG           2000
3         1958                       1956  IT_PROG           3000
4         1959             10        1956  AD_ASST           4000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1960             10              IT_PROG           1000
2         1961             10        1960  IT_PROG           2000
3         1962                       1960  IT_PROG           3000
4         1963             10        1960  AD_ASST           4000
5         1964             20        1960  IT_PROG           5000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1965             10              IT_PROG           1000
2         1966             10        1965  IT_PROG           2000
3         1967                       1965  IT_PROG           3000
4         1968             10        1965  AD_ASST           4000
5         1969             20        1965  IT_PROG           5000

Employees created in setup: DS-3 - As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600
================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1970             10              IT_PROG            100
2         1971             10        1970  IT_PROG            200
3         1972                       1970  IT_PROG            300
4         1973             10        1970  AD_ASST            400
5         1974             20        1970  IT_PROG            500

SCENARIO 1: DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep {
===============================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1956  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1957  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1956             10  22-OCT-2016
                   1958  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1956                 22-OCT-2016
                   1959  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1956             10  22-OCT-2016

        }
        =

        GROUP Where {
        =============

            Where
            -----


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                      .4
                LN_2  Administration  LN_1       2000                1.33                      .8

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 2: DS-2, testing same as 1 but with extra emp in another dep {
=======================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1960  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1961  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1960             10  22-OCT-2016
                   1962  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1960                 22-OCT-2016
                   1963  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1960             10  22-OCT-2016
                   1964  LN_5       EM_5   22-OCT-2016  IT_PROG    5000        1960             20  22-OCT-2016

        }
        =

        GROUP Where {
        =============

            Where
            -----


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 3, Expected = 3 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                     .33
                LN_2  Administration  LN_1       2000                1.33                     .67
                LN_5  Marketing       LN_1       5000                   1                    1.67

        } 0 failed, of 3: SUCCESS
        =========================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 3: DS-2, passing 'WHERE dep=10' {
==========================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1965  LN_1       EM_1   22-OCT-2016  IT_PROG    1000                         10  22-OCT-2016
                   1966  LN_2       EM_2   22-OCT-2016  IT_PROG    2000        1965             10  22-OCT-2016
                   1967  LN_3       EM_3   22-OCT-2016  IT_PROG    3000        1965                 22-OCT-2016
                   1968  LN_4       EM_4   22-OCT-2016  AD_ASST    4000        1965             10  22-OCT-2016
                   1969  LN_5       EM_5   22-OCT-2016  IT_PROG    5000        1965             20  22-OCT-2016

        }
        =

        GROUP Where {
        =============

            Where
            --------------------------------
            department_name='Administration'

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                     .33
                LN_2  Administration  LN_1       2000                1.33                     .67

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 4: DS-3, Salaries total 1500 (< threshold of 1600, so return nothing) {
================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date    Job      Salary  Manager Id  Department Id  Updated
            -----------  ---------  -----  -----------  -------  ------  ----------  -------------  -----------
                   1970  LN_1       EM_1   22-OCT-2016  IT_PROG     100                         10  22-OCT-2016
                   1971  LN_2       EM_2   22-OCT-2016  IT_PROG     200        1970             10  22-OCT-2016
                   1972  LN_3       EM_3   22-OCT-2016  IT_PROG     300        1970                 22-OCT-2016
                   1973  LN_4       EM_4   22-OCT-2016  AD_ASST     400        1970             10  22-OCT-2016
                   1974  LN_5       EM_5   22-OCT-2016  IT_PROG     500        1970             20  22-OCT-2016

        }
        =

        GROUP Where {
        =============

            Where
            -----


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 0, Expected = 0: SUCCESS
        =======================================================

} 0 failed, of 1: SUCCESS
=========================

TIMING: Actual = 1, Expected <= 1: SUCCESS
==========================================

SUMMARY for TT_View_Drivers.tt_HR_Test_View_V
=============================================

Scenario                                                                           # Failed  # Tests  Status
---------------------------------------------------------------------------------  --------  -------  -------
DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep         0        2  SUCCESS
DS-2, testing same as 1 but with extra emp in another dep                                 0        3  SUCCESS
DS-2, passing 'WHERE dep=10'                                                              0        2  SUCCESS
DS-3, Salaries total 1500 (< threshold of 1600, so return nothing)                        0        1  SUCCESS
Timing                                                                                    0        1  SUCCESS
---------------------------------------------------------------------------------  --------  -------  -------
Total                                                                                     0        9  SUCCESS
---------------------------------------------------------------------------------  --------  -------  -------

Timer Set: TT_View_Drivers.tt_HR_Test_View_V, Constructed at 22 Oct 2016 15:05:16, written at 15:05:16
======================================================================================================
[Timer timed: Elapsed (per call): 0.03 (0.000033), CPU (per call): 0.03 (0.000030), calls: 1000, '***' denotes corrected line below]

Timer       Elapsed         CPU         Calls       Ela/Call       CPU/Call
-------  ----------  ----------  ------------  -------------  -------------
Setup          0.03        0.03             4        0.00825        0.00750
Caller         0.01        0.00             4        0.00125        0.00000
(Other)        0.07        0.08             1        0.07000        0.08000
-------  ----------  ----------  ------------  -------------  -------------
Total          0.11        0.11             9        0.01200        0.01222
-------  ----------  ----------  ------------  -------------  -------------

TRAPIT TEST: TT_Emp_Batch.tt_AIP_Load_Emps
==========================================

SCENARIO 1: NV/OV/OU/NI/OI/EI: 1/0/0/0/0/0. Existing J/E: 0/0. [1 valid new record from scratch] {
==================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------
            ,LN 1,EM 1,01-JAN-2010,IT_PROG,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table (No records)
        ===================================

        GROUP Employees Table (No records)
        ==================================

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1976  LN 1  EM 1   01-JAN-2010  IT_PROG  10000   22-OCT-2016

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 1, Expected = 1 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              52  LOAD_EMPS     employees_20160801.dat               1                  0                  0  22-OCT-2016  22-OCT-2016  S

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 4: SUCCESS
=========================

SCENARIO 2: NV/OV/OU/NI/OI/EI: 1/1/1/0/0/0. Existing J/E: 1/0. [3 valid records of each kind] {
===============================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      3

        }
        =

        GROUP File {
        ============

            Line
            -----------------------------------------
            ,LN 1,EM 1,01-JAN-2010,IT_PROG,10000
            1978,LN 2,EM 2,01-JAN-2010,IT_PROG,20000
            1979,LN 3U,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          54  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1978  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010
                   1979  LN 3  EM 3   01-JAN-2010  IT_PROG   30000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 3, Expected = 3 {
        ==========================================

            F?  Employee Id  Name   Email  Hired        Job      Salary  Updated
            --  -----------  -----  -----  -----------  -------  ------  -----------
                       1978  LN 2   EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1979  LN 3U  EM 3   01-JAN-2010  IT_PROG  30000   22-OCT-2016
                       1980  LN 1   EM 1   01-JAN-2010  IT_PROG  10000   22-OCT-2016

        } 0 failed, of 3: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              54  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              55  LOAD_EMPS     employees_20160801.dat               2                  0                  0  22-OCT-2016  22-OCT-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 7: SUCCESS
=========================

SCENARIO 3: NV/OV/OU/NI/OI/EI: 0/0/0/0/1/0. Existing J/E: 1/1. Uid not found [1 invalid old - exception] {
==========================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            --------------------------------------
            99,LN 1,EM 1,01-JAN-2010,IT_PROG,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          57  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1982  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1982  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$       ORA_ERR_OPTYP$  Employee Id  Name  Email  Hired        Job      Salary
            --  ----------------  ------------  ------------------  --------------  -----------  ----  -----  -----------  -------  ------
                              58                Employee not found  PK                       99  LN 1  EM 1   01-JAN-2010  IT_PROG  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              57  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              58  LOAD_EMPS     employees_20160801.dat               0                  0                  1  22-OCT-2016  22-OCT-2016  F

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  ------------------------------------------------------
                ORA-20001: Batch failed with too many invalid records!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 5: SUCCESS
=========================

SCENARIO 4: NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Email too long [1 invalid new - exception] {
===========================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------------------------------------
            ,LN 1,EM 1123456789012345678901234567890,01-JAN-2010,IT_PROG,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          60  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1984  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1984  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                             ORA_ERR_OPTYP$  Employee Id  Name  Email                               Hired        Job      Salary
            --  ----------------  ------------  ----------------------------------------------------------------------------------------  --------------  -----------  ----  ----------------------------------  -----------  -------  ------
                              61                ORA-12899: value too large for column "HR"."EMPLOYEES"."EMAIL" (actual: 34, maximum: 25)  I                      1985  LN 1  EM 1123456789012345678901234567890  01-JAN-2010  IT_PROG  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              60  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              61  LOAD_EMPS     employees_20160801.dat               0                  0                  1  22-OCT-2016  22-OCT-2016  F

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  ------------------------------------------------------
                ORA-20001: Batch failed with too many invalid records!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 5: SUCCESS
=========================

SCENARIO 5: NV/OV/OU/NI/OI/EI: 1/0/0/0/1/0. Existing J/E: 1/1. Name too long [1 valid new, 1 invalid old - no exception] {
==========================================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      2

        }
        =

        GROUP File {
        ============

            Line
            ----------------------------------------------------------------------
            1987,LN 1123456789012345678901234567890,EM 1,01-JAN-2010,IT_PROG,10000
            ,LN 3,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          63  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1987  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 2 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1987  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1988  LN 3  EM 3   01-JAN-2010  IT_PROG  30000   22-OCT-2016

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                                 ORA_ERR_OPTYP$  Employee Id  Name                                Email  Hired        Job      Salary
            --  ----------------  ------------  --------------------------------------------------------------------------------------------  --------------  -----------  ----------------------------------  -----  -----------  -------  ------
                              64                ORA-12899: value too large for column "HR"."EMPLOYEES"."LAST_NAME" (actual: 34, maximum: 25)  U                      1987  LN 1123456789012345678901234567890  EM 1   01-JAN-2010  IT_PROG  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              63  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              64  LOAD_EMPS     employees_20160801.dat               1                  0                  1  22-OCT-2016  22-OCT-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 6: SUCCESS
=========================

SCENARIO 6: NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Invalid job [1 invalid new - exception] {
========================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------
            ,LN 1,EM 1,01-JAN-2010,NON_JOB,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          66  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1990  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       1990  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 1, Expected = 1 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                    ORA_ERR_OPTYP$  Employee Id  Name  Email  Hired        Job      Salary
            --  ----------------  ------------  -------------------------------------------------------------------------------  --------------  -----------  ----  -----  -----------  -------  ------
                              67                ORA-02291: integrity constraint (HR.EMP_JOB_FK) violated - parent key not found  I                      1991  LN 1  EM 1   01-JAN-2010  NON_JOB  10000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              66  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              67  LOAD_EMPS     employees_20160801.dat               0                  0                  1  22-OCT-2016  22-OCT-2016  F

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  ------------------------------------------------------
                ORA-20001: Batch failed with too many invalid records!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 5: SUCCESS
=========================

SCENARIO 7: NV/OV/OU/NI/OI/EI: 0/1/0/1/1/0. Existing J/E: 1/2. 2 invalid jobs [1 valid old, 2 invalid: old and new - no exception] {
====================================================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      3

        }
        =

        GROUP File {
        ============

            Line
            -----------------------------------------
            ,LN 1,EM 1,01-JAN-2010,NON_JOB,10000
            1993,LN 2,EM 2,01-JAN-2010,NON_JOB,20000
            1994,LN 3U,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          69  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1993  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010
                   1994  LN 3  EM 3   01-JAN-2010  IT_PROG   30000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 2 {
        ==========================================

            F?  Employee Id  Name   Email  Hired        Job      Salary  Updated
            --  -----------  -----  -----  -----------  -------  ------  -----------
                       1993  LN 2   EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1994  LN 3U  EM 3   01-JAN-2010  IT_PROG  30000   22-OCT-2016

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Error: Actual = 2, Expected = 2 {
        =======================================

            F?  Job Statistic Id  ORA_ERR_TAG$  ORA_ERR_MESG$                                                                    ORA_ERR_OPTYP$  Employee Id  Name  Email  Hired        Job      Salary
            --  ----------------  ------------  -------------------------------------------------------------------------------  --------------  -----------  ----  -----  -----------  -------  ------
                              70                ORA-02291: integrity constraint (HR.EMP_JOB_FK) violated - parent key not found  U                      1993  LN 2  EM 2   01-JAN-2010  NON_JOB  20000
                              70                ORA-02291: integrity constraint (HR.EMP_JOB_FK) violated - parent key not found  I                      1995  LN 1  EM 1   01-JAN-2010  NON_JOB  10000

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              69  LOAD_EMPS     employees_20160101.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S
                              70  LOAD_EMPS     employees_20160801.dat               1                  0                  2  22-OCT-2016  22-OCT-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 7: SUCCESS
=========================

SCENARIO 8: NV/OV/OU/NI/OI/EI: 0/1/0/0/0/1. Existing J/E: 1/2. Name 4001ch [1 valid old, 1 invalid new for external table - no exception; also file had previously failed] {
============================================================================================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      2

        }
        =

        GROUP File {
        ============

            Line
            --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
            ,123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567
            1998,LN 3U,EM 3,01-JAN-2010,IT_PROG,30000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          72  LOAD_EMPS     employees_20160801.dat               0                  0                  2  01-JAN-2010  01-JAN-2010  F

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   1997  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010
                   1998  LN 3  EM 3   01-JAN-2010  IT_PROG   30000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 2 {
        ==========================================

            F?  Employee Id  Name   Email  Hired        Job      Salary  Updated
            --  -----------  -----  -----  -----------  -------  ------  -----------
                       1997  LN 2   EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010
                       1998  LN 3U  EM 3   01-JAN-2010  IT_PROG  30000   22-OCT-2016

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 2, Expected = 2 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              72  LOAD_EMPS     employees_20160801.dat               0                  0                  2  01-JAN-2010  01-JAN-2010  F
                              73  LOAD_EMPS     employees_20160801.dat               1                  1                  0  22-OCT-2016  22-OCT-2016  S

        } 0 failed, of 2: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 6: SUCCESS
=========================

SCENARIO 9: NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. [File already processed - exception] {
=====================================================================================================

    INPUTS
    ======

        GROUP Parameter {
        =================

            File Name               Count
            ----------------------  -----
            employees_20160801.dat      1

        }
        =

        GROUP File {
        ============

            Line
            ------------------------------------
            ,LN 1,EM 1,01-JAN-2010,NON_JOB,10000

        }
        =

        GROUP Batch Job Table {
        =======================

            Name       Fail Percent
            ---------  ------------
            LOAD_EMPS            70

        }
        =

        GROUP Statistics Table {
        ========================

            Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                          75  LOAD_EMPS     employees_20160801.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        }
        =

        GROUP Employees Table {
        =======================

            Employee Id  Name  Email  Hired        Job      Salary  Manager Id  Department Id  Updated
            -----------  ----  -----  -----------  -------  ------  ----------  -------------  -----------
                   2000  LN 2  EM 2   01-JAN-2010  IT_PROG   20000                             01-JAN-2010

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee Id  Name  Email  Hired        Job      Salary  Updated
            --  -----------  ----  -----  -----------  -------  ------  -----------
                       2000  LN 2  EM 2   01-JAN-2010  IT_PROG  20000   01-JAN-2010

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Error: Actual = 0, Expected = 0: SUCCESS
        ==============================================

        GROUP Job Statistic: Actual = 1, Expected = 1 {
        ===============================================

            F?  Job Statistic Id  Batch job Id  File Name               Records Loaded  Records Failed ET  Records Failed DB  Start Time   End Time     Status
            --  ----------------  ------------  ----------------------  --------------  -----------------  -----------------  -----------  -----------  ------
                              75  LOAD_EMPS     employees_20160801.dat              10                  0                  2  01-JAN-2010  01-JAN-2010  S

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Message
            --  --------------------------------------------------------
                ORA-20002: File has already been processed successfully!

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 4: SUCCESS
=========================

TIMING: Actual = 254, Expected <= 2: FAILURE
============================================

SUMMARY for TT_Emp_Batch.tt_AIP_Load_Emps
=========================================

Scenario                                                                                                                                                        # Failed  # Tests  Status
--------------------------------------------------------------------------------------------------------------------------------------------------------------  --------  -------  -------
NV/OV/OU/NI/OI/EI: 1/0/0/0/0/0. Existing J/E: 0/0. [1 valid new record from scratch]                                                                                   0        4  SUCCESS
NV/OV/OU/NI/OI/EI: 1/1/1/0/0/0. Existing J/E: 1/0. [3 valid records of each kind]                                                                                      0        7  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/0/1/0. Existing J/E: 1/1. Uid not found [1 invalid old - exception]                                                                           0        5  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Email too long [1 invalid new - exception]                                                                          0        5  SUCCESS
NV/OV/OU/NI/OI/EI: 1/0/0/0/1/0. Existing J/E: 1/1. Name too long [1 valid new, 1 invalid old - no exception]                                                           0        6  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. Invalid job [1 invalid new - exception]                                                                             0        5  SUCCESS
NV/OV/OU/NI/OI/EI: 0/1/0/1/1/0. Existing J/E: 1/2. 2 invalid jobs [1 valid old, 2 invalid: old and new - no exception]                                                 0        7  SUCCESS
NV/OV/OU/NI/OI/EI: 0/1/0/0/0/1. Existing J/E: 1/2. Name 4001ch [1 valid old, 1 invalid new for external table - no exception; also file had previously failed]         0        6  SUCCESS
NV/OV/OU/NI/OI/EI: 0/0/0/1/0/0. Existing J/E: 1/1. [File already processed - exception]                                                                                0        4  SUCCESS
Timing                                                                                                                                                                 1        1  FAILURE
--------------------------------------------------------------------------------------------------------------------------------------------------------------  --------  -------  -------
Total                                                                                                                                                                  1       50  FAILURE
--------------------------------------------------------------------------------------------------------------------------------------------------------------  --------  -------  -------

Timer Set: TT_Emp_Batch.tt_AIP_Load_Emps, Constructed at 22 Oct 2016 15:05:16, written at 15:05:20
==================================================================================================
[Timer timed: Elapsed (per call): 0.03 (0.000033), CPU (per call): 0.03 (0.000030), calls: 1000, '***' denotes corrected line below]

Timer           Elapsed         CPU         Calls       Ela/Call       CPU/Call
-----------  ----------  ----------  ------------  -------------  -------------
Setup              0.97        0.23             9        0.10811        0.02556
Caller             1.27        0.99             5        0.25400        0.19800
Get_Tab_Lis        0.73        0.59             9        0.08122        0.06556
Get_Err_Lis        0.00        0.00             9        0.00011        0.00000
Get_Jbs_Lis        0.01        0.02             9        0.00056        0.00222
(Other)            0.34        0.35             1        0.34200        0.35000
-----------  ----------  ----------  ------------  -------------  -------------
Total              3.32        2.18            42        0.07910        0.05190
-----------  ----------  ----------  ------------  -------------  -------------

Suite Summary
=============

Package.Procedure                  Tests  Fails         ELA         CPU
---------------------------------  -----  -----  ----------  ----------
TT_Emp_WS.tt_AIP_Save_Emps            17      2        0.09        0.08
TT_Emp_WS.tt_AIP_Get_Dept_Emps         8      0        0.14        0.14
TT_View_Drivers.tt_HR_Test_View_V      9      0        0.11        0.11
TT_Emp_Batch.tt_AIP_Load_Emps         50      1        3.32        2.18
---------------------------------  -----  -----  ----------  ----------
Total                                 84      3        3.66        2.51
---------------------------------  -----  -----  ----------  ----------
Others error in (): ORA-20001: Suite BRENDAN returned error status: ORA-06512: at "DP_3.UTILS_TT", line 152
ORA-06512: at "DP_3.UTILS_TT", line 819
ORA-06512: at line 5


1687 rows selected.






 

Design Patterns for Database API Testing 2: Views 2 – Code

Last October I gave a presentation on database unit testing with utPLSQL, Oracle Unit Testing with utPLSQL. I mentioned design patterns as a way of reducing the effort of building unit tests and outlined some strategies for coding them effectively.

In the current set of articles, I develop the ideas further, starting from the idea that all database APIs can be considered in terms of the axes:

  • direction (i.e. getter or setter, noting that setters can also ‘get’)
  • mode (i.e. real time or batch)

For each cell in the implied matrix, I construct an example API (or view) with specified requirements against Oracle’s HR demo schema, and use this example to construct a testing program with appropriate scenarios as a design pattern. Concepts and common patterns and anti-patterns in automated API testing are discussed throughout, and these are largely independent of testing framework used. However, the examples use my own lightweight independent framework that is designed to help avoid many API testing anti-patterns. The code is available on GitHub here, BrenPatF/trapit_oracle_tester, and includes both framework and design pattern examples.

Behind the four examples, there is an underlying design pattern that involves wrapping the API call in a ‘pure’ procedure, called once per scenario, with the output ‘actuals’ array including everything affected by the API, whether as output parameters, or on database tables, etc. The inputs are also extended from the API parameters to include any other effective inputs. Assertion takes place after all scenarios and is against the extended outputs, with extended inputs also listed. This concept of the ‘pure’ function, central to Functional Programming, has important advantages in automated testing. I explained the concepts involved in a presentation at the Oracle User Group Ireland Conference in March 2018:

The Database API Viewed As A Mathematical Function: Insights into Testing


In the first part of this two part article,Design Patterns for Database API Testing 2: Views 1 – Design I presented a design pattern for unit testing views, using an example based on Oracle’s HR demo schema, and here I list the code for the main test procedure and a couple of the utility procedures, with notes.

A structure diagram shows how the PL/SQL packages relate to each other, and sections of the code are listed with notes.

Package Structure Diagram

Call Structure Table

TT_View_Drivers.tt_HR_Test_View_V – View Test Procedure

Declare section

PROCEDURE tt_HR_Test_View_V IS

  c_view_name           CONSTANT VARCHAR2(61) := 'HR_Test_View_V';
  c_proc_name           CONSTANT VARCHAR2(61) := 'TT_View_Drivers.tt_' || c_view_name;
  c_dep_id_1            CONSTANT PLS_INTEGER := 10;
  c_dep_id_2            CONSTANT PLS_INTEGER := 20;
  c_dep_nm_1            CONSTANT VARCHAR2(100) := 'Administration';
  c_dep_nm_2            CONSTANT VARCHAR2(100) := 'Marketing';
  c_job_bad             CONSTANT VARCHAR2(100) := 'AD_ASST';
  c_job_good            CONSTANT VARCHAR2(100) := 'IT_PROG';
  c_base_sal            CONSTANT PLS_INTEGER := 1000;

  c_ln_pre              CONSTANT VARCHAR2(10) := DML_API_TT_HR.c_ln_pre;

  c_sel_lis             CONSTANT L1_chr_arr := L1_chr_arr ('last_name', 'department_name', 'manager', 'salary', 'sal_rat', 'sal_rat_g');
  c_where_lis           CONSTANT L1_chr_arr := L1_chr_arr (NULL, NULL, 'department_name=''Administration''', NULL);

  c_dataset_3lis        CONSTANT L3_chr_arr := L3_chr_arr (
                             L2_chr_arr (L1_chr_arr ('4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job'),
--                                         dep           job          salary
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '1000'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '2000'),
                               L1_chr_arr (NULL,         c_job_good,  '3000'),
                               L1_chr_arr (c_dep_id_1,   c_job_bad,   '4000')
                                             ),
                             L2_chr_arr (L1_chr_arr ('As dataset 1 but with extra emp-5, in second dep (20)'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '1000'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '2000'),
                               L1_chr_arr (NULL,         c_job_good,  '3000'),
                               L1_chr_arr (c_dep_id_1,   c_job_bad,   '4000'),
                               L1_chr_arr (c_dep_id_2,   c_job_good,  '5000')
                                             ),
                             L2_chr_arr (L1_chr_arr ('As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '100'),
                               L1_chr_arr (c_dep_id_1,   c_job_good,  '200'),
                               L1_chr_arr (NULL,         c_job_good,  '300'),
                               L1_chr_arr (c_dep_id_1,   c_job_bad,   '400'),
                               L1_chr_arr (c_dep_id_2,   c_job_good,  '500')
                                             )
                        );

  c_exp_2lis            CONSTANT L2_chr_arr := L2_chr_arr (
                                               L1_chr_arr (
                                       Utils.List_Delim (c_ln_pre || '1',   c_dep_nm_1, NULL,            '1000', '.67',   '.4'),
                                       Utils.List_Delim (c_ln_pre || '2',   c_dep_nm_1, c_ln_pre || '1', '2000',  '1.33', '.8')
                                               ),
                                               L1_chr_arr (
                                       Utils.List_Delim (c_ln_pre || '1',   c_dep_nm_1, NULL,            '1000', '.67',  '.33'),
                                       Utils.List_Delim (c_ln_pre || '2',   c_dep_nm_1, c_ln_pre || '1', '2000',  '1.33', '.67'),
                                       Utils.List_Delim (c_ln_pre || '5',   c_dep_nm_2, c_ln_pre || '1', '5000',  '1',    '1.67')
                                               ),
                                               L1_chr_arr (
                                       Utils.List_Delim (c_ln_pre || '1',   c_dep_nm_1, NULL,            '1000', '.67',   '.33'),
                                       Utils.List_Delim (c_ln_pre || '2',   c_dep_nm_1, c_ln_pre || '1', '2000',  '1.33', '.67')
                                               ),
                                               tt_Utils.c_empty_list
                        );

  c_scenario_ds_lis     CONSTANT L1_num_arr := L1_num_arr (1, 2, 2, 3);
  c_scenario_lis        CONSTANT L1_chr_arr := L1_chr_arr (
                               'DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep',
                               'DS-2, testing same as 1 but with extra emp in another dep',
                               'DS-2, passing ''WHERE dep=10''',
                               'DS-3, Salaries total 1500 (< threshold of 1600)');

  c_inp_group_lis       CONSTANT L1_chr_arr := L1_chr_arr ('Employee', 'Where');
  c_inp_field_2lis      CONSTANT L2_chr_arr := L2_chr_arr (
                                                        L1_chr_arr (
                                                                '*Employee Id',
                                                                'Last Name',
                                                                'Email',
                                                                'Hire Date',
                                                                'Job',
                                                                '*Salary',
                                                                '*Manager Id',
                                                                '*department Id'),
                                                        L1_chr_arr (
                                                                'Where')
  );
  c_out_field_2lis      CONSTANT L2_chr_arr :=  L2_chr_arr ( L1_chr_arr (
                                'Name',
                                'Department',
                                'Manager',
                                '*Salary',
                                '*Salary Ratio (dep)',
                                '*Salary Ratio (overall)'));

  l_act_2lis                      L2_chr_arr := L2_chr_arr();
  c_ms_limit            CONSTANT PLS_INTEGER := 1;
  l_timer_set                    PLS_INTEGER;
  l_inp_3lis                     L3_chr_arr := L3_chr_arr();

Notes on declare section

  • Data sets, scenarios, expected values etc. are stored in generic arrays, where:
    • L1_chr_arr is type of array of VARCHAR2(4000), same as standard type SYS.ODCIVarchar2List
    • L2_chr_arr is a type of array of L1_chr_arr
    • L3_chr_arr is a type of array of L2_chr_arr

Setup section

  PROCEDURE Setup_Array IS
  BEGIN

    l_act_2lis.EXTEND (c_exp_2lis.COUNT);
    l_inp_3lis.EXTEND (c_exp_2lis.COUNT);

    FOR i IN 1..c_exp_2lis.COUNT LOOP

      l_inp_3lis (i) := L2_chr_arr();
      l_inp_3lis (i).EXTEND(2);
      l_inp_3lis(i)(2) := L1_chr_arr (c_where_lis(i));

    END LOOP;

  END Setup_Array;

  /***************************************************************************************************

  Setup_DB: Create test records for a given scenario for testing view

  ***************************************************************************************************/
  PROCEDURE Setup_DB (p_call_ind           PLS_INTEGER,   -- scenario index
                      x_inp_lis        OUT L1_chr_arr) IS -- input list, first group, employees

    l_emp_id            PLS_INTEGER;
    l_mgr_id            PLS_INTEGER;
    l_len_lis           L1_num_arr := L1_num_arr (1, -11, -13, -10, 10, -10);

  BEGIN

    Utils.Heading ('Employees created in setup: DS-' || p_call_ind || ' - ' || c_dataset_3lis (p_call_ind)(1)(1));
    Utils.Col_Headers (L1_chr_arr ('#', 'Employee id', 'Department id', 'Manager', 'Job id', 'Salary'), l_len_lis);
    x_inp_lis := L1_chr_arr();
    x_inp_lis.EXTEND (c_dataset_3lis (p_call_ind).COUNT - 1);
    FOR i IN 2..c_dataset_3lis (p_call_ind).COUNT LOOP

      l_emp_id := DML_API_TT_HR.Ins_Emp (
                            p_emp_ind  => i - 1,
                            p_dep_id   => c_dataset_3lis (p_call_ind)(i)(1),
                            p_mgr_id   => l_mgr_id,
                            p_job_id   => c_dataset_3lis (p_call_ind)(i)(2),
                            p_salary   => c_dataset_3lis (p_call_ind)(i)(3),
                            x_rec      => x_inp_lis(i - 1));
      Utils.Pr_List_As_Line (L1_chr_arr ((i-1), l_emp_id, Nvl (c_dataset_3lis (p_call_ind)(i)(1), ' '), Nvl (To_Char(l_mgr_id), ' '), c_dataset_3lis (p_call_ind)(i)(2), c_dataset_3lis (p_call_ind)(i)(3)), l_len_lis);
      IF i = 2 THEN
        l_mgr_id := l_emp_id;
      END IF;

    END LOOP;

  END Setup_DB;

Notes on setup section

  • c_dataset_3lis contains the data for all data sets indexed by (data set, record, field)
  • Setup_Array is called once to do some setup on the arrays
  • Setup_DB is called for a single data set at a time in each scenario
  • Description of the data set is contained in the array and printed out
  • Data set is printed out in tabular format. In the most recent version of the utility code, this is not strictly necessary, because all the input data is printed out before the outputs

'Pure' API wrapper procedure

  PROCEDURE Purely_Wrap_API (p_scenario_ds      VARCHAR2,      -- index of input dataset
                             p_where            VARCHAR2,      -- input where clause for
                             x_inp_lis_1    OUT L1_chr_arr,    -- first input group, employees
                             x_act_lis      OUT L1_chr_arr) IS -- generic actual values list (for scenario)
  BEGIN

    Setup_DB (p_scenario_ds, x_inp_lis_1);

    Timer_Set.Increment_Time (l_timer_set, Utils_TT.c_setup_timer);
 
    x_act_lis := Utils_TT.Get_View (
                            p_view_name         => c_view_name,
                            p_sel_field_lis     => c_sel_lis,
                            p_where             => p_where,
                            p_timer_set         => l_timer_set);
    ROLLBACK;

  END Purely_Wrap_API;

Notes on 'Pure' API wrapper procedure

  • Setup_DB is called to create the data set for the scenario
  • Get_View returns the results of the query on the view as 2-level array
  • Get_View rolls back after getting the results, so the inserted test records are removed from the database

Main section

BEGIN
--
-- Every testing main section should be similar to this, with array setup, then loop over scenarios
-- making a 'pure'(-ish) call to specific, local Purely_Wrap_API, with single assertion call outside
-- the loop
--
  l_timer_set := Utils_TT.Init (c_proc_name);
  Setup_Array;

  FOR i IN 1..c_exp_2lis.COUNT LOOP

    Purely_Wrap_API (c_scenario_ds_lis(i), c_where_lis(i), l_inp_3lis(i)(1), l_act_2lis(i));

  END LOOP;

  Utils_TT.Is_Deeply (c_proc_name, c_scenario_lis, l_inp_3lis, l_act_2lis, c_exp_2lis, l_timer_set, c_ms_limit,
                      c_inp_group_lis, c_inp_field_2lis, c_out_group_lis, c_out_field_2lis);

EXCEPTION

  WHEN OTHERS THEN
    Utils.Write_Other_Error;
    RAISE;

END tt_HR_Test_View_V;

Notes on main section

  • It's quite short isn't it 🙂
  • Setup is called to do array setup
  • Main section loops over the scenarios calling Purely_Wrap_API
  • Is_Deeply is called to do all the assertions within nested loops, then print the results

Utils_TT - Test Utility Procedures
We will include only one procedure from this package in the body of the article. See gitHub link for the full code.
Is_Deeply - to check results from testing

PROCEDURE Is_Deeply (p_proc_name                 VARCHAR2,      -- calling procedure
                     p_test_lis                  L1_chr_arr,    -- test descriptions
                     p_inp_3lis                  L3_chr_arr,    -- actual result strings
                     p_act_3lis                  L3_chr_arr,    -- actual result strings
                     p_exp_3lis                  L3_chr_arr,    -- expected result strings
                     p_timer_set                 PLS_INTEGER,   -- timer set index
                     p_ms_limit                  PLS_INTEGER,   -- call time limit in ms
                     p_inp_group_lis             L1_chr_arr,    -- input group names
                     p_inp_fields_2lis           L2_chr_arr,    -- input fields descriptions
                     p_out_group_lis             L1_chr_arr,    -- output group names
                     p_fields_2lis               L2_chr_arr) IS -- test fields descriptions

  l_num_fails_sce                L1_num_arr :=  L1_num_arr();
  l_num_tests_sce                L1_num_arr :=  L1_num_arr();
  l_tot_fails                    PLS_INTEGER := 0;
  l_tot_tests                    PLS_INTEGER := 0;

.
.
.
(private procedures - see gitHub project, https://github.com/BrenPatF/trapit_oracle_tester, for full code listings)
.
.
.
BEGIN

  Detail_Section (l_num_fails_sce, l_num_tests_sce);
  Summary_Section (l_num_fails_sce, l_num_tests_sce, l_tot_fails, l_tot_tests);
  Set_Global_Summary (l_tot_fails, l_tot_tests + 1);

END Is_Deeply;

Notes on Is_Deeply

  • This is the base version of Is_Deeply with 3-level arrays of expected and actuals
  • An inner loop asserts actual values (which are records) against expected
  • The final assertion is against average call time
  • It is expected that all assertion within a test procedure will be via a single call to one of the versions of this procedure, making a big reduction in code compared with traditional unit testing approaches
  • After final assertion a call is made to write out all the results, by scenario, with all inputs printed first, followed by actuals (and expected, where they differ); this means that the test outputs now become precise and accurate documents of what the program does

Get_View - run a query dynamically on a view and return result set as array of strings

FUNCTION Get_View (p_view_name         VARCHAR2,               -- name of view
                   p_sel_field_lis     L1_chr_arr,             -- list of fields to select
                   p_where             VARCHAR2 DEFAULT NULL,  -- optional where clause
                   p_timer_set         PLS_INTEGER)            -- timer set handle
                   RETURN              L1_chr_arr IS           -- list of delimited result records

  l_cur            SYS_REFCURSOR;
  l_sql_txt        VARCHAR2(32767) := 'SELECT Utils.List_Delim (L1_chr_arr (';
  l_result_lis     L1_chr_arr;
  l_len            PLS_INTEGER;

BEGIN

  FOR i IN 1..p_sel_field_lis.COUNT LOOP

    l_sql_txt := l_sql_txt || p_sel_field_lis(i) || ',';

  END LOOP;

  l_sql_txt := RTrim (l_sql_txt, ',') || ')) FROM ' || p_view_name || ' WHERE ' || Nvl (p_where, '1=1 ') || 'ORDER BY 1';

  OPEN l_cur FOR l_sql_txt;

  FETCH l_cur BULK COLLECT -- ut, small result set, hence no need for limit clause
   INTO l_result_lis;

  CLOSE l_cur;

  Timer_Set.Increment_Time (p_timer_set, Utils_TT.c_call_timer);
  ROLLBACK;
  RETURN tt_Utils.List_or_Empty (l_result_lis);

END Get_View;

Notes on Get_View

  • A query string is constructed from the input list of fields and optional where clause
  • The fields are concatenated with delimiters and returned into an array of strings
  • A rollback occurs to remove any test data created, so as not to interfere with any subsequent call
  • If no data was returned from the query, we return a default 1-record listing containing the string 'EMPTY'

Package block section

BEGIN

  DBMS_Application_Info.Set_Client_Info (client_info => 'TT');
  Utils.c_session_id_if_TT := SYS_Context ('userenv', 'sessionid');

END Utils_TT;

Notes on package block section

  • client_info is set to 'TT', meaning the session operates in test mode
  • The session id is stored in a package variable
  • This id is referenced in the testing views and in insertion of test records with utid column

Conclusions






 

Design Patterns for Database API Testing 2: Views 1 – Design

Last October I gave a presentation on database unit testing with utPLSQL, Oracle Unit Testing with utPLSQL. I mentioned design patterns as a way of reducing the effort of building unit tests and outlined some strategies for coding them effectively.

In the current set of articles, I develop the ideas further, starting from the idea that all database APIs can be considered in terms of the axes:

  • direction (i.e. getter or setter, noting that setters can also ‘get’)
  • mode (i.e. real time or batch)

For each cell in the implied matrix, I construct an example API (or view) with specified requirements against Oracle’s HR demo schema, and use this example to construct a testing program with appropriate scenarios as a design pattern. Concepts and common patterns and anti-patterns in automated API testing are discussed throughout, and these are largely independent of testing framework used. However, the examples use my own lightweight independent framework that is designed to help avoid many API testing anti-patterns. The code is available on GitHub here, BrenPatF/trapit_oracle_tester, and includes both framework and design pattern examples.

Behind the four examples, there is an underlying design pattern that involves wrapping the API call in a ‘pure’ procedure, called once per scenario, with the output ‘actuals’ array including everything affected by the API, whether as output parameters, or on database tables, etc. The inputs are also extended from the API parameters to include any other effective inputs. Assertion takes place after all scenarios and is against the extended outputs, with extended inputs also listed. This concept of the ‘pure’ function, central to Functional Programming, has important advantages in automated testing. I explained the concepts involved in a presentation at the Oracle User Group Ireland Conference in March 2018:

The Database API Viewed As A Mathematical Function: Insights into Testing


In this 2-part article, I present a design pattern for testing views. I start by discussing when and how to test views. Unlike in the design paatern of the first article in the series, Design Patterns for Database API Testing 1: Web Service Saving – Design, test data has to be created during testing of views, and a very general approach to creating and selecting the test data is proposed. The use case for the design pattern is described, and scenarios and sub-scenarios are defined conceptually. Finally, the output from the testing is presented, with notes.

The second post provides some code extracts, with notes: Design Patterns for Database API Testing 2: Views 2 – Code.

When to Test Views

Views can be simple or complex, or, as I categorise them in Brendan’s 2-Page Oracle Programming Standards, thin or thick, where thick views include table joins and thinviews don’t. Thin views do not normally require testing while it may or may not be appropriate to test thick views.

As explained in the second part of the first article mentioned above, method-based testing is a bad idea, and occurs when the test suite is based on testing all the methods in a package, rather than units of (external) behaviour (often corresponding to procedures prefixed AIP in a common naming convention). Similarly, we can consider views in the same way as methods and ask whether they represent testable units of behaviour or are merely internal code structures, which should not normally have individual automated tests for the reasons given there.

Good examples of views that should be tested would be those that form the basis of complex data extraction to file, by ETL tools such as Informatica, or those that form the basis of reporting tools such as Business Objects. In fact, it is very good practice to place SQL for these tools into views precisely so that they can be tested.

How to Test Views Using a PL/SQL Testing Framework

In order to leverage a PL/SQL API testing framework to also test views, the API test package procedures call a library procedure passing the name of the relevant view: The library procedure returns the result of querying the view as an array of delimited strings, and the API test procedures then compare the results against their own expected results.

Each API test procedure will have its own setup local procedure to create test data, and we need to discuss the issue of distinguishing test data from pre-existing data.

Test Data

In the earlier article on database save procedures, we did not create any test data within the testing code itself, but the base procedure did create data, and those were queried back for assertion. In order to select only the data created by the procedure call a prefix was used in one of the string fields which was assumed not to exist already. This is a workable approach in some cases, but may not be possible in general. Let us consider the different types of database data that may affect our testing:

  • Data created by the base code being tested
  • Data created by test code to be read by the base code
  • Data not created by test code to be read by base code

In order to verify that the program calls are giving results as expected, the test code needs to know all the data that influence the results, not necessarily just directly created data. Our view testing use case described below has an example where the results depend on an aggregate of all the records on the database. This is a problem when we have a shared database, where we cannot freeze the data at the time of test development. In order to handle this problem, we propose to borrow a technique used in Oracle’s ebusinees applications.

Partitioning Views with System Contexts
In Oracle ebusiness’s multi-org implementations, transactions are partitioned by a numeric identifier for the organization owning the transaction. This org_id value is stored in a column in the base table on transaction creation. Within the application code the base table is not queried directly, but through a view that restricts records returned to those for the organization corresponding to the current role of the application user, which is stored in the userenv system context (this is true up to release 11.5, but the mechanism changed in release 12.1).

See SYS_CONTEXT for information on the system context database feature, and Oracle E-Business Suite Multiple Organizations Implementation Guide (12.1) for release 12.1 multi-org implementation in Oracle ebusiness.

Partitioning Views for Testing
We propose to use views in a similar way to the multi-org views, to restrict records to those created in the testing session, by means of a ttid column on the base table that will hold the session id. The new optional column is added to those tables where this approach is required, and view are created on the tables. Our testing utility package Utils_TT sets a context variable to the value ‘TT’ to signify testing mode, and the session id is set to a package variable in the general utilities package Utils.

Any base code that inserts data into the tables has to check for test mode, and if set, put the session id into the ttid field, and if not, leave it blank. The views use the following clause:

 WHERE (ttid = SYS_Context ('userenv', 'sessionid') OR 
        Substr (Nvl (SYS_Context ('userenv', 'client_info'), 'XX'), 1, 2) != 'TT')

Both test code and base code now query the views instead of the base tables. As the base code to write to the tables has to account for the new column, it is necessary for the column to be added in all instances including production. If this seems a little drastic, consider the importance that you attach to testing, and bear in mind that the earlier, less general, approaches may suffice in many cases. In these design pattern demos I use the general solution.

Schema Structure

In the earlier articles, the base code and test packages were created in the HR schema, with utility packages kept in the custom brendan schema. However, it is more common to use separate schemas for code and data, so we will now place all packages and supporting objects in the brendan schema, and create the testing views there.

Design Pattern Use Case for Testing Views

Modern Oracle SQL is very powerful and can apply complex logic within a single statement, reducing the need for more complex procedural code. In order to show how to test SQL, we will devise a test view, HR_Test_V, having a range of features that we might want to test in general:

  • Inner joins suppress driving records where there is no joining record
  • Outer joins return driving records where there is no joining record
  • Analytic functions that partition by some key, and return aggregates on the returned record set
  • Functions based on aggregates over records that include those not in the returned record set
  • Constraints based on aggregates over records that include those not in the returned record set
  • Constraints on column values

The view functionality can be described in words as:

  • Selected values
    • Employee name, department name, and salary
    • Manager’s name
    • Ratio of employee’s salary to the department average (returned employees only)
    • Ratio of employee’s salary to the average salary of all employees
  • Constraints
    • Exclude employees in job ‘AD_ASST’
    • Exclude employees without a department
    • Do not return any records if the total salary of all employees is below 1600
  • Outer join
    • Include employees both with and without a manager

The view SQL is:

CREATE OR REPLACE VIEW hr_test_view_v AS
WITH all_emps AS (
        SELECT Avg (salary) avg_sal, SUM (salary) sal_tot_g
          FROM employees e
)
SELECT e.last_name, d.department_name, m.last_name manager, e.salary,
       Round (e.salary / Avg (e.salary) OVER (PARTITION BY e.department_id), 2) sal_rat,
       Round (e.salary / a.avg_sal, 2) sal_rat_g
  FROM all_emps a
 CROSS JOIN employees e
  JOIN departments d
    ON d.department_id = e.department_id
  LEFT JOIN employees m
    ON m.employee_id = e.manager_id
 WHERE e.job_id != 'AD_ASST'
   AND a.sal_tot_g >= 1600

Scenarios and Sub-scenarios

Scenario definition
Following our earlier article, we may define a scenario as being the set of all relevant records, both on the database and passed as parameters, to a single program call. API or view testing involves creating one or more scenarios, calling the program (or executing the process) for each scenario, and verifying that the output records are as expected.

Good testing is achieved when the scenarios are chosen to validate as wide a range of behaviours as possible. It is not always, or usually, necessary to create a new scenario for each aspect of behaviour to be tested.

Sub-scenario definition
Often, several features can be tested in the same program call by setting up different records in the scenario that will independently test the different features. For example, in our use case above we can create employees with and without a department, and with and without a manager in the same scenario to test the different types of join.

It may be helpful to think of these separate records, or fields within a record, as corresponding to sub-scenarios, and try to construct scenarios as efficiently as possible without making more calls than necessary.

View Test Output

Data setup section

SCENARIO 1: DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep, Employees created in setup: DS-1 - 4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job
=========================================================================================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1493             10              IT_PROG           1000
2         1494             10        1493  IT_PROG           2000
3         1495                       1493  IT_PROG           3000
4         1496             10        1493  AD_ASST           4000

SCENARIO 2: DS-2, testing same as 1 but with extra emp in another dep, Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
===============================================================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1497             10              IT_PROG           1000
2         1498             10        1497  IT_PROG           2000
3         1499                       1497  IT_PROG           3000
4         1500             10        1497  AD_ASST           4000
5         1501             20        1497  IT_PROG           5000

SCENARIO 3: DS-2, passing 'WHERE dep=10', Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
==================================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1502             10              IT_PROG           1000
2         1503             10        1502  IT_PROG           2000
3         1504                       1502  IT_PROG           3000
4         1505             10        1502  AD_ASST           4000
5         1506             20        1502  IT_PROG           5000

SCENARIO 4: DS-3, Salaries total 1500 (< threshold of 1600), Employees created in setup: DS-3 - As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600
=============================================================================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1507             10              IT_PROG            100
2         1508             10        1507  IT_PROG            200
3         1509                       1507  IT_PROG            300
4         1510             10        1507  AD_ASST            400
5         1511             20        1507  IT_PROG            500

Notes on data setup section

  • There are three data sets, and four scenarios, each of which references a data set
  • The call to set up the data for a scenario writes out all the data created
  • A header provides a description of the features (or sub-scenarios) in the data set
  • In the output above scenarios 2 and 3 use the same data set, DS-2

Results section

SQL>BEGIN

  Utils.Clear_Log;
  Utils_TT.Run_Suite (Utils_TT.c_tt_suite_bren);

EXCEPTION
  WHEN OTHERS THEN
    Utils.Write_Other_Error;
END;
/
SQL> @L_Log_Default

TEXT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------


TRAPIT TEST: TT_View_Drivers.tt_HR_Test_View_V
==============================================

Employees created in setup: DS-1 - 4 emps, 1 dep (10), emp-3 has no dep, emp-4 has bad job
==========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1518             10              IT_PROG           1000
2         1519             10        1518  IT_PROG           2000
3         1520                       1518  IT_PROG           3000
4         1521             10        1518  AD_ASST           4000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1522             10              IT_PROG           1000
2         1523             10        1522  IT_PROG           2000
3         1524                       1522  IT_PROG           3000
4         1525             10        1522  AD_ASST           4000
5         1526             20        1522  IT_PROG           5000

Employees created in setup: DS-2 - As dataset 1 but with extra emp-5, in second dep (20)
========================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1527             10              IT_PROG           1000
2         1528             10        1527  IT_PROG           2000
3         1529                       1527  IT_PROG           3000
4         1530             10        1527  AD_ASST           4000
5         1531             20        1527  IT_PROG           5000

Employees created in setup: DS-3 - As dataset 2 but with salaries * 0.1, total below reporting threshold of 1600
================================================================================================================

#  Employee id  Department id     Manager  Job id          Salary
-  -----------  -------------  ----------  ----------  ----------
1         1532             10              IT_PROG            100
2         1533             10        1532  IT_PROG            200
3         1534                       1532  IT_PROG            300
4         1535             10        1532  AD_ASST            400
5         1536             20        1532  IT_PROG            500

SCENARIO 1: DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep {
===============================================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date  Job      Salary  Manager Id  department Id
            -----------  ---------  -----  ---------  -------  ------  ----------  -------------
                   1518  LN_1       EM_1   09-JUL-16  IT_PROG    1000                         10
                   1519  LN_2       EM_2   09-JUL-16  IT_PROG    2000        1518             10
                   1520  LN_3       EM_3   09-JUL-16  IT_PROG    3000        1518
                   1521  LN_4       EM_4   09-JUL-16  AD_ASST    4000        1518             10

        }
        =

        GROUP Where {
        =============

            Where
            -----


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                      .4
                LN_2  Administration  LN_1       2000                1.33                      .8

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 2: DS-2, testing same as 1 but with extra emp in another dep {
=======================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date  Job      Salary  Manager Id  department Id
            -----------  ---------  -----  ---------  -------  ------  ----------  -------------
                   1522  LN_1       EM_1   09-JUL-16  IT_PROG    1000                         10
                   1523  LN_2       EM_2   09-JUL-16  IT_PROG    2000        1522             10
                   1524  LN_3       EM_3   09-JUL-16  IT_PROG    3000        1522
                   1525  LN_4       EM_4   09-JUL-16  AD_ASST    4000        1522             10
                   1526  LN_5       EM_5   09-JUL-16  IT_PROG    5000        1522             20

        }
        =

        GROUP Where {
        =============

            Where
            -----


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 3, Expected = 3 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                     .33
                LN_2  Administration  LN_1       2000                1.33                     .67
                LN_5  Marketing       LN_1       5000                   1                    1.67

        } 0 failed, of 3: SUCCESS
        =========================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 3: DS-2, passing 'WHERE dep=10' {
==========================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date  Job      Salary  Manager Id  department Id
            -----------  ---------  -----  ---------  -------  ------  ----------  -------------
                   1527  LN_1       EM_1   09-JUL-16  IT_PROG    1000                         10
                   1528  LN_2       EM_2   09-JUL-16  IT_PROG    2000        1527             10
                   1529  LN_3       EM_3   09-JUL-16  IT_PROG    3000        1527
                   1530  LN_4       EM_4   09-JUL-16  AD_ASST    4000        1527             10
                   1531  LN_5       EM_5   09-JUL-16  IT_PROG    5000        1527             20

        }
        =

        GROUP Where {
        =============

            Where
            --------------------------------
            department_name='Administration'

        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 2, Expected = 2 {
        ================================================

            F?  Name  Department      Manager  Salary  Salary Ratio (dep)  Salary Ratio (overall)
            --  ----  --------------  -------  ------  ------------------  ----------------------
                LN_1  Administration             1000                 .67                     .33
                LN_2  Administration  LN_1       2000                1.33                     .67

        } 0 failed, of 2: SUCCESS
        =========================

} 0 failed, of 2: SUCCESS
=========================

SCENARIO 4: DS-3, Salaries total 1500 (< threshold of 1600) {
=============================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Employee Id  Last Name  Email  Hire Date  Job      Salary  Manager Id  department Id
            -----------  ---------  -----  ---------  -------  ------  ----------  -------------
                   1532  LN_1       EM_1   09-JUL-16  IT_PROG     100                         10
                   1533  LN_2       EM_2   09-JUL-16  IT_PROG     200        1532             10
                   1534  LN_3       EM_3   09-JUL-16  IT_PROG     300        1532
                   1535  LN_4       EM_4   09-JUL-16  AD_ASST     400        1532             10
                   1536  LN_5       EM_5   09-JUL-16  IT_PROG     500        1532             20

        }
        =

        GROUP Where {
        =============

            Where
            -----


        }
        =

    OUTPUTS
    =======

        GROUP Select results: Actual = 0, Expected = 0: SUCCESS
        =======================================================

} 0 failed, of 1: SUCCESS
=========================

TIMING: Actual = 48, Expected <= 1: FAILURE
===========================================

SUMMARY for TT_View_Drivers.tt_HR_Test_View_V
=============================================

Scenario                                                                           # Failed  # Tests  Status
---------------------------------------------------------------------------------  --------  -------  -------
DS-1, testing inner, outer joins, analytic over dep, and global ratios with 1 dep         0        2  SUCCESS
DS-2, testing same as 1 but with extra emp in another dep                                 0        3  SUCCESS
DS-2, passing 'WHERE dep=10'                                                              0        2  SUCCESS
DS-3, Salaries total 1500 (< threshold of 1600)                                           0        1  SUCCESS
Timing                                                                                    1        1  FAILURE
---------------------------------------------------------------------------------  --------  -------  -------
Total                                                                                     1        9  FAILURE
---------------------------------------------------------------------------------  --------  -------  -------

Timer Set: TT_View_Drivers.tt_HR_Test_View_V, Constructed at 09 Jul 2016 13:32:42, written at 13:32:42
======================================================================================================
[Timer timed: Elapsed (per call): 0.01 (0.000013), CPU (per call): 0.02 (0.000020), calls: 1000, '***' denotes corrected line below]

Timer       Elapsed         CPU         Calls       Ela/Call       CPU/Call
-------  ----------  ----------  ------------  -------------  -------------
Setup          0.11        0.00             4        0.02675        0.00000
Caller         0.19        0.06             4        0.04750        0.01500
(Other)        0.03        0.03             1        0.02700        0.03000
-------  ----------  ----------  ------------  -------------  -------------
Total          0.32        0.09             9        0.03600        0.01000
-------  ----------  ----------  ------------  -------------  -------------

Notes on results section

  • In a view test there is only one group, namely the selected data set

The second part of the article is here: Design Patterns for Database Unit Testing 2: Views 2 - Code






 

A Template Script for JDBC Calling of Oracle Procedures with Object Array Parameters

Some time ago I wrote an Oracle database package for a web service. The Java developer for the service told me that it was throwing an error when called from Java, although I had unit tested it from PL/SQL. He gave me a small Java driver script to demonstrate the issue, and this allowed the issue to be quickly identified: As both Java and PL/SQL have boolean data types I had considered that a boolean parameter would make sense to pass a boolean value. However, it turns out that this does not work in JDBC, and so I replaced it with an integer parameter.

It occurred to me then that it would be nice if the database developer was able in general to test JDBC compatibility of his or her procedure as a final step after unit testing. To this end I created a more generic example script based on a simple procedure that I wrote against Oracle’s HR demo schema, the same procedure that I used as an example of a unit testing design pattern Design Patterns for Database Unit Testing 2: Web Service Saving – Code

Update, 4 November 2017: I have made a self-contained project on GitHub with both Java and Oracle code, avoiding dependency on my testing project. JDBC Calling of Oracle Procedures with Object Array Parameters on GitHub. I have also added the Oracle code below.

The code below runs against any Oracle instance in which the standard Oracle HR demo schema has been installed. There is a video demonstration of how to install and run it at the end of this article. The procedure has one input and one output object array parameter, and can easily be extended as desired. It requires one jar file in the classpath, ojdbc6.jar, which is available in Oracle client or database installations, and can be run from an IDE such as Eclipse.

Java Code

package jdbcdemo;
/***************************************************************************************************
Name:        Driver.java
Description: This is a Java driver script for Brendan's HR demo web service procedure. It is
             designed to serve as a template for other web service procedures to allow a database
             developer to do a JDBC integration test easily, and can also be used as a starting point
             for Java development.

             The template procedure takes an input array of objects and has an output array of 
             objects. It is easy to update for any named object and array types, procedure and
             Oracle connection. Any other signature types would need additional changes.

	     See 'A Template Script for JDBC Calling of Oracle Procedures with Object Array Parameters'
             
A Template Script for JDBC Calling of Oracle Procedures with Object Array Parameters
Modification History Who When Which What -------------------- ----------- ----- ------------------------------------------------------------- Brendan Furey 14-May-2016 1.0 Created Brendan Furey 04-Nov-2017 1.1 Put into new GitHub project along with Oracle code ***************************************************************************************************/ import java.sql.DriverManager; import java.sql.SQLException; import java.sql.Array; import java.sql.Struct; import oracle.jdbc.OracleTypes; import oracle.jdbc.OracleCallableStatement; import oracle.jdbc.OracleConnection; public class Driver { // Change section 1/2: Replace these constants with your own values private static final String DB_CONNECTION = "jdbc:oracle:thin:hr/hr@localhost:1521/orclpdb"; private static final String TY_IN_OBJ = "TY_EMP_IN_OBJ"; private static final String TY_IN_ARR = "TY_EMP_IN_ARR"; private static final String TY_OUT_ARR = "TY_EMP_OUT_ARR"; private static final String PROC_NAME = "Emp_WS.AIP_Save_Emps"; private static OracleConnection conn; public static void main(String[] argv) { try { getDBConnection (); prOutArray (callProc (inArray ())); } catch (SQLException e) { System.out.println(e.getMessage()); } } private static Array inArray () throws SQLException { // Change section 2/2: Replace [2] with number of test records, and the arrays with their values Struct[] struct = new Struct[2]; struct[0] = conn.createStruct (TY_IN_OBJ, new Object[] {"LN 1", "EM 1", "IT_PROG", 1000}); struct[1] = conn.createStruct (TY_IN_OBJ, new Object[] {"LN 2", "EM 2", "IT_PROG", 2000}); return conn.createARRAY (TY_IN_ARR, struct); } private static Array callProc (Array objArray) throws SQLException { OracleCallableStatement ocs = (OracleCallableStatement) conn.prepareCall ("BEGIN "+PROC_NAME+"(:1, :2); END;"); ocs.setArray (1, objArray); ocs.registerOutParameter (2, OracleTypes.ARRAY, TY_OUT_ARR); ocs.execute (); return ocs.getARRAY (2); } private static void prOutArray (Array arr) throws SQLException { Object[] objArr = (Object[]) arr.getArray(); int j = 0; for (Object rec : objArr) { Object[] objLis = ((Struct)rec).getAttributes (); int i = 0; String recStr = ""; for (Object fld : objLis) { if (i++ > 0) recStr = recStr + '/'; recStr = recStr + fld.toString(); } System.out.println ("Record "+(++j)+": "+recStr); } } private static void getDBConnection () throws SQLException { conn = (OracleConnection) DriverManager.getConnection (DB_CONNECTION); conn.setAutoCommit (false); System.out.println ("Connected..."); } }

Example output

Connected...
Record 1: 239/TWO HUNDRED THIRTY-NINE
Record 2: 240/TWO HUNDRED FORTY

Google Java Style

Oracle Code

/***************************************************************************************************

Author:      Brendan Furey
Description: Script to create objects to demo JDBC procedure calls with object array parameters

         See 'A Template Script for JDBC Calling of Oracle Procedures with Object Array Parameters'
             
A Template Script for JDBC Calling of Oracle Procedures with Object Array Parameters
Modification History Who When Which What -------------------- ----------- ----- ------------------------------------------------------------- Brendan Furey 04-May-2016 1.0 Created Brendan Furey 04-Nov-2017 1.1 Extracted the JDBC demo code from the unit testing project, and put into new GitHub project along with Java code ***************************************************************************************************/ REM Run this script from Oracle's standard HR schema to create objects to demo JDBC procedure calls COLUMN "Database" FORMAT A20 COLUMN "Time" FORMAT A20 COLUMN "Version" FORMAT A30 COLUMN "Session" FORMAT 9999990 COLUMN "OS User" FORMAT A10 COLUMN "Machine" FORMAT A20 SELECT 'Start: ' || dbs.name "Database", To_Char (SYSDATE,'DD-MON-YYYY HH24:MI:SS') "Time", Replace (Substr(ver.banner, 1, Instr(ver.banner, '64')-4), 'Enterprise Edition Release ', '') "Version" FROM v$database dbs, v$version ver WHERE ver.banner LIKE 'Oracle%'; PROMPT Input types creation DROP TYPE ty_emp_in_arr / CREATE OR REPLACE TYPE ty_emp_in_obj AS OBJECT ( last_name VARCHAR2(25), email VARCHAR2(25), job_id VARCHAR2(10), salary NUMBER ) / CREATE TYPE ty_emp_in_arr AS TABLE OF ty_emp_in_obj / PROMPT Output types creation DROP TYPE ty_emp_out_arr / CREATE OR REPLACE TYPE ty_emp_out_obj AS OBJECT ( employee_id NUMBER, description VARCHAR2(500) ) / CREATE TYPE ty_emp_out_arr AS TABLE OF ty_emp_out_obj / CREATE OR REPLACE PACKAGE Emp_WS AS /*************************************************************************************************** Description: HR demo web service code. Procedure saves new employees list and returns primary key plus same in words, or zero plus error message in output list ***************************************************************************************************/ PROCEDURE AIP_Save_Emps (p_ty_emp_in_lis ty_emp_in_arr, x_ty_emp_out_lis OUT ty_emp_out_arr); END Emp_WS; / CREATE OR REPLACE PACKAGE BODY Emp_WS AS PROCEDURE AIP_Save_Emps (p_ty_emp_in_lis ty_emp_in_arr, -- list of employees to insert x_ty_emp_out_lis OUT ty_emp_out_arr) IS -- list of employee results l_ty_emp_out_lis ty_emp_out_arr; bulk_errors EXCEPTION; PRAGMA EXCEPTION_INIT (bulk_errors, -24381); n_err PLS_INTEGER := 0; BEGIN FORALL i IN 1..p_ty_emp_in_lis.COUNT SAVE EXCEPTIONS INSERT INTO employees ( employee_id, last_name, email, hire_date, job_id, salary ) VALUES ( employees_seq.NEXTVAL, p_ty_emp_in_lis(i).last_name, p_ty_emp_in_lis(i).email, SYSDATE, p_ty_emp_in_lis(i).job_id, p_ty_emp_in_lis(i).salary ) RETURNING ty_emp_out_obj (employee_id, To_Char(To_Date(employee_id,'J'),'JSP')) BULK COLLECT INTO x_ty_emp_out_lis; EXCEPTION WHEN bulk_errors THEN l_ty_emp_out_lis := x_ty_emp_out_lis; FOR i IN 1 .. sql%BULK_EXCEPTIONS.COUNT LOOP IF i > x_ty_emp_out_lis.COUNT THEN x_ty_emp_out_lis.Extend; END IF; x_ty_emp_out_lis (SQL%Bulk_Exceptions (i).Error_Index) := ty_emp_out_obj (0, SQLERRM (- (SQL%Bulk_Exceptions (i).Error_Code))); END LOOP; FOR i IN 1..p_ty_emp_in_lis.COUNT LOOP IF i > x_ty_emp_out_lis.COUNT THEN x_ty_emp_out_lis.Extend; END IF; IF x_ty_emp_out_lis(i).employee_id = 0 THEN n_err := n_err + 1; ELSE x_ty_emp_out_lis(i) := l_ty_emp_out_lis(i - n_err); END IF; END LOOP; END AIP_Save_Emps; END Emp_WS; /

Here is a demo of installing and running the code, Oracle and Java:






 

Design Patterns for Database API Testing 1: Web Service Saving 1 – Design

Last October I gave a presentation on database unit testing with utPLSQL, Oracle Unit Testing with utPLSQL. I mentioned design patterns as a way of reducing the effort of building unit tests and outlined some strategies for coding them effectively.

In the current set of articles, I develop the ideas further, starting from the idea that all database APIs can be considered in terms of the axes:

  • direction (i.e. getter or setter, noting that setters can also ‘get’)
  • mode (i.e. real time or batch)

For each cell in the implied matrix, I construct an example API (or view) with specified requirements against Oracle’s HR demo schema, and use this example to construct a testing program with appropriate scenarios as a design pattern. Concepts and common patterns and anti-patterns in automated API testing are discussed throughout, and these are largely independent of testing framework used. However, the examples use my own lightweight independent framework that is designed to help avoid many API testing anti-patterns. The code is available on GitHub here, BrenPatF/db_unit_test, and includes both framework and design pattern examples.

Behind the four examples, there is an underlying design pattern that involves wrapping the API call in a ‘pure’ procedure, called once per scenario, with the output ‘actuals’ array including everything affected by the API, whether as output parameters, or on database tables, etc. The inputs are also extended from the API parameters to include any other effective inputs. Assertion takes place after all scenarios and is against the extended outputs, with extended inputs also listed. This concept of the ‘pure’ function, central to Functional Programming, has important advantages in automated testing. I explained the concepts involved in a presentation at the Oracle User Group Ireland Conference in March 2018:

The Database API Viewed As A Mathematical Function: Insights into Testing


In this first example, I present a design pattern for web service ‘save’ procedures by means of a conceptual discussion, together with a working example of base code and test code for a procedure to save new employees in Oracle’s well-known HR demonstration schema.

Design patterns involve abstraction and conceptual separation of general features of a situation from the particular. Therefore we will start with a fairly abstract discussion of automated API testing for the database, before proceeding to discuss the use case in question, describe the test cases, and show the results. The code is presented in another article, Design Patterns for Database API Testing 1: Web Service Saving 2 – Code. That article lists various extremely common antipatterns in database unit testing, and how they may be avoided. The code itself centralises as much as possible in order to make specific test code as small as possible, and is structured very differently from most unit testing code that I have seen.

This is the first in a series of four design patterns, with the second described again in two parts, starting here, Design Patterns for Database API Testing 2: Views 1 – Design. For an overview of the testing framework see TRAPIT – TRansactional API Testing in Oracle.

General Discussion of Database Unit Testing

The underlying functionality for unit testing could be described logically as:

  • Given a list of test inputs, X and a list of expected outputs, E, for function F:
  • For each x in X, with e in E:
    • Apply y = F(x)
    • Assert y = e

As the Functional Programming community knows well, functions having well-defined parameter inputs, returning values as outputs, and with no ‘side-effects’, are the easiest to test reliably. The difficulty with database unit testing is that most use cases do not fall into that category; instead, database procedures can read from and write to the database as well as using input and output parameters. This means that theoretically the inputs and outputs could include the whole database (at least); furthermore the database is a shared resource, so other parties can alter the data we are dealing with. One important consequence of these facts is that much of the thinking on best practices for unit testing, coming as it does from the non-database world, is not applicable here. So what to do?

Pragmatic testing

To make progress, we note that our purpose with unit testing is not to formally prove that our programs work, but rather includes the following aims:

  • Improve code quality within a Test Driven Development (TDD) approach
  • Provide regression tests to allow safe code-refactoring
  • Detect quickly changes external to the code that cause it to fail, such as reference data changes

That being so, we can note the following guidelines:

  • Testing code is written, as part of TDD, by the developer of the base code, who can identify the relevant database inputs and outputs
  • Some, but not necessarily all, test data may be created in a setup step; static reference data that are required for the code to work usually should not be created in setup
  • Testing code should be instrumented and logged liberally
  • The base code should be timed and a time limit included in the testing; this will help to quickly identify issues such as necessary indexes being dropped
  • Consideration should be given to running the unit test suites in performance and other instances

Design Pattern Use Case for Web Service Save Procedure

  • Purpose of procedure is to save a set of new records to a database table
  • Surrogate primary key is generated
  • Input is an array of objects with the records to be saved
  • Output is an array of objects containing the new primary key plus a description
  • For records failing validation, zero is returned, plus the error message, and the valid records will still be saved

ERD of Input and Output Data Structures in Relation to Scenarios

Unit Testing - ERD

  • In the diagram, a scenario corresponds to a web service call with a set of input records
  • The result of the call can be described as a set of output groups, each having a set of records
  • In our case the output array and the base table form two output groups, with a global group for average call timing
  • The logical diagram in terms of sets of records can be translated into an array structure diagram

Unit Testing - ASD

If we follow a similarly generic approach at the coding level, it becomes very easy to extend a simple example by adding groups, fields and records as necessary.

General Unit Test Design Process

The design process involves two high level steps

  • Identify a good set of scenarios with corresponding input records
  • Identify the expected outputs for each output group identified, for each scenario (there will also be a global group, for timing)

Design Pattern Scenarios

Unit Testing - HR

The procedure inserts records in Oracle’s HR employees table, and we identify four test scenarios:

  1. Passing a single valid record
  2. Passing a single invalid record
  3. Trying to pass a record with an invalid type
  4. Passing multiple valid records, and one invalid record

Design Pattern Output Groups

  • Records inserted into table employees
  • Records returned in output parameter array
  • Timing of average call

Test Results Output

The output below is for a failing run, where the time limit is breached, and I also have deliberately entered incorrect expected values for two records, to show the difference in formatting between success and failure output group records. I like to include the output tables on completion of development in my technical design document.

TRAPIT TEST: TT_Emp_WS.tt_AIP_Save_Emps
=======================================

SCENARIO 1: 1 valid record {
============================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 1  EM 1   IT_PROG    1000

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 1, Expected = 1 {
        ==========================================

            F?  Employee id  Name  Email  Job      Salary
            --  -----------  ----  -----  -------  ------
                       1513  LN 1  EM 1   IT_PROG    1000

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Output array: Actual = 1, Expected = 1 {
        ==============================================

            F?  Employee id  Description
            --  -----------  ----------------------------------
                       1513  ONE THOUSAND FIVE HUNDRED THIRTEEN

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 2: 1 invalid job id {
==============================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 2  EM 2   NON_JOB    1500

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 0, Expected = 0: SUCCESS
        =================================================

        GROUP Output array: Actual = 1, Expected = 1 {
        ==============================================

            F?  Employee id  Description
            --  -----------  -------------------------------------------------------------------
                          0  ORA-02291: integrity constraint (.) violated - parent key not found

        } 0 failed, of 1: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 3: 1 invalid number {
==============================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 3  EM 3   IT_PROG   2000x

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 0, Expected = 0: SUCCESS
        =================================================

        GROUP Output array: Actual = 0, Expected = 0: SUCCESS
        =====================================================

        GROUP Exception: Actual = 1, Expected = 1 {
        ===========================================

            F?  Error message
            --  -------------------------------------------------------------------------------
                ORA-06502: PL/SQL: numeric or value error: character to number conversion error

        } 0 failed, of 1: SUCCESS
        =========================

} 0 failed, of 3: SUCCESS
=========================

SCENARIO 4: 2 valid records, 1 invalid job id (2 deliberate errors) {
=====================================================================

    INPUTS
    ======

        GROUP Employee {
        ================

            Name  Email  Job      Salary
            ----  -----  -------  ------
            LN 4  EM 4   IT_PROG    3000
            LN 5  EM 5   NON_JOB    4000
            LN 6  EM 6   IT_PROG    5000

        }
        =

    OUTPUTS
    =======

        GROUP Employee: Actual = 2, Expected = 3 {
        ==========================================

            F?  Employee id  Name  Email  Job      Salary
            --  -----------  ----  -----  -------  ------
            F          1515  LN 4  EM 4   IT_PROG    3000
            >          1515  LN 4  EM 4   IT_PROG    1000
                       1517  LN 6  EM 6   IT_PROG    5000
            F
            >          1517  LN 6  EM 6   IT_PROG    5000

        } 2 failed, of 3: FAILURE
        =========================

        GROUP Output array: Actual = 3, Expected = 3 {
        ==============================================

            F?  Employee id  Description
            --  -----------  -------------------------------------------------------------------
                       1515  ONE THOUSAND FIVE HUNDRED FIFTEEN
                          0  ORA-02291: integrity constraint (.) violated - parent key not found
                       1517  ONE THOUSAND FIVE HUNDRED SEVENTEEN

        } 0 failed, of 3: SUCCESS
        =========================

        GROUP Exception: Actual = 0, Expected = 0: SUCCESS
        ==================================================

} 2 failed, of 7: FAILURE
=========================

TIMING: Actual = 222, Expected <= 2: FAILURE
============================================

SUMMARY for TT_Emp_WS.tt_AIP_Save_Emps
======================================

Scenario                                                 # Failed  # Tests  Status
-------------------------------------------------------  --------  -------  -------
1 valid record                                                  0        3  SUCCESS
1 invalid job id                                                0        3  SUCCESS
1 invalid number                                                0        3  SUCCESS
2 valid records, 1 invalid job id (2 deliberate errors)         2        7  FAILURE
Timing                                                          1        1  FAILURE
-------------------------------------------------------  --------  -------  -------
Total                                                           3       17  FAILURE
-------------------------------------------------------  --------  -------  -------

Timer Set: TT_Emp_WS.tt_AIP_Save_Emps, Constructed at 09 Jul 2016 13:32:40, written at 13:32:42
===============================================================================================
[Timer timed: Elapsed (per call): 0.01 (0.000013), CPU (per call): 0.02 (0.000020), calls: 1000, '***' denotes corrected line below]

Timer       Elapsed         CPU         Calls       Ela/Call       CPU/Call
-------  ----------  ----------  ------------  -------------  -------------
Setup          0.04        0.01             1        0.03700        0.01000
Caller         0.67        0.03             3        0.22200        0.01000
SELECT         0.60        0.00             3        0.19933        0.00000
(Other)        0.41        0.03             1        0.40500        0.03000
-------  ----------  ----------  ------------  -------------  -------------
Total          1.71        0.07             8        0.21325        0.00875
-------  ----------  ----------  ------------  -------------  -------------

Notes on output

  • In the event of the suite failing, as here, the utility code ensures that an Oracle error is generated. This can then be trapped by a calling Unix script from a scheduled Jenkins job to send out emails

Conclusions

  • A design pattern has been presented for database web service save procedures, with scenarios and output results
  • The implementation (presented in the part 2 article) was against an Oracle database publicly available demonstration schema, and used Brendan's database API testing framework
  • The main ideas could be applied with any database technology and any testing framework
  • It is suggested that any proposed alternative testing framework can be compared by implementing this design pattern, or similar
  • Three more design patterns are presented in further articles in this series, the second being for testing of views (Design Patterns for Database Unit Testing 2: Views 1 - Design)