Scribd
Upload
47 views

How Things Work - ISD Development Approach

The document outlines an approach for developing an ISD interface. It involves analyzing existing interfaces, designing ETL processes, developing tasks for data synchronization and mapping, unit testing tasks, and combining tasks into schedules. Examples of tasks involving data synchronization and complex transformations are provided.

Uploaded by

abraham_sse1594
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
47 views

How Things Work - ISD Development Approach

The document outlines an approach for developing an ISD interface. It involves analyzing existing interfaces, designing ETL processes, developing tasks for data synchronization and mapping, unit testing tasks, and combining tasks into schedules. Examples of tasks involving data synchronization and complex transformations are provided.

Uploaded by

abraham_sse1594
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

ISD Development Approach:

The purpose of this document is to outline how to develop an ISD interface. A


high-level guideline is given below:

Analyze Reyes Holdings inventory of existing UltiPro interfaces.

Design each UltiPro interface in to multiple units of Extract Transform


Load [ETL] process which may a Data Synchronization task or Mapping
Design/ Mapping configuration (DS or MC).

Develop each identified units of ETL process (DS or MC) in ISD cloud
and combine multiple ETL process units in to one Schedule called
{task flow} . Each task flow is equivalent of existing sql interfaces.

Unit test each {task}.

The approach has been discussed on 8/2/2016 with Reyes team and has
been implemented for about 45 interfaces as of 9/2/2016. Some details
regarding the approach is given below with examples:

Data Synchronization Tasks:


When the integration step need to synchronize data between source and
target with no or minimal transformations or joins without any complex left
outer joins then these kind of mapping approach followed. Target can be
either table and or flat file.

Mapping Configuration Tasks:


When the integration step involves source with minimal to complex
transformations and with minimal to complex outer joins then this kind of
mapping approach is followed. Target can be table or flat file with appended
dates.
In this approach we have researched on two kinds of approach:
1. Query Override Approach:

In this approach we have made use of the power of sql server to


perform all the transformations at the database end and passed the
transformed values to expression transformation and to target tables/target
files. This approach proves to be an advantage for support team to trouble
shoot and fix when there arise any issue in future,
Example: reyes_verifyx_export
Mapping Configuration:

Mapping Design:

Define Source and select source type Query

Source DML:
SELECT
RTRIM(ssn) [ssn]
, RTRIM(firstname) [fname]
, RTRIM(REPLACE(lastname,',','') ) [lname]
, CASE eestatus WHEN 'A' THEN 'A'
WHEN 'L' THEN 'A'
WHEN 'R' THEN 'I'
WHEN 'T' THEN 'I'
ELSE 'I'
END empstatus
, CONVERT(CHAR(10),dateoflasthire, 101) lasthired
, CASE WHEN dateoftermination IS NOT NULL THEN
CONVERT(VARCHAR(10),dateoftermination, 101) ELSE '' END termdate
, CASE WHEN RTRIM(jobdesc) like 'Conv%' THEN 'TERM'
ELSE RTRIM(jobdesc) END jobtitle
, CASE eetype WHEN 'REG' THEN 'F'
ELSE 'P'
END [emptype]
, CAST(CASE WHEN salaryorhourly = 'H' THEN hourlyrate
WHEN salaryorhourly = 'S' THEN annualsalary
ELSE ''
END as numeric(12,2)) salrate
, salaryorhourly salmode
, CASE WHEN left(payfrequency,1) = 'B' THEN 'BI'
WHEN left(payfrequency,1) = 'W' THEN 'W'
ELSE ''
END payfreq
, '' column_l
, ISNULL(CAST(current_year AS INT) ,YEAR(GETDATE())) current_year
, ISNULL(CAST(cur_yr_base AS INT) ,0) cur_yr_base
, ISNULL(CAST(cur_yr_ovt AS INT) ,0) cur_yr_ovt
, ISNULL(CAST(cur_yr_bonus AS INT) ,0) cur_yr_bonus
, ISNULL(CAST(cur_yr_gross AS INT),0) cur_yr_gross

, ISNULL(CAST([previous_year] AS INT) , YEAR(GETDATE()) -1) [previous_year]


, ISNULL(CAST([prev_yr_base] AS INT),0) [prev_yr_base]
, ISNULL(CAST([prev_yr_ovt] AS INT),0) [prev_yr_ovt]
, ISNULL(CAST([prev_yr_bonus] AS INT),0) [prev_yr_bonus]
, ISNULL(CAST([prev_yr_gross] AS INT),0) [prev_yr_gross]
, ISNULL(CAST([prev_year_2] AS INT),YEAR(GETDATE()) -2) [prev_year_2]
, ISNULL(CAST([prev_yr_2_base] AS INT),0) [prev_yr_2_base]
, ISNULL(CAST([prev_yr_2_ovt] AS INT),0) [prev_yr_2_ovt]
, ISNULL(CAST([prev_yr_2_bonus] AS INT),0) [prev_yr_2_bonus]
, ISNULL(CAST(prev_yr_2_gross AS INT),0) prev_yr_2_gross
, '' block
, '' column_ac
, '' column_ad
, CONVERT(CHAR(10),dateoforiginalhire, 101) orig_hiredate
, RTRIM(eenumber) [empnum]
, '' [svc_date]
, '' column_ah
, '' column_ai
, '' column_aj
, '' column_ak
, CASE WHEN company = 'MB' THEN 'MAR1084'
WHEN company = 'MBP' THEN 'MAR1101'
WHEN company = 'MBN' THEN 'MAR1105'
WHEN company IN ('RH','FWM','RHF') THEN 'REY1114'
WHEN company = 'CBS' THEN 'CHI1089'
WHEN company = 'HD' AND location IN ('114','115') THEN 'GAT1090'
WHEN company = 'HD' AND location IN ('116') THEN 'MES1093'
WHEN company = 'HD' AND location IN ('118','119') THEN 'ALL1118'
WHEN company = 'HD' THEN 'HAR1091'
WHEN company = 'HLD' THEN 'HEN1092'
WHEN company = 'PDC' THEN 'PRE1094'
WHEN company = 'PDM' THEN 'PRE1095'
WHEN company = 'PVA' THEN 'PRE1096'
WHEN company = 'CB' THEN 'CRE1103'
WHEN company = 'RFS' THEN 'REI1104'
WHEN company = 'RFL' THEN 'RFL1109'
WHEN company = 'FDC' THEN 'FLO1108'
WHEN company = 'CBY' THEN 'CHE1112'
WHEN company = 'WCD' THEN 'WIN1113'
WHEN company ='RSS' THEN 'REI1119'
WHEN company ='FGC' THEN 'GOL1120'
WHEN company ='GLCC' THEN 'GRE1121'
ELSE ''
END vx_comp_code
FROM
v_ee v
LEFT JOIN (
SELECT prgeeid
, prgcoid
, MAX(CASE WHEN pehpaydate = YEAR(GETDATE()) THEN [pehpaydate] END)

current_year
, MAX(CASE WHEN pehpaydate = YEAR(GETDATE()) THEN [Base Amount] END)
cur_yr_base
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) THEN [Overtime Amount]
END ) cur_yr_ovt
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) THEN [Bonus Amount] END )
cur_yr_bonus
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) THEN [Gross Amount] END)
cur_yr_gross
, MAX(CASE WHEN pehpaydate = YEAR(GETDATE()) - 1 THEN pehpaydate END)
previous_year
, MAX(CASE WHEN pehpaydate = YEAR(GETDATE()) - 1 THEN [Base Amount] END)
[prev_yr_base]
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) - 1 THEN [Overtime Amount]
END ) [prev_yr_ovt]
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) - 1 THEN [Bonus Amount]
END ) [prev_yr_bonus]
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) - 1 THEN [Gross Amount] END)
[prev_yr_gross]
, MAX(CASE WHEN pehpaydate = YEAR(GETDATE()) - 2 THEN [pehpaydate] END)
[prev_year_2]
, MAX(CASE WHEN pehpaydate = YEAR(GETDATE()) - 2 THEN [Base Amount] END)
[prev_yr_2_base]
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) - 2 THEN [Overtime Amount]
END ) [prev_yr_2_ovt]
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) - 2 THEN [Bonus Amount]
END ) [prev_yr_2_bonus]
, MAX(CASE WHEN PehPaydate = YEAR(GETDATE()) - 2 THEN [Gross Amount] END)
[prev_yr_2_gross]
FROM
(
SELECT YEAR(pehpaydate) PehPaydate
, pehEEID prgeeid
, pehCOID prgcoid
, SUM( CASE WHEN RTRIM(pehearncode) = PayCode AND category = 'Base
Amount' THEN ISNULL(pehcuramt,0) END ) [Base Amount]
, SUM( CASE WHEN RTRIM(pehearncode) = PayCode AND category = 'overtime'
THEN ISNULL(pehcuramt,0) END ) [Overtime Amount]
, SUM( CASE WHEN RTRIM(pehearncode) = PayCode AND category = 'Bonus' THEN
ISNULL(pehcuramt,0) END) [Bonus Amount]
, SUM( CASE WHEN PehReportCategory <> 'NCH' AND PehUseDedOffSet = 'N'
THEN ISNULL(pehcuramt,0) END ) [Gross Amount]
FROM pearhist
INNER JOIN Reyes_Verifyx_Codes ON Paycode = pehearncode
WHERE YEAR(pehPayDate) IN (YEAR(GETDATE()), YEAR(GETDATE()) -1,
YEAR(GETDATE()) - 2 )
GROUP BY YEAR(pehpaydate)
, pehEEID
, pehCOID
)t

GROUP BY prgeeid, prgcoid


) S ON s.prgeeid = v.eeid AND s.prgcoid = v.coid
WHERE(dateoftermination IS NULL OR dateoftermination > DATEADD(yy,5,GETDATE()))
AND company IN
('RH','MB','MBP','MBN','CBS','HD','HLD','PDC','PDM','PVA','CB','RFS','RFL','FDC','CBY','
FWM','RHF','WCD','RSS','FGC','GLCC')
AND eenumber NOT IN
('R143','36321','R641','36338','36321','36338','18062','18199','18298','18297','180
89','31918','18299', '14322','22496','21536')

2. Transformations at ISD Approach:


Mapping Configuration:

Implement Business logics in expression transformation in ISD:

3. Split
Source
Query
Approach:
Dissecting
complex
source
query with
multiple
joins would
take increased amount of efforts during the Analysis, development, testing
and release phases. This approach was not preferred because this approach
would take increased amount of time in transition to support personal.

You might also like