how to use an extracted row as a pattern to check other extracted rows

[stvn]

Hi

I'd like to use an extracted row as a pattern in order to check others rows and eliminate duplicates variable in order to keep only variable which does not appear in my previous extraction. I have a csv file as an input with several path.

/Default/Usines/Documents/Articles/Mois/
/Default/Usines/Documents/Articles/Diverses/
/Default/Usines/Documents/Articles/Factures/Payees
/Default/Usines/Documents/Articles/Factures/Factures_Comptable/ClintsPrivees

1.- I extract the first row as: 

/Default/Usines/Documents/Articles/Mois/

/Default/Usines/

/Default/Usines/Documents/

/Default/Usines/Documents/Articles/

/Default/Usines/Documents/Articles/Mois/

2.- I'd like to use that extract in order to check the followings rows extraction. In order to remove duplicates variables. So I will have as a result:

/Default/Usines/

/Default/Usines/Documents/

/Default/Usines/Documents/Articles/

/Default/Usines/Documents/Articles/Mois/

/Diverses/   -> the second row

/Factures/
/Factures/Payees/  -> Third row

/Factures_Comptable/
/Factures_Comptable/ClintsPrivees   -> Fourth row

the attached pictured show how I think my job should be. But the filter on tMap doesn't really work.

Thank you for your help.

Warm regards

[akumar2301 · Accepted solution]

Hello @stvn 

follow below steps ,

Step 1) Normalise all records like you want to do for Line 1 and add a record number e;g;

1;/Default/Usines/

1;/Default/Usines/Documents/

1;/Default/Usines/Documents/Articles/

1;/Default/Usines/Documents/Articles/Mois/

2;/Default/Usines/

----

Step2 ) 2.1° Using tMap , Read from 2nd record and compare it will all record created from Step 1 untill Preveous record ID 

e.g. Read 4th Line , compare 4th line with all the record  till ID is 3 from Step1.

2.2° if  Input record contains  lookup path ,  get the rest of sting 

e.g. for /Default/Usines/Documents/Articles/Factures/Factures_Comptable/ClintsPrivees

contains /Default/Usines/Documents/ so rest of sting is /Articles/Factures/Factures_Comptable/ClintsPrivees

contains /Default/Usines/so rest of sting is  /Documents/Articles/Factures/Factures_Comptable/ClintsPrivees

contains /Default/Usines/Documents/Articles//Factures  so rest of sting is /Factures_Comptable/ClintsPrivees

2.3° get the record having smallest of {rest of string } 

for above example it will be /Factures_Comptable/ClintsPrivees

So at the end of Step 2 you should have 

/Diverses/
/Factures/Payees
/Factures_Comptable/ClintsPrivees

Step 3 ) Normalize step2 output as you did for  step 1

/Diverses/

/Factures
/Factures/Payees

/Factures_Comptable
/Factures_Comptable/ClintsPrivees

Hope this will help