Hi all,
So I am trying to create a workflow in order to compare 2 databases with 120 columns and multiple 1000s of rows and try to flag the changes.
Starts with 2 data input tools. Then 2 transpose tools connected to each input file. Then joining the data on the basis of a common identifier (ID number) and then filter tool to filter out as per [Value] = [Right_Value]. All the unchanged cells coming out of True anchor and changed ones coming out of False anchor. (664000+ cells with a change)
Now I want to revert back the whole data from both the anchors (T and F) to the previous format by using cross tab or any other way so I can see the changes in columns corresponding to 1 ID number in all the columns simultaneously. But not getting the desired result, maybe not getting the configuration right.
Have all the company IDs under company ID column header right now. All the column headers are in 2nd column under "Name" column and the corresponding cell values are in 3rd column with column name "Value". For dataset A.
Then side by side to this has xyz ID column header which is actually company ID only in dataset B. All the column headers are in 2nd column under "Right_Name" column and the corresponding cell values are in 3rd column with column name "Right_Value".
Example after filter:
Company ID Name Value xyz ID Right_Name Right_Value
1000 status inactive 1000 status active
1000 security yes 1000 security no
1000 delivery completed 1000 delivery incomplete
1000 Kname xbnmz 1000 Kname sdww
1000 valuation 6227 1000 valuation 3232
This must go in format like:
company ID status security delivery Kname valuation
1000 inactive yes completed xbnmz 6227
1000 active no incomplete sdww 3232
1001
1001
Basically have to group back together as per company ID or xyz ID either right next to each other or top of each other with all the columns and values getting organized together so that we can reconcile data.
If anybody has some better workflow or solution for flagging data by comparing each and every column based on some common identifier (company ID and xyz ID in this case) for such data sets are also welcomed....
