Learning Tabulify - Step 10 - How to compare data resources ?

Learning Tabulify - Step 10 - How to compare data resources ?

Comparison

Data comparison is the corner stone of every development because it validates the data processing.

Tabulify ships with a Data comparison operation that allows you to compare:

The comparison operation is implemented with the data compare command.

Data to compare

For the sake of simplicty, we will compare two csv data resource but you may compare any content data resource such as a SQL table or SQL query

  • The source data resource (also known as the left resource)
Last Name,First Name,Birth Date,Wikipedia Page
Schwarzenegger,Arnold,1947-07-30,https://en.wikipedia.org/wiki/Arnold_Schwarzenegger
Norman,Don,1935-12-25,https://en.wikipedia.org/wiki/Don_Norman
Harbison Carnagey,Dale,1888-11-24,https://en.wikipedia.org/wiki/Dale_Carnegie
Kahneman,Daniel,1934-03-05,https://en.wikipedia.org/wiki/Daniel_Kahneman
  • The target data resource (also known as the right resource)
Last Name,First Name,Birth Date,Wikipedia Page
Schwarzenegger,Arnold,1947-07-30,https://en.wikipedia.org/wiki/Arnold_Schwarzenegger
Norman,Don,1935-12-23,https://en.wikipedia.org/wiki/Don_Norman
Kahneman,Daniel,1934-03-05,https://en.wikipedia.org/wiki/Daniel_Kahneman

The fact that a record is missing can be easily spotted but there is a more subtle difference.

The comparison report

All comparison can be executed with the data compare command by giving as argument the source and target data resource.

tabli data compare [email protected] [email protected]
Comparison Report
Source                 Target                        Equals   DiffCount   RecordCount   
--------------------   ---------------------------   ------   ---------   -----------   
[email protected]   [email protected]   false            3             4

The default report is to show a high level that lists the source and target resources and the result of the comparison:

  • Equals: Does the source and target are equals
  • DiffCount: How many records are differents
  • RecordCount: How many records were compared.

There is a total of 4 records that were compared and 3 were differents. It doesn't seem quite right, let's take a look at it by asking a comparison report at the record level to see the difference in details.

The record comparison report

To get a comparison on the record level, you set the report-level option to the record value.

tabli data compare --report-level record [email protected] [email protected]
Comparison result between source ([email protected]) and target ([email protected]).
Values that are not equals are encapsulated with the characters `***`.
comp_id   comp_origin   comp_comment                                                     comp_diff_id   comp_origin_id   Last Name                 First Name     Birth Date         Wikipedia Page                                        
-------   -----------   --------------------------------------------------------------   ------------   --------------   -----------------------   ------------   ----------------   ---------------------------------------------------   
1                                                                                                       1                Schwarzenegger            Arnold         1947-07-30         https://en.wikipedia.org/wiki/Arnold_Schwarzenegger   
2         Source        Value Diff (Birth Date)                                          1              2                Norman                    Don            ***1935-12-25***   https://en.wikipedia.org/wiki/Don_Norman              
3         Target        Value Diff (Birth Date)                                          1              2                Norman                    Don            ***1935-12-23***   https://en.wikipedia.org/wiki/Don_Norman              
4         Source        Value Diff (Last Name, First Name, Birth Date, Wikipedia Page)   2              3                ***Harbison Carnagey***   ***Dale***     ***1888-11-24***   ***https://en.wikipedia.org/wiki/Dale_Carnegie***     
5         Target        Value Diff (Last Name, First Name, Birth Date, Wikipedia Page)   2              3                ***Kahneman***            ***Daniel***   ***1934-03-05***   ***https://en.wikipedia.org/wiki/Daniel_Kahneman***   
6         Source        Record not found in target                                       3              4                Kahneman                  Daniel         1934-03-05         https://en.wikipedia.org/wiki/Daniel_Kahneman

Data Comparison Report explained

Columns

The report shows 3 columns that starts with the prefix comp:

  • comp_id is a record sequence to be able to locate the record
  • comp_origin gives the origin of the record.
    • Source means that record comes from the source
    • Target means that record comes from the target
    • Blank means that the record is the same in the source and in the target
  • comp_comment is a short explanation of the difference if any.
  • comp_diff_id is the id of a difference. The last number on the report would show the total number of difference found.
  • comp_record_id is the record number (ie row id, line id). This column is present only if a unique colum cannot be defined.

All the columns afterwards are column names from the source.

Values

  • The value that differs are encapsulated with three stars ***
  • The two records (source and target) are inserted when there is a difference.

Report Analysis

In this report, we can see that on the record_id 3, the target record Kahneman was compared to the source record Harbison Carnagey.

This is not right because the source has also a Kahneman record and they should have been compared together.

This is because the comparison was executed by record id and not by last name. The next step will show you how to define the unique column used to drive the comparison.

Data Comparison Report with Unique Column Definition

To improve our comparison, we will define the unique column of our data resource with the --unique-column column option.

tabli data compare --report-level record --unique-column "Last Name" [email protected] [email protected]
Comparison result between source ([email protected]) and target ([email protected]).
Values that are not equals are encapsulated with the characters `***`.
comp_id   comp_origin   comp_comment                 comp_diff_id   Last Name           First Name   Birth Date         Wikipedia Page                                        
-------   -----------   --------------------------   ------------   -----------------   ----------   ----------------   ---------------------------------------------------   
1                                                                   Schwarzenegger      Arnold       1947-07-30         https://en.wikipedia.org/wiki/Arnold_Schwarzenegger   
2         Source        Value Diff (Birth Date)      1              Norman              Don          ***1935-12-25***   https://en.wikipedia.org/wiki/Don_Norman              
3         Target        Value Diff (Birth Date)      1              Norman              Don          ***1935-12-23***   https://en.wikipedia.org/wiki/Don_Norman              
4         Source        Record not found in target   2              Harbison Carnagey   Dale         1888-11-24         https://en.wikipedia.org/wiki/Dale_Carnegie           
5                                                                   Kahneman            Daniel       1934-03-05         https://en.wikipedia.org/wiki/Daniel_Kahneman

This time the comparison has:

  • not compared the source record Harbison Carnagey with the target record Kahneman
  • compared the source record Kahneman with the target record Kahneman

We can see that:

  • we have 2 difference via the value in the comp_diff_id column.
  • the first difference is on the norman record where the birth date differs.
  • the record Harbison Carnagey was not found in the target.

Conclusion

This page end the learning guide of Tabulify at the command line with Tabli where we have learned to perform data operation one command at a time.

Powered by ComboStrap