Merge pandas dataframe on a combination of conditions with AND and OR not Equal

Question

Given: two dataframes below

df1:
| Company  | Package | Badge Number | Work Date  |
|----------|---------|--------------|------------|
| Compnay1 | X       | 1            | 2020-01-01 |
| Company2 | X       | 2            | 2020-01-01 |

df2:
| Company  | Package | Badge Number | Work Date  |
|----------|---------|--------------|------------|
| Compnay1 | X       | 1            | 2020-01-01 |
| Compnay1 | Y       | 1            | 2020-01-01 |
| Company2 | X       | 1            | 2020-01-01 |
| Company2 | Y       | 1            | 2020-01-01 |
| Company2 | X       | 2            | 2020-01-01 |

What's needed: I need to write python code which will be similar to this SQL statement.

SELECT * 
FROM df1
INNER JOIN df2
ON df1.[Badge Number] = df2.[Badge Number]
AND df1.[Work Date] = df2.[Work Date]
AND (df1.[Company] != df2.[Company] OR df1.[Package] != df2.[Package])

result:

| df1.Company | df1.Package | df1.Badge Number | df1.Work Date | df2.Company | df2.Package | df2.Badge Number | df2.Work Date |
|-------------|-------------|------------------|---------------|-------------|-------------|------------------|---------------|
| Compnay1    | X           | 1                | 2020-01-01    | Compnay1    | Y           | 1                | 2020-01-01    |
| Compnay1    | X           | 1                | 2020-01-01    | Company2    | X           | 1                | 2020-01-01    |
| Compnay1    | X           | 1                | 2020-01-01    | Company2    | Y           | 1                | 2020-01-01    |

Can this be done purely in pandas without needed to write SQL queries in the python code?

Answer 1

One idea is use DataFrame.merge :

df = df1.merge(df2, on=['Badge Number','Work Date'])

Ane then filter:

df [(df['Company_x'] != df['Company_y']) | (df['Package_x'] != df['Package_y'])]