如何在 MySQL 中复制 pandas function？

Question

Im new to SQL and trying to unlearn what I know in python. I have a script where I connect to the odbc of SSMS to work with data in Python:

import pyodbc
import pandas as pd
#odbc
conn = pyodbc.connect('Driver={SQL Server};'
                      'Server=PMZZ315\RION;'
                      'Database=Warehouse;'
                      'Trusted_Connection=yes;')

cursor = conn.cursor()

df = pd.read_sql_query("SELECT [LetId],[StreetAddressLine1],[CompanyName] FROM Dim.Let", conn)
df

df.head()
#print(df.columns)


# Select duplicate rows except first occurrence based on all columns
duplicateRowsDF = df[df.duplicated(['CompanyName','StreetAddressLine1'])]

#print("Duplicate Rows except first occurrence based on all columns are :")
print(duplicateRowsDF)
duplicateRowsDF.to_csv("duplicateRowsDFodbc.csv")

What function in SQL can substitute the df.duplicated function? All I am trying to do is detect duplicate records ignoring the first instance if the company name and street address are repeated

Reprex of output dataset:

LetId   StreetAddressLine1           CompanyName
32  1451 West Brimson View Court    Palmer 
405 1808 North Lonion Ave           Ozark 
465 4223 Monty Hwy              Alabama 

Answer 1

SQL tables represent unordered sets. Ordering is only provided by columns in the data. There is no "first" without an ordering. Let me assume that letid defines the ordering.

The canonical way in SQL uses row_number() :

select t.*
from (select t.*,
             row_number() over (partition by CompanyName, StreetAddressLine1 order by letid) as seqnum
      from t
     ) t
where seqnum = 1;