Good design for modifying various combination of columns in a table

问题

I have a database table with multiple (~ 15) columns.

I have requirement where where data is updated in a row, in combinations of columns, for the given primary key.

Eg.

Update Col1,Col2,Col3 where PK is something

Update Col4 where PK is something

Update Col5,Col7,Col9,Col10 where PK is something

Update Col5,Col10 where PK is something

I used Python and wrote raw queries to handle such cases. I wrote separate functions to manage each requirements. They look like this :

requirement1( Col5,val1,Col7,val2,Col9,val3,Col10,val4 )
{
    Update TABLE SET (Col5,Col7,Col9,Col10) VALUES (val1,val2,val3,val4)
}

requirement2( Col5,val1,Col10,val2 )
{
    Update TABLE SET (Col5,Col10) VALUES (val1,val2)
}

etc.

But the number of such cases has increased to too many. Is there a good design to handle this ?

Something like a general function :

UpdateTableValue(ColumnName1,ColumnValue1,ColumnName2,ColumnValue2, and so on )
{
    Update TABLE SET (ColumnName1,ColumnName2.. ) VALUES (ColumnValue1,ColumnValue2 .. )
}

And then user could just call :

UpdateTableValue(Col5,val1,Col7,val2,Col9,val3,Col10,val4)
UpdateTableValue(Col7,val,Col3,val2)

P.S. I am not asking for a good DB design to handle this (so no OR mapping etc.). I want to understand a good design from the query wrapper point of view which could handle this.

回答1:

What about something like:

def update_table_value(*args):
    sql = "Update TABLE SET ({}) VALUES ({}) "
    sql.format(",".join([args[0::2]]),
               ",".join([args[1::2]]))
return sql

note that:

args takes all the arguments that you put into the function and creates a list out of them named args
args[0::2] picks ever 2nd entry in the list

来源：https://stackoverflow.com/questions/34943407/good-design-for-modifying-various-combination-of-columns-in-a-table

标签

python

design-patterns

database-design