ENH/SQL: provide single value to 'dtype' argument in to_sql as dtype for all columns

[jorisvandenbossche]

Idea from http://stackoverflow.com/questions/34383000/pandas-to-sql-all-columns-as-nvarchar

You can specify the sql type using the dtype argument in to_sql. Now, this argument has to be a dict (mapping of column name -> type).
Letting to_sql recognize a single value as the sql type for all columns is not that difficult to add.

So instead of this:

df.to_sql(...., dtype={col_name: sqlalchemy.types.NVARCHAR for col_name in df})

you can do this:

df.to_sql(...., dtype=sqlalchemy.types.NVARCHAR)

[RahulHP]

Will it be alright if I:

1. First check if the dtype passed in is an instance of dict.
2. If Not, create a dictionary the same was as above ie. {col_name: sqlalchemy.types.NVARCHAR for col_name in df}
3. Pass this new dictionary onwards.

[oguzhanogreden]

It seems that this is not documented here: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.to_sql.html

[jorisvandenbossche]

@oguzhanogreden would you like to do a PR to add it?