SNOW-687991: Align the table_name quotation in save_as_table() and write_pandas()

[ChuliangXiao]

What is the current behavior?

• snowpark_df.write.save_as_table('abc') creates a table ABC in snowflake
• session.write_pandas(pandas_df, 'abc') creates a table abc in snowflake
• session.write_pandas(pandas_df, 'abc', quote_identifiers=False) creates a table ABC in snowflake

What is the desired behavior?

• session.write_pandas(pandas_df, 'abc') creates a table ABC in snowflake
• session.write_pandas(pandas_df, 'abc', quote_identifiers=True) creates a table abc in snowflake

How would this improve snowflake-snowpark-python?

Make writing Snowflake Table consistent

References, Other Background

As explained in write_pandas docs

quote_identifiers – By default, identifiers, specifically database, schema, table and column names (from DataFrame.columns) will be quoted. If set to False, identifiers are passed on to Snowflake without quoting, i.e. identifiers will be coerced to uppercase by Snowflake.

[sfc-gh-yixie]

Thanks @ChuliangXiao pandas columns are case sensitive. For instance, ab and aB are two different column names. If they're not quoted, the would both become AB after writing to a table. So write_pandas quote the column names by default and therefore quote the table name too.

Aside from dealing with pandas DataFrame, we try to follow the Snowflake object identifier requirements.

But this is indeed confusing to users. We'll think of a way to improve it.