Reputation: 123
I am trying to create a table (in Snowflake db) with exactly the same column names as I keep in the R data.frame object:
'data.frame': 1 obs. of 26 variables:
$ Ship_To : chr "0002061948"
$ Del_Coll_Indicator : chr "D"
$ Currency : chr "GBP"
$ Total_Volume : num 0
$ Total_Quantity : num 0
...
There is no problem with the table creation:
dbWriteTable(con = my_db$con, name = "test5", value = df)
but all column names in the database are converted to upper cases:
'data.frame': 1 obs. of 26 variables:
$ SHIP_TO : chr "0002061948"
$ DEL_COLL_INDICATOR : chr "D"
$ CURRENCY : chr "GBP"
...
Is there any way to keep in the table original names from R's data frame?
Upvotes: 2
Views: 702
Reputation: 696
As covered by Snowflake's SQL reference docs, when identifiers (such as column names) are unquoted at creation, Snowflake will upper case them, and treat them as case-insensitive. Any quoted identifiers will be kept as-is and treated as a case-sensitive identifier.
Alter the data frame column names (colnames(df)
) to use a quoted identifier format via the dbQuoteIdentifier(my_db$con, each_column_name)
DBI function. This should help preserve the casing.
Upvotes: 3