Numeric Value out of Range when Inserting to SQL from R - sql

Question

Edit: Here is the column file for you to try to insert to your database: https://easyupload.io/jls3mk
So I narrowed my problem down to 1 column in my dataframe. It's a numeric column from 0-260000 with NaNs in it.
When I try to insert pred_new_export[46] (only column 46) using this statement:
dbWriteTable(conn = con,
name = SQL("ML.CreditLineApplicationOutputTemp"),
value = pred_new_export[46], overwrite=TRUE) ## x is any data frame
I get the issue:
Error in result_insert_dataframe(rs#ptr, values, batch_rows) :
nanodbc/nanodbc.cpp:1655: 22003: [Microsoft][SQL Server Native Client 11.0]Numeric value out of range
I've looked at this for 2 hours and it's been driving me insane. I can't figure out why it wouldn't insert into a fresh SQL table. The column only contains numbers.
The numbers are within range of the column:
This is the SQL schema create statement.
USE [EDWAnalytics]
GO
/****** Object: Table [ML].[CreditLineApplicationOutputTemp] Script Date: 4/20/2022 9:26:22 AM ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [ML].[CreditLineApplicationOutputTemp](
[MedianIncomeInAgeBracket] [float] NULL
) ON [PRIMARY]
GO

Answer 1

You said it has NaNs, which many DBMSes do not understand. I suggest you replace all NaN with NA.
Reprex:
# con <- DBI::dbConnect(..)
DBI::dbExecute(con, "create table quux (num float)")
# [1] 0
df <- data.frame(num=c(1,NA,NaN))
DBI::dbAppendTable(con, "quux", df)
# Error in result_insert_dataframe(rs#ptr, values, batch_rows) :
# nanodbc/nanodbc.cpp:1655: 42000: [Microsoft][ODBC Driver 17 for SQL Server][SQL Server]The incoming tabular data stream (TDS) remote procedure call (RPC) protocol stream is incorrect. Parameter 1 (""): The supplied value is not a valid instance of data type float. Check the source data for invalid values. An example of an invalid value is data of numeric type with scale greater than precision.
df$num[is.nan(df$num)] <- NA
DBI::dbAppendTable(con, "quux", df)
DBI::dbGetQuery(con, "select * from quux")
# num
# 1 1
# 2 NA
# 3 NA
FYI, the version of SQL Server ODBC you are using is rather antiquated: even the most recent release of 11 was in 2017. For many reasons, I suggest you upgrade to ODBC Driver for SQL Server 17 (the 17 has nothing to do with the version of SQL Server to which you are connecting).
FYI, my DBMS/version:
cat(DBI::dbGetQuery(con, "select ##version")[[1]], "\n")
# Microsoft SQL Server 2019 (RTM-CU14) (KB5007182) - 15.0.4188.2 (X64)
# Nov 3 2021 19:19:51
# Copyright (C) 2019 Microsoft Corporation
# Developer Edition (64-bit) on Linux (Ubuntu 20.04.3 LTS) <X64>
though this is also the case with SQL Server 2016 (and likely other versions).