2 Answers
- Newest
- Most votes
- Most comments
0
Hello,
Thanks for reaching out. For this scenario, a workaround is to add an intermediate node called "Change Schema" in Glue studio, which converts the input column into "string" before writing data to Redshift.
So the data flow is as below:
Data source, S3 => Change schema => Data target, Redshift.
The generated code for "Change schema" looks like below, for sample only:
# Script generated for node Change Schema
ChangeSchema_node1702948869240 = ApplyMapping.apply(
frame=AmazonS3_node1702946282328,
mappings=[
("invoiceno", "string", "invoiceno", "string"),
("amount", "bingint", "amount", "string"),
],
transformation_ctx="ChangeSchema_node1702948869240",
)
Hope it helps.
answered 4 months ago
0
You may need to leverage CONVERT function to convert from scientific number format to float: https://docs.aws.amazon.com/redshift/latest/dg/r_CONVERT_function.html
Relevant content
- Accepted Answerasked 3 months ago
- asked 2 years ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated a year ago