1

I am trying to directly export a table from redshift to my local computer. I am successful in getting data from redshift however it doesn't differentiate any of the data. When I do pandas.dtypes they all come out as objects and not various data types such as string or date timestamps. I would also like to add the headers of the columns straight from the export.

I've successfully exported to my local using PSQL commands from my terminal to access the redshift.

psql -h omaha-prod-cluster.example.us-east-1.redshift.amazonaws.com -d prod -U <username> -p 5439 -A -t -c "select * from l2_survey.survey_customerinsight" -F ',' -o Downloads/survey_customerInsights.csv

I am then running the panda command to read the kinds of data types

data.dtypes()

and it is returning every column with the data type of object. It also doesn't give me the headers of the columns with the psql command above

Loyal_Burrito
  • 115
  • 1
  • 13

1 Answers1

1

There is problem with your command where you are explicitly asking the export command to skip the column names by supplying the argument -t, which tells the command to just export tuples without column names. Just change it like below and it will provide you the header.

psql -h <host-values>.redshift.amazonaws.com -U <user> -d <database> -p 5439 -c "select * from your_schema.your_table" > out.txt

Hope it helps you.

Red Boy
  • 4,333
  • 2
  • 18
  • 35