Question

I have a csv file with dump data of table and I would like to import it directly into my database using rails.

I am currently having this code:

csv_text = File.read("public/csv_fetch/#{model.table_name}.csv")
ActiveRecord::Base.connection.execute("TRUNCATE TABLE #{model.table_name}")
puts "\nUpdating table #{model.table_name}"
csv = CSV.parse(csv_text, :headers => true)
csv.each do |row|
  row = row.to_hash.with_indifferent_access
  ActiveRecord::Base.record_timestamps = false
  model.create!(row.to_hash.symbolize_keys)
end

with help from here..

Consider my Sample csv:

id,code,created_at,updated_at,hashcode
10,00001,2012-04-12 06:07:26,2012-04-12 06:07:26,
2,00002,0000-00-00 00:00:00,0000-00-00 00:00:00,temphashcode
13,00007,0000-00-00 00:00:00,0000-00-00 00:00:00,temphashcode
43,00011,0000-00-00 00:00:00,0000-00-00 00:00:00,temphashcode
5,00012,0000-00-00 00:00:00,0000-00-00 00:00:00,temphashcode

But problem with this code is :

• It is generating `id’ as autoincrement 1,2,3,.. instead of what in
  csv file.
• The timestamps for records where there is 0000-00-00 00:00:00 defaults to null automatically and throws error as the column created_at cannot be null…

Is there any way I can do it in generic way to import from csv to models?
or would i have to write custom code for each model to manipulate the attributes in each row manually??

Answer 1

providing fields like ‘id’ and for timestamps fields too manually solved it…

model.id = row[:id]

and similar for created_at,updated_at if these exists in model..