I came up with two ways to append a column(s) to an existing CSV file.
Method 1 late merges the new column by reading the file into an array of hashes, then appending the columns to the end of each row. This method can exhibit anomalies if run multiple times.
require 'csv'
filename = 'test.csv'
# Load the original CSV file
rows = CSV.read(filename, headers: true).collect do |row|
row.to_hash
end
# Original CSV column headers
column_names = rows.first.keys
# Array of the new column headers
additional_column_names = ['flag']
# Append new column name(s)
column_names += additional_column_names
s = CSV.generate do |csv|
csv << column_names
rows.each do |row|
# Original CSV values
values = row.values
# Array of the new column(s) of data to be appended to row
additional_values_for_row = ['1']
values += additional_values_for_row
csv << values
end
end
# Overwrite csv file
File.open(filename, 'w') { |file| file.write(s) }
Method 2 early merges the new column(s) into the row hash. The nicety of this method is it is more compact and avoids duplicate column names if run more than once. This method can also be used to change any existing values in the CSV.
require 'csv'
filename = 'test.csv'
# Load the original CSV file
rows = CSV.read(filename, headers: true).collect do |row|
hash = row.to_hash
# Merge additional data as a hash.
hash.merge('flag' => '0')
# BONUS: Change any existing data here too!
hash.merge('a1' => hash['a1'].to_i + 1 )
end
# Extract column names from first row of data
column_names = rows.first.keys
txt = CSV.generate do |csv|
csv << column_names
rows.each do |row|
# Extract values for row of data
csv << row.values
end
end
# Overwrite csv file
File.open(filename, 'w') { |file| file.write(txt) }