Skip to main content

select

The select command manipulates columns in CSV data, allowing you to reorder, duplicate, reverse, or drop them. Columns can be referenced by index or by name if there is a header row, with duplicate column names disambiguated using indexing. Column ranges can also be specified, and columns can be selected using regular expressions.

Syntax​

qsv select [options] [--] <selection> [<input_file>] [<output_file>]

Description​

The select command offers a wide range of column manipulation capabilities, including reordering, duplicating, reversing, and dropping columns. It supports selection by index, name, range, and regular expressions, making it a powerful tool for data processing.

Options​

  • -R, --random: Randomly shuffle the columns in the selection.
  • --seed <number>: Seed for the random number generator.
  • -S, --sort: Sort the selected columns lexicographically, i.e., by their byte values.
  • -h, --help: Display this message.
  • -o, --output <file>: Write output to <file> instead of stdout.
  • -n, --no-headers: When set, the first row will not be interpreted as headers.
  • -d, --delimiter <arg>: The field delimiter for reading CSV data. Must be a single character. (default: ,)

Examples​

Basic Selection​

Select the first and fourth columns:

qsv select 1,4

Selecting Columns by Index and Name​

Select the first 4 columns by index and by name:

qsv select 1-4
qsv select Header1-Header4

Ignoring Columns​

Ignore the first 2 columns by range and by omission:

qsv select 3-
qsv select '!1-2'

Selecting Columns by Name​

Select the third column named 'Foo':

qsv select 'Foo[2]'

Selecting First and Last Columns​

Select the first and last columns, using _ as a special character for the last column:

qsv select 1,_

Reversing Column Order​

Reverse the order of columns:

qsv select _-1

Sorting Columns​

Sort the columns lexicographically:

qsv select 1- --sort

Selecting and Sorting Columns​

Select some columns and then sort them:

qsv select 1,4,5-7 --sort

Shuffling Columns​

Randomly shuffle the columns:

qsv select 1- --random

Shuffling Columns with a Seed​

Randomly shuffle the columns with a seed:

qsv select 1- --random --seed 42

Selecting Columns with a Regex​

Select columns using a regex:

# select columns starting with 'a'
qsv select /^a/
# select columns with a digit
qsv select '/^.*\d.*$/'
# remove SSN, account_no and password columns
qsv select '!/SSN|account_no|password/'

Reordering and Duplicating Columns​

Reorder and duplicate columns arbitrarily using different types of selectors:

qsv select 3-1,Header3-Header1,Header1,Foo[2],Header1

Quoting Column Names​

Quote column names that conflict with selector syntax:

qsv select '\"Date - Opening\",\"Date - Actual Closing\"'

Common Use Cases​

  • Focusing on specific data columns for analysis or reporting
  • Reducing dataset size by excluding unnecessary columns
  • Preparing data for import into systems with specific column requirements
  • Data cleaning and preprocessing

Tips​

  • Use select in combination with other qsv commands for more complex data processing
  • Verify the selected columns by inspecting the output

See Also​

  • table - for displaying selected columns in a tabular format
  • sort - for sorting selected columns