You might be interested in xsv (https://github.com/BurntSushi/xsv), A fast CSV command line toolkit written in Rust :)

As a warning I’d like to point out that using awk and jq splitting on just , will subtly break if you have strings in your CSV files that contain that character as part of the string. Good stuff though!

csvkit is much better for handling CSV files. If you try and use awk -F, to parse CSVs somebody will throw a field containing a comma and you’re in for a world of hurt.

Miller (mlr) is a great tool for working with CSV: https://miller.readthedocs.io/en/latest/

SQLite is great for handling CSV too

CSV is quite nice and simple data format, but it is not as simple as some people expect. For example this is valid CSV (RFC 4180):

name,age
Joe,25
"Weird
Freak",96
Jenny,18
"Marry, PhD",32

Parsing it using the classic text tools fails because such tools does not understand the particular format. (see also Classic pipeline example). We get e.g. a single value as two values (Weird + Freak) or only a part of a single value (missing , PhD ):

$ awk -F, '{print $1}' test.csv 
name
Joe
"Weird
Freak"
Jenny
"Marry

We also get quotes in values and scrambled content:

$ jq -R -r 'split(",") | {name:.[0],age:.[1]}' test.csv 
{
"name": "name",
"age": "age\r"
}
{
"name": "Joe",
"age": "25\r"
}
{
"name": "\"Weird\r",
"age": null
}
{
"name": "Freak\"",
"age": "96\r"
}
{
"name": "Jenny",
"age": "18\r"
}
{
"name": "\"Marry",
"age": " PhD\""
}

CSV also uses CRLF line ends, not only LF, which these tools also do not understand and pollute the output with \r (escaped) or 0x0d bytes.

This „simple“ approach gives quite random and unreliable results. Reliable solution is to use a software that has a parser for given format (even in cases of formats that might looks „simple“, „plain-text“ and „obvious“). Many programs have this parser.

Such issues are also one of reasons why Relational pipes are being developed.

Few examples:

Read a CSV and display it as a table in console:

$ cat test.csv | relpipe-in-csv | relpipe-out-tabular 
csv:
╭───────────────┬──────────────╮
│ name (string) │ age (string) │
├───────────────┼──────────────┤
│ Joe           │ 25           │
│ Weird↲Freak   │ 96           │
│ Jenny         │ 18           │
│ Marry, PhD    │ 32           │
╰───────────────┴──────────────╯
Record count: 4

Filter only the first attribute (name):

$ cat test.csv | relpipe-in-csv | relpipe-tr-cut --relation '.*' --attribute name | relpipe-out-tabular 
csv:
╭───────────────╮
│ name (string) │
├───────────────┤
│ Joe           │
│ Weird↲Freak   │
│ Jenny         │
│ Marry, PhD    │
╰───────────────╯
Record count: 4

Filter records by value:

$ cat test.csv | relpipe-in-csv | relpipe-tr-awk --relation '.*' --where 'age < 30' | relpipe-out-tabular 
csv:
╭───────────────┬──────────────╮
│ name (string) │ age (string) │
├───────────────┼──────────────┤
│ Joe           │ 25           │
│ Jenny         │ 18           │
╰───────────────┴──────────────╯
Record count: 2