Matching and printing regex groups in sed

Overview

A CLI one-liner using regular expressions and sed to capture a group of information and print it. I typically use python to build complex regular expression pattern matching and grouping. In this example, a simple one-liner pulls data out of a text file.

Details

The input file is a list of Markdown links:

jemurray@jasons-mbp:~ $ head -1 scratch.txt
[Schedule requirements gathering meeting for WSA](https://example.com)

The sed command pulls all the text between the square-brackets [...].

• -n - Surpasses output; by default, sed prints all data after a match is made.
• s - Search
• \[ - Match a left-bracket
• \( - Start the group match
• .* - Match anything up to the next more specific match character
• \) - End the group match (this group match will be printed to the console)
• \](.*$ - Match the right-bracket, left-parentheses, and any text to the end of the line.
• /\1/ - print the information stored in the first group match ()
• p - print the match

jemurray@jasons-mbp:~ $ cat scratch.txt | sed -n "s/\[\(.*\)\](.*$/\1/p" | head -1
Schedule requirements gathering meeting for WSA

I love the command line!