Im currently using regex101 to try and work out the following, id like to be able to capture a full items data for example name_template_2 and its associated description, define and write data

Here's my data model

templates:
  name_template:
    description: test_description
    define: yes
    write: true
  name_template_2:
    description: test_description2
    define: false
    write: true

I can capture the lines I need with the following

^[[:space:]][[:space:]][[:space:]][[:space:]].*

and

^[[:space:]][[:space:]]name_template_2:

but I am unable to join both patterns together to filter just the key and data related to name_template_2. The more I read online the more I understand it less. Has anyone achieved this before or is there a much more efficient way of doing this?

CodePudding user response：

I suggest using a structure-aware tool like yq to manipulate YAML and not using regular expressions.

#!/bin/bash

INPUT='
templates:
  name_template:
    description: test_description
    define: yes
    write: true
  name_template_2:
    description: test_description2
    define: false
    write: true
'

echo "$INPUT" | yq '{"name_template_2": .templates.name_template_2}'

Output

name_template_2:
  description: test_description2
  define: false
  write: true

CodePudding user response：

Using 2 capture groups:

^[^\S\n]{2}(name_template_2:)((?:\n[^\S\n]{4}\S.*) )

Explanation

• ^ Start of string
• [^\S\n]{2} Match 2 spaces without newlines
• (name_template_2:) Match the string and capture in group 1
• ( Capture group 2
  • (?: Non capture group
    • \n Match a newline
    • [^\S\n]{4} Match 4 spaces without newlines
    • \S.* Match a non whitespace char and the rest of the line
  • ) Repeat the non capture group 1 or more times
• ) Close group 2

Regex demo