Non-indented lines are ignored when calculating minimum indentation

[d07RiV]

Consider this example

dedent`
test
  test2
`

Outputs

test
test2

Regex in line 43 (l.match(/^(\s+)\S+/);) should probably read \s* instead of \s+.

[JoshuaKGoldberg]

Hmm, I'm not sure that this is correct actually. Arguably, having a line without any indentation indicates that the string is already dedented, no?

Does anybody have a good use case of why this should be changed?

[JoshuaKGoldberg]

Thinking more deeply - seems reasonable that many would consider this a bug. Now that we have options (#65) we could take in an option to preserve the old behavior. ignoreNonIndentedLines, maybe?

import dedent from 'dedent';

// test
//   test2
dedent`
test
  test2
`;

// test
// test2
dedent.withOptions({ ignoreNonIndentedLines })`
test
  test2
`;

cc @G-Rath & @Haroenv since you've been active in this repo recently 👋

[Haroenv]

Makes sense to me, the default behaviour (maybe breaking change again) should be what's described in issue and consider that when something isn't indented, we shouldn't dedent anything (although at that point, I'm not even sure why someone would still use the template string, function could make sense)