This package provides `ivy-read' as an alternative to `completing-read' and similar functions. There's no intricate code to determine the best candidate. Instead, the user can navigate to it with `ivy-next-line' and `ivy-previous-line'. The matching is done by splitting the input text by spaces and re-building it into a regex. So "for example" is transformed into "\\(for\\).*\\(example\\)".