reformat_spaces_to_tabs.bash

#!/usr/bin/env bash

# OK, why tabs? Why resurrect this age-old nerd debate again?
# Very simple: It's an accessibility issue:
# https://adamtuttle.codes/blog/2021/tabs-vs-spaces-its-an-accessibility-issue/
# To set the default tab width on your terminal (which unfortunately defaults to 8),
# run: `tabs -2` (or whatever width you want), possibly in your dotfiles.

# is AWK set to anything? If not, prefer frawk, then gawk, then awk
[ -z "${AWK}" ] && export AWK=$(command -v frawk || command -v gawk || command -v awk)

reformat_spaces_to_tabs() {
	local file="$1"
	function determine_tab_size_in_spaces() {
		$AWK '
			{
				# Match the leading spaces and count the occurrences
				match($0, /^ */);
				indents[RLENGTH]++;
			}
			END {
				# Find the 3 most common indentation levels
				max_count = 0;
				max_indent = 0;
				second_max_indent = 0;
				third_max_indent = 0;
				for (indent in indents) {
					if (indents[indent] > max_count && indent > 0) {
						third_max_indent = second_max_indent;
						second_max_indent = max_indent;
						max_indent = indent;
						max_count = indents[indent];
					}
				}
				# Determine what the greatest common divisor is for the most common indentation levels,
				# handling up to 8 spaces (which is what my terminal outputs, which is ridiculous)
				# There is undoubtedly a less janky way to do this
				if (max_indent % 8 == 0 && second_max_indent % 8 == 0 && third_max_indent % 8 == 0) {
					print 8;
				} else if (max_indent % 7 == 0 && second_max_indent % 7 == 0 && third_max_indent % 7 == 0) {
					print 7;
				} else if (max_indent % 6 == 0 && second_max_indent % 6 == 0 && third_max_indent % 6 == 0) {
					print 6;
				} else if (max_indent % 5 == 0 && second_max_indent % 5 == 0 && third_max_indent % 5 == 0) {
					print 5;
				} else if (max_indent % 4 == 0 && second_max_indent % 4 == 0 && third_max_indent % 4 == 0) {
					print 4;
				} else if (max_indent % 3 == 0 && second_max_indent % 3 == 0 && third_max_indent % 3 == 0) {
					print 3;
				} else if (max_indent % 2 == 0 && second_max_indent % 2 == 0 && third_max_indent % 2 == 0) {
					print 2;
				} else if (max_indent == 0) {
					print 0;
				} else {
					print 1;
				}
			}
		' "$1"
	}
	local indent=$(determine_tab_size_in_spaces "$file")
	if (( indent > 0 )); then
		# >&2 echo "Indent size: $indent"
		indent_spaces=$(printf "%${indent}s")
		$AWK -v indent="$indent_spaces" '
		{
			while (match($0, "^\t*"indent)) {
				$0 = gensub("^(\t*)"indent, "\\1\t", "g");
			}
			print;
		}' "$file" #> "$file.tmp" && mv "$file.tmp" "$file"
	else
		# >&2 echo "Indent size: 0"
		cat "$file"
	fi
}

# Export the function so it can be used by find -exec
export -f reformat_spaces_to_tabs

# Run the function, passing along any args, if this file was run directly (such as via sudo) instead of as an include
# Sometimes, $0 contains a leading dash to indicate an interactive (or is it login?) shell,
# which is apparently an old convention (which also broke the basename call on OS X)
_me=$(basename "${0##\-}")
if [ "$_me" = "reformat_spaces_to_tabs" ]; then
	$_me "$@"
fi
unset _me