require "set"
#
# Remove duplicate case‑insensitive words separated by multiple delimiters.
#
# Features:
# - Case‑insensitive comparison (String#downcase)
# - Preserves original casing of the first occurrence
# - Trims whitespace around tokens
# - Supports ANY number of delimiters (including multi‑character)
# - Preserves original order
# - Efficient O(n) lookup using Set
#
# ---------------------------------------------------------------
# Build a regex that matches ANY delimiter
# ---------------------------------------------------------------
def build_delimiter_regex(delimiters)
# Escape delimiters so characters like "|" or "*" are treated literally
escaped = delimiters.map { |d| Regexp.escape(d) }
# Combine into one regex: (delim1|delim2|...)
Regexp.new("(?:#{escaped.join('|')})+")
end
# ---------------------------------------------------------------
# Split input string using multiple delimiters
# ---------------------------------------------------------------
def split_by_delimiters(input, delimiters)
regex = build_delimiter_regex(delimiters)
input
.split(regex)
.map(&:strip)
.reject(&:empty?)
end
# ---------------------------------------------------------------
# Remove duplicates (case‑insensitive)
# ---------------------------------------------------------------
def remove_duplicates_case_insensitive(tokens)
seen = Set.new
unique = []
tokens.each do |token|
key = token.downcase
unless seen.include?(key)
seen.add(key)
unique << token
end
end
unique
end
# ---------------------------------------------------------------
# Join tokens with a chosen delimiter
# ---------------------------------------------------------------
def join_tokens(tokens, delimiter)
tokens.join(delimiter)
end
# ---------------------------------------------------------------
# Main function
# ---------------------------------------------------------------
def remove_duplicates_multi_delimiter_ci(input, delimiters, output_delimiter)
tokens = split_by_delimiters(input, delimiters)
unique = remove_duplicates_case_insensitive(tokens)
join_tokens(unique, output_delimiter)
end
# ---------------------------------------------------------------
# Main
# ---------------------------------------------------------------
s = "AAA | aaa , aAA * aaA | AAa | AAA | BBB | ccc ---- CCC | AAA ; aaa | bbb"
delimiters = [" ", "|", ",", "*", "-", ";"]
result = remove_duplicates_multi_delimiter_ci(s, delimiters, " | ")
puts result
=begin
run:
AAA | BBB | ccc
=end
