class Kanjidic2

# File lib/kanjidic2.rb, line 7
def initialize(filename)
        @kanjidic2_file = File.open(filename) { |f| Nokogiri::XML(f) }
end

# File lib/kanjidic2.rb, line 11
def each_character()
        @kanjidic2_file.css("character").each do |character|
                yield(parse_chr(character))
        end
end

# File lib/kanjidic2.rb, line 17
def header
        return parse_header(@kanjidic2_file)
end

# File lib/kanjidic2.rb, line 32
def parse_chr(character)

        # This will be the parsed character.
        this_character = {}

        # character -> literal
        this_character["literal"] = character.css("literal").text

        # character -> codepoint
        this_character["codepoint"] = {}

        character.css("codepoint").css("cp_value").each do |cp_value|
                this_character["codepoint"][cp_value["cp_type"]] = cp_value.text.strip
        end

        # character -> radical
        this_character["radical"] = {}

        character.css("radical").css("rad_value").each do |rad_value|
                this_character["radical"][rad_value["rad_type"]] = rad_value.text.strip
        end

        # character -> misc

        misc_data = character.css("misc")
        this_character["misc"] = {}

        # character -> misc -> grade
        this_character["misc"]["grade"] = misc_data.css("grade").text

        # character -> misc -> stroke_count
        this_character["misc"]["stroke_count"] = misc_data.css("stroke_count").text

        # character -> misc -> variant
        this_character["misc"]["variant"] = {}

        misc_data.css("misc").css("variant").each do |variant|
                this_character["misc"]["variant"][variant["var_type"]] = variant.text.strip
        end

        # character -> misc -> freq
        this_character["misc"]["freq"] = misc_data.css("freq").text

        # character -> misc -> rad_name
        this_character["misc"]["rad_name"] = misc_data.css("rad_name").text

        # character -> misc -> jlpt
        this_character["misc"]["jlpt"] = misc_data.css("jlpt").text

        # character -> dic_number
        this_character["dic_number"] = {}

        character.css("dic_number").css("dic_ref").each do |dic_ref|
                unless dic_ref["dr_type"] == "moro"
                        this_character["dic_number"][dic_ref["dr_type"]] = dic_ref.text.strip
                else
                        this_character["dic_number"]["moro"] = {}
                        this_character["dic_number"]["moro"]["m_vol"] = dic_ref["m_vol"]
                        this_character["dic_number"]["moro"]["m_page"] = dic_ref["m_page"]
                        this_character["dic_number"]["moro"]["value"] = dic_ref.text.strip
                end
        end

        # character -> query_code
        this_character["query_code"] = {}
        character.css("query_code").css("q_code").each do |q_code|
                this_character["query_code"][q_code["qc_type"]] = q_code.text.strip
        end

        # character -> reading_meaning
        reading_meaning_data = character.css("reading_meaning")
        this_character["reading_meaning"] = {}

        # character -> reading_meaning -> rmgroup
        this_character["reading_meaning"]["rmgroup"] = {}
        this_character["reading_meaning"]["rmgroup"]["reading"] = {}
        this_character["reading_meaning"]["rmgroup"]["meaning"] = {}

        # character -> reading_meaning -> rmgroup -> reading
        reading_meaning_data.css("rmgroup").css("reading").each do |reading|
                unless ["ja_on", "ja_kun"].include? reading["r_type"]
                        this_character["reading_meaning"]["rmgroup"]["reading"][reading["r_type"]] = reading.text.strip
                else
                        if reading["r_type"] == "ja_on"
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_on"] = {}
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_on"]["on_type"] = reading["on_type"]
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_on"]["r_status"] = reading["r_status"]
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_on"]["value"] = reading.text.strip
                        else
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_kun"] = {}
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_kun"]["r_status"] = reading["r_status"]
                                this_character["reading_meaning"]["rmgroup"]["reading"]["ja_kun"]["value"] = reading.text.strip
                        end
                end
        end

        # character -> reading_meaning -> rmgroup -> meaning
        this_character["reading_meaning"]["rmgroup"]["meaning"]["en"] = []
        this_character["reading_meaning"]["rmgroup"]["meaning"]["fr"] = []
        this_character["reading_meaning"]["rmgroup"]["meaning"]["es"] = []
        this_character["reading_meaning"]["rmgroup"]["meaning"]["pt"] = []

        reading_meaning_data.css("rmgroup").css("meaning").each do |meaning|

                if meaning["m_lang"].nil?
                        this_character["reading_meaning"]["rmgroup"]["meaning"]["en"] << meaning.text.strip
                else
                        this_character["reading_meaning"]["rmgroup"]["meaning"][meaning["m_lang"]] << meaning.text.strip
                end
        end

        # character -> reading_meaning -> nanori
        this_character["reading_meaning"]["nanori"] = []

        reading_meaning_data.css("nanori").each do |nanori|
                this_character["reading_meaning"]["nanori"] << nanori.text.strip
        end

        return this_character
end

# File lib/kanjidic2.rb, line 23
def parse_header(kanjidic2_file)
        header = kanjidic2_file.css("header")
        parsed_header = {}
        ["file_version", "database_version", "date_of_creation"].each do |header_elem|
                parsed_header[header_elem] = header.css("#{header_elem}").text
        end
        return parsed_header
end