diff --git a/app/models/tag.rb b/app/models/tag.rb index aad38d8ca..9cbe716f1 100644 --- a/app/models/tag.rb +++ b/app/models/tag.rb @@ -243,14 +243,15 @@ class Tag < ApplicationRecord concerning :WordMethods do # Characters that delimit words in tags. - WORD_DELIMITERS = " _+:;!.\/()-" - WORD_DELIMITER_REGEX = /([#{WORD_DELIMITERS}]+)/ + WORD_DELIMITERS = " _+:;!./()-" + WORD_DELIMITER_REGEX = /([#{WORD_DELIMITERS}]+[^a-zA-Z0-9]*|[^a-zA-Z0-9]*[#{WORD_DELIMITERS}]+|\A[^[a-zA-Z0-9]]+|[^[a-zA-Z0-9]]+\z)/ class_methods do # Split the tag at word boundaries. # # Tag.split_words("jeanne_d'arc_alter_(fate)") => ["jeanne", "_", "d'arc", "_", "alter", "_(", "fate", ")"] - # Tag.split_words("k-on!") => ["k", "-", "on!"] + # Tag.split_words(%q{don't_say_"lazy"}) => ["don't", "_", "say", '_"', "lazy", '"'] + # Tag.split_words("jack-o'-lantern") => ["jack", "-", "o", "'-", "lantern"] # Tag.split_words("_") => ["_"] def split_words(name) return [name] if !parsable_into_words?(name) @@ -261,7 +262,8 @@ class Tag < ApplicationRecord # Parse the tag into plain words, removing punctuation and delimiters. # # Tag.parse_words("jeanne_d'arc_alter_(fate)") => ["jeanne", "d'arc", "alter", "fate"] - # Tag.parse_words("k-on!") => ["k", "on"] + # Tag.parse_words(%q{don't_say_"lazy"}) => ["don't", "say", "lazy"] + # Tag.parse_words("jack-o'-lantern") => ["jack", "o", "lantern"] # Tag.parse_words("_") => ["_"] def parse_words(name) return [name] if !parsable_into_words?(name) diff --git a/test/unit/tag_test.rb b/test/unit/tag_test.rb index 2e6c3b38c..47dbf21dc 100644 --- a/test/unit/tag_test.rb +++ b/test/unit/tag_test.rb @@ -143,15 +143,41 @@ class TagTest < ActiveSupport::TestCase should "parse tag names into words" do assert_equal(%w[very long hair], Tag.new(name: "very_long_hair").words) + assert_equal(%w[k - on !], Tag.split_words("k-on!")) + assert_equal(%w[. hack //], Tag.split_words(".hack//")) + assert_equal(%w[re : zero], Tag.split_words("re:zero")) + assert_equal(%w[# compass], Tag.split_words("#compass")) + assert_equal(%w[. hack // g . u .], Tag.split_words(".hack//g.u.")) + assert_equal(%w[me ! me ! me !], Tag.split_words("me!me!me!")) + assert_equal(%w[d . gray - man], Tag.split_words("d.gray-man")) + assert_equal(%w[steins ; gate], Tag.split_words("steins;gate")) + assert_equal(%w[tiger _&_ bunny], Tag.split_words("tiger_&_bunny")) + assert_equal(%w[ssss . gridman], Tag.split_words("ssss.gridman")) + assert_equal(%w[yu - gi - oh !_ 5d's], Tag.split_words("yu-gi-oh!_5d's")) + assert_equal(%w[don't _ say _" lazy "], Tag.split_words(%q{don't_say_"lazy"})) + assert_equal(%w[jack - o '- lantern], Tag.split_words("jack-o'-lantern")) + assert_equal(%w[d . va _( overwatch )], Tag.split_words("d.va_(overwatch)")) + assert_equal(%w[rosario + vampire], Tag.split_words("rosario+vampire")) + assert_equal(%w[girls '_ frontline], Tag.split_words("girls'_frontline")) + assert_equal(%w[fate / grand _ order], Tag.split_words("fate/grand_order")) + assert_equal(%w[yorha _ no ._ 2 _ type _ b], Tag.split_words("yorha_no._2_type_b")) + assert_equal(%w[love _ live !_ sunshine !!], Tag.split_words("love_live!_sunshine!!")) + assert_equal(%w[jeanne _ d'arc _ alter _( ver ._ shinjuku _ 1999 )_( fate )], Tag.split_words("jeanne_d'arc_alter_(ver._shinjuku_1999)_(fate)")) + assert_equal(%w[kaguya - sama _ wa _ kokurasetai _~ tensai - tachi _ no _ renai _ zunousen ~], Tag.split_words("kaguya-sama_wa_kokurasetai_~tensai-tachi_no_renai_zunousen~")) + assert_equal(%w[k on], Tag.new(name: "k-on!").words) assert_equal(%w[hack], Tag.new(name: ".hack//").words) + assert_equal(%w[nyoro~n], Tag.new(name: "nyoro~n").words) assert_equal(%w[re zero], Tag.new(name: "re:zero").words) assert_equal(%w[compass], Tag.new(name: "#compass").words) + assert_equal(%w[hack g u], Tag.new(name: ".hack//g.u.").words) assert_equal(%w[me me me], Tag.new(name: "me!me!me!").words) assert_equal(%w[d gray man], Tag.new(name: "d.gray-man").words) assert_equal(%w[steins gate], Tag.new(name: "steins;gate").words) + assert_equal(%w[tiger bunny], Tag.new(name: "tiger_&_bunny").words) assert_equal(%w[ssss gridman], Tag.new(name: "ssss.gridman").words) assert_equal(%w[yu gi oh 5d's], Tag.new(name: "yu-gi-oh!_5d's").words) + assert_equal(%w[don't say lazy], Tag.new(name: %q{don't_say_"lazy"}).words) assert_equal(%w[jack o lantern], Tag.new(name: "jack-o'-lantern").words) assert_equal(%w[d va overwatch], Tag.new(name: "d.va_(overwatch)").words) assert_equal(%w[rosario vampire], Tag.new(name: "rosario+vampire").words) @@ -160,6 +186,7 @@ class TagTest < ActiveSupport::TestCase assert_equal(%w[yorha no 2 type b], Tag.new(name: "yorha_no._2_type_b").words) assert_equal(%w[love live sunshine], Tag.new(name: "love_live!_sunshine!!").words) assert_equal(%w[jeanne d'arc alter ver shinjuku 1999 fate], Tag.new(name: "jeanne_d'arc_alter_(ver._shinjuku_1999)_(fate)").words) + assert_equal(%w[kaguya sama wa kokurasetai tensai tachi no renai zunousen], Tag.new(name: "kaguya-sama_wa_kokurasetai_~tensai-tachi_no_renai_zunousen~").words) assert_equal(%w[:o], Tag.new(name: ":o").words) assert_equal(%w[o_o], Tag.new(name: "o_o").words)