diff options
Diffstat (limited to 'packages/meshbay-node/tests/test_title_parse.py')
| -rw-r--r-- | packages/meshbay-node/tests/test_title_parse.py | 50 |
1 files changed, 50 insertions, 0 deletions
diff --git a/packages/meshbay-node/tests/test_title_parse.py b/packages/meshbay-node/tests/test_title_parse.py index 1a277f2..5f0977c 100644 --- a/packages/meshbay-node/tests/test_title_parse.py +++ b/packages/meshbay-node/tests/test_title_parse.py @@ -6,6 +6,7 @@ is a manual acceptance step (§11), not something this repo's corpus holds. from meshbay_node.indexer.title_parse import ( ParsedName, + leading_episode_number, naive_title, parse_episode_filename, parse_movie_filename, @@ -111,6 +112,31 @@ def test_season_folder_roman_numeral(): assert season_from_folder_name("Saison IV") == 4 +def test_season_folder_book_word_roman_numeral(): + # Some shows number their seasons by "book" (Roman numerals) rather + # than "season"/"saison" — real, observed vocabulary (§3.4). + assert season_from_folder_name("Livre I") == 1 + assert season_from_folder_name("Livre VI") == 6 + + +def test_season_folder_bare_s_abbreviation(): + # A common abbreviated convention distinct from SEASON_WORDS' full + # words — found live: a show with "S1"/"S2"/"S3" folders instead of + # "Season 1" etc, exactly the shape that grouped its later seasons as + # loose individual entries rather than under the show. + assert season_from_folder_name("S1") == 1 + assert season_from_folder_name("S2") == 2 + assert season_from_folder_name("S02") == 2 + + +def test_season_folder_bare_s_abbreviation_does_not_match_inside_a_longer_name(): + # Anchored to the whole folder name — a real folder that merely starts + # with "S" followed by digits somewhere in a longer, unrelated name + # must not be read as a season abbreviation. + assert season_from_folder_name("S1 Extended Cut") is None + assert season_from_folder_name("Something2") is None + + def test_specials_folder_maps_to_season_zero(): assert season_from_folder_name("Specials") == 0 assert season_from_folder_name("Bonus") == 0 @@ -121,6 +147,30 @@ def test_non_season_folder_name_returns_none(): assert season_from_folder_name("Some Show Name") is None +# ── §3.4c: a bare leading episode number, guessit's 3-digit blind spot ─────── + +def test_leading_episode_number_reads_the_whole_number(): + assert leading_episode_number("001 Episode's Own Title.mkv") == 1 + assert leading_episode_number("099 Episode's Own Title.mkv") == 99 + + +def test_leading_episode_number_not_split_by_guessit_at_three_digits(): + # The real bug this guards: guessit itself reads a bare "100" as season=1, + # episode=0 (a concatenated SxxE guess), not episode=100 — silently, with + # nothing in its output telling apart a real 2-digit episode from this. + assert leading_episode_number("100 Episode's Own Title.mkv") == 100 + + +def test_leading_episode_number_ignores_a_leading_year(): + # Four digits, not the 1-3 an episode-number prefix can be — this is a + # normal movie filename shape, not an episode number. + assert leading_episode_number("2010 - Some Movie.mkv") is None + + +def test_leading_episode_number_none_without_a_leading_number(): + assert leading_episode_number("Some Show.S01E01.mkv") is None + + def test_parsed_name_is_a_plain_dataclass(): # sanity: constructible with just the one required field, per the # "None => caller must supply from elsewhere" contract. |