Initial commit: CookIT recipe collection and week planner

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

tools/excel-import/tests/test_parse.py

@@ -0,0 +1,135 @@
+import pytest
+
+from cookit_import import parse
+
+
+@pytest.mark.parametrize(
+    "value, expected",
+    [
+        (275, (275, None)),
+        (1.5, (1.5, None)),
+        (0.08333333333333333, (0.083, None)),
+        ("0,5", (0.5, None)),
+        ("½", (0.5, None)),
+        ("1/2", (0.5, None)),
+        ("1 1/2", (1.5, None)),
+        ("3/4", (0.75, None)),
+        ("2-3", (2, 3)),
+        ("10 - 12", (10, 12)),
+        ("200-250", (200, 250)),
+        (None, (None, None)),
+        (0, (None, None)),
+    ],
+)
+def test_parse_amount(value, expected):
+    assert parse.parse_amount(value) == expected
+
+
+@pytest.mark.parametrize("value", ["opt. als Topping", "(2", "ODER"])
+def test_parse_amount_rejects_text(value):
+    assert parse.parse_amount(value) is None
+
+
+@pytest.mark.parametrize(
+    "value, key",
+    [
+        ("Gramm", "gram"),
+        ("Mililiter", "milliliter"),
+        ("milliliter", "milliliter"),
+        ("Zehe/n", "clove"),
+        ("Zehen", "clove"),
+        ("Kleine", "small"),
+        ("kleine", "small"),
+        ("Große", "large"),
+        ("Zweig (e)", "sprig"),
+        ("Msp.", "knifeTip"),
+        ("Etwas", "some"),
+        ("Tassen", "cup"),
+    ],
+)
+def test_parse_unit(value, key):
+    assert parse.parse_unit(value) == key
+
+
+def test_parse_unit_unknown():
+    assert parse.parse_unit("Fett") is None
+
+
+@pytest.mark.parametrize(
+    "value, expected",
+    [("100 g", (100, "gram")), ("125 ml", (125, "milliliter")), ("1 TL", (1, "teaspoon")), ("1 Pkt.", (1, "pack")), ("3", (3, None))],
+)
+def test_split_amount_unit(value, expected):
+    assert parse.split_amount_unit(value) == expected
+
+
+@pytest.mark.parametrize(
+    "value, expected",
+    [(30, 30), ("45 Minuten", 45), ("2 Stunden", 120), ("1h 30min", 90), ("0 Minuten", None), (None, None), ("über Nacht", "über Nacht")],
+)
+def test_parse_minutes(value, expected):
+    assert parse.parse_minutes(value) == expected
+
+
+@pytest.mark.parametrize(
+    "value, number, remark",
+    [
+        (4, 4, None),
+        (3.5, 3.5, None),
+        ("6 Portionen", 6, None),
+        ("4 Personen", 4, None),
+        ("8 Stk.", 8, "8 Stk."),
+        ("2 Rollen;\neine Rolle entspricht 8 Scheiben", 2, "2 Rollen;\neine Rolle entspricht 8 Scheiben"),
+        (None, None, None),
+    ],
+)
+def test_parse_servings(value, number, remark):
+    s = parse.parse_servings(value)
+    assert (s.value, s.remark) == (number, remark)
+
+
+def test_calories_basis_threshold():
+    assert parse.calories_basis(1500) == "perServing"
+    assert parse.calories_basis(1501) == "wholeRecipe"
+
+
+def test_extract_urls():
+    urls, rest = parse.extract_urls("Quelle: https://sallys-blog.de/rezepte/focaccia")
+    assert urls == ["https://sallys-blog.de/rezepte/focaccia"]
+    assert rest == "Quelle:"
+
+
+class TestSplitSteps:
+    def test_blank_lines_separate_steps(self):
+        sections = parse.split_steps("\nZutaten mischen.\n\nTeig gehen lassen.\n\n\nBacken.")
+        assert [(s.title, s.steps) for s in sections] == [(None, ["Zutaten mischen.", "Teig gehen lassen.", "Backen."])]
+
+    def test_heading_line_starts_a_section(self):
+        sections = parse.split_steps("Vorteig\nVerrühre die Zutaten.\n\nTeig\nFüge das Mehl hinzu.")
+        assert [(s.title, s.steps) for s in sections] == [
+            ("Vorteig", ["Verrühre die Zutaten."]),
+            ("Teig", ["Füge das Mehl hinzu."]),
+        ]
+
+    def test_lone_heading_paragraph(self):
+        sections = parse.split_steps("Soße:\n\nAlles verrühren.\n\nAbschmecken.")
+        assert [(s.title, s.steps) for s in sections] == [("Soße", ["Alles verrühren.", "Abschmecken."])]
+
+    def test_inline_heading(self):
+        sections = parse.split_steps("Salat:Die Zwiebel schneiden.\n\nSoße:Mayo verrühren.")
+        assert [(s.title, s.steps) for s in sections] == [("Salat", ["Die Zwiebel schneiden."]), ("Soße", ["Mayo verrühren."])]
+
+    def test_single_paragraph_splits_on_lines(self):
+        sections = parse.split_steps("    Mehl und Salz mischen.\n    Wasser zugeben und kneten.")
+        assert sections[0].steps == ["Mehl und Salz mischen.", "Wasser zugeben und kneten."]
+
+    def test_short_last_line_stays_a_step(self):
+        sections = parse.split_steps("Alles anbraten.\n\nServieren")
+        assert sections[0].steps == ["Alles anbraten.", "Servieren"]
+
+    def test_strips_enumeration(self):
+        sections = parse.split_steps("1. Wasser kochen.\n\n2) Nudeln garen.")
+        assert sections[0].steps == ["Wasser kochen.", "Nudeln garen."]
+
+    def test_empty(self):
+        assert parse.split_steps("\n\n  ") == []

tools/excel-import/tests/test_recipe.py

@@ -0,0 +1,178 @@
+from cookit_import.recipe import Reason, build_recipe, is_recipe_sheet, recipe_id
+from cookit_import.workbook import Sheet
+
+TEMPLATE = {
+    "C1": "Testrezept",
+    "F1": "Hauptgericht",
+    "G1": "Rezeptindex",
+    "C2": "Menge",
+    "D2": "Maßeinheit",
+    "E2": "Zutat",
+    "F2": "Anleitungen",
+    "G2": "Vorbereitungszeit",
+    "B5": "Kalorien",
+    "G6": "Kochzeit",
+    "B8": "Rezept-Stichwörter",
+    "G10": "Ergebnis (Portionen)",
+    "B11": "Kommentare",
+    "B12": "Löschen Sie dieses Rezeptarbeitsblatt nicht.",
+}
+
+
+def sheet(title="Testrezept", comments=None, **cells) -> Sheet:
+    return Sheet(title, {**TEMPLATE, **cells}, comments or {})
+
+
+def lines(recipe):
+    return [[(l["amount"], l["unit"], l["name"]) for l in g["lines"]] for g in recipe["ingredientGroups"]]
+
+
+def test_recognises_recipe_sheets_and_skips_template_ones():
+    assert is_recipe_sheet(sheet())
+    assert not is_recipe_sheet(sheet(title="Leeres Rezept"))
+    assert not is_recipe_sheet(sheet(title="Spinat-Feta-Pizza"))
+    assert not is_recipe_sheet(Sheet("Speiseplan", {"C2": "Montag"}))
+
+
+def test_basic_fields():
+    r = build_recipe(sheet(G3=20, G7="15 Minuten", G11=4, B6=775, B9="Tomaten, Nudeln, tomaten, Käse", F3="Kochen.\n\nEssen."))
+    assert r["id"] == recipe_id("Testrezept")
+    assert r["name"] == "Testrezept"
+    assert r["course"] == "Hauptgericht"
+    assert r["tags"] == ["Tomaten", "Nudeln", "Käse"]
+    assert (r["prepMinutes"], r["cookMinutes"], r["servings"]) == (20, 15, 4)
+    assert r["calories"] == {"kcal": 775, "basis": "perServing"}
+    assert r["stepSections"] == [{"title": None, "steps": ["Kochen.", "Essen."]}]
+    assert r["notes"] is None
+    assert r["reviewReasons"] == []
+
+
+def test_ids_are_stable_per_sheet():
+    assert recipe_id("Focaccia") == recipe_id("Focaccia") != recipe_id("Brot")
+
+
+def test_placeholder_course_and_missing_servings():
+    r = build_recipe(sheet(F1="Gang"))
+    assert r["course"] is None
+    assert Reason.NO_SERVINGS in r["reviewReasons"]
+
+
+def test_longer_name_wins_and_template_title_is_ignored():
+    assert build_recipe(sheet(title="Flammkuchen, low carb", C1="Flammkuchen"))["name"] == "Flammkuchen, low carb"
+    assert build_recipe(sheet(title="Kohlrabi-Kartoffel-Auflauf", C1="Leeres Rezept"))["name"] == "Kohlrabi-Kartoffel-Auflauf"
+
+
+def test_blank_rows_separate_ingredient_groups_and_units_are_canonical():
+    r = build_recipe(sheet(G11=1, C3=300, D3="Gramm", E3="Wasser", C4=1, D4="Teelöffel", E4="Zucker", E6="Rosmarin"))
+    assert lines(r) == [[(300, "gram", "Wasser"), (1, "teaspoon", "Zucker")], [(None, None, "Rosmarin")]]
+    assert [g["name"] for g in r["ingredientGroups"]] == [None, None]
+
+
+def test_rescaling_helper_numbers_are_dropped_silently():
+    r = build_recipe(sheet(G11=4, C3=350, D3="Gramm", E3="Rigatoni", C33=350, D33=87.5, C34=1, D34=0.25, C10=125))
+    assert lines(r) == [[(350, "gram", "Rigatoni")]]
+    assert r["reviewReasons"] == []
+
+
+def test_oder_row_starts_alternative_group():
+    r = build_recipe(sheet(G11=8, C3=375, D3="Gramm", E3="Mehl", C5="ODER", C7=295, D7="Gramm", E7="Mehl", C8=80, D8="Gramm", E8="Proteinpulver"))
+    assert [g["name"] for g in r["ingredientGroups"]] == [None, "Alternative"]
+    assert Reason.ALTERNATIVE in r["reviewReasons"]
+
+
+def test_duplicate_ingredient_list_is_dropped():
+    first = dict(C3=100, D3="Gramm", E3="Mehl", C4=3, E4="Eier", C5=125, D5="Milliliter", E5="Milch")
+    copy = dict(D20="100 g", E20="Mehl", D21="3", E21="Eier", D22="125 ml", E22="Milch")
+    r = build_recipe(sheet(G11=2, **first, **copy))
+    assert len(r["ingredientGroups"]) == 1
+    assert Reason.DUPLICATE_LIST in r["reviewReasons"]
+
+
+def test_amount_typed_into_unit_column():
+    r = build_recipe(sheet(G11=2, D3="100 g", E3="Mehl", D4="1 TL", E4="Zucker"))
+    assert lines(r) == [[(100, "gram", "Mehl"), (1, "teaspoon", "Zucker")]]
+
+
+def test_ranges_fractions_and_unknown_amounts():
+    r = build_recipe(sheet(G11=2, C3="2-3", D3="Esslöffel", E3="Kakao", C4="½", D4="Bund", E4="Schnittlauch", C5="opt. als Topping", E5="Sesam"))
+    first, second, third = r["ingredientGroups"][0]["lines"]
+    assert (first["amount"], first["amountMax"]) == (2, 3)
+    assert second["amount"] == 0.5
+    assert (third["amount"], third["remark"]) == (None, "opt. als Topping")
+    assert Reason.UNKNOWN_AMOUNT.format(amount="opt. als Topping") in r["reviewReasons"]
+
+
+def test_parenthesised_line_is_optional():
+    r = build_recipe(sheet(G11=4, C3="(2", D3="Teelöffel", E3="Oregano)"))
+    line = r["ingredientGroups"][0]["lines"][0]
+    assert (line["amount"], line["unit"], line["name"], line["remark"]) == (2, "teaspoon", "Oregano", "optional")
+
+
+def test_unknown_unit_moves_into_name():
+    r = build_recipe(sheet(G11=2, C3=2, D3="Würfel", E3="Brühe"))
+    assert lines(r) == [[(2, None, "Würfel Brühe")]]
+    assert Reason.UNKNOWN_UNIT.format(unit="Würfel") in r["reviewReasons"]
+
+
+def test_comment_group_markers_name_groups_and_other_comments_become_remarks():
+    comments = {"C3": "Zutaten für die Rouladen", "C5": "Ab hier: \nZutaten für die Sauce", "E4": "kleiner Espresso-Löffel"}
+    r = build_recipe(sheet(G11=6, comments=comments, C3=6, E3="Rinderrouladen", C4=1, D4="Teelöffel", E4="Senf", C5=50, D5="Gramm", E5="Butter"))
+    assert [g["name"] for g in r["ingredientGroups"]] == ["Rouladen", "Sauce"]
+    assert r["ingredientGroups"][0]["lines"][1]["remark"] == "kleiner Espresso-Löffel"
+
+
+def test_pro_portion_comment_settles_calories_basis():
+    r = build_recipe(sheet(G11=4, B6=662, comments={"B6": "Pro Portion"}))
+    assert r["calories"] == {"kcal": 662, "basis": "perServing"}
+    assert r["reviewReasons"] == []
+
+
+def test_high_calories_mean_whole_recipe():
+    r = build_recipe(sheet(G11=1, B6=3831))
+    assert r["calories"]["basis"] == "wholeRecipe"
+
+
+def test_comment_b12_and_stray_cells_go_to_notes_and_urls_to_sources():
+    r = build_recipe(
+        sheet(
+            G11=1,
+            B12="Springform 26 cm",
+            I12="ca.",
+            J12="15",
+            K12="Dampfnudeln",
+            B17="https://sallys-blog.de/rezepte/focaccia",
+        )
+    )
+    assert r["notes"] == "Springform 26 cm\n\nca. 15 Dampfnudeln"
+    assert r["sources"] == ["https://sallys-blog.de/rezepte/focaccia"]
+    assert Reason.STRAY_TEXT in r["reviewReasons"]
+
+
+def test_template_comment_is_not_a_note():
+    r = build_recipe(sheet(G11=1))
+    assert r["notes"] is None
+    assert r["reviewReasons"] == []
+
+
+def test_nutrition_rows_in_ingredient_columns_become_notes():
+    r = build_recipe(sheet(G11=1, C3=3, E3="Eier", D18="Kalorien/ 100g", E18="263.47", D19="Fett", E19="17.92"))
+    assert lines(r) == [[(3, None, "Eier")]]
+    assert r["notes"] == "Kalorien/ 100g 263.47\nFett 17.92"
+
+
+def test_instructions_split_over_cells_are_merged_and_flagged():
+    r = build_recipe(sheet(G11=6, F3="Teig rühren.", F19="Spätzle abschrecken."))
+    assert r["stepSections"][0]["steps"] == ["Teig rühren.", "Spätzle abschrecken."]
+    assert Reason.MERGED_INSTRUCTIONS in r["reviewReasons"]
+
+
+def test_long_step_is_flagged():
+    r = build_recipe(sheet(G11=1, F3="Rühren. " * 100))
+    assert Reason.LONG_STEP in r["reviewReasons"]
+
+
+def test_unreadable_time_goes_to_notes():
+    r = build_recipe(sheet(G11=1, G3="über Nacht"))
+    assert r["prepMinutes"] is None
+    assert r["notes"] == "Vorbereitungszeit: über Nacht"
+    assert Reason.UNREADABLE_TIME in r["reviewReasons"]

tools/excel-import/tests/test_workbook.py

@@ -0,0 +1,63 @@
+"""Checks against the real (private, untracked) workbook. Skipped when it isn't there."""
+
+import io
+import json
+import os
+import zipfile
+
+import pytest
+from PIL import Image
+
+from cookit_import.bundle import MAX_PHOTO_EDGE, write_bundle
+from cookit_import.recipe import build_recipe, is_recipe_sheet, real_photos, template_photo_hashes
+from cookit_import.workbook import read_sheets
+
+WORKBOOK = os.environ.get("COOKIT_WORKBOOK", os.path.join(os.path.dirname(__file__), "..", "..", "..", "Wochenspeiseplan.xlsm"))
+
+pytestmark = pytest.mark.skipif(not os.path.exists(WORKBOOK), reason="Wochenspeiseplan.xlsm not available")
+
+
+@pytest.fixture(scope="module")
+def recipes():
+    all_sheets = read_sheets(WORKBOOK)
+    placeholders = template_photo_hashes(all_sheets)
+    return {s.title: (build_recipe(s), real_photos(s, placeholders)) for s in all_sheets if is_recipe_sheet(s)}
+
+
+def test_all_recipe_sheets_found(recipes):
+    assert len(recipes) == 214
+    assert "Leeres Rezept" not in recipes and "Spinat-Feta-Pizza" not in recipes
+
+
+def test_focaccia(recipes):
+    r, photos = recipes["Focaccia"]
+    assert [s["title"] for s in r["stepSections"]][:2] == ["Vorteig", "Teig"]
+    assert len(r["ingredientGroups"]) == 3
+    assert len(r["sources"]) == 2
+    assert r["notes"] == "22x22 cm eckige Form"
+    assert photos == []  # only the template's stock picture was on the sheet
+
+
+def test_template_stock_photo_is_dropped_but_real_photos_kept(recipes):
+    without = [title for title, (_, photos) in recipes.items() if not photos]
+    assert 25 <= len(without) <= 35
+    assert len(recipes["Melonen-Minze-Shake"][1]) == 3
+
+
+def test_rinderrouladen_groups_from_comments(recipes):
+    r, _ = recipes["Klassische Rinderrouladen"]
+    assert [g["name"] for g in r["ingredientGroups"]] == ["Rouladen", "Sauce", "Zusätzliche Zutaten"]
+
+
+def test_bundle_round_trip(recipes, tmp_path):
+    subset = [recipes[t] for t in ("Focaccia", "Melonen-Minze-Shake", "Hähnchen")]
+    out = tmp_path / "bundle.zip"
+    write_bundle(str(out), [(dict(r), photos) for r, photos in subset])
+    with zipfile.ZipFile(out) as z:
+        manifest = json.loads(z.read("bundle.json"))
+        assert manifest["format"] == "cookit.recipe-bundle" and manifest["version"] == 1
+        shake = next(r for r in manifest["recipes"] if r["name"] == "Melonen-Minze-Shake")
+        assert len(shake["photos"]) == 3
+        for name in shake["photos"]:
+            img = Image.open(io.BytesIO(z.read(name)))
+            assert img.format == "JPEG" and max(img.size) <= MAX_PHOTO_EDGE