moved from urllib to requests

README.md

@@ -88,13 +88,9 @@ docker-compose -p recipe-test up
 
 running tests
 ```sh
-pytest --cov=src/recipe_graph --cov-report lcov --cov-report html
+pytest
 ```
 
-The html report is under `htmlcov/` and can be viewed through any browser.
-The `lcov` file can be used for the [Coverage Gutters](https://marketplace.visualstudio.com/items?itemName=ryanluker.vscode-coverage-gutters)
-plugin for VS Code to view coverage in your editor.
-
 **WARNINING**: If you get `ERROR at setup of test_db_connection` and 
 `ERROR at setup of test_db_class_creation`, please check if testing database is
 already initiated. Testing is destructive and should be done on a fresh database.
@@ -109,6 +105,16 @@ docker-compose -p recipe-test down
 Test are written in pytest framework. Currently focused on unittest and code
 coverage. Integration tests to come.
 
+To run test use:
+```sh
+pytest --cov=src/recipe_graph --cov-report lcov --cov-report html
+```
+
+The html report is under `htmlcov/` and can be viewed through any browser.
+The `lcov` file can be used for the [Coverage Gutters](https://marketplace.visualstudio.com/items?itemName=ryanluker.vscode-coverage-gutters)
+plugin for VS Code to view coverage in your editor.
+
+
 ## TODO
  > ☑ automate scraping\
  > ☑ extracting quantity and name (via regex)\

pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.0"]
 build-backend = "setuptools.build_meta"
 
 [project]
-name = "recipe_graph"
+name = "recepie_graph"
 version = "0.0.1"
 description = "mapping out recipes relations"
 dependencies = [

src/recipe_graph/scrape.py

@@ -9,53 +9,40 @@ from urllib.parse import urljoin
 import logging
 from argparse import ArgumentParser
 
-
-def ingredient_regex(units: list[str], instructions: list[str]) -> re.Pattern:
-    number_regex = "((?:[\d\\./\\u00BC-\\u00BE\\u2150-\\u215E]*\s?(?:\(.+\))?)*)"
-    ingredient_regex = "([a-zA-Z '\-]+)"
-    supplement_regex = ",?(.*)"
-    units_regex = "|".join(
-        [f"[{unit[0]}{unit[0].capitalize()}]{unit[1:]}" for unit in units]
-    )
+def parse_ingredient(ingredient_text):
+    units = ['teaspoon', 'tablespoon', 'gram', 'ounce', 'jar', 'cup', 'pinch',
+             'container', 'slice', 'package', 'pound', 'can', 'dash', 'spear',
+             'bunch', 'quart', 'cube', 'envelope', 'square', 'sprig', 'bag',
+             'box', 'drop', 'fluid ounce', 'gallon', 'head', 'link', 'loaf',
+             'pint', 'pod', 'sheet', 'stalk', 'whole', 'bar', 'bottle', 'bulb',
+             'year', 'fillet', 'litter', 'packet', 'slices']
+    instructions = ['and', 'or', 'chopped', 'diced', 'brewed', 'chilled',
+                    'chunky', 'small', 'medium', 'large', 'couarse', 'cracked',
+                    'crushed', 'ground', 'cooked', 'cubed', 'crumbled', 'cut',
+                    'cold', 'hot', 'warm', 'day', 'old', 'drained', 'canned',
+                    'dried', 'dry', 'fine', 'firm', 'fresh', 'frozen',
+                    'grated', 'grilled', 'hard', 'hot', 'juliened?', 'leftover',
+                    'light', 'lite', 'mashed', 'melted', 'minced', 'packed',
+                    'peeled', 'pitted', 'sliced', 'prepared', 'refrigerated',
+                    'rehydrated', 'seedless', 'shaved', 'shredded', 'sifted',
+                    'sieved', 'shucked', 'slivered', 'thick', 'sliced', 'thin',
+                    'toasted', 'trimmed', 'unbaked', 'uncooked', 'unpeeled',
+                    'unopened', 'unseasoned']
+    number_regex = '((?:[\d\\./\\u00BC-\\u00BE\\u2150-\\u215E]*\s?(?:\(.+\))?)*)'
+    ingredient_regex = '([a-zA-Z \'\-]+)'
+    supplement_regex = ',?(.*)'
+    units_regex = "|".join([f'[{unit[0]}{unit[0].capitalize()}]{unit[1:]}' 
+                            for unit in units])
     units_regex = f"((?:(?:{units_regex})e?s?)?)"
-    instructions_regex = "|".join(
-        [f"[{inst[0]}{inst[0].capitalize()}]{inst[1:]}" for inst in instructions]
-    )
+    instructions_regex = "|".join([f'[{inst[0]}{inst[0].capitalize()}]{inst[1:]}'
+                                  for inst in instructions])
     instructions_regex = f"((?:(?:(?:{instructions_regex})(?:ly)?)| )*)"
     
-    return re.compile(
-        number_regex
-        + units_regex
-        + instructions_regex
-        + ingredient_regex
-        + supplement_regex
-    )
-
-# TODO: load units and instructions from config.
-# Moved data into optional parameters for the time being.
-def parse_ingredient(
-    ingredient_text: str,
-    units: list[str] = [ "teaspoon", "tablespoon", "gram", "ounce", "jar", 
-                        "cup", "pinch", "container", "slice", "package",
-                        "pound", "can", "dash", "spear", "bunch", "quart",
-                        "cube", "envelope", "square", "sprig", "bag", "box",
-                        "drop", "fluid ounce", "gallon", "head", "link",
-                        "loaf", "pint", "pod", "sheet", "stalk", "whole",
-                        "bar", "bottle", "bulb", "year", "fillet", "litter",
-                        "packet", "slices"],
-    instructions: list[str] = [
-        "and", "or", "chopped", "diced", "brewed", "chilled", "chunky", "small",
-        "medium", "large", "couarse", "cracked", "crushed", "ground", "cooked",
-        "cubed", "crumbled", "cut", "cold", "hot", "warm", "day", "old",
-        "drained", "canned", "dried", "dry", "fine", "firm", "fresh", "frozen",
-        "grated", "grilled", "hard", "hot", "juliened?", "leftover", "light",
-        "lite", "mashed", "melted", "minced", "packed", "peeled", "pitted",
-        "sliced", "prepared", "refrigerated", "rehydrated", "seedless", "shaved",
-        "shredded", "sifted", "sieved", "shucked", "slivered", "thick", "sliced",
-        "thin", "toasted", "trimmed", "unbaked", "uncooked", "unpeeled",
-        "unopened", "unseasoned"],
-):
-    regex = ingredient_regex(units, instructions) 
+    regex = re.compile(number_regex +
+                       units_regex + 
+                       instructions_regex +
+                       ingredient_regex + 
+                       supplement_regex)
     
     m = regex.match(ingredient_text)
     logging.info(f"Parsed {ingredient_text}, found: {m}")
@@ -64,120 +51,94 @@ def parse_ingredient(
 
     return [text.strip() if text else None for text in m.groups()]
 
-# this code is unused
-# TODO: add tests when this is used
-def missing_ingredients_query(session):
-    cte = (
-        except_(select(db.RecipeIngredient.id), select(db.RecipeIngredientParts.id))
-    ).alias("missing")
-    missing = (
-        session.query(db.RecipeIngredient).where(db.RecipeIngredient.id.in_(cte)).all()
-    )
-    return missing
 
-# this code is unused
-# TODO: add tests when this is used
-def parse_missing_ingredients(session):
-    missing = missing_ingredients_query(session)
+
+def reparse_ingredients(session):
+    cte = (except_(select(db.RecipeIngredient.id),
+                  select(db.RecipeIngredientParts.id))).\
+            alias('missing')
+    missing = session.query(db.RecipeIngredient).where(db.RecipeIngredient.id.in_(cte)).all()
+
     for ingredient in missing:
-        parts = ingredient_to_parts(ingredient)
-        session.add(parts)
+        parts = parse_ingredient(ingredient.text)
+        if not parts:
+            continue
+        quantity, unit, instruction, name, supplement = parts
+        session.add(db.RecipeIngredientParts(id = ingredient.id,
+                                             quantity = quantity,
+                                             unit = unit,
+                                             instruction = instruction,
+                                             ingredient = name,
+                                             supplement = supplement))
 
 
-def load_page(recipe_url: str) -> bs4.BeautifulSoup:
+
+def load_page(recipe_url):
     try:    
-        logging.info(f"Loading Page: {recipe_url}")
-        with req.get(recipe_url) as resp:
-            if resp.status_code == 404:
+        logging.info(f'Loading Page: {recipe_url}')
+        with req.get(recipe_url) as f:
+            if f.status_code == 404:
                 raise Exception(f"Page does not exist (404): {recipe_url}")
-            return bs4.BeautifulSoup(resp.text, "html.parser")
+            return bs4.BeautifulSoup(f.read().decode(), 'html.parser')
 
     except Exception as e:
         logging.warning(f"Could not download or parse recipe: {recipe_url}")
         logging.warning(e)
 
 
-def parse_recipe_name(
-    site: db.RecipeSite,
-    page: bs4.BeautifulSoup,
-    recipe: db.Recipe,
-    url: str = None,
-) -> db.Recipe:
-    if not url:
-        url = {"site": site.base_url, "recipe": recipe.identifier}
-    name_candidates = page.find_all(class_=site.name_class)
-    if len(name_candidates) == 0:
-        raise Exception(f"Could not extract recipe name: {url}")
-    name_div = name_candidates[0]
-    recipe.name = name_div.text
-
-    logging.info(f"Adding Recipe {recipe.name} from {url}")
-
-    return recipe
-
-def ingredient_to_parts(
-    ingredient: db.Ingredient
-) -> db.RecipeIngredientParts:
-    parts = parse_ingredient(ingredient.text)
-    if parts:
-        quantity, unit, instruction, ingredient_name, supplement = parts
-        return db.RecipeIngredientParts(
-            id=ingredient.id,
-            quantity=quantity,
-            unit=unit,
-            instruction=instruction,
-            ingredient=ingredient_name,
-            supplement=supplement,
-        )
-
-
 def parse_recipe(session, recipe, site):
     recipe_url = urljoin(site.base_url, str(recipe.identifier))
     recipe_page = load_page(recipe_url)
     if not recipe_page:
         return None
 
-    recipe = parse_recipe_name(site, recipe_page, recipe, recipe_url)
+    name_candidates = recipe_page.find_all(class_=site.name_class)
+    if len(name_candidates) == 0:
+        raise Exception(f"Could not extract recipe name: {recipe_url}")
+    name_div = name_candidates[0]
+    recipe.name = name_div.text
+
+    logging.info(f"Adding Recipe {recipe.name} from {recipe_url}")   
+
     session.add(recipe)
     session.flush()
 
-    candidates = recipe_page.find_all(class_=site.ingredient_class)
-    for candidate in candidates:
-        ingredient = db.RecipeIngredient(text=candidate.text, recipe_id=recipe.id)
-        session.add(ingredient)
+    ingred_candidates = recipe_page.find_all(class_=site.ingredient_class)
+    for candidate in ingred_candidates:
+        ingred = db.RecipeIngredient(text=candidate.text,
+                                         recipe_id=recipe.id)
+        session.add(ingred)  
         session.flush()
                            
-        parts = ingredient_to_parts(ingredient)
+        parts = parse_ingredient(ingred.text)
         if parts:
-            session.add(parts)
+            quantity, unit, instruction,ingredient, supplement = parts
+            ingred_parts = db.RecipeIngredientParts(id = ingred.id,
+                                                 quantity = quantity,
+                                                 unit = unit,
+                                                 instruction = instruction,
+                                                 ingredient = ingredient,
+                                                 supplement = supplement)
+            session.add(ingred_parts)
         
-    logging.info(f"{len(candidates)} ingredients found. Inserting into DB")
+    logging.info(f"{len(ingred_candidates)} ingredients found. Inserting into DB")
     
     return recipe
     
-
 def main(): # pragma: no cover
     parser = ArgumentParser(description="Scrape a recipe site for recipies")
-    parser.add_argument("site", help="Name of site")
-    parser.add_argument(
-        "-id",
-        "--identifier",
-        dest="id",
-        help="url of recipe(reletive to base url of site) or commma seperated list",
-    )
-    parser.add_argument(
-        "-a",
-        "--auto",
-        action="store",
-        dest="n",
-        help="automaticaly generate identifier(must supply number of recipies to scrape)",
-    )
-    parser.add_argument("-v", "--verbose", action="store_true")
+    parser.add_argument('site',
+                        help='Name of site')
+    parser.add_argument('-id', '--identifier', dest='id',
+                        help='url of recipe(reletive to base url of site) or commma seperated list')
+    parser.add_argument('-a', '--auto', action='store', dest='n',
+                        help='automaticaly generate identifier(must supply number of recipies to scrape)')
+    parser.add_argument('-v', '--verbose', action='store_true')
 
     args = parser.parse_args(sys.argv)
     if args.verbose:
         logging.basicConfig(level=logging.INFO)
-        logging.getLogger("sqlalchemy.engine").setLevel(logging.INFO)
+        logging.getLogger('sqlalchemy.engine').setLevel(logging.INFO)
 
     eng = db.get_engine()
     S = sessionmaker(eng)
@@ -190,29 +151,27 @@ def main():  # pragma: no cover
         starting_id = 0
         if args.id and not args.n:
             recipe_ids.append(args.id)
-            logging.info(f"Retreiving single recipe: {args.id}")
+            logging.info(f'Retreiving single recipe: {args.id}')
         elif args.n:
             if not args.id:
-                last_recipe = (
-                    sess.query(db.Recipe)
-                    .where(db.Recipe.recipe_site_id == site.id)
-                    .order_by(desc(db.Recipe.identifier))
-                    .limit(1)
-                    .scalar()
-                )
+                last_recipe =  sess.query(db.Recipe).\
+                                where(db.Recipe.recipe_site_id == site.id).\
+                                order_by(desc(db.Recipe.identifier)).\
+                                limit(1).\
+                                scalar()
                 starting_id = int(last_recipe.identifier) + 1
             else:
                 starting_id = int(args.id)
-            recipe_ids = range(starting_id, starting_id + int(args.n))
-            logging.info(
-                f"Retreving {args.n} recipes from {site.base_url} starting at {starting_id}"
-            )
+            recipe_ids = range(starting_id, starting_id+int(args.n))
+            logging.info(f'Retreving {args.n} recipes from {site.base_url} starting at {starting_id}')
+        
+        
         
         for recipe_id in recipe_ids:            
             try: 
                 savepoint = sess.begin_nested()
 
-                recipe = db.Recipe(identifier=recipe_id, recipe_site_id=site.id)
+                recipe = db.Recipe(identifier = recipe_id, recipe_site_id = site.id)
                 parse_recipe(sess, recipe, site)
 
                 savepoint.commit()

test/test_scrape.py

@@ -1,115 +1,12 @@
 from recipe_graph import scrape
 from bs4 import BeautifulSoup
-from recipe_graph.db import RecipeSite, Recipe, RecipeIngredient, RecipeIngredientParts
 
-from pytest import fixture
-
-
-@fixture
-def mock_site():
-    return RecipeSite(
-        name="mock-site",
-        ingredient_class="mock-ing",
-        name_class="mock-name",
-        base_url="example-site/mock-site",
-    )
-
-# TODO: should probably load HTML from file
-@fixture
-def mock_page():
-    return BeautifulSoup(
-        """
-                    <header></header><body>
-                        <div class="mock-name">test_recipe</div>
-                        <div class="mock-ing">test_ingredient</div>
-                    </body>
-                  """,
-        "html.parser",
-    )
-
-
-@fixture
-def mock_blank_page():
-    return BeautifulSoup(""" <header></header><body> </body> """, "html.parser")
-
-
-@fixture
-def mock_recipe():
-    return Recipe(name="test_recipe", identifier="mock_1")
-
-
-@fixture
-def mock_ingredient():
-    return RecipeIngredient(text="1 ounce water")
-
-@fixture
-def mock_url():
-    return "example-site/mock-site"
+import pytest
 
 
 def test_load_page():
-    page = scrape.load_page("https://www.google.com")
+    page = scrape.load_recipe("https://hs.andreistoica.ca:4943")
     assert type(page) == BeautifulSoup
 
-    page = scrape.load_page("https://www.google.com/some-nonsense")
+    page = scrape.load_recipe("https://hs.andreistoica.ca:4943/some-nonesense")
     assert page == None
-
-
-def test_ingredient_regex():
-    regex = scrape.ingredient_regex(["cup"], ["crushed"])
-    assert (
-        regex.pattern
-        == "((?:[\\d\\./\\u00BC-\\u00BE\\u2150-\\u215E]*\\s?(?:\\(.+\\))?)*)((?:(?:[cC]up)e?s?)?)((?:(?:(?:[cC]rushed)(?:ly)?)| )*)([a-zA-Z '\\-]+),?(.*)"
-    )
-    regex = scrape.ingredient_regex(["cup", "ounce"], ["crushed", "ground"])
-    assert (
-        regex.pattern
-        == "((?:[\\d\\./\\u00BC-\\u00BE\\u2150-\\u215E]*\\s?(?:\\(.+\\))?)*)((?:(?:[cC]up|[oO]unce)e?s?)?)((?:(?:(?:[cC]rushed|[gG]round)(?:ly)?)| )*)([a-zA-Z '\\-]+),?(.*)"
-    )
-
-def test_parse_ingredient(mock_ingredient):
-    parts = scrape.parse_ingredient(mock_ingredient.text)
-    assert len(parts) > 0
-    assert parts == ['1', 'ounce', '', 'water', None]
-
-    parts = scrape.parse_ingredient("Water")
-    assert len(parts) > 0
-    assert parts == [None, None, None, 'Water', None]
-
-    parts = scrape.parse_ingredient("")
-    assert parts == None
-
-def test_parse_recipe_name(mock_site, mock_page, mock_recipe, mock_url, mock_blank_page,):
-    expected_name = mock_recipe.name
-    mock_recipe.name = None
-
-    mock_recipe = scrape.parse_recipe_name(
-        mock_site,
-        mock_page,
-        mock_recipe,
-    )
-    assert mock_recipe.name == expected_name
-
-    ex = None
-    try:
-        mock_recipe = scrape.parse_recipe_name(
-            mock_site,
-            mock_blank_page,
-            mock_recipe,
-        )
-    except Exception as e:
-        ex = e
-        url = {"site": mock_site.base_url, "recipe": mock_recipe.identifier}
-        assert str(e) == f"Could not extract recipe name: {url}"
-    assert ex
-
-
-
-def test_ingredient_to_parts(mock_ingredient):
-    parts = scrape.ingredient_to_parts(mock_ingredient)
-    assert parts.quantity == "1"
-    assert parts.unit == "ounce"
-    assert parts.instruction == ""
-    assert parts.ingredient == "water"
-    assert parts.supplement == None
-