Normalize recipe instructions, yield, and image url

Including tests and example recipe data.

mealie/services/scrape_services.py

@@ -1,3 +1,5 @@
+from typing import List
+
 import json
 from pathlib import Path
 
@@ -12,10 +14,44 @@ CWD = Path(__file__).parent
 TEMP_FILE = CWD.parent.joinpath("data", "debug", "last_recipe.json")
 
 
-def normalize_data(recipe_data: dict) -> dict:
-    if type(recipe_data["recipeYield"]) == list:
-        recipe_data["recipeYield"] = recipe_data["recipeYield"][0]
+def normalize_image_url(image) -> str:
+    if type(image) == list:
+        return image[0]
+    elif type(image) == dict:
+        return image['url']
+    elif type(image) == str:
+        return image
+    else:
+        raise Exception(f"Unrecognised image URL format: {image}")
 
+
+def normalize_instructions(instructions) -> List[dict]:
+    # One long string split by (possibly multiple) new lines
+    if type(instructions) == str:
+        return [{"text": line.strip()} for line in filter(None, instructions.split("\n"))]
+
+    # Plain strings in a list
+    elif type(instructions) == list and type(instructions[0]) == str:
+        return [{"text": step.strip()} for step in instructions]
+
+    # Dictionaries (let's assume it's a HowToStep) in a list
+    elif type(instructions) == list and type(instructions[0]) == dict:
+        return [{"text": step['text'].strip()} for step in instructions if step['@type'] == 'HowToStep']
+
+    else:
+        raise Exception(f"Unrecognised instruction format: {instructions}")
+
+
+def normalize_yield(yld) -> str:
+    if type(yld) == list:
+        return yld[-1]
+    else:
+        return yld
+
+
+def normalize_data(recipe_data: dict) -> dict:
+    recipe_data["recipeYield"] = normalize_yield(recipe_data.get("recipeYield"))
+    recipe_data["recipeInstructions"] = normalize_instructions(recipe_data["recipeInstructions"])
     return recipe_data
 
 
@@ -52,7 +88,7 @@ def process_recipe_url(url: str) -> dict:
     new_recipe.update(mealie_tags)
 
     try:
-        img_path = scrape_image(new_recipe.get("image"), slug)
+        img_path = scrape_image(normalize_image_url(new_recipe.get("image")), slug)
         new_recipe["image"] = img_path.name
     except:
         new_recipe["image"] = None