Add nosalty.hu parser

Extracts ingredients (with groups), instructions (with section
headers), tags, and story-as-description from nosalty.hu recipe pages.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

app/scraper.py

@@ -298,6 +298,123 @@ def _parse_streetkitchen(soup: BeautifulSoup, url: str) -> dict:
     }
 
 
+# ---------------------------------------------------------------------------
+# nosalty.hu
+# ---------------------------------------------------------------------------
+
+
+@_register("nosalty")
+def _parse_nosalty(soup: BeautifulSoup, url: str) -> dict:
+    title = _og(soup, "og:title") or _text(soup.find("title"))
+    if title:
+        title = re.sub(r"\s*\|.*$", "", title).strip()
+
+    # Story as description (no dedicated description on nosalty)
+    description = ""
+    story = soup.find("div", id="recipe-story")
+    if story:
+        paragraphs = [p.get_text(strip=True) for p in story.find_all("p")
+                      if p.get_text(strip=True)]
+        description = " ".join(paragraphs)
+
+    image_url = _og(soup, "og:image")
+
+    # --- Ingredients ---
+    # Scoped to div#ingredients to avoid per-serving / nutrition duplicates.
+    # Structure: h3.m-list__title = group header, ul.m-list__list = ingredient rows.
+    ingredients = []
+    ing_container = soup.find("div", id="ingredients")
+    if ing_container:
+        for el in ing_container.find_all(["h3", "ul"]):
+            cls = el.get("class") or []
+            if el.name == "h3" and "m-list__title" in cls:
+                group_name = el.get_text(strip=True)
+                if group_name:
+                    ingredients.append({"group": group_name})
+            elif el.name == "ul" and "m-list__list" in cls:
+                for li in el.find_all("li", class_="m-list__item"):
+                    _parse_nosalty_ingredient(li, ingredients)
+
+    # --- Instructions ---
+    # Container: div#select inside div.p-recipe__directions.
+    # h4.m-list__title = section header, ol.m-list__list = steps.
+    instructions = []
+    dir_container = soup.find("div", id="select")
+    if dir_container:
+        for el in dir_container.find_all(["h4", "ol"]):
+            cls = el.get("class") or []
+            if el.name == "h4" and "m-list__title" in cls:
+                section_name = el.get_text(strip=True)
+                if section_name:
+                    instructions.append(f"--- {section_name} ---")
+            elif el.name == "ol" and "m-list__list" in cls:
+                for li in el.find_all("li", class_="m-list__item"):
+                    txt = li.get_text(strip=True)
+                    if txt:
+                        instructions.append(txt)
+
+    # --- Tags ---
+    tags = []
+    for a in soup.find_all("a", class_="m-tags__tagItem"):
+        tag_text = a.get_text(strip=True)
+        if tag_text:
+            tags.append(tag_text)
+
+    return {
+        "title": title or "Ismeretlen recept",
+        "description": description,
+        "image_url": image_url,
+        "ingredients": ingredients,
+        "instructions": instructions,
+        "tags": tags,
+        "original_url": url,
+    }
+
+
+def _parse_nosalty_ingredient(li, ingredients: list):
+    """Parse a single nosalty ingredient <li> into the ingredients list."""
+    inner = li.find("div")
+    if not inner:
+        return
+
+    food_el = inner.find("a", class_="a-link")
+    if not food_el:
+        return
+
+    food = food_el.get_text(strip=True)
+    if not food:
+        return
+
+    # Walk children of inner div in order.
+    # Spans before the <a> link = quantity, spans after = extra/note.
+    qty_raw = ""
+    extra_parts = []
+    before_link = True
+    for child in inner.children:
+        if child is food_el:
+            before_link = False
+            continue
+        if not hasattr(child, "get_text"):
+            continue
+        text = child.get_text(strip=True)
+        if not text:
+            continue
+        if before_link:
+            qty_raw = text
+        else:
+            extra_parts.append(text.strip("() "))
+
+    extra = "; ".join(p for p in extra_parts if p)
+    qty, unit = _split_qty_unit(qty_raw)
+
+    ingredients.append({
+        "quantity": qty,
+        "unit": unit,
+        "food": food,
+        "extra": extra,
+    })
+
+
 def _split_qty_unit(raw: str) -> tuple[str, str]:
     """Split a merged quantity+unit string like '200g' into ('200', 'g')."""
     raw = raw.strip()