Lists & tables

Lists of itemsCopy Anchor Link

Pages repeat the same block: products, reviews, search results. items returns one object per block, with fields read inside it:

"reviews": [
  { "author": "Ana", "stars": 5, "date": "2026-10-02", "text": "Bright and sturdy." },
  { "author": "Ben", "stars": 3, "date": "2026-09-18", "text": "Good, a bit heavy." }
]

• selector (or xpath) picks the blocks; each item's fields are searched inside its block only. A field missing from a block is null, so every object has the same keys.
• items can be an object (named values) or an array, like fields.
• A list returns every block. "multiple": false returns only the first one, as a single object.
• In XPath, start item fields with . to stay inside the block: .//h2. //h2 searches the whole page.
• Lists can contain lists, three levels deep.

TablesCopy Anchor Link

"output": "table" reads a <table>, or the first table inside the element, as rows. With a header row (<th> cells), each row is an object keyed by the headers:

"specs": [
  { "Property": "Power", "Value": "8 W" },
  { "Property": "Height", "Value": "45 cm" },
  { "Property": "Weight", "Value": "1,2 kg" }
]

Without headers, each row is an array of cells. Empty headers become column_1, column_2…, and repeated ones get a suffix. With "multiple": true, every matching table is returned.