{
    "id": "http://trec-kba.org/schemas/v1.0/corpus-item.json",
    "description": "An item in a corpus of text documents.  Typically has one or more properties of type content-item.",
    "type": "object",
    "$schema": "http://json-schema.org/draft-03/schema#",
    "properties": {
        "$schema": {
            "description": "URI of this JSON schema document; provided as a property in every instance document.",
            "type": "string",
            "enum": [
                "http://trec-kba.org/schemas/v1.0/corpus-item.json",
                "http://trec-kba.org/schemas/v1.0/stream-item.json"
                ],
            "required": true,
            "default": "http://trec-kba.org/schemas/v1.0/corpus-item.json"
            },
        "doc_id": {
            "description": "md5 hash of abs_url",
            "title": "Document ID",
            "type": "string",
            "pattern": "^[a-f0-9]{32}$",
            "required": true
            },
	"abs_url": {
            "description": "Normalized absolute URL derived from original_url.",
            "title": "Normalized URL",
            "type": "string",
            "required": true
            },
	"schost": {
            "description": "scheme://netloc from abs_url",
            "title": "scheme://netloc",
            "type": "string",
            "required": true
            },
	"original_url": {
            "description": "URL provided by source, only non-null if differs from abs_url.",
            "title": "Original URL",
            "type": ["string", "null"],
            "required": true
            },
        "source": {
            "description": "Identifier of source providing URL and possibly other metadata.",
            "title": "Source",
            "type": "string"
            },
        "title": {
            "description": "Headline for article, often HTML title.",
            "type": [{"$ref": "http://trec-kba.org/schemas/v1.0/content-item.json"}]
            },
        "body": {
            "description": "Primary content of article.",
            "type": [{"$ref": "http://trec-kba.org/schemas/v1.0/content-item.json"}]
            },
        "anchor": {
            "description": "Short commentary for article, typically the anchor text of an HTML in-link.",
            "type": [{"$ref": "http://trec-kba.org/schemas/v1.0/content-item.json"}]
            },
        "source_metadata": {
            "description": "Info provided by source.",
            "type": [
                {"$ref": "http://trec-kba.org/schemas/v1.0/news-metadata.json"},
                {"$ref": "http://trec-kba.org/schemas/v1.0/social-metadata.json"},
                {"$ref": "http://trec-kba.org/schemas/v1.0/linking-metadata.json"}
                ]
            }
        }
    }