{ "id": "http://trec-kba.org/schemas/v1.0/content-item.json", "description": "Raw data, such as text, used by stream-item for 'title', 'body', and 'anchor'.", "type": "object", "$schema": "http://json-schema.org/draft-03/schema#", "properties": { "$schema": { "description": "URI of this JSON schema document.", "type": "string", "enum": ["http://trec-kba.org/schemas/v1.0/content-item.json"], "required": false, "default": "http://trec-kba.org/schemas/v1.0/content-item.json" }, "raw": { "description": "Raw data downloaded from remote source. This is always treated as a byte array and converted to a string literal using python v2.7.2 String.encode('string-escape'). See http://trec-kba.org/corpus.shtml for more info.", "type": "string", "required": true }, "encoding": { "description": "Character encoding extracted from headers, and if that fails then the first