Files
GEWU_agent/contracts/schemas/SourceRevisionRecord.schema.json

205 lines
3.9 KiB
JSON

{
"$defs": {
"ReliabilityGrade": {
"enum": [
"low",
"medium",
"high"
],
"title": "ReliabilityGrade",
"type": "string"
}
},
"additionalProperties": false,
"properties": {
"archive_status": {
"enum": [
"archived",
"reference-only",
"capture-failed"
],
"title": "Archive Status",
"type": "string"
},
"artifact_sha256": {
"anyOf": [
{
"pattern": "^[a-f0-9]{64}$",
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Artifact Sha256"
},
"authors": {
"default": [],
"items": {
"type": "string"
},
"maxItems": 50,
"title": "Authors",
"type": "array"
},
"byte_length": {
"minimum": 0,
"title": "Byte Length",
"type": "integer"
},
"capture_method": {
"default": "unspecified",
"enum": [
"unspecified",
"local-file",
"url-fetch",
"pasted-text",
"notebooklm-export",
"transcript"
],
"title": "Capture Method",
"type": "string"
},
"captured_at": {
"format": "date-time",
"title": "Captured At",
"type": "string"
},
"final_url": {
"anyOf": [
{
"maxLength": 4000,
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Final Url"
},
"ingested_for": {
"items": {
"maxLength": 128,
"minLength": 1,
"pattern": "^[a-zA-Z0-9][a-zA-Z0-9._:-]*$",
"type": "string"
},
"maxItems": 50,
"minItems": 1,
"title": "Ingested For",
"type": "array"
},
"license": {
"maxLength": 500,
"minLength": 1,
"title": "License",
"type": "string"
},
"media_type": {
"maxLength": 128,
"minLength": 1,
"title": "Media Type",
"type": "string"
},
"original_url": {
"anyOf": [
{
"maxLength": 4000,
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Original Url"
},
"publication_date": {
"anyOf": [
{
"format": "date-time",
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Publication Date"
},
"publisher": {
"anyOf": [
{
"maxLength": 300,
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Publisher"
},
"reliability_grade": {
"$ref": "#/$defs/ReliabilityGrade"
},
"revision_id": {
"maxLength": 128,
"minLength": 1,
"pattern": "^[a-zA-Z0-9][a-zA-Z0-9._:-]*$",
"title": "Revision Id",
"type": "string"
},
"schema_version": {
"const": 1,
"default": 1,
"title": "Schema Version",
"type": "integer"
},
"source_id": {
"maxLength": 128,
"minLength": 1,
"pattern": "^[a-zA-Z0-9][a-zA-Z0-9._:-]*$",
"title": "Source Id",
"type": "string"
},
"source_type": {
"enum": [
"paper",
"review",
"press",
"vendor",
"internal",
"talk",
"export",
"other"
],
"title": "Source Type",
"type": "string"
},
"title": {
"maxLength": 500,
"minLength": 1,
"title": "Title",
"type": "string"
}
},
"required": [
"source_id",
"revision_id",
"media_type",
"byte_length",
"captured_at",
"title",
"license",
"source_type",
"reliability_grade",
"ingested_for",
"archive_status"
],
"title": "SourceRevisionRecord",
"type": "object"
}