```json { "classifications": [ { "column": "batch_id", "role": "identifier", "confidence": 1.0, "reason": "Primary key with unique values per row (304 distinct in 304 rows), used to identify batches" }, { "column": "dataset_id", "role": "identifier", "confidence": 1.0, "reason": "Foreign key referencing datasets, used for JOINs (96 distinct values, repeating)" }, { "column": "tenant_id", "role": "dimension", "confidence": 1.0, "reason": "Categorical field for filtering/grouping by tenant (only 1 value 'default' currently)" }, { "column": "started_at", "role": "dimension", "confidence": 1.0, "reason": "Timestamp for filtering, grouping by time periods, or ordering" }, { "column": "completed_at", "role": "dimension", "confidence": 1.0, "reason": "Timestamp for filtering, grouping by time periods, or ordering" }, { "column": "status", "role": "dimension", "confidence": 1.0, "reason": "Low-cardinality categorical field (3 values: completed, ready, running) for grouping/filtering" }, { "column": "error_text", "role": "attribute", "confidence": 1.0, "reason": "Free-form descriptive text for display, not for grouping or aggregation (currently all null)" }, { "column": "total_rows", "role": "measure", "confidence": 1.0, "reason": "Numeric quantity to aggregate (sum/avg) representing row counts" }, { "column": "total_bytes", "role": "measure", "confidence": 1.0, "reason": "Numeric quantity to aggregate (sum/avg) representing byte counts" }, { "column": "elapsed_ms", "role": "measure", "confidence": 1.0, "reason": "Numeric quantity to aggregate (sum/avg) representing elapsed time in milliseconds" }, { "column": "engine_versions_json", "role": "attribute", "confidence": 0.95, "reason": "JSON text describing engine versions, primarily for display; low cardinality (7) but complex structure not suitable for grouping" }, { "column": "nettle_version", "role": "dimension", "confidence": 1.0, "reason": "Low-cardinality categorical field (2 values) for grouping/filtering by version" }, { "column": "job_id", "role": "identifier", "confidence": 1.0, "reason": "Identifier for jobs (295 distinct values), used for JOINs or lookups" }, { "column": "content_hash", "role": "identifier", "confidence": 0.95, "reason": "Hash value used to identify unique content (85 distinct values), high null rate but serves as identifier when present" } ] } ```