dq-doctor data quality report
contract v{{ outcome.schema_version }} · generated {{ outcome.finished_at }}
· datasource {{ outcome.datasource }}
· dq-doctor {{ outcome.dqdoctor_version }}
{% if outcome.gate_policy %}· gate: max-failures {{ outcome.gate_policy.max_failures }},
fail-on {{ outcome.gate_policy.fail_on }}{% endif %}
{% for table in outcome.tables %}
{{ table.table_name }}
{% if table.score is none %}
score n/a (no executed checks)
{% elif table.score >= 80 %}
{{ table.score }}/100
{% elif table.score >= 50 %}
{{ table.score }}/100
{% else %}
{{ table.score }}/100
{% endif %}
score v{{ table.score_version }} ·
{{ table.profile.row_count }} rows · {{ table.profile.columns|length }} columns
{% if table.profile.sampling.method == 'sample' %}
· sampled: {{ table.profile.sampling.sample_size }}/{{ table.profile.sampling.row_count }} rows
{% endif %}
{% if table.pii_findings %}
PII findings (informational, not part of the score)
| Column | Type | Evidence |
{% for p in table.pii_findings %}
| {{ p.column }} | {{ p.pii_type }} |
{{ "%.0f"|format(p.matched_fraction * 100) }}% of {{ p.sample_size }} samples |
{% endfor %}
{% endif %}
| Rule | Status | Severity | Detail | Source |
{% for r in table.rule_results %}
{{ r.rule_id }} |
{{ r.status.value }}{% if r.exempted %} (baseline-exempted){% endif %} |
{{ r.severity }} |
{{ r.message }} |
{{ r.source }} |
{% endfor %}
{% if table.profile.columns %}
Column profile
| Column | Type | Nulls | Distinct |
Min | Max | Semantic | PII |
{% for c in table.profile.columns %}
{{ c.name }} |
{{ c.dtype }} |
{{ "%d (%.1f%%)"|format(c.null_count, c.null_rate * 100) }} |
{{ c.distinct_count }} |
{{ redact(c.min_value) if c.min_value is not none else "-" }} |
{{ redact(c.max_value) if c.max_value is not none else "-" }} |
{{ c.inferred_semantic_type }} |
{% if c.pii_type %}{{ c.pii_type }}{% else %}-{% endif %} |
{% endfor %}
Text min/max values are fingerprinted; numeric and
temporal ranges are shown as-is. Sample values are never serialized.
{% endif %}
{% endfor %}
{% if outcome.fk_issues %}
Referential integrity (each relationship counted once)
| Relationship | Source | Orphans |
{% for fk in outcome.fk_issues %}
{{ fk.from_table }}.{{ fk.from_column }} → {{ fk.to_table }}.{{ fk.to_column }} |
{{ 'declared' if fk.declared else 'heuristic' }} |
{{ fk.orphan_rows }}/{{ fk.total_rows }}
{% if fk.error %}{{ fk.error }}{% endif %}
|
{% endfor %}
{% endif %}
Raw sample values are never included in reports; text
min/max values are fingerprinted. Generated by dq-doctor.