Python API Tutorial¶

%%bash
# Create test schema
cat > /tmp/disease_schema.yaml << 'EOF'
id: https://example.org/disease
name: disease_schema
prefixes:
  linkml: https://w3id.org/linkml/
imports:
  - linkml:types
default_range: string

classes:
  Disease:
    attributes:
      id:
        identifier: true
      name:
        required: true
      description:
        recommended: true
      synonyms:
        multivalued: true
        recommended: true
      ontology_id:
        recommended: true
EOF

# Create test data
cat > /tmp/disease_good.yaml << 'EOF'
id: DISEASE:001
name: Asthma
description: A chronic respiratory condition
synonyms:
  - bronchial asthma
ontology_id: MONDO:0004979
EOF

cat > /tmp/disease_poor.yaml << 'EOF'
id: DISEASE:002
name: Unknown Disease
EOF

echo "Test files created!"

from linkml_data_qc import ComplianceAnalyzer

# Create an analyzer with your schema
analyzer = ComplianceAnalyzer("/tmp/disease_schema.yaml")

# Analyze a file
report = analyzer.analyze_file("/tmp/disease_good.yaml", "Disease")

print(f"Global Compliance: {report.global_compliance}%")
print(f"Total Checks: {report.total_checks}")
print(f"Total Populated: {report.total_populated}")

# Summary by slot shows compliance for each recommended field
print("Summary by Slot:")
for slot, percentage in report.summary_by_slot.items():
    print(f"  {slot}: {percentage}%")

# Path scores show per-object compliance
print("Path Scores:")
for ps in report.path_scores:
    print(f"  {ps.path}: {ps.overall_percentage}%")
    for ss in ps.slot_scores:
        print(f"    {ss.slot_name}: {ss.populated}/{ss.total} ({ss.percentage}%)")

from linkml_data_qc import SchemaIntrospector

introspector = SchemaIntrospector("/tmp/disease_schema.yaml")

# Get all recommended slots in schema
print(f"Recommended slots: {introspector.recommended_slots}")

# Get class-specific info
class_info = introspector.get_class_slots("Disease")
print(f"\nDisease class recommended: {class_info.recommended_slots}")

from linkml_data_qc import QCConfig, SlotQCConfig

# Create configuration with weights and thresholds
config = QCConfig(
    default_weight=1.0,
    slots={
        "ontology_id": SlotQCConfig(weight=2.0, min_compliance=80.0),
        "description": SlotQCConfig(weight=0.5),
    }
)

# Create analyzer with configuration
analyzer = ComplianceAnalyzer("/tmp/disease_schema.yaml", config)
report = analyzer.analyze_file("/tmp/disease_good.yaml", "Disease")

print(f"Global Compliance: {report.global_compliance}%")
print(f"Weighted Compliance: {report.weighted_compliance}%")

# Analyze poor compliance file with strict threshold
config = QCConfig(
    slots={
        "description": SlotQCConfig(min_compliance=50.0),
    }
)

analyzer = ComplianceAnalyzer("/tmp/disease_schema.yaml", config)
report = analyzer.analyze_file("/tmp/disease_poor.yaml", "Disease")

if report.threshold_violations:
    print("Threshold Violations:")
    for v in report.threshold_violations:
        print(f"  {v.path}.{v.slot_name}: {v.actual_compliance}% < {v.min_required}%")
else:
    print("No violations!")

from linkml_data_qc import JSONFormatter, TextFormatter

analyzer = ComplianceAnalyzer("/tmp/disease_schema.yaml")
report = analyzer.analyze_file("/tmp/disease_good.yaml", "Disease")

# Text format
print(TextFormatter.format(report))

# JSON format
import json
json_output = JSONFormatter.format(report)
print(json.dumps(json.loads(json_output), indent=2))

from linkml_data_qc import analyze_directory, create_multi_file_report

# Analyze all matching files in a directory
reports = analyze_directory(
    schema_path="/tmp/disease_schema.yaml",
    data_dir="/tmp",
    target_class="Disease",
    pattern="disease_*.yaml"
)

# Create aggregated report
multi_report = create_multi_file_report(reports)

print(f"Files Analyzed: {multi_report.files_analyzed}")
print(f"Overall Compliance: {multi_report.global_compliance}%")
print("\nSummary by Slot:")
for slot, pct in multi_report.summary_by_slot.items():
    print(f"  {slot}: {pct}%")