API Reference¶

TracePipe provides a minimal API surface for maximum functionality.

Quick Reference¶

Core Functions (6)¶

Function	Purpose
`tp.enable()`	Start tracking
`tp.disable()`	Stop tracking
`tp.reset()`	Clear lineage data
`tp.check()`	Health audit
`tp.trace()`	Row journey
`tp.why()`	Cell provenance

Additional Functions¶

Function	Purpose
`tp.stage()`	Label pipeline stages
`tp.register()`	Manually register DataFrames
`tp.report()`	Generate HTML report
`tp.snapshot()`	Capture DataFrame state
`tp.diff()`	Compare snapshots

Namespaces¶

Namespace	Purpose
`tp.contracts`	Data quality contracts
`tp.debug`	Advanced debugging tools

Import¶

import tracepipe as tp

All public functions are available directly from the tp namespace.

Result Objects¶

All query functions return structured result objects:

Function	Returns
`tp.check()`	`CheckResult`
`tp.trace()`	`TraceResult`
`tp.why()`	`WhyResult`
`tp.contract().check()`	`ContractResult`
`tp.snapshot()`	`Snapshot`
`tp.diff()`	`DiffResult`

All result objects support:

print(result) — Pretty-printed output
result.to_dict() — Dictionary representation
Boolean evaluation — if result.passed:

Type Hints¶

TracePipe is fully typed. IDE autocompletion works out of the box:

import tracepipe as tp

result = tp.check(df)  # Autocomplete shows CheckResult methods
result.retention       # Type: float
result.warnings        # Type: list[str]