Description

Processors for data validation.

Functions
template<typename Event0, typename Event1, typename Downstream>
auto	tcspc::check_alternating (Downstream downstream)
	Create a processor that checks that events of two types appear in alternation.
template<typename DataTypes = default_data_types, bool RequireStrictlyIncreasing = false, typename Downstream>
auto	tcspc::check_monotonic (Downstream downstream)
	Create a processor that checks that abstime is monotonically increasing or nondecreasing.

Function Documentation

◆ check_alternating()

template<typename Event0, typename Event1, typename Downstream>

auto tcspc::check_alternating ( Downstream downstream )

Create a processor that checks that events of two types appear in alternation.

The processor passes through all events. It examines events of types Event0 and Event1, and checks that they alternate, starting with Event0. If a violation is detected, a tcspc::warning_event is emitted just before the offending event.

Template Parameters

Event0	event type expected first
Event1	event type expected second
Downstream	downstream processor type (usually deduced)

Parameters

downstream downstream processor

Returns: processor

Events handled

Event0, Event1: if not strictly alternating, starting with Event0, emit tcspc::warning_event; pass through
All other types: pass through with no action
Flush: pass through with no action

◆ check_monotonic()

template<typename DataTypes = default_data_types, bool RequireStrictlyIncreasing = false, typename Downstream>

auto tcspc::check_monotonic ( Downstream downstream )

Create a processor that checks that abstime is monotonically increasing or nondecreasing.

The processor passes through time-tagged events and checks that their abstime is monotonic (that is, increasing or non-decreasing). The event's abstime field type must match DataTypes::abstime_type. If a violation is detected, a tcspc::warning_event is emitted just before the offending event.

Checking abstime monotonicity is often a good way to detect gross issues in the data, such as reading data in an incorrect format or using text mode to read binary data.