Namespace for general, non-LArSoft-specific utilities. More...

Namespaces
	details

	flags
	Classes and functions to manage bit masks and flags.

	manip

	pre_std

	quantities
	Types of variables with a unit.

Classes
struct	add_tag

struct	AddressTaker
	Functor returning the address in memory of the operand. More...

struct	always_false_type
	A `std::false_type` with a template argument. More...

struct	always_true_type
	A `std::true_type` with a template argument. More...

struct	collection_from_reference_type
	Trait with the type of collection referenced by `collRef`. More...

struct	collection_reference_type
	Trait of a type that can be used to reference the collection `Coll`. More...

struct	collection_value_access_type
	Trait of type obtained by access to element of collection `Coll`. More...

struct	collection_value_constant_access_type
	Trait of type obtained by constant access to element of collection `Coll`. More...

struct	collection_value_type
	Trait of value contained in the template collection `Coll`. More...

struct	count_extracted_types
	Counts the elements of a tuple-like type containing a `Target` type. More...

class	count_iterator
	An iterator dereferencing to a counter value. More...

struct	count_type_in_tuple
	Holds whether the `Target` type is element of the specified `std::tuple`. More...

struct	count_type_in_tuple< Target, std::tuple< T... > >

class	DatabaseUtil

class	DataProductChangeTracker_t
	Detects the presence of a new event or data product. More...

struct	Dereferencer
	Functor dereferencing the operand. More...

class	EventChangeTracker_t
	Detects the presence of a new event. More...

struct	extract_to_tuple_type
	Returns type `TargetClass<U...>` from a `SrcTuple<T...>`. More...

struct	find_next_type
	Trait: index of the first occurrence of `T` among the specified `Types`, starting from the one with index `StartFrom`. More...

class	GaussianEliminationAlg

class	GeometryUtilities

class	GridContainerBase1D
	Base class for a container of data arranged on a 1D-grid. More...

class	GridContainerBase2D
	Base class for a container of data arranged on a 2D-grid. More...

class	GridContainerBase3D
	Base class for a container of data arranged on a 3D-grid. More...

class	GridContainerIndicesBase1D
	Index manager for a container of data arranged on a >=1-dim grid. More...

class	GridContainerIndicesBase2D
	Index manager for a container of data arranged on a >=2-dim grid. More...

class	GridContainerIndicesBase3D
	Index manager for a container of data arranged on a >=3-dim grid. More...

struct	has_duplicate_extracted_types
	Traits holding whether elements of `Tuple` have duplicate types. More...

struct	has_extracted_type
	Trait holding whether an element in `Tuple` type contains `Target`. More...

struct	index_of_extracted_type
	Returns the index of the element in `Tuple` with the specified type. More...

struct	is_any_of
	Trait: whether `T` is among the specified `Types`. More...

struct	is_basic_string_type
	Trait: whether type `T` is a STL string type. More...

struct	is_basic_string_view_type
	Trait: whether type `T` is a `std::string_view` type. More...

struct	is_character_type
	Trait: whether type `T` is a character type. More...

struct	is_instance_of
	Trait describing whether `T` is a template instance of `Template`. More...

struct	is_STLarray
	Identifies whether the specified type is a STL array. More...

struct	is_STLarray< std::array< T, N > >

struct	is_string_type
	Trait: whether type `T` is a character string type. More...

struct	KeepByPositionFilterTag
	Tag for filters. More...

class	LArFFT

class	LArFFTW

class	LArFFTWPlan

class	LazyVector
	A contiguous data container expanded on write. More...

struct	lvalue_reference_into_wrapper_type
	Trait with type `T` into `std::reference_wrapper` if reference. More...

class	MappedContainer
	A meta-container providing transparent mapping on top of another. More...

struct	MappedContainerBase
	Non-template base class for `MappedContainer`. More...

class	MultipleChoiceSelection
	Helper to select one among multiple choices via strings. More...

class	MultipleChoiceSelectionBase
	Base class of `util::MultipleChoiceSelection` with basics independent of the option type. More...

class	NormalDistribution

class	PlaneDataChangeTracker_t
	Detects the presence of a new event, data product or wire plane. More...

class	PositionInVolumeFilter
	Use to keep particles with at least part of trajectory in a volume. More...

class	PxHit

class	PxHitConverter

class	PxLine

class	PxPoint

class	Range
	represents a "Range" w/ notion of ordering. A range is defined by a pair of "start" and "end" values. This is stored in std::pair attribute util::Range::_window. This attribute is protected so that the start/end cannot be changed w/o a check that start is always less than end. Note the specialization requires a template class T to have less operator implemented. More...

struct	RangeForWrapperTag
	Tag marking the use of RangeForWrapperBox. More...

struct	reference_addresser
	Functor applying the proper `referenced_address()` function. More...

struct	remove_tag
	Trait holding the type contained in a `TaggedType` (or the type itself). More...

struct	remove_tag< TaggedType< T, Tag > >

struct	self_type
	Trait returning the very same type as in the template argument. More...

class	SignalShaper

class	SignalShaping

struct	span
	Simple class with a begin and an end. More...

struct	span_base
	Non-templated base class for `span`. More...

struct	strip_referenceness_type
	Trait with type `T` stripped of all known reference types. More...

struct	SumSecondFunction

struct	TagExtractor
	Extracts the tag from a type. More...

struct	TaggedType
	A type with a specified tag. More...

struct	TagN
	Tag class parametrized by a sequence of numbers. More...

class	TensorIndices
	Converts a tensor element specification into a linear index. More...

class	TensorIndices< 1U >

struct	TensorIndicesBasicTypes
	Types for TensorIndices class. More...

class	UBDaqID

class	UniqueRangeSet
	std::set of util::Range, which does not allow any overlap in contained element. std::set<Range> w/ modified insert/emplace function. Original std::set does not allow modification of element. I assume what we're interested in is "find if the range already \n exists, and merge if it exists". The insert function does that by recursively looking up overlapping elements w.r.t. input argument of insert function. More...

class	UtilException

class	VectorMap

struct	with_const_as
	Trait with type `Base`, plus the constantness as in `Key`. More...

Typedefs
template<typename Coll >
using	collection_value_t = typename collection_value_type< Coll >::type
	Type contained in the collection `Coll`. More...

template<typename Coll >
using	collection_value_access_t = typename collection_value_access_type< Coll >::type
	Type obtained by constant access to element of collection `Coll`. More...

template<typename Coll >
using	collection_value_constant_access_t = typename collection_value_constant_access_type< Coll >::type
	Type obtained by constant access to element of collection `Coll`. More...

template<typename Coll >
using	collection_reference_t = typename collection_reference_type< Coll >::type
	The type contained in `util::collection_reference_type` trait. More...

template<typename Cont >
using	collection_from_reference_t = typename collection_from_reference_type< Cont >::type
	Type contained in `util::collection_from_reference_type` trait. More...

template<typename T >
using	self_t = typename self_type< T >::type
	The very same type as in the template argument. More...

template<typename A , typename B >
using	is_not_same = std::negation< std::is_same< A, B >>
	The negation of `std::is_same`. More...

template<typename T , typename... Types>
using	find_type = find_next_type< T, 0U, Types... >
	Trait: index of the first occurrence of `T` among the specified `Types`. More...

template<typename T >
using	is_reference_wrapper = is_instance_of< std::reference_wrapper, T >
	Identifies whether the specified type is a `std::reference_wrapper`. More...

template<typename T >
using	is_unique_ptr = is_instance_of< std::unique_ptr, T >
	Identifies whether the specified type is a `std::unique_ptr`. More...

template<typename Base , typename Key >
using	with_const_as_t = typename with_const_as< Base, Key >::type
	The type `Base`, plus the constantness as in `Key`. More...

template<typename T >
using	strip_referenceness_t = typename strip_referenceness_type< T >::type
	The type `T` stripped of all known reference types. More...

template<typename T >
using	lvalue_reference_into_wrapper_t = typename lvalue_reference_into_wrapper_type< T >::type
	The type `T` stripped of all known reference types. More...

typedef int	UBLArSoftCh_t

typedef std::map< UBDaqID, UBLArSoftCh_t >	UBChannelMap_t

typedef std::map< UBLArSoftCh_t, UBDaqID >	UBChannelReverseMap_t

using	GridContainer2DIndices = GridContainerIndicesBase2D<>
	Index manager for a container of data arranged on a 2D grid. More...

using	GridContainer3DIndices = GridContainerIndicesBase3D<>
	Index manager for a container of data arranged on a 3D grid. More...

template<typename DATUM >
using	GridContainer2D = GridContainerBase2D< DATUM, GridContainer2DIndices >
	Container allowing 2D indexing. More...

template<typename DATUM >
using	GridContainer3D = GridContainerBase3D< DATUM, GridContainer3DIndices >
	Container allowing 3D indexing. More...

using	MatrixIndices = TensorIndices< 2U >
	Type for indexing a 2D-tensor (matrix) More...

template<typename SrcTuple , template< typename T, typename... > class Extractor, template< typename... > class TargetClass = std::tuple>
using	extract_to_tuple_type_t = typename extract_to_tuple_type< SrcTuple, Extractor, TargetClass >::type
	Direct access to the type in `extract_to_tuple_type`. More...

template<typename Tuple , template< typename... > class TargetClass = std::tuple>
using	to_tuple = extract_to_tuple_type< Tuple, self_type, TargetClass >

template<typename Tuple , template< typename... > class TargetClass = std::tuple>
using	to_tuple_t = typename to_tuple< Tuple, TargetClass >::type
	Direct access to the type in `to_tuple`. More...

template<typename Target , typename Tuple >
using	index_of_type = index_of_extracted_type< self_type, Target, Tuple >

template<template< typename T, typename... > class Extractor, typename Target , typename Tuple >
using	type_with_extracted_type = std::tuple_element< index_of_extracted_type_v< Extractor, Target, Tuple >, Tuple >
	Returns the element type in `Tuple` with the specified type. More...

template<template< typename T, typename... > class Extractor, typename Target , typename Tuple >
using	type_with_extracted_type_t = typename type_with_extracted_type< Extractor, Target, Tuple >::type
	Direct access to the value in `type_with_extracted_type`. More...

template<typename Target , typename Tuple >
using	has_type = has_extracted_type< self_type, Target, Tuple >

template<typename Tuple >
using	has_duplicate_types = has_duplicate_extracted_types< self_type, Tuple >

template<typename Target , typename Tuple >
using	count_types = count_extracted_types< self_type, Target, Tuple >

template<typename T , typename Tag >
using	add_tag_t = typename add_tag< T, Tag >::type

template<typename Tagged >
using	remove_tag_t = typename remove_tag< Tagged >::type
	Direct access to the type contained in `remove_tag`. More...

template<typename Tagged >
using	tag_of = TagExtractor< Tagged >
	Trait holding the tag of `Tagged` as `type`. More...

template<typename Tagged >
using	tag_of_t = typename tag_of< Tagged >::type
	Direct access to the type in `tag_of`. More...

template<typename SrcTuple >
using	extract_tags = extract_to_tuple_type< SrcTuple, TagExtractor >
	Returns a tuple with all the tags from `SrcTuple`. More...

template<typename SrcTuple >
using	extract_tags_t = typename extract_tags< SrcTuple >::type
	Direct access to the type in `extract_tags`. More...

template<typename Tag , typename Tuple >
using	index_of_tag = index_of_extracted_type< TagExtractor, Tag, Tuple >
	Trait holding the index of the element of `Tuple` with tag `Tag`. More...

template<typename Tag , typename Tuple >
using	type_with_tag = type_with_extracted_type< TagExtractor, Tag, Tuple >
	Trait holding the type of the element of `Tuple` with tag `Tag`. More...

template<typename Tag , typename Tuple >
using	type_with_tag_t = typename type_with_tag< Tag, Tuple >::type
	Direct access to the value in `type_with_tag`. More...

template<typename Tag , typename Tuple >
using	has_tag = has_extracted_type< TagExtractor, Tag, Tuple >
	Trait informing if there are elements in `Tuple` with tag `Tag`. More...

template<typename Tag , typename Tuple >
using	count_tags = count_extracted_types< TagExtractor, Tag, Tuple >
	Trait counting the elements in `Tuple` with tag `Tag`. More...

template<typename Tuple >
using	has_duplicate_tags = has_duplicate_extracted_types< TagExtractor, Tuple >
	Trait reporting if multiple elements in `Tuple` have the same tag. More...

Functions
template<typename Coll >
auto	make_collection_reference (Coll &&coll)
	Returns an object referencing to the data contained in `coll`. More...

template<typename CollRef >
decltype(auto)	collection_from_reference (CollRef &collRef)
	Returns the object referenced by `collRef` as a C++ reference. More...

template<typename Coll , typename Extractor >
decltype(auto)	makeValueIndex (Coll const &coll, Extractor getter)
	Returns a map of value to index. More...

template<typename Coll >
auto	makeValueIndex (Coll const &coll)

template<typename Ref >
auto	referenced_address (Ref &&ref)
	Returns the address of the referenced object. More...

template<typename T >
auto	lvalue_reference_into_wrapper (T &&obj)
	Converts a l-value reference object into a `std::reference_wrapper`. More...

decltype(auto)	takeAddress ()
	Returns a functor that returns the address of its argument. More...

decltype(auto)	dereference ()
	Returns a functor that returns `*ptr` of its argument `ptr`. More...

template<typename Coll >
auto	makePointerVector (Coll &coll)
	Creates a STL vector with pointers to data from another collection. More...

template<typename Coll , typename PtrColl >
void	MoveFromPointers (Coll &dest, PtrColl &src)
	Moves the content from a collection of pointers to one of data. More...

template<typename Coll , typename Sorter >
void	SortByPointers (Coll &coll, Sorter sorter)
	Applies sorting indirectly, minimizing data copy. More...

template<typename Coll , typename Sorter >
void	SortUniquePointers (Coll &coll, Sorter &&sorter)
	Sorts a vector of unique pointers using a C pointer sorter. More...

template<typename IterB , typename IterE , typename Adaptor >
	span (IterB &&b, IterE &&e, Adaptor &&adaptor) -> span< decltype(adaptor(std::forward< IterB >(b))), decltype(adaptor(std::forward< IterE >(e))) >

template<typename T = double>
constexpr T	pi ()
	Returns the constant pi (up to 35 decimal digits of precision) More...

template<typename T >
constexpr T	DegreesToRadians (T angle)
	Converts the argument angle from degrees into radians. More...

template<typename T >
constexpr T	RadiansToDegrees (T angle)
	Converts the argument angle from radians into degrees ( $\pi \rightarrow 180$ ) More...

template<typename Stream , typename Left , typename Right , typename Data >
void	DumpAssociationsIntro (Stream &&out, art::Assns< Left, Right, Data > const &assns)
	Dumps a short introduction about specified association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, std::vector< T > const &a, art::Ptr< U > const &b, art::Assns< U, T > &assn, std::string a_instance, size_t index=UINT_MAX)
	Creates a single one-to-one association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, std::vector< T > const &a, art::Ptr< U > const &b, art::Assns< U, T > &assn, size_t index=UINT_MAX)
	Creates a single one-to-one association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, art::Ptr< T > const &a, art::Ptr< U > const &b, art::Assns< U, T > &assn)
	Creates a single one-to-one association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, std::vector< T > const &a, art::PtrVector< U > const &b, art::Assns< T, U > &assn, size_t index=UINT_MAX)
	Creates a single one-to-many association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, art::Ptr< T > const &a, std::vector< art::Ptr< U >> const &b, art::Assns< T, U > &assn)
	Creates a single one-to-many association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, std::vector< T > const &a, std::vector< art::Ptr< U >> const &b, art::Assns< T, U > &assn, size_t index=UINT_MAX)
	Creates a single one-to-many association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, std::vector< T > const &a, std::vector< U > const &b, art::Assns< T, U > &assn, size_t startU, size_t endU, size_t index=UINT_MAX)
	Creates a single one-to-many association. More...

template<class T , class U >
bool	CreateAssn (art::Event &evt, std::vector< T > const &a, std::vector< U > const &b, art::Assns< T, U > &assn, std::vector< size_t > const &indices, size_t index=UINT_MAX)
	Creates a single one-to-many association. More...

template<typename T , typename U , typename Iter >
bool	CreateAssn (art::Event &evt, art::Assns< T, U > &assn, size_t first_index, Iter from_second_index, Iter to_second_index)
	Creates a single one-to-many association. More...

template<class T , class U >
std::vector< const U * >	FindUNotAssociatedToT (art::Handle< U > b, art::Event const &evt, std::string const &label)

template<class T , class U >
std::vector< art::Ptr< U > >	FindUNotAssociatedToTP (art::Handle< U > b, art::Event const &evt, std::string const &label)

template<class T , class U >
std::vector< size_t >	GetAssociatedVectorOneI (art::Handle< art::Assns< T, U >> h, art::Handle< std::vector< T >> index_p)

template<class T , class U >
std::vector< const U * >	GetAssociatedVectorOneP (art::Handle< art::Assns< T, U >> h, art::Handle< std::vector< T >> index_p)

template<class T , class U >
std::vector< std::vector< size_t > >	GetAssociatedVectorManyI (art::Handle< art::Assns< T, U >> h, art::Handle< std::vector< T >> index_p)

template<class T , class U >
std::vector< std::vector< const U * > >	GetAssociatedVectorManyP (art::Handle< art::Assns< T, U >> h, art::Handle< std::vector< T >> index_p)

template<typename Producer , typename... Args>
std::enable_if_t< std::is_base_of_v< art::EDProducer, Producer >, bool >	CreateAssn (Producer const &, Args &&...args)

template<typename Producer , typename... Args>
std::enable_if_t< std::is_base_of_v< art::EDProducer, Producer >, bool >	CreateAssnD (Producer const &, Args &&...args)

template<typename Range , typename Pred >
auto	filterRangeFor (Range &&range, Pred &&pred) -> decltype(auto)
	Provides iteration only through elements passing a condition. More...

template<class A , class F >
void	for_each_associated_group (A const &assns, F &func)
	Helper functions to access associations in order. More...

template<class A >
auto	associated_groups (A const &assns)
	Helper functions to access associations in order. More...

template<class A >
auto	associated_groups_with_left (A const &assns)
	Helper functions to access associations in order, also with key. More...

template<typename Groups >
auto	groupByIndex (Groups &&groups, std::size_t index) -> decltype(auto)
	Returns the group within `groups` with the specified index. More...

template<typename Coll , typename KeyOf >
std::vector< size_t >	MakeIndex (Coll const &data, KeyOf key_of=KeyOf())
	Creates a map of indices from an existing collection. More...

template<typename Coll , typename KeyOf >
auto	MakeMap (Coll const &data, KeyOf key_of=KeyOf()) -> std::vector< decltype(key_of((data.begin()))) const >
	Creates a map of objects from an existing collection. More...

template<typename Range >
auto	wrapRangeFor (Range &&range) -> decltype(auto)
	Wraps an object for use in a range-for loop. More...

template<typename Range >
auto	operator\| (Range &&range, RangeForWrapperTag) -> decltype(auto)
	Transforms a range so that it can be used in a range-for loop. More...

template<unsigned int RANK1, unsigned int RANK2, typename = std::enable_if_t<(RANK1 != RANK2), bool>>
bool	operator== (TensorIndices< RANK1 > const &a, TensorIndices< RANK2 > const &b)
	Comparison operator with tensors of different rank. More...

template<unsigned int RANK1, unsigned int RANK2, typename = std::enable_if_t<(RANK1 != RANK2), bool>>
bool	operator!= (TensorIndices< RANK1 > const &a, TensorIndices< RANK2 > const &b)
	Comparison operator with tensors of different rank. More...

template<typename... DIMS>
auto	makeTensorIndices (DIMS...dims)
	Instantiates a TensorIndices class with the specified dimensions. More...

template<template< typename T, typename... > class Extractor, typename Target , typename Tuple >
auto	getByExtractedType (Tuple const &data) -> decltype(auto)
	Returns the value of the element containing the specified type. More...

template<typename Tag , typename T >
auto	makeTagged (T &obj) -> decltype(auto)
	"Converts" `obj` to an object with tag `Tag`. More...

template<typename Tag , typename T >
auto	makeTagged (T const &obj) -> decltype(auto)
	"Converts" `obj` to an object with tag `Tag`. More...

template<typename Tag , typename T >
auto	makeTagged (T const &&obj) -> decltype(auto)
	"Converts" `obj` to an object with tag `Tag`. More...

template<typename Tag , typename T >
auto	makeTagged (T &&obj) -> decltype(auto)
	"Converts" `obj` to an object with tag `Tag`. More...

template<typename Tagged >
auto	removeTag (Tagged &tagged) -> decltype(auto)
	"Converts" a tagged type back to its original type. More...

template<typename Tagged >
auto	removeTag (Tagged const &tagged) -> decltype(auto)
	"Converts" a tagged type back to its original type. More...

template<typename Tagged >
auto	removeTag (Tagged const &&tagged) -> decltype(auto)
	"Converts" a tagged type back to its original type. More...

template<typename Tagged >
auto	removeTag (Tagged &&tagged) -> decltype(auto)
	"Converts" a tagged type back to its original type. More...

template<typename Tag , typename Tuple >
auto	getByTag (Tuple const &data) -> decltype(auto)
	Returns the object with the specified tag. More...

template<typename Cont , typename Mapping >
auto	mapContainer (Cont cont, Mapping mapping)
	Returns a container-like object mapping the content of `cont`. More...

template<typename Cont , typename Mapping , typename Container , typename Reference >
MappedContainer< Cont, Mapping >::template IteratorBase< Container, Reference >	operator+ (typename MappedContainer< Cont, Mapping >::template IteratorBase< Container, Reference >::difference_type n, typename MappedContainer< Cont, Mapping >::template IteratorBase< Container, Reference > const &it)

template<typename _Key , typename _Tp , typename _Compare >
bool	operator== (const VectorMap< _Key, _Tp, _Compare > &__x, const VectorMap< _Key, _Tp, _Compare > &__y)

template<typename _Key , typename _Tp , typename _Compare >
bool	operator< (const VectorMap< _Key, _Tp, _Compare > &__x, const VectorMap< _Key, _Tp, _Compare > &__y)

template<typename _Key , typename _Tp , typename _Compare >
bool	operator!= (const VectorMap< _Key, _Tp, _Compare > &__x, const VectorMap< _Key, _Tp, _Compare > &__y)
	Based on operator==. More...

template<typename _Key , typename _Tp , typename _Compare >
bool	operator> (const VectorMap< _Key, _Tp, _Compare > &__x, const VectorMap< _Key, _Tp, _Compare > &__y)
	Based on operator<. More...

template<typename _Key , typename _Tp , typename _Compare >
bool	operator<= (const VectorMap< _Key, _Tp, _Compare > &__x, const VectorMap< _Key, _Tp, _Compare > &__y)
	Based on operator<. More...

template<typename _Key , typename _Tp , typename _Compare >
bool	operator>= (const VectorMap< _Key, _Tp, _Compare > &__x, const VectorMap< _Key, _Tp, _Compare > &__y)
	Based on operator<. More...

template<typename _Key , typename _Tp , typename _Compare >
void	swap (VectorMap< _Key, _Tp, _Compare > &__x, VectorMap< _Key, _Tp, _Compare > &__y)
	See VectorMap::swap(). More...

std::ostream &	operator<< (std::ostream &out, EventChangeTracker_t const &trk)

std::ostream &	operator<< (std::ostream &out, DataProductChangeTracker_t const &trk)

std::ostream &	operator<< (std::ostream &out, PlaneDataChangeTracker_t const &trk)

Counted iterations
template<typename T >
auto	counter (T begin, T end)
	Returns an object to iterate values from `begin` to `end` in a range-for loop. More...

template<typename T >
auto	counter (T end)

template<typename T = std::size_t>
auto	infinite_counter (T begin=T{})
	Version of `util::counter()` starting at `begin` and never ending. More...

Enumerated iterations
template<std::size_t Lead, typename... Iterables>
auto	enumerate (Iterables &&...iterables)
	Range-for loop helper tracking the number of iteration. More...

template<typename... Iterables>
auto	enumerate (Iterables &&...iterables)
	This version of `enumerate` implicitly uses the first iterable as lead. More...

Transformed iterations
template<std::size_t... Indices, typename Coll >
decltype(auto)	get_elements (Coll &&coll)
	Range-for loop helper iterating across some of the element of each value in the specified collection. More...

template<std::size_t... Indices, typename Coll >
decltype(auto)	get_const_elements (Coll &&coll)
	Range-for loop helper iterating across the constant values of the specified collection. More...

template<typename Coll >
decltype(auto)	values (Coll &&coll)
	Range-for loop helper iterating across the values of the specified collection. More...

template<typename Coll >
decltype(auto)	const_values (Coll &&coll)
	Range-for loop helper iterating across the constant values of the specified collection. More...


template<typename T >
void	staticDumpClassName ()
	Helper to determine the type of a variable at compilation time. More...

template<typename T >
void	staticDumpClassName (T)
	Helper to determine the type of a variable at compilation time. More...

Span helper functions
template<typename BIter , typename EIter >
auto	make_span (BIter begin, EIter end)
	Creates a span from specified iterators (can use constructor instead). More...

template<typename Cont >
auto	make_span (Cont &cont)
	Creates a span from a container type. More...

template<typename Cont >
auto	make_const_span (Cont &cont)
	Creates a span with constant iterator access from a container type. More...

Adapted span helper functions
template<typename BIter , typename EIter , typename Adaptor >
auto	make_adapted_span (BIter begin, EIter end, Adaptor &&adaptor)

template<typename Cont , typename Adaptor >
auto	make_adapted_span (Cont &cont, Adaptor &&adaptor)
	Creates a span from specified collection via an adaptor. More...

template<typename Cont , typename Adaptor >
auto	make_adapted_const_span (Cont &cont, Adaptor &&adaptor)
	Creates constant iteration span from specified collection via an adaptor. More...

Transformed span helper functions
template<typename BIter , typename EIter , typename Op >
auto	make_transformed_span (BIter begin, EIter end, Op &&op)

template<typename Cont , typename Op >
auto	make_transformed_span (Cont &cont, Op &&op)
	Creates a span from specified collection via an adaptor. More...

template<typename Cont , typename Op >
auto	make_transformed_const_span (Cont &cont, Op &&op)

C++ standard library customization for user-defined classes.
template<typename T >
decltype(auto) constexpr	to_string (T &&obj)
	ADL-aware version of `std::to_string`. More...

template<typename T >
decltype(auto) constexpr	begin (T &&obj)
	ADL-aware version of `std::begin`. More...

template<typename T >
decltype(auto) constexpr	end (T &&obj)
	ADL-aware version of `std::end`. More...

template<typename T >
decltype(auto) constexpr	cbegin (T &&obj)
	ADL-aware version of `std::cbegin`. More...

template<typename T >
decltype(auto) constexpr	cend (T &&obj)
	ADL-aware version of `std::cend`. More...

template<typename T >
decltype(auto) constexpr	size (T &&obj)
	ADL-aware version of `std::size`. More...

template<typename T >
decltype(auto) constexpr	empty (T &&obj)
	ADL-aware version of `std::empty`. More...

template<std::size_t I, typename T >
decltype(auto)	get (T &&obj)
	ADL-aware version of `std::to_string`. More...

Parallel iterations
template<std::size_t Lead, typename... Iterables>
auto	zip (Iterables &&...iterables)
	Range-for loop helper iterating across many collections at the same time. More...

template<typename... Iterables>
auto	zip (Iterables &&...iterables)
	Version of `zip()` with first iterator implicitly leading the iteration. More...


template<typename T , typename U , typename D >
bool	CreateAssnD (art::Event &evt, art::Assns< T, U, D > &assn, size_t first_index, size_t second_index, typename art::Assns< T, U, D >::data_t &&data)
	Creates a single one-to-one association with associated data. More...

template<typename T , typename U , typename D >
bool	CreateAssnD (art::Event &evt, art::Assns< T, U, D > &assn, size_t first_index, size_t second_index, typename art::Assns< T, U, D >::data_t const &data)
	Creates a single one-to-one association with associated data. More...

simple mathematical functions
template<typename T >
constexpr auto	abs (T v)
	Returns the absolute value of the argument. More...

Variables
template<typename >
constexpr bool	always_false_v = false
	A templated constant, always false. More...

template<typename >
constexpr bool	always_true_v = true
	A template constant always true. More...

template<typename T , std::size_t StartFrom, typename... Types>
constexpr std::size_t	find_next_type_v = find_next_type<T, StartFrom, Types...>::value

template<typename T , typename... Types>
constexpr std::size_t	find_type_v = find_type<T, Types...>::value

template<typename T , typename... Types>
constexpr bool	is_any_of_v = is_any_of<T, Types...>::value
	Whether `T` is among the specified `Types` (see `util::is_any_of`). More...

template<typename T , typename U >
constexpr auto	is_same_decay_v = std::is_same_v<std::decay_t<T>, std::decay_t<U>>
	Whether `T` and `U` are the same type, after being applied `std::decay`. More...

template<template< typename... > typename Template, typename T >
constexpr bool	is_instance_of_v = is_instance_of<Template, T>::value

template<typename T >
constexpr bool	is_STLarray_v = is_STLarray<T>::value
	A constant describing whether the specified type is a STL array. More...

template<typename T >
constexpr bool	is_reference_wrapper_v = is_reference_wrapper<T>::value
	A constant describing whether the specified type is a `std::reference_wrapper`. More...

template<typename T >
constexpr bool	is_unique_ptr_v = is_unique_ptr<T>::value
	A constant describing whether the specified type is a `std::unique_ptr`. More...

template<typename T >
constexpr bool	is_character_type_v = is_character_type<T>::value
	Whether type `T` is a character type (see `util::is_character_type`). More...

template<typename T >
constexpr bool	is_string_type_v = is_string_type<T>::value
	Whether type `T` is a character string type (see `util::is_string_type`). More...

template<typename T >
constexpr bool	is_basic_string_type_v = is_basic_string_type<T>::value

template<typename T >
constexpr bool	is_basic_string_view_type_v = is_basic_string_view_type<T>::value

constexpr double	kGeVToElectrons = 4.237e7
	23.6eV per ion pair, 1e9 eV/GeV More...

constexpr double	kc = 29.9792458
	Speed of light in vacuum in LArSoft units [cm/ns]. More...

constexpr double	kMeterToCentimeter = 1.e2
	1 m = 100 cm More...

constexpr double	kCentimeterToMeter = 1. / kMeterToCentimeter

constexpr double	kMeterToKilometer = 1.e-3
	1000 m = 1 km More...

constexpr double	kKilometerToMeter = 1. / kMeterToKilometer

constexpr double	keVToMeV = 1.e-6
	1e6 eV = 1 MeV More...

constexpr double	kMeVToeV = 1. / keVToMeV

constexpr double	kBogusD = -999.
	obviously bogus double value More...

constexpr int	kBogusI = -999
	obviously bogus integer value More...

constexpr float	kBogusF = -999.
	obviously bogus float value More...

constexpr double	quietCompiler

constexpr double	kINVALID_DOUBLE = std::numeric_limits<double>::max()

constexpr RangeForWrapperTag	range_for

template<template< typename T, typename... > class Extractor, typename Target , typename Tuple >
constexpr std::size_t	index_of_extracted_type_v
	Direct access to the value in `index_of_extracted_type`. More...

template<typename Target , typename Tuple >
constexpr std::size_t	index_of_type_v = index_of_type<Target, Tuple>()
	Direct access to the value in `index_of_type`. More...

template<template< typename T, typename... > class Extractor, typename Target , typename Tuple >
constexpr bool	has_extracted_type_v = has_extracted_type<Extractor, Target, Tuple>()
	Direct access to the value in `has_extracted_type`. More...

template<typename Target , typename Tuple >
constexpr bool	has_type_v = has_type<Target, Tuple>()
	Direct access to the value in `has_type`. More...

template<template< typename T, typename... > class Extractor, typename Tuple >
constexpr bool	has_duplicate_extracted_types_v
	Direct access to the value in `has_duplicate_extracted_types`. More...

template<typename Tuple >
constexpr bool	has_duplicate_types_v = has_duplicate_types<Tuple>()
	Direct access to the value in `has_duplicate_types`. More...

template<template< typename T, typename... > class Extractor, typename Target , typename Tuple >
constexpr unsigned int	count_extracted_types_v
	Direct access to the value in `count_extracted_types`. More...

template<typename Target , typename Tuple >
constexpr unsigned int	count_types_v = count_types<Target, Tuple>()
	Direct access to the value in `count_extracted_types`. More...

template<typename Tag , typename Tuple >
constexpr std::size_t	index_of_tag_v = index_of_tag<Tag, Tuple>()
	Direct access to the value in `index_of_tag`. More...

template<typename Tag , typename Tuple >
constexpr bool	has_tag_v = has_tag<Tag, Tuple>()
	Direct access to the value in `has_tag`. More...

template<typename Tag , typename Tuple >
constexpr unsigned int	count_tags_v = count_tags<Tag, Tuple>()
	Direct access to the value in `count_tags`. More...

template<typename Tuple >
constexpr bool	has_duplicate_tags_v = has_duplicate_tags<Tuple>()
	Direct access to the value in `has_duplicate_tags`. More...

const float	SQRT_TWO_PI = 2.506628

Recombination factor coefficients (NIM).
See also sim::ISCalculationSeparate::CalculateIonizationAndScintillation() Recombination factor coefficients come from Nucl.Instrum.Meth.A523:275-286,2004 is given by the voxel energy deposition, but have to convert it to MeV/cm from GeV/voxel width electric field: in kV/cm $A = 0.800 \pm 0.003$ needs to be scaled with Electric field
constexpr double	kRecombA = 0.800
	A constant. More...

constexpr double	kRecombk = 0.0486

Recombination factor coefficients (modified box, ArguNeuT JINST).
See also sim::ISCalculationSeparate::CalculateIonizationAndScintillation() Recombination factor coefficients come from Nucl.Instrum.Meth.A523:275-286,2004 is given by the voxel energy deposition, but have to convert it to MeV/cm from GeV/voxel width electric field: in kV/cm `kModBoxB` needs to be scaled with the electric field.
constexpr double	kModBoxA = 0.930
	Modified Box Alpha. More...

constexpr double	kModBoxB = 0.212
	Modified Box Beta in g/(MeV cm²)*kV/cm. More...

template<typename Target , typename... T>
using	count_type_in_list = details::count_type_in_list_impl< Target, T... >
	Returns how many of the types in `T` exactly match `Target`. More...

template<std::size_t N, typename... T>
using	typelist_element_type = std::tuple_element< N, std::tuple< T... >>
	Returns the `N` type of the type list. More...

template<std::size_t N, typename... T>
using	typelist_element_t = typename typelist_element_type< N, T... >::type
	Direct access to the value in `typelist_element_type`. More...

template<typename Target , typename... T>
using	type_is_in = details::type_is_in_impl< Target, T... >
	Holds whether the `Target` type is among the ones in the `T` pack. More...

template<typename Target , typename... T>
constexpr unsigned int	count_type_in_list_v = count_type_in_list<Target, T...>()
	Direct access to the value in `count_type_in_list`. More...

template<typename Target , typename... T>
constexpr bool	type_is_in_v = type_is_in<Target, T...>()
	Direct access to the value in `type_is_in`. More...

Detailed Description

Namespace for general, non-LArSoft-specific utilities.

Namespace for general, not LArSoft-specific utilities.

General LArSoft Utilities.

Generic namespace of utility functions generally independent of LArSoft.

Some physical constants are also included here. As a reminder, the "standard" units in LArSoft are:

energy: GeV
time: ns
space: cm

SumSecondFunction 28-Jul-2009 William Seligman selig.nosp@m.man@.nosp@m.nevis.nosp@m..col.nosp@m.umbia.nosp@m..edu When using STL, maps, and the std::accumulate function, there's one function I keep coding over and over again: accumulate the second member of a pair.

To save myself time (and others who know and use STL), here's a complete STL-compatible implementation of that function. To use it, assume you have an object:

std::map<K,V> myMap;

To sum all the V's in the map:

V sum = std::accumulate( myMap.begin(),myMap.end(),V(),SumSecondFunction<K,V>() );

(Yes, I know there are other, better ways to do this, if one has access to BOOST. Unfortunately, we're not supposed to use BOOST in LArSoft, since as of Jul-2009 it's not universally installed on FNAL machines.)

VectorMap 17-Apr-2008 William Seligman selig.nosp@m.man@.nosp@m.nevis.nosp@m..col.nosp@m.umbia.nosp@m..edu

This class is an implementation of a concept discussion in "Effective STL" by Scott Meyers:

STL maps are useful because their contents are always sorted, so they're effective for fast searches. However, in almost every other respect vectors are superior: They take up less space, and they use random-access iterators.

This class implements "sorted vector maps," that is, an STL-style map implemented as a sorted STL vector of pairs. I've done my best to implement all aspects of the std::map interface in this class, with some additions; if you've defined the following:

VectorMap<key_type, data_type> svm;

svm(i) will return the "i-th" value in the map; that is, "i" is a numeric index instead of a key. (Note the use of parenthesis instead of square brackets.) This is a boon to physicists, most of whom couldn't tell an iterator from a hole in the wall.
svm.Key(i) will return the "i-th" key in the map.
svm.Data(i) will return the same result as svm(i).
svm[key_type] will now return the corresponding data_type in both const and non-const contexts. However, if you ask for svm[key] and the key isn't in the map, and you're in a const context, the routine will throw an out-of-range exception.

INCREDIBLY IMPORTANT NOTE: The "key type" of a VectorMap cannot be a "const" type (unlike maps); it won't even compile. When you do an insert, the underlying vector has to move things around within its list, and it uses the assignment operator=() (or "vector{i+1)=vector(i)" if you like). You can't do that if either the key or the data is const.

As with a map, there's no way to insert items at a specific location in a VectorMap. The insertion methods (including operator[]) all operate on a sorted sequence according to the key. Because of this, insertions take a long time.

However, for our processing, this doesn't matter much; for almost all our maps, we typically have:

Initialization, where the time doesn't matter (e.g., tracks in a Monte Carlo).
Access, where efficient or "simple" access to the map's contents are important. In general, we access a map many, many more times that we create one.
After we create/initialize a map, we never change its contents.

For this usage, a sorted vector is generally superior to a map.

This class just implements the equivalent of an STL map, not a multimap, set, nor a multiset. If there's a need, I may create additional classes.

Is there any map feature that's not implemented? Yes:

equal_range in a const context (which causes some weird ROOT dictionary problem); this isn't likely to be used for a map anyway (multimaps or multisets would be a different story).

Advanced implementation note: Depending on the application, it might be possible to speed up this class by using "lazy evaluation"; that is, we wouldn't actually sort the vector until the user actually tries to access its contents. I'm not going to do this, because:

A) I don't think my programming skills are up to the task.

B) In the primary application for which I plan to use this class (Monte-Carlo particle tracks), we're performing at least one search after every insert; lazy evaluation wouldn't be much of a speed improvement.

Feb-2011 WGS: VectorMap mostly looks like a map, but there are some memory-management issues that relate to it being a vector. Include the vector-based routines reserve() and capacity().

Title: GaussianEliminationAlg Class Author: Wes Ketchum (wketc.nosp@m.hum@.nosp@m.lanl..nosp@m.gov)

Description: Class that solves system of linear equations via Gaussian Elimination. Intended for use with RFFHitFitter

Typedef Documentation

template<typename DATUM >

using util::GridContainer2D = typedef GridContainerBase2D<DATUM, GridContainer2DIndices>

Container allowing 2D indexing.

Template Parameters

DATUM type of contained data

See also: GridContainer2DIndices

This is an alias for GridContainerBase2D, with a proper index manager. See the documentation of GridContainerBase2D.

Definition at line 248 of file GridContainers.h.

using util::GridContainer2DIndices = typedef GridContainerIndicesBase2D<>

Index manager for a container of data arranged on a 2D grid.

Definition at line 170 of file GridContainerIndices.h.

template<typename DATUM >

using util::GridContainer3D = typedef GridContainerBase3D<DATUM, GridContainer3DIndices>

Container allowing 3D indexing.

Template Parameters

DATUM type of contained data

See also: GridContainer3DIndices

This is an alias for GridContainerBase3D, with a proper index manager. See the documentation of GridContainerBase3D.

Definition at line 259 of file GridContainers.h.

using util::GridContainer3DIndices = typedef GridContainerIndicesBase3D<>

Index manager for a container of data arranged on a 3D grid.

Definition at line 173 of file GridContainerIndices.h.

using util::MatrixIndices = typedef TensorIndices<2U>

Type for indexing a 2D-tensor (matrix)

Definition at line 535 of file TensorIndices.h.

typedef std::map<UBDaqID, UBLArSoftCh_t> util::UBChannelMap_t

Definition at line 46 of file DatabaseUtil.h.

typedef std::map<UBLArSoftCh_t, UBDaqID> util::UBChannelReverseMap_t

Definition at line 47 of file DatabaseUtil.h.

typedef int util::UBLArSoftCh_t

Definition at line 44 of file DatabaseUtil.h.

Function Documentation

template<typename T >

constexpr auto util::abs ( T v )

Returns the absolute value of the argument.

Template Parameters

T	type of the argument

Parameters

v	value to be processed

Returns: the absolute value of v

If the value v is negative, its opposite is returned. Note that this implementation does not work with data types that are not comparable (like std::complex).

Requirements

constexpr construction of a T value from the literal 0
operator- (T) constexpr
operator< (T, T) constexpr convertible to bool

Definition at line 40 of file constexpr_math.h.

   {
     return (-T(0) < v) ? v : -v;
   }

template<class A >

auto util::associated_groups ( A const & assns )

Helper functions to access associations in order.

Template Parameters

A	type of association being read

Parameters

assns the association being read

See also: for_each_associated_group()

This function provides a functionality equivalent to art::for_each_group(), but it grants the caller additional control on the external loop and on the function.

Example: assuming that a module with input tag stored in fTrackTag has created associations of each track to its hits, the total charge for each track can be extracted by:

auto assns = art::getValidHandle<art::Assns<recob::Track, recob::Hit>>
  (fTrackTag);
std::vector<double> totalCharge;
for (auto const& hits: util::associated_groups(*assns)) {
  double total = 0.;
  for (art::Ptr<recob::Hit> const& hit: hits)
    total += hit->Integral();
  totalCharge.push_back(total);
} // for

A number of important points need to be realised about this example:

the requirements of this function on its input association are the same as for art::for_each_group()
we can code the action on each group of hits directly in a loop, if like in this case the code is succinct
again, there is one outer loop iteration for every track;
the value of hits is an object representing a range of art pointers (art::Ptr<recob::Hit>) which can be navigated with the begin()/end() free functions, or in a range-for loop;
on each iteration, the information of which track the hits are associated to is not available; if that is also needed, use util::associated_groups_with_left() instead.

Definition at line 93 of file ForEachAssociatedGroup.h.

References a1, a2, range_for, and values().

Referenced by TrackProducerFromTrack::produce(), TrackProducerFromTrackTrajectory::produce(), and TrackProducerFromPFParticle::produce().

   {
     return assns | ranges::views::all |
            ranges::views::chunk_by([](auto a1, auto a2) { return a1.first == a2.first; }) |
            ranges::views::transform(
              [](auto pairs) { return pairs | ranges::views::values | util::range_for; }) |
            util::range_for;
   } // associated_groups()

template<class A >

auto util::associated_groups_with_left ( A const & assns )

Helper functions to access associations in order, also with key.

Template Parameters

A	type of association being read

Parameters

assns the association being read

See also: for_each_associated_group()

This function provides a functionality equivalent to art::for_each_group_with_left(), but it grants the caller additional control on the external loop and on the function.

Example: assuming that a module with input tag stored in fTrackTag has created associations of each track to its hits, the total charge for each track can be extracted by:

auto assns = art::getValidHandle<art::Assns<recob::Track, recob::Hit>>
  (fTrackTag);
std::map<int, double> totalCharge;
for (auto const& trackWithHits: util::associated_groups_with_left(*assns))
{
  art::Ptr<recob::Track> const& track = trackWithHits.first;
  auto const& hits = trackWithHits.second;
  if (totalCharge.count(track->ID()) > 0) {
    throw art::Exception(art::errors::LogicError)
      << "Multiple tracks have ID " << track->ID() << "!\n";
  }
  double& total = totalCharge[track->ID()];
  total = 0.0;
  for (art::Ptr<recob::Hit> const& hit: hits)
    total += hit->Integral();
} // for

A number of important points need to be realised about this example:

the requirements of this function on its input association are the same as for art::for_each_group_with_left()
we can code the action on each group of hits directly in a loop, if like in this case the code is succinct
again, there is one outer loop iteration for every track;
the value of hits is an object representing a range of art pointers (art::Ptr<recob::Hit>) which can be navigated with the begin()/end() free functions, or in a range-for loop.

Definition at line 150 of file ForEachAssociatedGroup.h.

References a1, a2, range_for, and values().

   {
     return assns | ranges::views::all |
            ranges::views::chunk_by([](auto a1, auto a2) { return a1.first == a2.first; }) |
            ranges::views::transform([](auto pairs) {
              return std::make_pair(pairs.front().first, // assuming they're all the same, pick first
                                    pairs | ranges::views::values | util::range_for);
            }) |
            util::range_for;
   } // associated_groups_with_left()

template<typename Coll >

decltype(auto) util::const_values ( Coll && coll )

Range-for loop helper iterating across the constant values of the specified collection.

See also: util::values()

This function is equivalent to util::values() but the values are extracted as if the specified collection were constant.

template<typename T >

auto util::counter	(	T	begin,
		T	end
	)

Returns an object to iterate values from begin to end in a range-for loop.

Template Parameters

T	type of counter value

Returns: a control object for range-for loop

See also: util::count_iterator

An example of usage:

std::vector<std::size_t> data;
for (auto i: util::counter(4, 8)) {
  data.push_back(i);
}

will insert in data the numbers from 4 to 7, just like:

for (std::size_t i = 4; i < 8; ++i) {
  data.push_back(i);
}

would.

Definition at line 295 of file counter.h.

References util::count_iterator< T >::count_iterator(), and span().

Referenced by opdet::SimPhotonCounterAlg::AddSimPhotonCollection(), opdet::SimPhotonCounterAlg::AddSimPhotonsVector(), lar::details::bulk_allocator::ReferenceCounter::AddUser(), NestAlg::CalculateIonizationAndScintillation(), opdet::SimPhotonCounterAlg::ClearCounters(), lar::details::bulk_allocator::ReferenceCounter::Count(), counter(), larg4::OpFastScintillation::detectedDirectHits(), phot::SemiAnalyticalModel::detectedDirectVisibilities(), larg4::OpFastScintillation::detectedReflecHits(), phot::SemiAnalyticalModel::detectedReflectedVisibilities(), larg4::AuxDetSD::EndOfEvent(), trkf::KGTrack::fillHits(), trkf::SeedFinderAlgorithm::FindSeedAtEnd(), phot::PropagationTimeModel::getVISTimes(), larg4::OpFastScintillation::getVISTimes(), lar::details::bulk_allocator::ReferenceCounter::hasUsers(), calo::TrackCalorimetryAlg::IsInvertedTrack(), lar::CountersMap< KEY, COUNTER, SIZE, ALLOC, SUBCOUNTERS >::CounterKey_t::Key(), ClusteringValidation::ClusterAnalyser::MakeHistograms(), phot::PropagationTimeModel::opDetCenters(), phot::PDFastSimPAR::opDetCenters(), phot::PropagationTimeModel::opDetOrientations(), art::detail::IPRHelper< ProdA, ProdB, Data, DATACOLL, EVENT >::operator()(), larg4::OpFastScintillation::OpFastScintillation(), phot::SemiAnalyticalModel::opticalDetectors(), sim::MergeSimSources::produce(), evgen::RadioGen::RadioGen(), pmtana::AlgoSiPM::RecoPulse(), pmtana::AlgoThreshold::RecoPulse(), larg4::OpFastScintillation::RecordPhotonsProduced(), lar::details::bulk_allocator::ReferenceCounter::RemoveUser(), shower::LArPandoraShowerAlg::RMSShowerGradient(), hit::HitAnaAlg::ROIInfo(), lar_content::TransverseMatrixVisualizationTool::Run(), lar_content::ShowerTensorVisualizationTool::Run(), and lar_content::TransverseTensorVisualizationTool::Run().

 {
   return util::span(count_iterator(begin), count_iterator(end));
 }

template<typename T >

auto util::counter ( T end )

Version of util::counter() starting at default-constructed T (usually some form of 0).

Definition at line 194 of file counter.h.

References counter(), end(), and infinite_counter().

   {
     return counter(T{}, end);
   }

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		std::vector< T > const &	a,
		art::Ptr< U > const &	b,
		art::Assns< U, T > &	assn,
		std::string	a_instance,
		size_t	index = `UINT_MAX`
	)

Creates a single one-to-one association.

Template Parameters

T	type of the new object to associate
U	type of the object already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	vector of data products that are in, or will be put into, evt
b	art::Ptr to the (new) object to be associated to the one in a
assn	reference to association object where the new one will be put
a_instance	name of the instance that will be used for a in evt
index	index of the element in a to be associated with b (default: the last element)

Returns: whether the operation was successful (can it ever fail??)

As example of usage: create a wire/raw digit association. This code should live in the art::EDProduce::produce() method. The raw::RawDigit product was created already by a DigitModuleLabel module. The code is supposed to produce one recob::Wire for each existing raw::RawDigit, and contextually associate the new wire to the source digit. We are also assuming that there might be different RawDigit sets produced by the same producer: we identify the one we care of by the string spill_name and we create wires and associations with the same label for convenience.

// this is the original list of digits, thawed from the event
art::Handle< std::vector<raw::RawDigit>> digitVecHandle;
evt.getByLabel(DigitModuleLabel, spill_name, digitVecHandle);

// the collection of wires that will be written as data product
std::unique_ptr<std::vector<recob::Wire>> wirecol(new std::vector<recob::Wire>);
// ... and an association set
std::unique_ptr<art::Assns<raw::RawDigit,recob::Wire>> WireDigitAssn
  (new art::Assns<raw::RawDigit,recob::Wire>);

for(size_t iDigit = 0; iDigit < digitVecHandle->size(); ++iDigit) {
  // turn the digit into a art::Ptr:
  art::Ptr<raw::RawDigit> digit_ptr(digitVecHandle, iDigit);

  // store the wire in its final position in the data product;
  // the new wire is currently the last of the list
  wirecol->push_back(std::move(wire));

  // add an association between the last object in wirecol
  // (that we just inserted) and digit_ptr
  if (!util::CreateAssn(evt, *wirecol, digit_ptr, *WireDigitAssn, spill_name)) {
    throw art::Exception(art::errors::ProductRegistrationFailure)
      << "Can't associate wire #" << (wirecol->size() - 1)
      << " with raw digit #" << digit_ptr.key();
  } // if failed to add association

} // for digits

evt.put(std::move(wirecol), spill_name);
evt.put(std::move(WireDigitAssn), spill_name);

Definition at line 558 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), e, and tca::evt.

 {
   if (index == UINT_MAX) index = a.size() - 1;
 
   try {
     assn.addSingle(b, art::PtrMaker<T>{evt, a_instance}(index));
     return true;
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
 } // util::CreateAssn() [01]

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		std::vector< T > const &	a,
		art::Ptr< U > const &	b,
		art::Assns< U, T > &	assn,
		size_t	index = `UINT_MAX`
	)

inline

Creates a single one-to-one association.

Template Parameters

T	type of the new object to associate
U	type of the object already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	vector of data products that are in, or will be put into, evt
b	art::Ptr to the (new) object to be associated to the one in a
assn	reference to association object where the new one will be put
index	index of the element in a to be associated with b (default: the last element)

Returns: whether the operation was successful (can it ever fail??)

The instance name of the product a will be in is assumed empty. Example of usage:

// this is the original list of digits, thawed from the event
art::Handle< std::vector<raw::RawDigit>> digitVecHandle;
evt.getByLabel(DigitModuleLabel, digitVecHandle);

// the collection of wires that will be written as data product
std::unique_ptr<std::vector<recob::Wire>> wirecol(new std::vector<recob::Wire>);
// ... and an association set
std::unique_ptr<art::Assns<raw::RawDigit,recob::Wire>> WireDigitAssn
  (new art::Assns<raw::RawDigit,recob::Wire>);

for(size_t iDigit = 0; iDigit < digitVecHandle->size(); ++iDigit) {
  // turn the digit into a art::Ptr:
  art::Ptr<raw::RawDigit> digit_ptr(digitVecHandle, iDigit);

  // store the wire in its final position in the data product;
  // the new wire is currently the last of the list
  wirecol->push_back(std::move(wire));

  // add an association between the last object in wirecol
  // (that we just inserted) and digit_ptr
  if (!util::CreateAssn(evt, *wirecol, digit_ptr, *WireDigitAssn)) {
    throw art::Exception(art::errors::ProductRegistrationFailure)
      << "Can't associate wire #" << (wirecol->size() - 1)
      << " with raw digit #" << digit_ptr.key();
  } // if failed to add association

} // for digits

evt.put(std::move(wirecol));
evt.put(std::move(WireDigitAssn));

Definition at line 211 of file AssociationUtil.h.

References CreateAssn(), CreateAssnD(), tca::evt, FindUNotAssociatedToT(), FindUNotAssociatedToTP(), GetAssociatedVectorManyI(), GetAssociatedVectorManyP(), GetAssociatedVectorOneI(), GetAssociatedVectorOneP(), geo::vect::indices(), and lar::dump::vector().

   {
     return CreateAssn(evt, a, b, assn, std::string(), index);
   }

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		art::Ptr< T > const &	a,
		art::Ptr< U > const &	b,
		art::Assns< U, T > &	assn
	)

Creates a single one-to-one association.

Template Parameters

T	type of one object to associate
U	type of the other object to associate

Parameters

evt	reference to the current event
a	art::Ptr to the first object in the association
b	art::Ptr to the object to be associated to the one in a
assn	reference to association object where the new one will be put

Returns: whether the operation was successful (can it ever fail??)

This is the simplest way ever. Neither the event not the producer references are used.

Definition at line 583 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   try {
     assn.addSingle(b, a);
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [03]

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		std::vector< T > const &	a,
		art::PtrVector< U > const &	b,
		art::Assns< T, U > &	assn,
		size_t	index = `UINT_MAX`
	)

Creates a single one-to-many association.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	vector of data products that are in, or will be put into, evt
b	art::PtrVector to the (new) objects to be associated to the one in a
assn	reference to association object where the new one will be put
index	index of the element in a to be associated with all the ones in b (default: the last element)

Returns: whether the operation was successful (can it ever fail??)

A "one-to-many" association is actually a number of one-to-one associations. If you want to keep the information of the order of the many, you may have to use an association with a data member (the third template parameter that we pretent not to exist).

Definition at line 604 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
   if (index == UINT_MAX) index = a.size() - 1;
 
   try {
     auto const aptr = art::PtrMaker<T>{evt}(index);
     for (art::Ptr<U> const& b_item : b)
       assn.addSingle(aptr, b_item);
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [04]

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		art::Ptr< T > const &	a,
		std::vector< art::Ptr< U >> const &	b,
		art::Assns< T, U > &	assn
	)

Creates a single one-to-many association.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	art::Ptr to the item to be associated with many
b	vector to art::Ptr to the (new) objects to be associated to a
assn	reference to association object where the new one will be put

Returns: whether the operation was successful (can it ever fail??)

A "one-to-many" association is actually a number of one-to-one associations. If you want to keep the information of the order of the many, you may have to use an association with a data member (the third template parameter that we pretent not to exist).

Definition at line 630 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   try {
     for (art::Ptr<U> const& b_item : b)
       assn.addSingle(a, b_item);
   }
   catch (cet::exception const& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [05]

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		std::vector< T > const &	a,
		std::vector< art::Ptr< U >> const &	b,
		art::Assns< T, U > &	assn,
		size_t	index = `UINT_MAX`
	)

Creates a single one-to-many association.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	vector of data products that are in, or will be put into, evt
b	vector to art::Ptr to the (new) objects to be associated to the one in a
assn	reference to association object where the new one will be put
index	index of the element in a to be associated with all the ones in b (default: the last element)

Returns: whether the operation was successful (can it ever fail??)

A "one-to-many" association is actually a number of one-to-one associations. If you want to keep the information of the order of the many, you may have to use an association with a data member (the third template parameter that we pretent not to exist).

Definition at line 652 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   if (index == UINT_MAX) index = a.size() - 1;
 
   try {
     auto const aptr = art::PtrMaker<T>{evt}(index);
     for (art::Ptr<U> const& b_item : b)
       assn.addSingle(aptr, b_item);
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [06]

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		std::vector< T > const &	a,
		std::vector< U > const &	b,
		art::Assns< T, U > &	assn,
		size_t	startU,
		size_t	endU,
		size_t	index = `UINT_MAX`
	)

Creates a single one-to-many association.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	vector of data products that are in, or will be put into, evt
b	vector of the (new) objects to be associated to the one in a
assn	reference to association object where the new one will be put
startU	index in b of the first element to be associated to the one in a
endU	index in b after the last element to be associated to the one in a
index	index of the element in a to be associated with all the ones in b (default: the last element)

Returns: whether the operation was successful (can it ever fail??)

Use this when the objects in b are not yet stored in the event and are in a std::vector collection instead.

The method gets the product id for those as well as for the element in a. Also specify the range of entries to use from the std::vector collection of U objects.

A "one-to-many" association is actually a number of one-to-one associations. If you want to keep the information of the order of the many, you may have to use an association with a data member (the third template parameter that we pretent not to exist).

Definition at line 679 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   if (index == UINT_MAX) index = a.size() - 1;
 
   try {
     auto const aptr = art::PtrMaker<T>{evt}(index);
     art::PtrMaker<U> const make_bptr{evt};
     for (size_t i = startU; i < endU; ++i) {
       assn.addSingle(aptr, make_bptr(i));
     }
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [07]

template<class T , class U >

bool util::CreateAssn	(	art::Event &	evt,
		std::vector< T > const &	a,
		std::vector< U > const &	b,
		art::Assns< T, U > &	assn,
		std::vector< size_t > const &	indices,
		size_t	index = `UINT_MAX`
	)

Creates a single one-to-many association.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr

Parameters

evt	reference to the current event
a	vector of data products that are in, or will be put into, evt
b	vector of the (new) objects to be associated to the one in a
assn	reference to association object where the new one will be put
indices	indices of the elements in b to be associated to the one in a
index	index of the element in a to be associated with all the ones in b (default: the last element)

Returns: whether the operation was successful (can it ever fail??)

Use this when the objects in b are not yet stored in the event and are in a std::vector collection instead.

The method gets the product id for those as well as for the element in a. Also specify the entries to use from the std::vector collection of U objects.

A "one-to-many" association is actually a number of one-to-one associations. If you want to keep the information of the order of the many, you may have to use an association with a data member (the third template parameter that we pretent not to exist).

Definition at line 710 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   if (index == UINT_MAX) index = a.size() - 1;
 
   try {
     auto const aptr = art::PtrMaker<T>{evt}(index);
     art::PtrMaker<U> const make_bptr{evt};
     for (size_t index : indices) {
       assn.addSingle(aptr, make_bptr(index));
     }
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [07a]

template<typename T , typename U , typename Iter >

bool util::CreateAssn	(	art::Event &	evt,
		art::Assns< T, U > &	assn,
		size_t	first_index,
		Iter	from_second_index,
		Iter	to_second_index
	)

Creates a single one-to-many association.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr
Iter	iterator to size_t-compatible elements

Parameters

evt	reference to the current event
assn	reference to association object where the new one will be put
first_index	index of the object of type T to be associated to all the others
from_second_index	iterator pointing to the first of the indices of U type objects to be associated to the one of the first type
to_second_index	iterator pointing after the last of the indices of U type objects to be associated to the one of the first type

Returns: whether the operation was successful (can it ever fail??)

A "one-to-many" association is actually a number of one-to-one associations. If you want to keep the information of the order of the many, you may have to use an association with a data member (the third template parameter that we pretent not to exist).

Use this if the objects that have to be associated to the one of type T are sparse, spread across a to-be-data-product, but you have a list of the indices in the data product of the elements to associate to the one of type T. In other words, given that you have a data product "a" of type std::vector<T> and a data product "b" of type std::vector<U>, this method creates an association betweena[first_index]and b[*(from_second_index)], another between a[first_index] and b[*(from_second_index + 1)], etc.

The surprising concept here is that you don't need to specify neither of the collections of T or U elements. The data product is uniquely defined by its type, producer, process and product label. Here we assume that the type of the products are std::vector<T> and std::vector<U>, and that the products have empty product labels, and that the producer is prod for both of them.

Definition at line 740 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   try {
     // we declare here that we want to associate the element first_index of the
     // (only) data product of type std::vector<T> with other objects.
     // This is the pointer to that element:
     auto const first_ptr = art::PtrMaker<T>{evt}(first_index);
 
     // we are going to associate that element in a with a number of elements
     // of the only data product of type std::vector<U>
     art::PtrMaker<U> const make_second_ptr{evt};
     for (; from_second_index != to_second_index; ++from_second_index) {
       assn.addSingle(first_ptr, make_second_ptr(*from_second_index));
     } // while
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssn() [08]

template<typename Producer , typename... Args>

std::enable_if_t<std::is_base_of_v<art::EDProducer, Producer>, bool> util::CreateAssn	(	Producer const &	,
		Args &&...	args
	)

Definition at line 930 of file AssociationUtil.h.

References CreateAssn().

   {
     return CreateAssn(std::forward<Args>(args)...);
   }

template<typename T , typename U , typename D >

bool util::CreateAssnD	(	art::Event &	evt,
		art::Assns< T, U, D > &	assn,
		size_t	first_index,
		size_t	second_index,
		typename art::Assns< T, U, D >::data_t &&	data
	)

Creates a single one-to-one association with associated data.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr
D	type of the "metadata" coupled to this pair association

Parameters

evt	reference to the current event
assn	reference to association object where the new one will be put
first_index	index of the object of type T to be associated
second_index	index of the object of type U to be associated
data	"metadata" to be store in this association

Returns: whether the operation was successful (can it ever fail??)

Use this if you want some metadata to travel together with the association. An example may be the order of the second element within a list:

size_t a_index = 2;
std::vector<size_t> b_indices{ 6, 9, 18, 12 };
for (size_t i = 0; i < b_indices.size(); ++i)
  CreateAssn(prod, evt, assn, a_index, b_index[i], i);

In this way, the association between the element #2 of "a" (a vector that is not specified – nor needed – in this snippet of code) and the element #18 will be remembered as being the third (metadata value of 2). In this example metadata is of type size_t the association would be declared as art::Assn<A, B, size_t>. A FindMany query of that association might look like:

art::Handle<std::vector<A>> a_list; // read this from the event

art::FindMany<B, size_t> Query(a_list, event, ModuleLabel);

// search for the last of the objects associated to the third element:
size_t a_index = 2; // this means third element

std::vector<size_t const*> const& metadata = Query.data(a_index);
size_t largest_index = 0, last_item = 0;
for (size_t iB = 0; iB < metadata.size(); ++iB) {
  if (largest_index >= *(metadata[iB])) continue;
  largest_index = *(metadata[iB]);
  last_item = iB;
} // for iB
B const& lastB = Query.at(last_item);

In alternative, the elements and their metadata can be fetched simultaneously with:

std::vector<art::Ptr<B>> const& Bs;
std::vector<size_t const*> const& metadata;

size_t a_index = 2; // this means third element
size_t nMatches = Query.get(a_index, Bs, metadata);

Definition at line 772 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

Referenced by CreateAssn(), CreateAssnD(), cluster::LineCluster::produce(), cluster::ClusterCrawler::produce(), and cluster::TrajCluster::produce().

 {
 
   try {
     // we declare here that we want to associate the element first_index of the
     // (only) data product of type std::vector<T> with the other object
     auto const first_ptr = art::PtrMaker<T>{evt}(first_index);
 
     // the same to associate the element second_index of the (only)
     // data product of type std::vector<U> with the first object.
     auto const second_ptr = art::PtrMaker<U>{evt}(second_index);
 
     assn.addSingle(first_ptr, second_ptr, std::move(data));
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssnD() [01a]

template<typename T , typename U , typename D >

bool util::CreateAssnD	(	art::Event &	evt,
		art::Assns< T, U, D > &	assn,
		size_t	first_index,
		size_t	second_index,
		typename art::Assns< T, U, D >::data_t const &	data
	)

Creates a single one-to-one association with associated data.

Template Parameters

T	type of the new object to associate
U	type of the many objects already in the data product or art::Ptr
D	type of the "metadata" coupled to this pair association

Parameters

evt	reference to the current event
assn	reference to association object where the new one will be put
first_index	index of the object of type T to be associated
second_index	index of the object of type U to be associated
data	"metadata" to be store in this association

Returns: whether the operation was successful (can it ever fail??)

Use this if you want some metadata to travel together with the association. An example may be the order of the second element within a list:

size_t a_index = 2;
std::vector<size_t> b_indices{ 6, 9, 18, 12 };
for (size_t i = 0; i < b_indices.size(); ++i)
  CreateAssn(prod, evt, assn, a_index, b_index[i], i);

In this way, the association between the element #2 of "a" (a vector that is not specified – nor needed – in this snippet of code) and the element #18 will be remembered as being the third (metadata value of 2). In this example metadata is of type size_t the association would be declared as art::Assn<A, B, size_t>. A FindMany query of that association might look like:

art::Handle<std::vector<A>> a_list; // read this from the event

art::FindMany<B, size_t> Query(a_list, event, ModuleLabel);

// search for the last of the objects associated to the third element:
size_t a_index = 2; // this means third element

std::vector<size_t const*> const& metadata = Query.data(a_index);
size_t largest_index = 0, last_item = 0;
for (size_t iB = 0; iB < metadata.size(); ++iB) {
  if (largest_index >= *(metadata[iB])) continue;
  largest_index = *(metadata[iB]);
  last_item = iB;
} // for iB
B const& lastB = Query.at(last_item);

In alternative, the elements and their metadata can be fetched simultaneously with:

std::vector<art::Ptr<B>> const& Bs;
std::vector<size_t const*> const& metadata;

size_t a_index = 2; // this means third element
size_t nMatches = Query.get(a_index, Bs, metadata);

Definition at line 800 of file AssociationUtil.h.

References art::Assns< L, R, D >::addSingle(), and e.

 {
 
   try {
     // we declare here that we want to associate the element first_index of the
     // (only) data product of type std::vector<T> with the other object
     auto const first_ptr = art::PtrMaker<T>{evt}(first_index);
 
     // the same to associate the element second_index of the (only)
     // data product of type std::vector<U> with the first object.
     auto const second_ptr = art::PtrMaker<U>{evt}(second_index);
 
     assn.addSingle(first_ptr, second_ptr, data);
   }
   catch (cet::exception& e) {
     mf::LogWarning("AssociationUtil")
       << "unable to create requested art:Assns, exception thrown: " << e;
     return false;
   }
 
   return true;
 } // util::CreateAssnD() [01b]

template<typename Producer , typename... Args>

std::enable_if_t<std::is_base_of_v<art::EDProducer, Producer>, bool> util::CreateAssnD	(	Producer const &	,
		Args &&...	args
	)

Definition at line 937 of file AssociationUtil.h.

References CreateAssnD().

   {
     return CreateAssnD(std::forward<Args>(args)...);
   }

template<typename T >

constexpr T util::DegreesToRadians ( T angle )

inline

Converts the argument angle from degrees into radians.

Definition at line 87 of file PhysicalConstants.h.

Referenced by cluster::LazyClusterParamsAlg::StartAngle(), and cluster::StandardClusterParamsAlg::StartAngle().

   {
     return angle / 180 * pi<T>();
   }

decltype(auto) util::dereference ( )

Returns a functor that returns *ptr of its argument ptr.

See also: util::Dereferencer

Example:

std::vector<int> values(ptrs.size());
std::transform
  (ptrs.cbegin(), ptrs.cend(), values.begin(), util::dereference());

will fill the vector values with the values pointed by the elements in ptrs.

Definition at line 128 of file operations.h.

   {
     return Dereferencer();
   }

template<typename Stream , typename Left , typename Right , typename Data >

void util::DumpAssociationsIntro	(	Stream &&	out,
		art::Assns< Left, Right, Data > const &	assns
	)

Dumps a short introduction about specified association.

Template Parameters

Stream	type of output stream
Left	first type in the association
Right	second type in the association
Data	metadata type in the association

Parameters

out	output stream
assns	the associations to be dumped

Definition at line 31 of file DumpAssociations.h.

References art::Assns< L, R, D >::size().

   {
     out << "Association between '" << cet::demangle_symbol(typeid(Left).name()) << "' and '"
         << cet::demangle_symbol(typeid(Right).name()) << "'";
     if (std::is_same<Data, void>()) {
       out << " with '" << cet::demangle_symbol(typeid(Data).name()) << "' metadata";
     }
     if (assns.size() > 0) { out << " contains " << assns.size() << " relations"; }
     else {
       out << " is empty";
     }
   } // DumpAssociationsIntro<Data>()

template<std::size_t Lead, typename... Iterables>

auto util::enumerate ( Iterables &&... iterables )

Range-for loop helper tracking the number of iteration.

Template Parameters

Lead	index of the parameter driving the start and end of the loop
Iterables	type of objects to be iterated together

Parameters

iterables all iterable objects to be iterated together

Returns: an object suitable for range-for loop

In the range-for loop, at each iteration this object yields a tuple of values, each of the type returned by dereferencing begin(iterable). For example:

constexpr std::size_t N = 4;
std::array<int, N> twice;
std::vector<double> thrice(N + 1);
for (auto&& [i, a, b]: util::enumerate(twice, thrice)) {
  a = 2 * i;
  b = 3.0 * i;
} // for

In this example, N iterations will be run because that is the size of the first iterable given to enumerate. If a different leading iterable is needed, that has to be specified as an argument. The following loop is completely equivalent to the former one:

for (auto&& [i, b, a]: util::enumerate<1U>(thrice, twice)) {
  a = 2 * i;
  b = 3.0 * i;
} // for

(the index is zero-based, so 1U refers to the second argument).

Definition at line 65 of file enumerate.h.

References infinite_counter().

Referenced by ophit::DumpOpHits::analyze(), sim::DumpSimEnergyDeposits::analyze(), ophit::DumpOpFlashes::analyze(), sim::MergeSimSources::dumpConfiguration(), larg4::ISTPC::ISTPC(), makeValueIndex(), larg4::OpFastScintillation::OpFastScintillation(), phot::PDFastSimPAR::PDFastSimPAR(), sim::MergeSimSources::produce(), and evgen::RadioGen::RadioGen().

   {
     return zip<Lead + 1>(infinite_counter(), std::forward<Iterables>(iterables)...);
   }

template<typename... Iterables>

auto util::enumerate ( Iterables &&... iterables )

This version of enumerate implicitly uses the first iterable as lead.

Definition at line 72 of file enumerate.h.

   {
     return enumerate<0U>(std::forward<Iterables>(iterables)...);
   }

template<typename Range , typename Pred >

auto util::filterRangeFor	(	Range &&	range,
		Pred &&	pred
	)		-> decltype(auto)

Provides iteration only through elements passing a condition.

Template Parameters

Range	the data to be iterated
Pred	the type of the predicate to be fulfilled

Parameters

range	the data to be iterated through
pred	the predicate to be tested

Returns: an object suitable to be used in a range-for loop

This adapter makes the range for loop iterate only through the elements of range which fulfil the predicate pred.

This example will print: "0 3 6 9 ":

std::vector<int> data = { 0, 1, 2, 3, 4, 5, 6 ,7, 8, 9 };
for (int v: util::filterRangeFor(data, [](int v){ return v % 3 == 0; })) {
  std::cout << v << " ";
} // for
std::cout << std::endl;

Note that pred may be copied (range will not be).

Requirements

Range is an object which can itself go through a range-for:

for (auto&& v: range);

is valid
Pred is a copiable unary function type, whose single argument can be converted from the value type of Range, and whose return value can be converted into a bool vaule

Definition at line 128 of file filterRangeFor.h.

Referenced by proxy::TrackCollectionProxyElement< CollProxy >::selectPoints().

   {
     return details::FilterRangeForStruct<Range, Pred>(std::forward<Range>(range),
                                                       std::forward<Pred>(pred));
   }

template<class T , class U >

std::vector< const U * > util::FindUNotAssociatedToT	(	art::Handle< U >	b,
		art::Event const &	evt,
		std::string const &	label
	)

inline

Definition at line 829 of file AssociationUtil.h.

References fa(), and art::Ptr< T >::get().

Referenced by CreateAssn().

 {
   // Do a FindOne for type T for each object of type U
   // If the FindOne returns an invalid maybe ref, add the pointer
   // of object type U to the return vector
 
   std::vector<const U*> notAssociated;
 
   art::FindOne<T> const fa(b, evt, label);
 
   for (size_t u = 0; u < b->size(); ++u) {
     cet::maybe_ref<T const> t(fa.at(u));
     if (!t.isValid()) {
       art::Ptr<U> ptr(b, u);
       notAssociated.push_back(ptr.get());
     }
   }
 
   return notAssociated;
 }

template<class T , class U >

std::vector< art::Ptr< U > > util::FindUNotAssociatedToTP	(	art::Handle< U >	b,
		art::Event const &	evt,
		std::string const &	label
	)

inline

Definition at line 854 of file AssociationUtil.h.

References fa().

Referenced by CreateAssn().

 {
   // Do a FindOneP for type T for each object of type U
   // If the FindOne returns an invalid maybe ref, add the pointer
   // of object type U to the return vector
 
   std::vector<art::Ptr<U>> notAssociated;
 
   art::FindOneP<T> const fa(b, evt, label);
 
   for (size_t u = 0; u < b->size(); ++u) {
     cet::maybe_ref<T const> t(fa.at(u));
     if (!t.isValid()) { notAssociated.emplace_back(b, u); }
   }
 
   return notAssociated;
 }

template<class A , class F >

void util::for_each_associated_group	(	A const &	assns,
		F &	func
	)

Helper functions to access associations in order.

Template Parameters

A	type of association being read
F	type of functor to be called on each associated group

Parameters

assns	the association being read
func	functor to be called on each associated group

See also: associated_groups() art::for_each_group()

Deprecated:: Moved into canvas: art::for_each_group().

Definition at line 46 of file ForEachAssociatedGroup.h.

References art::for_each_group().

   {
     art::for_each_group(assns, func);
   }

template<std::size_t... Indices, typename Coll >

decltype(auto) util::get_const_elements ( Coll && coll )

Range-for loop helper iterating across the constant values of the specified collection.

See also: util::values()

This function is equivalent to util::values() but the values are extracted as if the specified collection were constant.

Referenced by util::details::get_elements_impl< Coll, First, Others >::iterate().

template<std::size_t... Indices, typename Coll >

decltype(auto) util::get_elements ( Coll && coll )

Range-for loop helper iterating across some of the element of each value in the specified collection.

Template Parameters

Indices	indices of the elements to extract
Coll	type of the collection to iterate through

Parameters

coll	the collection to iterate through

Returns: an object suitable for range-for loop

See also: util::get_const_elements()

This function enables range-for loops with a selection of elements from a container of tuple (or anything responding to util::get()).

The following example fills a map using as key the first element (0U) of a tuple and as value the third element (2U):

std::vector<std::tuple<char, int, float>> data
  { { 'z', 0, 1.0F }, { 'o', 1, 3.0F }, { 't', 2, 9.0F } };
std::map<char, double> factors;
for (auto const& [ letter, factor ]: util::get_elements<0U, 2U>(data)) {
  factors.emplace(letter, factor);
}

If only one index is specified, the loop will not use structured binding, but rather a simple variable:

std::vector<int> exponents;
for (int exponent: util::get_elements<1U>(data)) {
  exponents.push_back(exponent);
}

While the examples do not demonstrate changing the values in the data collection, that is also supported.

Note: This function also works with associative containers based on std::pair (std::map and the likes).

Referenced by util::details::get_elements_impl< Coll, First, Others >::iterate().

template<class T , class U >

std::vector< std::vector< size_t > > util::GetAssociatedVectorManyI	(	art::Handle< art::Assns< T, U >>	h,
		art::Handle< std::vector< T >>	index_p
	)

inline

Definition at line 895 of file AssociationUtil.h.

Referenced by microboone::CosmicRemovalAna::analyze(), CreateAssn(), cosmic::BeamFlashTrackMatchTagger::produce(), and calo::TrackCalorimetry::produce().

 {
   std::vector<std::vector<size_t>> associated_indices(index_p->size());
   for (auto const& pair : *h)
     associated_indices.at(pair.first.key()).push_back(pair.second.key());
   return associated_indices;
 }

template<class T , class U >

std::vector< std::vector< const U * > > util::GetAssociatedVectorManyP	(	art::Handle< art::Assns< T, U >>	h,
		art::Handle< std::vector< T >>	index_p
	)

inline

Definition at line 906 of file AssociationUtil.h.

Referenced by CreateAssn().

 {
   std::vector<std::vector<const U*>> associated_pointers(index_p->size());
   for (auto const& pair : *h)
     associated_pointers.at(pair.first.key()).push_back(&(*(pair.second)));
   return associated_pointers;
 }

template<class T , class U >

std::vector< size_t > util::GetAssociatedVectorOneI	(	art::Handle< art::Assns< T, U >>	h,
		art::Handle< std::vector< T >>	index_p
	)

inline

Definition at line 875 of file AssociationUtil.h.

Referenced by CreateAssn().

 {
   std::vector<size_t> associated_index(index_p->size());
   for (auto const& pair : *h)
     associated_index.at(pair.first.key()) = pair.second.key();
   return associated_index;
 }

template<class T , class U >

std::vector< const U * > util::GetAssociatedVectorOneP	(	art::Handle< art::Assns< T, U >>	h,
		art::Handle< std::vector< T >>	index_p
	)

inline

Definition at line 885 of file AssociationUtil.h.

Referenced by microboone::CosmicRemovalAna::analyze(), and CreateAssn().

 {
   std::vector<const U*> associated_pointer(index_p->size());
   for (auto const& pair : *h)
     associated_pointer.at(pair.first.key()) = &(*(pair.second));
   return associated_pointer;
 }

template<typename Groups >

auto util::groupByIndex	(	Groups &&	groups,
		std::size_t	index
	)		-> decltype(auto)

Returns the group within groups with the specified index.

Template Parameters

Groups the type of collection of groups

Parameters

groups	the collection of all groups
index	the index of the group to be accessed

Returns: the group with specified index (may be a reference)

See also: associated_groups()

The groups argument is expected to be the one returned by associated_groups.

Definition at line 173 of file ForEachAssociatedGroup.h.

Referenced by TrackProducerFromPFParticle::produce().

   {
     return *(std::next(groups.begin(), index));
   }

template<typename T = std::size_t>

auto util::infinite_counter ( T begin = T{} )

Version of util::counter() starting at begin and never ending.

Template Parameters

T	type of counter value

Parameters

begin the count to start from (default-constructed, usually some form of 0, by default)

Returns: a control object for range-for loop

An example of usage:

std::vector<unsigned char> data;
for (auto ch: util::infinite_counter<unsigned char>()) {
  if (data.size() >= 512U) break;
  data.push_back(ch);
}

This loop runs through the full range of a character (8 bits, supposedly) twice before being broken.

Definition at line 302 of file counter.h.

References util::count_iterator< T >::count_iterator(), and span().

Referenced by counter(), and enumerate().

 {
   return util::span(count_iterator(begin), details::infinite_endcount_iterator<T>());
 } // util::infinite_counter()

template<typename Cont , typename Adaptor >

auto util::make_adapted_const_span	(	Cont &	cont,
		Adaptor &&	adaptor
	)

Creates constant iteration span from specified collection via an adaptor.

Definition at line 273 of file span.h.

References util::span_base::get_cbegin(), util::span_base::get_cend(), and make_adapted_span().

   {
     return make_adapted_span(
       span_base::get_cbegin(cont), span_base::get_cend(cont), std::forward<Adaptor>(adaptor));
   }

template<typename BIter , typename EIter , typename Adaptor >

auto util::make_adapted_span	(	BIter	begin,
		EIter	end,
		Adaptor &&	adaptor
	)

Creates a span from specified iterators via an adaptor.

See also: make_adapted_span(Cont&, Adaptor&&)

Definition at line 227 of file span.h.

References span().

Referenced by make_adapted_const_span(), make_adapted_span(), and make_transformed_span().

   {
     return util::span(adaptor(begin), adaptor(end));
   }

template<typename Cont , typename Adaptor >

auto util::make_adapted_span	(	Cont &	cont,
		Adaptor &&	adaptor
	)

Creates a span from specified collection via an adaptor.

Parameters

cont	collection to be iterated through
adapter	iterator transformation to be applied on `cont`

Returns: a util::span object iterating via an adapter to cont

See also: make_adapted_span(BIter, EIter, Adaptor&&), make_adapted_const_span(Cont, Adaptor&&)

This interface is just a little help on using iterator adapters. Note that adapter transforms the iterators of cont, not its values. The adapter needs to be written aside and it's in general not trivial. An example of iterating through a collection of objects (actually, just float for simplicity) stored as unique pointers in a collection:

float accumulate(std::vector<std::unique_ptr<float>> const& v) {
  using src_iterator = std::vector<std::unique_ptr<float>>::const_iterator;
  float sum = 0.0F;
  for (float v: util::make_adapted_span(v, boost::make_indirect_iterator<src_iterator>))
    sum += v;
  return sum;
} // accumulate()

This example shows the usage of util::make_adapted_span(). In the specific example, it would have been more explicit to use the constant counterpart, util::make_adapted_const_span(). The adaptor helper boost::make_indirect_iterator() is provided in boost/iterator/indirect_iterator.hpp.

Definition at line 264 of file span.h.

References util::span_base::get_begin(), util::span_base::get_end(), and make_adapted_span().

   {
     return make_adapted_span(
       span_base::get_begin(cont), span_base::get_end(cont), std::forward<Adaptor>(adaptor));
   }

template<typename Cont >

auto util::make_const_span ( Cont & cont )

Creates a span with constant iterator access from a container type.

Definition at line 212 of file span.h.

References util::span_base::get_cbegin(), and util::span_base::get_cend().

   {
     return span{span_base::get_cbegin(cont), span_base::get_cend(cont)};
   }

template<typename BIter , typename EIter >

auto util::make_span	(	BIter	begin,
		EIter	end
	)

Creates a span from specified iterators (can use constructor instead).

Definition at line 198 of file span.h.

References span().

   {
     return util::span(begin, end);
   }

template<typename Cont >

auto util::make_span ( Cont & cont )

Creates a span from a container type.

Definition at line 205 of file span.h.

References util::span_base::get_begin(), and util::span_base::get_end().

   {
     return span{span_base::get_begin(cont), span_base::get_end(cont)};
   }

template<typename Cont , typename Op >

auto util::make_transformed_const_span	(	Cont &	cont,
		Op &&	op
	)

Creates constant iteration span from specified collection via a transformation op.

Definition at line 349 of file span.h.

References make_transformed_span().

   {
     return make_transformed_span(std::as_const(cont), std::forward<Op>(op));
   }

template<typename BIter , typename EIter , typename Op >

auto util::make_transformed_span	(	BIter	begin,
		EIter	end,
		Op &&	op
	)

Creates a span from specified iterators via an adaptor.

See also: make_transformed_span(Cont&, Op&&)

Definition at line 294 of file span.h.

References make_adapted_span().

Referenced by util::details::get_elements_impl< Coll, First, Others >::iterate(), make_transformed_const_span(), and make_transformed_span().

   {
     auto adaptor = [&op](auto iter) { return boost::make_transform_iterator(iter, op); };
     return util::make_adapted_span(begin, end, adaptor);
   }

template<typename Cont , typename Op >

auto util::make_transformed_span	(	Cont &	cont,
		Op &&	op
	)

Creates a span from specified collection via an adaptor.

Parameters

cont	collection to be iterated through
adapter	iterator transformation to be applied on `cont`

Returns: a util::span object iterating via an adapter to cont

See also: make_transformed_span(BIter, EIter, Op&&), make_transformed_const_span(Cont, Op&&), make_adapted_span(),

This function transforms as directed by the unary operation op the result of each iteration cycle, so that instead of cont[i], a op(cont[i]) is assigned to the iteration control variable. An example of iterating through a collection of objects (actually, just float for simplicity) stored as unique pointers in a collection:

float accumulate(std::vector<std::unique_ptr<float>> const& v) {
  float sum = 0.0F;
  for (float v: util::make_transformed_span(v, [](auto& ptr){ return *ptr; }))
    sum += v;
  return sum;
} // accumulate()

This example shows the usage of util::make_transformed_span(). In the specific example, it would have been more explicit to use the constant counterpart, util::make_transformed_const_span(). Note that in the example the dereference result is copied into v. For have it by reference, the operation op must be properly written:

void scale(std::vector<std::unique_ptr<float>>& v, float factor) {
  for (float& v: util::make_transformed_span(v, [](auto& ptr) -> float& { return *ptr; }))
    v *= factor;
} // scale()

Definition at line 339 of file span.h.

References util::span_base::get_begin(), util::span_base::get_end(), and make_transformed_span().

   {
     return make_transformed_span(
       span_base::get_begin(cont), span_base::get_end(cont), std::forward<Op>(op));
   }

template<typename Coll , typename KeyOf >

std::vector<size_t> util::MakeIndex	(	Coll const &	data,
		KeyOf	key_of = `KeyOf()`
	)

Creates a map of indices from an existing collection.

Template Parameters

Coll	type of the collection
KeyOf	type of the extractor of the key

Parameters

data	the data collection
key_of	instance of a functor extracting a key value from a datum

Returns: a vector with indices corresponding to the data keys

This function maps the index of the items in data to an integral key extracted from each item. For example, if the items are wires and the key_of function extracts their channel ID, the resulting vector will contain for each channel ID the index in data of the wire with that channel ID.

The key is converted into a unsigned integer (size_t). If multiple items have the same key, the outcome for that key is undefined. If no items has a specific key, the index of that key is assigned as

std::numeric_limits<size_t>::max()

, i.e. an index larger than the size of the original data collection.

The returned vector is big enough to accommodate indices corresponding to the keys of all the items in data. It may contain "holes" (that is, some keys that have no corresponding items have a

std::numeric_limits<size_t>::max()

value). The memory allocated for the vector may be larger than necessary (if that is a problem, std::vector::shrink_to_fit() can be used, but it may create more problems than it solves).

Definition at line 43 of file MakeIndex.h.

Referenced by recob::HitCollectionAssociator::prepare_associations().

   {
 
     // we start the index with the best guess that all the items will have
     // a unique key and they are contiguous:
     // the index would have the same size as the data
     std::vector<size_t> Index(data.size(), std::numeric_limits<size_t>::max());
 
     size_t min_size = 0; // minimum size needed to hold all keys
 
     size_t iDatum = 0;
     for (auto const& datum : data) {
       size_t key = size_t(key_of(datum));
       if (key >= min_size) min_size = key + 1;
       if (Index.size() <= key) {
         // make room for the entry: double the size
         Index.resize(std::max(key + 1, Index.size() * 2), std::numeric_limits<size_t>::max());
       } // if expand index
       Index[key] = iDatum;
       ++iDatum;
     } // for datum
     Index.resize(min_size);
     return Index;
   } // MakeIndex()

template<typename Coll , typename KeyOf >

auto util::MakeMap	(	Coll const &	data,
		KeyOf	key_of = `KeyOf()`
	)		-> std::vector<decltype(key_of((data.begin()))) const>

Creates a map of objects from an existing collection.

Template Parameters

Coll	type of the collection
KeyOf	type of the extractor of the key

Parameters

data	the data collection
key_of	instance of a functor extracting a key value from a datum

Returns: a vector with pointers to data corresponding to their keys

This function maps the items in data to an integral key extracted from each of them. For example, if the items are wires and the key_of function extracts their channel ID, the resulting vector will contain for each channel ID the pointer to the wire with that channel ID.

The key is converted into a unsigned integer (size_t). If multiple items have the same key, the outcome for that key is undefined. If no items has a specific key, the index of that key is assigned a null pointer.

The returned vector is big enough to accommodate pointers corresponding to the keys of all the items in data. It may contain "holes" (that is, some keys that have no corresponding items have a null pointer value). The memory allocated for the vector may be larger than necessary (if that is a problem, std::vector::shrink_to_fit() can be used, but it may create more problems than it solves).

Definition at line 96 of file MakeIndex.h.

   {
     using Mapped_t = decltype(key_of(*(data.begin())));
     using Ptr_t = Mapped_t const*;
     using Map_t = std::vector<Ptr_t>;
 
     // we start the index with the best guess that all the items will have
     // a unique key and they are contiguous:
     // the index would have the same size as the data
     Map_t Index(data.size(), nullptr);
 
     size_t min_size = 0; // minimum size needed to hold all keys
 
     for (auto const& datum : data) {
       size_t key = size_t(key_of(datum));
       if (key >= min_size) min_size = key + 1;
       if (Index.size() <= key) {
         // make room for the entry: double the size
         Index.resize(std::max(key + 1, Index.size() * 2), nullptr);
       } // if expand index
       Index[key] = &datum;
     } // for datum
     Index.resize(min_size);
     return Index;
   } // MakeMap()

template<typename Coll >

auto util::makePointerVector ( Coll & coll )

Creates a STL vector with pointers to data from another collection.

Template Parameters

Coll	type of collection of data

Parameters

coll	data collection

Returns: a STL vector with pointers to coll data elements, with same order

Definition at line 186 of file SortByPointers.h.

References util::details::PointerVectorMaker< Coll, typename >::make().

Referenced by sim::DumpSimPhotons::DumpElement(), SortByPointers(), and SortUniquePointers().

 {
   return details::PointerVectorMaker<Coll>::make(coll);
 }

template<typename... DIMS>

auto util::makeTensorIndices ( DIMS... dims )

Instantiates a TensorIndices class with the specified dimensions.

Template Parameters

DIMS	types for each of the arguments

Parameters

dims	size of each of the dimensions

Returns: a TensorIndices object with properly initialised dimensions

The rank of the tensor is determined by the number of arguments; example:

auto indices = util::makeTensorIndices(3, 4);

will initialise a TensorIndices<2> (that's matrix indices), for a 3 x 4 (3 rows, 4 columns) disposition.

Definition at line 529 of file TensorIndices.h.

   {
     return TensorIndices<sizeof...(DIMS)>{TensorIndicesBasicTypes::DimSize_t(dims)...};
   }

template<typename Coll , typename Extractor >

decltype(auto) util::makeValueIndex	(	Coll const &	coll,
		Extractor	getter
	)

Returns a map of value to index.

Template Parameters

Coll	type of container
Extractor	type of value extraction function

Parameters

coll	container to get the map of
getter	function applied to each element to extract the value for map

Returns: a value-to-index associative container

Exceptions

std::runtime_error if multiple elements yield the same value

The collection coll is navigated in sequence from begin() to end(), and a map is created where to each key, getter(coll[i]), the index i is associated. The value returned by getter() is copied into the key. Therefore that value needs to satisfy all the requirements of the key of the STL associative container std::map. Duplicate values will trigger an exception.

Requirements

The collection type Coll must have:

value_type type defining the content of the container
support begin() and end() free functions returning input iterators, that is support a ranged-for loop

Referenced by makeValueIndex(), and SortUniquePointers().

template<typename Coll >

auto util::makeValueIndex ( Coll const & coll )

Definition at line 59 of file makeValueIndex.h.

References enumerate(), makeValueIndex(), and to_string().

   {
     return makeValueIndex(coll, util::pre_std::identity());
   }

template<typename Cont , typename Mapping >

auto util::mapContainer	(	Cont	cont,
		Mapping	mapping
	)

Returns a container-like object mapping the content of cont.

Template Parameters

Cont	type of the original container
Mapping	type of mapping object

Parameters

cont	the container
mapping	the mapping to be applied

Returns: a util::MappedContainer object transparently applying the mapping

See util::MappedContainer for the details on the format of the mapping, the ownership of the container and the supported operations.

Definition at line 415 of file MappedContainer.h.

   {
     return MappedContainer<Cont, Mapping>(cont, mapping);
   }

template<typename Coll , typename PtrColl >

void util::MoveFromPointers	(	Coll &	dest,
		PtrColl &	src
	)

Moves the content from a collection of pointers to one of data.

Template Parameters

Coll	type of collection of data
PtrColl	type of collection of pointers to data

Parameters

dest	collection to be filled
src	collection with the pointers to data to be moved

The data pointed from each pointer in src is moved into dest. The destination collection is cleared first, and Coll must support both clear() and push_back()

Definition at line 58 of file SortByPointers.h.

References util::details::MoveFromPointersImpl< Coll, PtrColl >::move(), SortByPointers(), and SortUniquePointers().

Referenced by SortByPointers().

   {
     details::MoveFromPointersImpl<Coll, PtrColl>::move(dest, src);
   }

template<typename _Key , typename _Tp , typename _Compare >

bool util::operator!=	(	const VectorMap< _Key, _Tp, _Compare > &	__x,
		const VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

Based on operator==.

Definition at line 437 of file VectorMap.h.

   {
     return !(__x == __y);
   }

template<unsigned int RANK1, unsigned int RANK2, typename = std::enable_if_t<(RANK1 != RANK2), bool>>

bool util::operator!=	(	TensorIndices< RANK1 > const &	a,
		TensorIndices< RANK2 > const &	b
	)

Comparison operator with tensors of different rank.

Definition at line 510 of file TensorIndices.h.

Referenced by util::TensorIndices< DIMS >::operator!=().

   {
     return true;
   }

template<typename Cont , typename Mapping , typename Container , typename Reference >

MappedContainer<Cont, Mapping>::template IteratorBase<Container, Reference> util::operator+	(	typename MappedContainer< Cont, Mapping >::template IteratorBase< Container, Reference >::difference_type	n,
		typename MappedContainer< Cont, Mapping >::template IteratorBase< Container, Reference > const &	it
	)

Definition at line 703 of file MappedContainer.h.

References n.

   {
     return it + n;
   }

template<typename _Key , typename _Tp , typename _Compare >

bool util::operator<	(	const VectorMap< _Key, _Tp, _Compare > &	__x,
		const VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

Definition at line 425 of file VectorMap.h.

   {
     return std::lexicographical_compare(__x.sortedVectorMap.begin(),
                                         __x.sortedVectorMap.end(),
                                         __y.sortedVectorMap.begin(),
                                         __y.sortedVectorMap.end(),
                                         __x.valueCompare);
   }

std::ostream& util::operator<<	(	std::ostream &	out,
		EventChangeTracker_t const &	trk
	)

inline

Definition at line 97 of file ChangeTrackers.h.

   {
     out << std::string(trk);
     return out;
   }

std::ostream& util::operator<<	(	std::ostream &	out,
		DataProductChangeTracker_t const &	trk
	)

inline

Definition at line 197 of file ChangeTrackers.h.

   {
     out << std::string(trk);
     return out;
   }

std::ostream& util::operator<<	(	std::ostream &	out,
		PlaneDataChangeTracker_t const &	trk
	)

inline

Definition at line 308 of file ChangeTrackers.h.

   {
     out << std::string(trk);
     return out;
   }

template<typename _Key , typename _Tp , typename _Compare >

bool util::operator<=	(	const VectorMap< _Key, _Tp, _Compare > &	__x,
		const VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

Based on operator<.

Definition at line 453 of file VectorMap.h.

   {
     return !(__y < __x);
   }

template<typename _Key , typename _Tp , typename _Compare >

bool util::operator==	(	const VectorMap< _Key, _Tp, _Compare > &	__x,
		const VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

Definition at line 418 of file VectorMap.h.

References util::VectorMap< _Key, _Tp, _Compare >::sortedVectorMap.

   {
     return __x.sortedVectorMap == __y.sortedVectorMap;
   }

template<unsigned int RANK1, unsigned int RANK2, typename = std::enable_if_t<(RANK1 != RANK2), bool>>

bool util::operator==	(	TensorIndices< RANK1 > const &	a,
		TensorIndices< RANK2 > const &	b
	)

Comparison operator with tensors of different rank.

Definition at line 501 of file TensorIndices.h.

Referenced by util::TensorIndices< DIMS >::operator==().

   {
     return false;
   }

template<typename _Key , typename _Tp , typename _Compare >

bool util::operator>	(	const VectorMap< _Key, _Tp, _Compare > &	__x,
		const VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

Based on operator<.

Definition at line 445 of file VectorMap.h.

   {
     return __y < __x;
   }

template<typename _Key , typename _Tp , typename _Compare >

bool util::operator>=	(	const VectorMap< _Key, _Tp, _Compare > &	__x,
		const VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

Based on operator<.

Definition at line 461 of file VectorMap.h.

   {
     return !(__x < __y);
   }

template<typename Range >

auto util::operator\|	(	Range &&	range,
		RangeForWrapperTag
	)		-> decltype(auto)

Transforms a range so that it can be used in a range-for loop.

Template Parameters

Range the type of range to be transformed

Parameters

range the range to be transformed

Returns: an equivalent range object to be used in a range-for loop

This is necessary only when the argument provides different types for the begin-of-range and end-of-range iterators. This is also superfluous for compilers adhering to C++ 2017 standard, which accepts iterators of different types by requirement. Example of usage:

Range data; // initialization

for (auto&& value: data | util::range_for) // ...

where data is supposed to gave begin and end iterators of different types.

Definition at line 470 of file RangeForWrapper.h.

References wrapRangeFor().

   {
     return wrapRangeFor(std::forward<Range>(range));
   }

template<typename T = double>

constexpr T util::pi ( )

inline

Returns the constant pi (up to 35 decimal digits of precision)

Definition at line 80 of file PhysicalConstants.h.

Referenced by util::GeometryUtilities::CalculatePitch(), art::detail::consumed_products_for_module(), larg4::OpFastScintillation::detectedReflecHits(), phot::SemiAnalyticalModel::detectedReflectedVisibilities(), phot::SemiAnalyticalModel::Disk_SolidAngle(), larg4::OpFastScintillation::Disk_SolidAngle(), art::PathManager::fillWorkers_(), apa::DisambigAlg::FindChanTimeEndPts(), phot::PropagationTimeModel::getVISTimes(), larg4::OpFastScintillation::getVISTimes(), art::detail::make_path_ordering_edges(), phot::SemiAnalyticalModel::Omega_Dome_Model(), larg4::OpFastScintillation::Omega_Dome_Model(), art::FileDumperOutput::printPrincipal(), phot::PropagationTimeModel::propagationTime(), larg4::OpFastScintillation::propagationTime(), art::ConsumesInfo::showMissingConsumes(), art::ConsumesCollector::sortConsumables(), geo::OpDetGeo::ThetaZ(), art::detail::triggerReport(), larg4::OpFastScintillation::VISHits(), phot::SemiAnalyticalModel::VISVisibility(), larg4::OpFastScintillation::VUVHits(), phot::SemiAnalyticalModel::VUVVisibility(), and geo::WireGeo::WireGeo().

   {
     return 3.14159265358979323846264338327950288L;
   }

template<typename T >

constexpr T util::RadiansToDegrees ( T angle )

inline

Converts the argument angle from radians into degrees ( $\pi \rightarrow 180$ )

Definition at line 94 of file PhysicalConstants.h.

Referenced by geo::OpDetGeo::ThetaZ(), and geo::WireGeo::ThetaZ().

   {
     return angle / pi<T>() * 180;
   }

template<typename Coll , typename Sorter >

void util::SortByPointers	(	Coll &	coll,
		Sorter	sorter
	)

Applies sorting indirectly, minimizing data copy.

Template Parameters

Coll	type of collection to be sorted
Sorter	type of sorter

Parameters

coll	collection to be sorted
sorter	functor sorting a vector of pointers (`makePointerVector()`)

The sorter functor can receive a reference to a vector as the one produced by makePointerVector(coll) (that is, a C++ STL vector of pointers to the value type of Coll), and sort it "in place". The container Comp must implement push_back() call in a std::vector fashion.

The algorithm is equivalent to the following:

create a parallel vector of pointers to the data
sort the data pointers (delegating to sorter)
move the data, sorted, from the original collection to a new one
replace the content of cont with the one from the sorted collection

Single elements are moved from the original collection to a new one.

The data elements of Coll must be moveable, as Coll itself must be.

Note: Use this algorithm only as a last resort, as there are usually better ways to sort collections than this one, which is not even particularly optimized.

Definition at line 193 of file SortByPointers.h.

References makePointerVector(), and MoveFromPointers().

Referenced by MoveFromPointers().

 {
 
   using coll_t = Coll;
 
   //
   // create the collection of pointers to data
   //
   auto ptrs = makePointerVector(coll);
 
   //
   // delegate the sorting by pointers
   //
   sorter(ptrs);
 
   //
   // create a sorted collection moving the content from the original one
   //
   coll_t sorted;
   MoveFromPointers(sorted, ptrs);
 
   //
   // replace the old container with the new one
   //
   coll = std::move(sorted);
 
 } // util::SortByPointers()

template<typename Coll , typename Sorter >

void util::SortUniquePointers	(	Coll &	coll,
		Sorter &&	sorter
	)

Sorts a vector of unique pointers using a C pointer sorter.

Template Parameters

Coll	type of collection to be sorted
Sorter	type of sorter function

Parameters

coll	collection to be sorted
sorter	sorting procedure

This adapter moves the unique pointers around to match a sorted version of source. This is an expensive procedure, implying the creation of a temporary vector and of additional supporting data: avoid it if at all possible.

Definition at line 223 of file SortByPointers.h.

References makePointerVector(), and makeValueIndex().

Referenced by MoveFromPointers().

 {
 
   using Collection_t = Coll;
   using UPtr_t = typename Collection_t::value_type;
 
   static_assert(util::is_unique_ptr_v<UPtr_t>);
 
   //
   // create the collection of pointers to data
   //
   auto ptrs = makePointerVector(coll);
 
   // data pointer -> index
   auto const ptrIndex = util::makeValueIndex(ptrs);
 
   //
   // delegate the sorting by pointers
   //
   sorter(ptrs);
 
   //
   // create a sorted collection moving the content from the original one
   //
   Collection_t sorted;
   for (auto const& dataPtr : ptrs) {
     std::size_t const originalIndex = ptrIndex.at(dataPtr);
     sorted.emplace_back(std::move(coll[originalIndex]));
   }
 
   //
   // replace the old container with the new one
   //
   coll = std::move(sorted);
 
 } // util::SortUniquePointers()

template<typename IterB , typename IterE , typename Adaptor >

util::span	(	IterB &&	b,
		IterE &&	e,
		Adaptor &&	adaptor
	)		-> span< decltype(adaptor(std::forward< IterB >(b))), decltype(adaptor(std::forward< IterE >(e))) >

Referenced by cluster::ClusterCrawlerAlg::ClusterLoop(), counter(), lar_content::NViewDeltaRayMatchingAlgorithm< T >::CreatePfos(), util::span< BIter, EIter >::empty(), sim::PhotonVoxelDef::GetVoxelStepCoordsUnchecked(), infinite_counter(), geo::vect::iterateCoords(), make_adapted_span(), make_span(), tca::OverlapFraction(), util::span< BIter, EIter >::span(), and zip().

template<typename _Key , typename _Tp , typename _Compare >

void util::swap	(	VectorMap< _Key, _Tp, _Compare > &	__x,
		VectorMap< _Key, _Tp, _Compare > &	__y
	)

inline

See VectorMap::swap().

Definition at line 469 of file VectorMap.h.

References util::VectorMap< _Key, _Tp, _Compare >::swap().

   {
     __x.swap(__y);
   }

decltype(auto) util::takeAddress ( )

Returns a functor that returns the address of its argument.

See also: util::AddressTaker

Example:

std::vector<int*> ptrs(data.size());
std::transform
  (data.begin(), data.end(), ptrs.begin(), util::takeAddress());

will fill the vector ptrs with pointers to the elements of data.

Why bother?

C++ already provides a tool to effectively take an address, std::addressof. The reason for takeAddress() is that std::addressof() is a function, with many overloads, and to use it in a STL algorithm the overload has to be resolved first. For example:

using addressof_t = int const*(*)(int const&);
std::transform(data.cbegin(), data.cend(), std::back_inserter(dataPtr),
  ((addressof_t) &std::addressof));

One important limit is that the type of the argument (in this case int const&) needs to be known or deduced in a quite precise way, in particular regarding constantness and referenceness. This is unconvenient and convoluted enough that one would rather create a new function, like:

auto takeAddress = [](auto&& ref){ return std::addressof(ref); };
std::vector<int const*> dataPtr;
std::transform(data.cbegin(), data.cend(), std::back_inserter(dataPtr),
  takeAddress);

This util::takeAddress() operates in a very similar way to the lambda in the last example.

Definition at line 85 of file operations.h.

   {
     return AddressTaker();
   }

template<typename Coll >

decltype(auto) util::values ( Coll && coll )

Range-for loop helper iterating across the values of the specified collection.

Template Parameters

Coll	type of the collection to iterate through

Parameters

coll	the collection to iterate through

Returns: an object suitable for range-for loop

See also: util::const_values()

This function is in most of cases a no-operation, returning the collection just as it was specified, to be iterated on directly. In case of mapping types, though, a different object is returned and the iteration will happen to the value type of the mapping instead than on the key-value pair.

std::map<int, float> data { { 1, 4.0F }, { 3, 12.0F }, { 2, 8.0F } };
std::vector<float> values;
for (float value: util::values(data))
  values.push_back(value);

will result in values vector being of size 3 and with values { 4.0F, 8.0F, 12.0F } (the order is the one of iterating through a std::map).

template<typename Range >

auto util::wrapRangeFor ( Range && range ) -> decltype(auto)

Wraps an object for use in a range-for loop.

Template Parameters

Range type of range object (anything with begin() and end())

Parameters

range instance of the range object to be wrapped

This is necessary only when the argument provides different types for the begin-of-range and end-of-range iterators. This is also superfluous for compilers adhering to C++ 2017 standard, which accepts iterators of different types by requirement. Example of usage:

Range data; // initialization

for (auto&& value: util::wrapRangeFor(data)) // ...

where data is supposed to gave begin and end iterators of different types.

Definition at line 440 of file RangeForWrapper.h.

Referenced by operator|().

   {
     return details::WrapRangeForDispatcher<Range>::wrap(std::forward<Range>(range));
   }

template<std::size_t Lead, typename... Iterables>

auto util::zip ( Iterables &&... iterables )

Range-for loop helper iterating across many collections at the same time.

Template Parameters

Lead	index of the parameter driving the start and end of the loop
Iterables	type of objects to be iterated together

Parameters

iterables all iterable objects to be iterated together

Returns: an object suitable for range-for loop

See also: util::enumerate()

In the range-for loop, at each iteration this object yields a tuple of values, each of the type returned by dereferencing begin(iterable). For example:

constexpr std::size_t N = 4;
std::array<int, N> twice;
std::vector<double> thrice(N + 1);
unsigned int i = 0;
for (auto&& [ a, b]: util::zip(twice, thrice)) {
  a = 2 * i;
  b = 3.0 * i;
  ++i;
} // for

In this example, N iterations will be run because that is the size of the first iterable given to enumerate. If a different leading iterable is needed, that has to be specified as an argument. The following loop is completely equivalent to the former one:

unsigned int i = 0;
for (auto&& [ b, a]: util::zip<1U>(thrice, twice)) {
  a = 2 * i;
  b = 3.0 * i;
  ++i;
} // for

(the index is zero-based, so 1U refers to the second argument).

Range-for loop helper iterating across many collections at the same time.

Definition at line 293 of file zip.h.

References span().

Referenced by detinfo::DetectorClocksStandard::debugReport(), and sim::MergeSimSources::produce().

 {
 
   return util::span(details::make_zip_begin_iterator<Lead>(iterables...),
                     details::make_zip_end_iterator<Lead>(iterables...));
 
 } // util::zip()

template<typename... Iterables>

auto util::zip ( Iterables &&... iterables )

Version of zip() with first iterator implicitly leading the iteration.

Range-for loop helper iterating across many collections at the same time.

Definition at line 78 of file zip.h.

References span().

Referenced by detinfo::DetectorClocksStandard::debugReport(), and sim::MergeSimSources::produce().

   {
     return zip<0U>(std::forward<Iterables>(iterables)...);
   }

Variable Documentation

constexpr double util::kBogusD = -999.

obviously bogus double value

Definition at line 71 of file PhysicalConstants.h.

Referenced by trkf::Geometric3DVertexFitter::addTrackToVertex(), ShowerRecoTools::ShowerExampleTool::CalculateElement(), recob::Vertex::chi2PerNdof(), recob::Event::Energy(), trkf::TrackKalmanFitter::fillResult(), trkf::Geometric3DVertexFitter::fitTwoTracks(), trkf::TrackState::mass(), trkf::TrajectoryMCSFitter::mass(), trkf::SpacePts::produce(), shwf::ShowerReco::produce(), calo::Calorimetry::produce(), calo::GnocchiCalorimetry::produce(), trkf::CosmicTracker::produce(), evd::TWireProjPad::SaveSeedList(), recob::Event::SigmaEnergy(), recob::Vertex::Vertex(), and trkf::Geometric3DVertexFitter::weightedAverageState().

constexpr float util::kBogusF = -999.

obviously bogus float value

Definition at line 73 of file PhysicalConstants.h.

Referenced by lar_pandora::LArPandoraTrackCreation::BuildTrack(), ShowerRecoTools::ShowerPandoraSlidingFitTrackFinder::CalculateElement(), recob::Track::Chi2PerNdof(), trkf::Geometric3DVertexFitter::computeMeta(), and lar_pandora::LArPandoraTrackCreation::produce().

constexpr int util::kBogusI = -999

obviously bogus integer value

Definition at line 72 of file PhysicalConstants.h.

Referenced by lar_pandora::LArPandoraShowerCreation::BuildPCAxis(), lar_pandora::LArPandoraTrackCreation::BuildTrack(), ShowerRecoTools::ShowerExampleTool::CalculateElement(), ShowerRecoTools::ShowerPandoraSlidingFitTrackFinder::CalculateElement(), reco::shower::LArPandoraModularShowerCreation::produce(), and recob::Vertex::Vertex().

constexpr double util::kc = 29.9792458

Speed of light in vacuum in LArSoft units [cm/ns].

Definition at line 59 of file PhysicalConstants.h.

Referenced by genf::GFEnergyLossBrems::energyLoss(), and genf::GFMaterialEffects::energyLossBrems().

constexpr double util::kCentimeterToMeter = 1. / kMeterToCentimeter

Definition at line 64 of file PhysicalConstants.h.

constexpr double util::keVToMeV = 1.e-6

1e6 eV = 1 MeV

Definition at line 68 of file PhysicalConstants.h.

constexpr double util::kGeVToElectrons = 4.237e7

23.6eV per ion pair, 1e9 eV/GeV

Definition at line 55 of file PhysicalConstants.h.

Referenced by detinfo::DetectorPropertiesStandard::BirksCorrection(), calo::CalorimetryAlg::BirksCorrection(), ShowerRecoTools::ShowerNumElectronsEnergy::CalculateEnergy(), lar_pandora::LArPandoraInput::GetMips(), sim::LArG4Parameters::GeVToElectrons(), detinfo::DetectorPropertiesStandard::ModBoxCorrection(), calo::CalorimetryAlg::ModBoxCorrection(), and showerreco::ShowerRecoAlg::RecoOneShower().

constexpr double util::kINVALID_DOUBLE = std::numeric_limits<double>::max()

Definition at line 35 of file GeometryUtilities.h.

Referenced by util::GeometryUtilities::Get3DaxisN().

constexpr double util::kKilometerToMeter = 1. / kMeterToKilometer

Definition at line 66 of file PhysicalConstants.h.

constexpr double util::kMeterToCentimeter = 1.e2

1 m = 100 cm

Definition at line 63 of file PhysicalConstants.h.

constexpr double util::kMeterToKilometer = 1.e-3

1000 m = 1 km

Definition at line 65 of file PhysicalConstants.h.

constexpr double util::kMeVToeV = 1. / keVToMeV

Definition at line 69 of file PhysicalConstants.h.

constexpr double util::kModBoxA = 0.930

Modified Box Alpha.

Definition at line 49 of file PhysicalConstants.h.

Referenced by sim::LArG4Parameters::LArG4Parameters().

constexpr double util::kModBoxB = 0.212

Modified Box Beta in g/(MeV cm²)*kV/cm.

Definition at line 50 of file PhysicalConstants.h.

Referenced by calo::CalorimetryAlg::CalorimetryAlg(), and sim::LArG4Parameters::LArG4Parameters().

constexpr double util::kRecombA = 0.800

A constant.

Definition at line 35 of file PhysicalConstants.h.

Referenced by detinfo::DetectorPropertiesStandard::BirksCorrection(), and sim::LArG4Parameters::LArG4Parameters().

constexpr double util::kRecombk = 0.0486

k constant, in g/(MeV cm²)*kV/cm.

Definition at line 36 of file PhysicalConstants.h.

Referenced by detinfo::DetectorPropertiesStandard::BirksCorrection(), calo::CalorimetryAlg::CalorimetryAlg(), and sim::LArG4Parameters::LArG4Parameters().

constexpr double util::quietCompiler

Initial value:

=

kBogusD * kBogusI * kBogusF * kRecombA * kRecombk * kGeVToElectrons

util::kBogusI

constexpr int kBogusI

obviously bogus integer value

Definition: PhysicalConstants.h:72

util::kGeVToElectrons

constexpr double kGeVToElectrons

23.6eV per ion pair, 1e9 eV/GeV

Definition: PhysicalConstants.h:55

util::kBogusF

constexpr float kBogusF

obviously bogus float value

Definition: PhysicalConstants.h:73

util::kRecombk

constexpr double kRecombk

Definition: PhysicalConstants.h:36

util::kBogusD

constexpr double kBogusD

obviously bogus double value

Definition: PhysicalConstants.h:71

util::kRecombA

constexpr double kRecombA

A constant.

Definition: PhysicalConstants.h:35

Definition at line 75 of file PhysicalConstants.h.

constexpr RangeForWrapperTag util::range_for

Constant to be used with operator|(Range&&, details::RangeForWrapperTag).

Definition at line 450 of file RangeForWrapper.h.

Referenced by associated_groups(), and associated_groups_with_left().

const float util::SQRT_TWO_PI = 2.506628

Definition at line 18 of file GaussianEliminationAlg.h.

Namespaces

Classes

Typedefs

Functions

Variables

Detailed Description

Typedef Documentation

Function Documentation

Requirements

Requirements

Requirements

Why bother?

Variable Documentation