FINQ

Lightweight conveyor data processing python framework, which allows to quickly write long queries without a fear that it'll become unreadable, because FINQ as opposed to standard library allows you to write each logical part of query at next line without tearing it and expanding logical block over whole function

Start

To start conveyor processing of your iterable you need to wrap it with FINQ() object which then allows you to call Basic methods

Basic methods

Method	IsTerminal	Description
concat(b:Iterable[T])	-	Concatenates two sequences, creating sequence that contains items of the first iterable then of second iterable.
map(*f:T -> T2)	-	Applies given function to every element of sequence.
zip(b:Iterable[T])	-	Pairs corresponding elements of two sequences in pairs.
flat_map(f:T -> Collection[T2] = Identity)	-	Applies given function to every element to get collection, then glues these collections.
flatten(f:T -> Collection[T] = Identity)	-	Applies given function to every element to get collection, then glues these collections. Repeats until all elements are non iterable.
filter(f:T -> bool)	-	Removes elements that doesn't satisfy predicate from sequence.
distinct(f:T -> T2)	-	Skips elements which f(element) repeated.
sort(f:T -> int)	-	Sorts sequence elements by key given by f.
skip(count:int)	-	Skips count elements from sequence.
take(count:int)	-	Limits sequence by count elements, dropping others.
cartesian_product(b:Iterable[T1], mapping:T×T1 -> T2)	-	Returns Cartesian product of two sequences after application of mapping if specified.
cartesian_power(pow:int, mapping:T^pow -> T2)	-	Returns Cartesian power of sequence after application of mapping if specified.
pairs()	-	Returns Cartesian square of sequence. Equivalent to Cartesian square with Identity mapping.
enumerate(start=0)	-	Maps sequence elements to pair which first value is index in sequence starting by start.
peek(f:T -> ())	-	Applies function to each element in sequence leaving sequence unchanged.
group_by(mapping:T -> T2 = Identity)	-	Splits sequence into sequence of lists of elements which f(x) is the same.
random(precentage:float)	-	Takes roughly percentage*100% of random elements of sequence.
sort_randomly()	-	Shuffles sequence.
join_str(delimiter:str)	+	Joins sequence by delimiter.
join(seq:Iterable[T2], cond: :T×T2 -> bool, aggr:T×T2 -> T3)	+	Joins two sequences. Two values are aggregated if condition is true.
for_each(f:T -> () = Consumer)	+	Calls f for every element of a sequence. Equivalent to: for e in collection: f(e).
all(f:T -> bool = IdentityTrue)	+	Checks if all elements in sequence satisfy predicate.
any(f:T -> bool = IdentityTrue)	+	Checks if there exist element in sequence that satisfies predicate.
none(f:T -> bool = IdentityTrue)	+	Checks if there no element in sequence that satisfies predicate.
first()	+	Takes first element of sequence.
to_list()	+	Creates default python-list containing all sequence elements.
to_set()	+	Creates default python-set containing all distinct sequence elements.
to_counter()	+	Creates Counter containing all sequence elements.
to_dict(key:T -> TKey = First, value:T -> TValue = Second)	+	Creates default python-dict containing mapping (key(x), value(x)) for every x in sequence.
count()	+	Returns count of elements in sequence.
min()	+	Finds minimal element in sequence.
max()	+	Finds maximal element in sequence.
sum()	+	Sums all elements of sequence. Works only for summable types.
max_diff()	+	Counts maximal difference between elements. Equal to difference between max and min for sequence.
reduce(f:T×T -> T, /, first:T)	+	Applies function to first two elements, then to result and next element until elements end. Allows to specify first element.
fold(m:T -> T2, g:T2×T2 -> T2)	+	Applies mapper to each element, then aggregates pairs of T2 into single T2 until elements end. Equivalent to finq.map(mapper).reduce(aggregator)

Constant functions

These functions aren't intended to be called manually. Instead you have to pass them as an arguments to FINQ methods as mappings, reducers, predicates. Ordered Collection here is any collection which provides __get_item__ based on index (Tuple, List)

Method		Returns
Identity	T -> T	Given argument
Consumer	T -> None	None
IdentityTrue	T -> bool	True for any argument
IdentityFalse	T -> bool	False for any value
Sum	T×T -> T	Sum of two given values
PairSum	T² -> T	Sum of first two values of Ordered Collection
First	Tⁿ -> T	First value of Ordered Collection
Second	Tⁿ -> T	Second value of Ordered Collection
Multiply	T×T -> T	Product of two given value
Square	T -> T	Square of given value
OneArgRandom	T -> float	Random value independent of given value
TupleSum	Tⁿ -> T	Sum of given Ordered Collection
PairWith	Fⁿ -> (T -> T²)	Function that, when applied to value e returns e, f1(f2(...fn(e)...))
RPairWith	Fⁿ -> (T -> T²)	Function that, when applied to value e returns f1(f2(...fn(e)...)), e
Compose	Fⁿ -> F	Function that, when applied to value e returns f1(f2(...fn(e)...))