Miou_vectorSourceVectors (aka resizable arrays, growing arrays, dynamic arrays, etc.)
This module implements arrays that automatically expand as necessary. Its implementation uses a traditional array and replaces it with a larger array when needed (and elements are copied from the old array to the new one). The current implementation doubles the capacity when growing the array (and shrinks it whenever the number of elements comes to one fourth of the capacity).
The unused part of the internal array is filled with a dummy value, which is user-provided at creation time (and referred to below as ``the dummy value''). Consequently, vectors do not retain pointers to values that are not used anymore after a shrinking.
Vectors provide an efficient implementation of stacks, with a better locality of reference than list-based implementations (such as standard library Stack). A stack interface is provided, similar to that of Stack (though push have arguments in the other way round). Inserting n elements with push has overall complexity O(n) i.e. each insertion has amortized constant time complexity.
The polymorphic type of vectors. This is a mutable data type.
Arraycreate returns a fresh vector of length 0. All the elements of this new vector are initially physically equal to dummy (in the sense of the == predicate). When capacity is omitted, it defaults to 0.
make dummy n x returns a fresh vector of length n with all elements initialized with x. If dummy is omitted, x is also used as a dummy value for this vector.
init n f returns a fresh vector of length n, with element number i initialized to the result of f i. In other terms, init n f tabulates the results of f applied to the integers 0 to n-1.
Raise Invalid_argument if n < 0 or n > Sys.max_array_length.
resize a n sets the length of vector a to n.
The elements that are no longer part of the vector, if any, are internally replaced by the dummy value of vector a, so that they can be garbage collected when possible.
Raise Invalid_argument if n < 0 or n > Sys.max_array_length.
Discard all elements from a vector. This is equivalent to setting the size to 0 with resize.
Return the length (number of elements) of the given vector. Note: the number of memory words occupied by the vector can be larger.
get a n returns the element number n of vector a. The first element has number 0. The last element has number length a - 1.
Raise Invalid_argument "Vector.get" if n is outside the range 0 to length a - 1.
set a n x modifies vector a in place, replacing element number n with x.
Raise Invalid_argument "Vector.set" if n is outside the range 0 to length a - 1.
sub a start len returns a fresh vector of length len, containing the elements number start to start + len - 1 of vector a.
fill a ofs len x modifies the vector a in place, storing x in elements number ofs to ofs + len - 1.
Raise Invalid_argument "Vector.fill" if ofs and len do not designate a valid subvector of a.
blit a1 o1 a2 o2 len copies len elements from vector a1, starting at element number o1, to vector a2, starting at element number o2. It works correctly even if a1 and a2 are the same vector, and the source and destination chunks overlap.
Raise Invalid_argument "Vector.blit" if o1 and len do not designate a valid subvector of a1, or if o2 and len do not designate a valid subvector of a2.
append a1 a2 returns a fresh vector containing the concatenation of the elements of a1 and a2.
It works correctly even if a1 and a2 are the same vector.
merge_right a1 a2 moves all elements of a2 to the end of a1. Empties a2. Assumes a1 and a2 to be disjoint.
map f a applies function f to all the elements of a, and builds a fresh vector with the results returned by f.
Note: the dummy value of the returned vector is obtained by applying f to the dummy value of a. If this is not what you want, first create a new vector and then fill it with the value f (get a 0), f (get a 1), etc.
Same as map, but the function is applied to the index of the element as first argument, and the element itself as second argument.
Note: the dummy value of the returned vector is obtained by applying f 0 to the dummy value of a.
copy a returns a copy of a, that is, a fresh vector containing the same elements as a.
fold_left f x a computes f (... (f (f x (get a 0)) (get a 1)) ...) (get a (n-1)), where n is the length of the vector a.
fold_right f a x computes f (get a 0) (f (get a 1) ( ... (f (get a (n-1)) x) ...)), where n is the length of the vector a.
iter f a applies function f in turn to all the elements of a. It is equivalent to f (get a 0); f (get a 1); ...; f (get a (length a - 1)).
Same as iter, but the function is applied to the index of the element as first argument, and the element itself as second argument.
Contrary to standard library's Stack, module Vector uses less space (between N and 2N words, instead of 3N) and has better data locality.
push a x appends x at the end of vector a, i.e., increases the size of a by one and stores x at the rightmost position.
Note: the order of the arguments is not that of Stack.push.
pop a removes and returns the rightmost element in vector a, or raises Empty if the stack is empty.
top a returns the rightmost element in vector a, or raises Empty if the vector is empty.