A sequence (abbreviated to seq) is a group of adjacent
parenthesized elements. For example,
...is a seq of 3 elements--a, b, and c.
Sequences are data structures that merge the properties of
both lists and tuples with the exception that a seq, like
a tuple, cannot be empty. Therefore, an "empty" seq
is considered a special case scenario that must be handled separately in
#define SEQ (x)(y)(z)
#define REVERSE(s, state, elem) (elem) state
// append to head ^
BOOST_PP_SEQ_FOLD_LEFT(REVERSE, BOOST_PP_EMPTY, SEQ)()
// #1 #2
// 1) placeholder for "empty" seq
// 2) remove placeholder
#define SEQ_B (1)(2)(3)
#define INC(s, state, elem) state (BOOST_PP_INC(elem))
// append to tail ^
BOOST_PP_SEQ_FOLD_RIGHT(INC, BOOST_PP_SEQ_NIL, SEQ)
// special placeholder that will be "eaten"
// by appending to the tail
Sequences are extremely efficient. Element access speed
approaches random access--even with seqs of up to 256
elements. This is because element access (among other things) is
implemented iteratively rather than recursively. Therefore, elements
can be accessed at extremely high indices even on preprocessors with low
maximum expansion depths.
Elements of a seq can be extracted with BOOST_PP_SEQ_ELEM.