This commit should reduce the amount of iterations in concatenate to N
where N is the sum of the lengths of the input lists.
The previous implementation iterated from the beginning in each
concatenation because of `last-pair`.
This implementation is significantly faster in this extreme case:
(concatenate! `(,(iota 50000) ,@(map list (iota 500))))
>> Previous implementation:
real 0m0.671s
user 0m0.658s
sys 0m0.013s
>> This implementation:
real 0m0.175s
user 0m0.174s
sys 0m0.001s
The tests is done using `time`, which is not reliable at all, but using
`(trace last-pair)` shows accurately what happens with the iterations.
lib/srfi/231/base.scm:
1. Add argument checks for index-first, index-last, index-rotate, and index-swap.
lib/srfi/231/test.sld:
1. 1,$s/179/231/g (in vim notation)
2. Add tests for index-first, index-last, index-rotate, and index-swap.
3. Remove trailing spaces.
Makefile:
When installing chibi, make installation directory and install files for SRFI 231.
When uninstalling chibi, remove installation directory and files for SRFI 231.
lib/srfi/231/transforms.scm:
Move definition of vector-iota from here to ...
lib/srfi/231/base.scm:
Here.
lib/srfi/231/base.sld:
Export vector-iota to use in both base.scm and transforms.scm.
SRFI-144 requires that (flmin) returns +inf.0 and that
(flmax) returns -inf.0, so these procedures can't really
be aliases to the Chibi implementation of R7RS max and min.
This restores third-party (ab)users of the Chibi macro system such
as in https://gist.github.com/baguette/2632464, while allowing us
to break those uses in more interesting ways.
It also keeps the core slightly smaller (both in C and Scheme)
and speeds up the macro expansion process.
