about summary refs log tree commit diff
path: root/emacs/.emacs.d/wpc/sequence.el
diff options
context:
space:
mode:
authorWilliam Carroll <wpcarro@gmail.com>2020-01-30T16·00+0000
committerWilliam Carroll <wpcarro@gmail.com>2020-01-30T16·00+0000
commit578ed1ba98510058cf48f897a9bf4e3391684120 (patch)
tree5eadabd3d232151c75ac2424eaa8771ab25e927e /emacs/.emacs.d/wpc/sequence.el
parent3684adf23f847114e1beeaab162398628ba571aa (diff)
Move move .emacs.d out of configs/shared
Moving all of my Emacs-related files into their own directory at the root of
this repository.
Diffstat (limited to 'emacs/.emacs.d/wpc/sequence.el')
-rw-r--r--emacs/.emacs.d/wpc/sequence.el105
1 files changed, 105 insertions, 0 deletions
diff --git a/emacs/.emacs.d/wpc/sequence.el b/emacs/.emacs.d/wpc/sequence.el
new file mode 100644
index 000000000000..a5428ef04448
--- /dev/null
+++ b/emacs/.emacs.d/wpc/sequence.el
@@ -0,0 +1,105 @@
+;;; sequence.el --- Working with the "sequence" types -*- lexical-binding: t -*-
+;; Author: William Carroll <wpcarro@gmail.com>
+
+;;; Commentary:
+;; Elisp supports a typeclass none as "sequence" which covers the following
+;; types:
+;; - list: '(1 2 3 4 5)
+;; - vector: ["John" 27 :blue]
+;; - string: "To be or not to be..."
+
+;; TODO: Document the difference between a "reduce" and a "fold".  I.e. - reduce
+;; has an initial value whereas fold uses the first element in the sequence as
+;; the initial value.
+;;
+;; Note: This should be an approximation of Elixir's Enum protocol albeit
+;; without streams.
+;;
+;; Elisp has done a lot of this work already and these are mostly wrapper
+;; functions.
+;; See the following list for reference:
+;; - sequencep
+;; - elt
+;; - copy-sequence
+;; - reverse
+;; - nreverse
+;; - sort
+;; - seq-elt
+;; - seq-length
+;; - seqp
+;; - seq-drop
+;; - seq-take
+;; - seq-take-while
+;; - seq-drop-while
+;; - seq-do
+;; - seq-map
+;; - seq-mapn
+;; - seq-filter
+;; - seq-remove
+;; - seq-reduce
+;; - seq-some
+;; - seq-find
+;; - seq-every-p
+;; - seq-empty-p
+;; - seq-count
+;; - seq-sort
+;; - seq-contains
+;; - seq-position
+;; - seq-uniq
+;; - seq-subseq
+;; - seq-concatenate
+;; - seq-mapcat
+;; - seq-partition
+;; - seq-intersection
+;; - seq-difference
+;; - seq-group-by
+;; - seq-into
+;; - seq-min
+;; - seq-max
+;; - seq-doseq
+;; - seq-let
+
+;;; Code:
+
+;; Perhaps we can provide default implementations for `filter' and `map' derived
+;; from the `reduce' implementation.
+;; (defprotocol sequence
+;;   :functions (reduce))
+;; (definstance sequence list
+;;   :reduce #'list/reduce
+;;   :filter #'list/filter
+;;   :map    #'list/map)
+;; (definstance sequence vector
+;;   :reduce #'vector/reduce)
+;; (definstance sequence string
+;;   :reduce #'string)
+
+(defun sequence/classify (xs)
+  "Return the type of `XS'."
+  (cond
+   ((listp xs) 'list)
+   ((vectorp xs) 'vector)
+   ((stringp xs) 'string)))
+
+(defun sequence/reduce (acc f xs)
+  "Reduce of `XS' calling `F' on x and `ACC'."
+  (seq-reduce
+   (lambda (acc x)
+     (funcall f x acc))
+   xs
+   acc))
+
+;; Elixir also turned everything into a list for efficiecy reasons.
+
+(defun sequence/filter (p xs)
+  "Filter `XS' with predicate, `P'.
+Returns a list regardless of the type of `XS'."
+  (seq-filter p xs))
+
+(defun sequence/map (f xs)
+  "Maps `XS' calling `F' on each element.
+Returns a list regardless of the type of `XS'."
+  (seq-map f xs))
+
+(provide 'sequence)
+;;; sequence.el ends here