branch: externals/trie
commit a8bc50f997973eb0f0679ec197ae293ecf05bcf8
Author: Toby Cubitt <toby-predict...@dr-qubit.org>
Commit: tsc25 <toby-predict...@dr-qubit.org>
Minor code tidying
---
trie.el | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/trie.el b/trie.el
index 0cd0760..f2d5aa1 100644
--- a/trie.el
+++ b/trie.el
@@ -1202,12 +1202,12 @@ element stored in the trie.)"
;; For queries ranked in anything other than lexical order, we use a partial
;; heap-sort to find the k=MAXNUM highest ranked matches among the n possibile
;; matches. This has worst-case time complexity O(n log k), and is both simple
-;; and elegant. An optimal algorithm (e.g. partial quick-sort where the
-;; irrelevant partition is discarded at each step) would have complexity O(n +
-;; k log k), but is probably not worth the extra coding effort, and would have
-;; worse space complexity unless coded to work "in-place", which would be
-;; highly non-trivial. (I haven't done any benchmarking, though, so feel free
-;; to do so and let me know the results!)
+;; and elegant. An optimal algorithm (e.g. partial quick-sort discarding the
+;; irrelevant partition at each step) would have complexity O(n + k log k),
+;; but is probably not worth the extra coding effort, and would have worse
+;; space complexity unless coded to work "in-place", which would be highly
+;; non-trivial. (I haven't done any benchmarking, though, so feel free to do
+;; so and let me know the results!)
(defmacro trie--construct-accumulator (maxnum filter resultfun)
;; Does what it says on the tin! | sed -e 's/on/in/' -e 's/tin/macro name/'
