diff options
author | Kartik K. Agaram <vc@akkartik.com> | 2016-02-11 16:13:10 -0800 |
---|---|---|
committer | Kartik K. Agaram <vc@akkartik.com> | 2016-02-11 17:29:50 -0800 |
commit | 5d67fac7966b6f05611d014420eb4971b8016c31 (patch) | |
tree | 4ad84631596f09a994a2154b77e80118e7189b37 /070text.mu | |
parent | e4b03c6f574ade8c3fe4df18b958da981ac58acb (diff) | |
download | mu-5d67fac7966b6f05611d014420eb4971b8016c31.tar.gz |
2646 - redo static dispatch algorithm
The old approach of ad hoc boosts and penalties based on various features was repeatedly running into exceptions and bugs. New organization: multiple tiered scores interleaved with tie-breaks. The moment one tier yields one or more candidates, we stop scanning further tiers. Just break ties and return.
Diffstat (limited to '070text.mu')
-rw-r--r-- | 070text.mu | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/070text.mu b/070text.mu index 56291b75..a37e8a6e 100644 --- a/070text.mu +++ b/070text.mu @@ -297,7 +297,8 @@ recipe to-text n:number -> result:address:shared:array:character [ # add sign { break-unless negate-result:boolean - tmp <- append tmp, 45 # '-' + minus:character <- copy 45/- + tmp <- append tmp, minus } # reverse buffer into text result len:number <- get *tmp, length:offset |