Enhanced ranking algorithm

Based on the patch by Matt Westcott (@mjwestcott).
But with a more conservative approach:
- Does not use linearly increasing penalties; It is agreed upon that we
  should prefer matching characters at the beginnings of the words, but
  it's not always clear that the relevance is inversely proportional to
  the distance from the beginning.
- The approach here is more conservative in that the bonus is never
  large enough to override the matchlen, so it can be thought of as the
  first implicit tiebreak criterion.
- One may argue the change breaks the contract of --tiebreak, but the
  judgement depends on the definition of "tie".
This commit is contained in:
Junegunn Choi
2016-04-16 14:02:43 +09:00
parent 5f63a7b587
commit 2f6d23b91e
6 changed files with 214 additions and 94 deletions

View File

@@ -70,10 +70,10 @@ func TestExact(t *testing.T) {
clearPatternCache()
pattern := BuildPattern(true, true, CaseSmart, true,
[]Range{}, Delimiter{}, []rune("'abc"))
sidx, eidx := algo.ExactMatchNaive(
res := algo.ExactMatchNaive(
pattern.caseSensitive, pattern.forward, []rune("aabbcc abc"), pattern.termSets[0][0].text)
if sidx != 7 || eidx != 10 {
t.Errorf("%s / %d / %d", pattern.termSets, sidx, eidx)
if res.Start != 7 || res.End != 10 {
t.Errorf("%s / %d / %d", pattern.termSets, res.Start, res.End)
}
}
@@ -82,11 +82,11 @@ func TestEqual(t *testing.T) {
clearPatternCache()
pattern := BuildPattern(true, true, CaseSmart, true, []Range{}, Delimiter{}, []rune("^AbC$"))
match := func(str string, sidxExpected int, eidxExpected int) {
sidx, eidx := algo.EqualMatch(
match := func(str string, sidxExpected int32, eidxExpected int32) {
res := algo.EqualMatch(
pattern.caseSensitive, pattern.forward, []rune(str), pattern.termSets[0][0].text)
if sidx != sidxExpected || eidx != eidxExpected {
t.Errorf("%s / %d / %d", pattern.termSets, sidx, eidx)
if res.Start != sidxExpected || res.End != eidxExpected {
t.Errorf("%s / %d / %d", pattern.termSets, res.Start, res.End)
}
}
match("ABC", -1, -1)