Precedence optimization fails when applied to spans which cover more than one token #221
Comments
ghost
assigned
thomaskrause
added a commit
to thomaskrause/ANNIS
that referenced
this issue
Sep 29, 2013
Precedence optimization fails when applied to spans which cover more than one token Take e.g. this query on pcc2 NP & NP & NP & #1 . #2 & #2 . #3 In ANNIS 2 this gave us 2 results, but since ANNIS 3 incorrectly applies the precedence optimization the query gets translated to NP & NP & NP & #1 . #2 & #2 . #3 & #1 . #3 and has only 1 match. The correct optimization would be NP & NP & NP & #1 . #2 & #2 . #3 & #1 .* #3 This commit adds proper test cases for this situation and gives a fix
Merged
|
The correct optimization also provides only one match in ANNIS 3.0.0: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Take e.g. this query on pcc2
In ANNIS 2 this gave us 2 results, but ANNIS 3 incorrectly applies the precedence optimization and the query gets translated to
which results in only 1 match. The correct optimization would be
The text was updated successfully, but these errors were encountered: