Ankou76ers opened a new issue, #12210:
URL: https://github.com/apache/lucene/issues/12210
### Description
When calling [preceding
](https://github.com/apache/lucene/blob/0782535017c9e737350e96fb0f53457c7b8ecf03/lucene/analysis/opennlp/src/java/org/apache/lucene/analysis/opennlp/OpenNL
Ankou76ers commented on issue #12210:
URL: https://github.com/apache/lucene/issues/12210#issuecomment-1479267433
[OpenNLPSentenceBreakIterator.patch.txt](https://github.com/apache/lucene/files/11038808/OpenNLPSentenceBreakIterator.patch.txt)
--
This is an automated message from the Ap
rmuir commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479280860
> Hi @rmuir, first of all, I deeply appreciate the time you are taking to
help us on this issue, thank you for that.
When I said "'ll leave the BoostAttribute discussion for another time
alessandrobenedetti commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479292833
Hi @rmuir, we definitely don't want to ignore improvement recommendations,
rest assured.
Sorry if I am pedantic, I just want to understand why it shouldn't be used
for add
rmuir commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479301623
> Hi @rmuir, we definitely don't want to ignore improvement recommendations,
rest assured.
> Sorry if I am pedantic, I just want to understand why it shouldn't be used
for additional th
rmuir commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479306693
> Allowing negative numbers shouldn't be an issue, the vector similarity
score is 0<=x<=1.
You don't understand. Some day, someone may want to add such safety as a
check to the attr
romseygeek commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479323710
Would your worries be assuaged if we created a separate
`QueryBoostAttribute` class @rmuir? Then `QueryBuilder` can use that, and we
can add checks for negative boosts, javadocs that
rmuir commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479355672
> Would your worries be assuaged if we created a separate
`QueryBoostAttribute` class @rmuir? Then `QueryBuilder` can use that, and we
can add checks for negative boosts, javadocs that say
alessandrobenedetti commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479357302
Ok, let me rephrase my question then:
Let's assume:
- we don't care why that class was originally created and its JavaDocs
comments
- we don't intend to modify it
rmuir commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479361208
i explained it to you multiple times on this issue and i feel the javadoc
explanation is already good. Please read it!
--
This is an automated message from the Apache Git Service.
To res
rmuir commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479362609
i'm -1 for this PR. analyzers shouldnt be mixing with query boosts at all.
it is a mixing of concerns that we should avoid: please refactor to be
something other than an analyzer.
--
Th
alessandrobenedetti commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479386226
@rmuir bear in mind a veto should be motivated and I have seen honestly zero
practical motivation so far.
Given that, I definitely don't have time for sterile discussions a
mkhludnev commented on PR #12169:
URL: https://github.com/apache/lucene/pull/12169#issuecomment-1479396426
> i'm -1 for this PR. analyzers shouldnt be mixing with query boosts at all.
it is a mixing of concerns that we should avoid: please refactor to be
something other than an analyzer.
magibney commented on PR #12207:
URL: https://github.com/apache/lucene/pull/12207#issuecomment-1479836168
In fact, even for the `wikimedium10m` case, benefits are apparent for
broader matches. The improvements are modest, but demonstrable even for these
generic cases. And again, in general
david-sitsky commented on issue #12185:
URL: https://github.com/apache/lucene/issues/12185#issuecomment-1480390389
As an aside, in some standard benchmark tests I run with our product, I have
found the final optimisation of Lucene indexes after all the data has been
indexed took 36 seconds
15 matches
Mail list logo
