always multiply query score to function score. For script score
functions, this means that boost_mode has to be set to `plain` if
'function_score' should behave like 'custom_score'
This commit fixes inconsistencies in `function_score` and `filters_function_score`
using scripts, see issue #3464
The method 'ScoreFunction.factor(docId)' is removed completely, since the name
suggests that this method actually computes a factor which was not the case.
Multiplying the computed score is now handled by 'FiltersFunctionScoreQuery'
and 'FunctionScoreQuery' and not implicitely performed in
'ScoreFunction.factor(docId, subQueryScore)' as was the case for 'BoostScoreFunction'
and 'DecayScoreFunctions'.
This commit also fixes the explain function for FiltersFunctionScoreQuery. Here,
the influence of the maxBoost was never printed. Furthermore, the queryBoost was
printed as beeing multiplied to the filter score.
Closes#3464
There is no need to write the pidfile in the bin/elasticsearchshell script
as this happens already in the java code.
Also cleaning up the bin/elasticsearch shell script a bit (no need to return
an error code when exec is called, as this forks and exits the shell script
immediately).
Closes#3529Closes#1745
This looks like a copy/paste issue where onCached was being called
rather than onRemoval. This should fix the ID cache stats not being
correct after a call to /_cache/clear?id_cache=true
During segment merges FieldConsumers for all fields from the source
segments are created. Yet, if all documents that have a value in a
certain field are deleted / pruned during the merge the FieldConsumer
will not be called at all causing the internally used FST Builder to
return `null` from it's `build()` method. This means we can't store
it and run potentially into a NPE. This commit adds handling for
this corner case both during merge and during suggest phase since
we also don't have a Lookup instance for this segments field.
Closes#3555
We use number of processors to choose default thread pool sizes, and number of workers in networking (for HTTP and transport). Bound it to max at 32 by default as a safety measure to create too many threads.
This relates to #3478, where we set the default to 24, but 32 is probably a better default.
closes#3545
Terminate phrase searches early if max phrase window is exceeded in
FastVectorHighlighter to prevent very long running phrase
extraction if phrase terms are high frequent. See LUCENE-5182
Closes#3543
The blocking thread pool type is not recommended to be used, since it will end up blocking the IO thread most times when executing, which is not recommended (other operations will then stall as well).
closes#3531
If a static cached version of MatchAllDocsQuery is used through for
instanst the `query_string` together with a boost like `*:*^2.0` the
globally used version is modified since queries are not immutable and it's
boost variable can change at any time. Holding on to queries that are modifiable
is risky and should not be done in a global scope.
This commit also adds tests for constant scores from `constant_score` query.
Closes#3521
Unless the field is not mapped phrase suggester should return
empty results or skip candidate generation if a field in not in
the index rather than failing hard with an illegal argument exception.
Some shards might not have a value in a certain field.
Closes#3469
From #3469.
When running suggest on empty shards, it raises an error like:
```
"failures" : [ {
"status" : 400,
"reason" : "ElasticSearchIllegalArgumentException[generator field [title] doesn't exist]"
} ]
```
We should ignore empty shards.
Closes#3473.
Britta Weber
Alexander Reelsen
Shay Banon
Louis Gueye
Martijn van Groningen
Dan Everton
uboness
Boaz Leskes
Simon Willnauer
David Pilato
Benjamin Devèze
Luca Cavanna
Lee Hinman