Commit Graph

4628 Commits

Author SHA1 Message Date
Simon Willnauer b3356d9f8d remove dead code 2013-03-31 10:19:17 +02:00
Simon Willnauer 2a09342405 remove Bytes.java in favor of BytesRef / ArrayUtils 2013-03-31 08:54:39 +02:00
Simon Willnauer e864d5785e optimize matcher reset to not create unnecessary string objects 2013-03-30 17:35:23 +01:00
Simon Willnauer fefa8da2ea remove StringValues in favor of BytesValues 2013-03-30 17:35:23 +01:00
Simon Willnauer dff2a9279c clean-up double values 2013-03-30 17:35:23 +01:00
Simon Willnauer d5c271acf5 clean-up long values 2013-03-30 17:35:22 +01:00
Simon Willnauer 5aedf74fb0 Remove getValues from numeric and string field data & clean up geo field data 2013-03-30 17:35:22 +01:00
Simon Willnauer 7f81469137 Refactor BytesValues to be reused as the interface for HashedBytesValues and remove HashBytesValues. 2013-03-30 17:35:22 +01:00
Simon Willnauer 129f02623b Added FST based FieldData implementation holding all data in a per segment FST.
This commit factors our a common API for BytesValues based impl to shared code and reduce code duplication.
2013-03-30 17:35:22 +01:00
Shay Banon 72c76c2799 fail on malformed sort 2013-03-30 13:58:39 +01:00
Shay Banon 6a1cb8f61b {sort: "field"} throws misleading errors
fixes #2835
2013-03-30 13:46:53 +01:00
Martijn van Groningen 2e93329e23 If match then go to next doc 2013-03-29 16:57:42 +01:00
Martijn van Groningen a89dde8bac Fixed `bool` filter bugs:
* In the case only should clauses were specified with specific type of filters, the first clause determined which documents matched.
 * In some cases the minimum at least 1 should clause should match behaviour was broken.
2013-03-29 16:48:36 +01:00
Igor Motov b657bdfa1a Optimize aliases processing
Closes# 2832
2013-03-29 10:44:45 -04:00
Martijn van Groningen 941aa17a43 Added sort mode to geo distance sorting. Closes #1846 2013-03-28 17:04:42 +01:00
Igor Motov 9bc50ea609 Fix LeastUsedDistributor and ensure random distribution for multiple non-fs directories
If we cannot determine available space the fallback scenario is now to use random distribution instead of always using the last directory.

Fixes #2820
2013-03-28 11:08:54 -04:00
Shay Banon 1fc37e5954 Segments API: Add version & compound for each segment
closes #2823
2013-03-28 15:34:38 +01:00
Shay Banon 473473e867 remove the field settings for query parser cache, not really relevant 2013-03-27 20:39:36 +01:00
Shay Banon c18c609af1 Date math in query_string caches now()
fixes #2808
2013-03-27 20:32:38 +01:00
Igor Motov 5bb75f9da3 Move applying alias filter to ContextSearch#preProcess() 2013-03-27 09:23:54 -04:00
Simon Willnauer 17f83f33bb Terminate early when no terms left in the suggest string.
Closes #2817
2013-03-26 17:44:34 +01:00
Igor Motov 9ae421a8b2 Fix filtering aliases with non-empty sort options
Fixes #2816
2013-03-26 07:23:44 -04:00
Shay Banon d35a3b03c8 Warmers: Have an explicit warmer thread pool
add 1 in case there is 1 core...
closes #2815
2013-03-25 23:34:52 +01:00
Simon Willnauer aa97c031f2 Don't reset tokenstream before passing to the MemoryIndex, otherwise some tokenizer might swallow tokens.
Closes #2814
2013-03-25 22:46:11 +01:00
Shay Banon b7106622d8 Warmers: Have an explicit warmer thread pool
Have an explicit threadpool warmer that is dedicated to execute warmers. Currently, it uses the search threadpool, which does not work well since the number of concurrent searches should be separate from the number of concurrent warmers allows, also the characteristics of the search pool (for example, bounded queue_size) might not fit well with how warmers should be executed (they should not be "rejected").

closes #2815
2013-03-25 16:46:37 +01:00
Shay Banon 0e815ce11c add 0.20.7 2013-03-25 12:33:55 +01:00
adavis 6a93fbcf07 Adding parsing for zero terms query for multi match
Tests for multi-match zero_terms_query and making references to the ZeroTermsQuery enum consistent to others used in MultiMatchQueryBuilder
2013-03-23 08:59:39 +01:00
adavis 3f83904680 Fixes java6_u31 compile error w.r.t. type inference 2013-03-22 16:46:42 -07:00
Simon Willnauer 560d2c094e Fix issue where entire shards are lost due to too many open files exceptions and a but in Lucene's
IndexWriter / DirectoryReader where an existsing index was not detected and then deleted due to a wrong
creation mode. See LUCENE-4870

Closes #2812
2013-03-22 17:18:55 +01:00
Florian Schilling 1a67793a4b Added Script test for geo distance tests and modified GeoUtils.normalizePoint() 2013-03-22 13:34:18 +01:00
Simon Willnauer 075779a397 Call onMissing if doc has no value in the field.
Closes #2807
2013-03-21 22:45:17 +01:00
Simon Willnauer 064d272916 Respect offset and length when iterating over BytesRef in Uid. The length is starting at offset
Closes #2806
2013-03-21 19:29:05 +01:00
Simon Willnauer 5f05c2106f Use more efficient StemmerOverrideFilter from Lucene trunk
Closes #2800
2013-03-21 07:58:51 +01:00
Shay Banon ea698add72 move to 0.90.0.RC2 snap 2013-03-20 19:06:30 +01:00
Shay Banon a2f14b68e8 release 0.90.0.RC1 2013-03-20 19:05:08 +01:00
Florian Schilling f08d458545 # GeoShape Precision
The `geo_shape` precision could be only set via `tree_levels` so far. A new option `precision` now allows to set the levels of the underlying tree structures to be set by distances like `50m`. The def

## Example
```json
curl -XPUT 'http://127.0.0.1:9200/myindex/' -d '{
  "mappings" : {
      "type1": {
          "dynamic": "false",
          "properties": {
              "location" : {
                  "type" : "geo_shape",
                  "geohash" : "true",
                  "store" : "yes",
                  "precision":"50m"
              }
          }
      }
  }
}'
```

## Changes
- GeoUtils defines the [WGS84](http://en.wikipedia.org/wiki/WGS84) reference ellipsoid of earth
- DistanceUnits refer to a more precise definition of earth circumference
- DistanceUnits for inch, yard and meter have been defined
- Set default levels in GeoShapeFieldMapper to 50m precision

Closes #2803
2013-03-20 14:52:47 +01:00
Simon Willnauer 4705eb2959 Lazily initialize the delegate in BloomFilteredPostingsFormat to prevent unnecessary loading if bloomfilter terminates early 2013-03-20 12:43:17 +01:00
Simon Willnauer 747ce36915 Specialise the default codec to reuse Lucene41 files in the common case.
Closes #2799
2013-03-20 12:43:17 +01:00
Shay Banon 54e7e309a5 better comment... 2013-03-19 14:36:13 +01:00
Shay Banon d5beea4bba if multicast socket closes, try and restart it
also, throttle on socket failures, so it won't spin out of control...
relates to #2783
2013-03-19 11:20:47 +01:00
Shay Banon f4a212420b multicastSocket should be volatile as well... 2013-03-19 10:23:39 +01:00
Shay Banon c92207f483 broadcast API to by default ignore missing / illegal shard state
this happens for example because we list assigned shards, and they might not have been allocated on the relevant node yet, no need to list those as actual failures in some APIs
2013-03-19 10:22:43 +01:00
Shay Banon bea7bdde4c upgrade to guava 14.0.1 2013-03-18 23:10:14 +01:00
Shay Banon aca713d68e tar.gz distro by mistake include a windows lib 2013-03-18 22:46:04 +01:00
Shay Banon 566d1d13f7 fix javadoc 2013-03-18 22:04:31 +01:00
Clinton Gormley 2123ab591c Correct filter strategy opt: random_access_random to random_access_always 2013-03-18 20:17:26 +01:00
Shay Banon 7d9cef904b Field Data: optimize long type to use narrowest possible type automatically
closes #2795
2013-03-18 12:37:15 +01:00
Shay Banon 82072fc47f make ES compile with java 8
- that isAnnotationPresent bug is known, and probably will be fixed in later versions, but it costs us nothing to not use it now
- some tests fail, mainly due to consistent ordering expected from Map (within versions) which does not seem to be preserved, need to fix those tests to be agnostic to it
2013-03-18 01:33:09 +01:00
Shay Banon e347a626da use ImmutableList.Builder instead of ArrayList 2013-03-17 21:55:07 +01:00
Shay Banon 2ed6ea25cc fix logging message to include the index
also add the list of current indices
2013-03-16 22:58:45 +01:00