Commit Graph

4311 Commits

Author SHA1 Message Date
Shay Banon f17ad829ac remove snappy support
relates to #2459
2012-12-03 12:30:13 +01:00
Shay Banon 677e6ce4ef Deprecate Shared Gateway
closes #2458
2012-12-03 11:44:17 +01:00
Shay Banon a2a8553faf Indexing Slow Log
closes #2457
2012-12-03 10:21:59 +01:00
Shay Banon b10cec1908 Upgrade to Netty 3.5.11
closes #2456
2012-12-02 22:29:49 +01:00
Shay Banon b4f85ee422 no need to check for log levels
we already do that when we log, and those are set to TRACE most times for slow log (since logging is based on thresholds)
2012-12-02 22:03:51 +01:00
Shay Banon a274d9386f Add types and stats to search slow log
closes #2455
2012-12-02 22:01:17 +01:00
Igor Motov 6021515567 The relevancy score in explanation should match the actual score in custom_filters_query
Fixes #2441
2012-11-27 10:13:16 -08:00
Shay Banon 69ef822da6 cleanup docsets
- remove the DocSet abstraction, and use Bits where we can by getting it from DocIdSet
- better handling of acceptDocs, though still need to properly apply them when caching is involved
2012-11-27 10:04:21 -08:00
Igor Motov fb9143aac1 fix sporadically disappearing fields during concurrent dynamic mapping updates 2012-11-24 14:02:58 +01:00
Simon Willnauer 4ab78bc537 Add basic javadocs for o.e.cluster.rounting package and related classes 2012-11-23 15:14:30 +01:00
Simon Willnauer 32a0772821 #2436 expose KeepWordTokenFilter by default 2012-11-23 10:11:30 +01:00
Igor Motov 65a43d3ad4 Fix handling of stop word _lang_ notation
Fixes #2412
2012-11-23 09:54:02 +01:00
Shay Banon 2094207bf1 add completed count to thread pools 2012-11-22 15:55:25 +01:00
Shay Banon e1679b89bb fix failed test that were using the wrong form match query 2012-11-22 15:14:02 +01:00
Shay Banon 192cf5298a fix failed test that were using the wrong form match query 2012-11-22 14:44:03 +01:00
Shay Banon f4d6d8139d Match query should fail when trying to provide several fields in its simplified form
fixes #2432
2012-11-22 10:23:48 +01:00
Chris Male 2541847945 Added control over Query used by MatchQuery with there are zero terms after analysis 2012-11-22 22:13:29 +13:00
Shay Banon 9a90c1c3b5 conservative timeouts on internal recovery actions
safe guards against cases where intenral recovery actions take too long (possibly due to a bug)
2012-11-22 00:31:57 +01:00
Shay Banon f5a3261e15 only log that we delete unused shard if it exists 2012-11-21 20:45:31 +01:00
Shay Banon d9b78000b1 Setting logger levels using cluster update settings does not work
fixes #2428
2012-11-21 13:44:54 +01:00
Chris Male 9e2469e04f Add per-field Similarity support 2012-11-21 12:44:59 +13:00
Shay Banon 4e8a9008b7 second phase at optimizing merging/parsing large new mappings
apply the new mappings only after the parsing/merging of a full doc/mapping is done
2012-11-19 17:40:13 +01:00
Shay Banon 303752d78a first phase at optimizing merging large mappings
bulk them the same level ones when traversing and introduce them
2012-11-19 17:13:45 +01:00
Shay Banon 6e597ffccb allow to associate a payload with bulk requests 2012-11-19 16:16:35 +01:00
David Pilato 83257c8af8 Add constructor IndexRequest(String index, String type) and fix javadoc 2012-11-19 13:52:55 +01:00
David Pilato b2597b5316 Add a toString() method to MultiSearchResponse 2012-11-19 13:50:58 +01:00
Simon Willnauer 840eaf983d Add JavaDocs for Codecs, PostingsFormat and related services/modules 2012-11-19 10:25:26 +01:00
Shay Banon c09ee82ef5 keep the uidField around so we don't have to look it up 2012-11-15 12:52:58 +01:00
Shay Banon e2e25ffea3 uid to use bloom filter posting by default 2012-11-15 11:57:48 +01:00
Martijn van Groningen 3577d826f2 Removed old file. 2012-11-15 11:54:51 +01:00
Martijn van Groningen be70722de7 Renamed pulsing40 and Lucene40 postings format providers to pulsing and default respectively for more consistent naming in settings. 2012-11-15 09:54:00 +01:00
Martijn van Groningen 20c6085852 changed test method names. 2012-11-15 09:40:24 +01:00
Martijn van Groningen e80f74584b Added licence header. 2012-11-15 00:18:49 +01:00
Martijn van Groningen fd5bd102aa lucene 4: Exposed Lucene's codec api
This feature adds the option to configure a `PostingsFormat` and assign it to a field in the mapping. This feature is very expert and in almost all cases Elasticsearch's defaults will suite your needs.

## Configuring a postingsformat per field

There're several default postings formats configured by default which can be used in your mapping:
a* `direct` - A codec that wraps the default postings format during write time, but loads the terms and postinglists into memory directly in memory during read time as raw arrays. This postings format is exceptional memory intensive, but can give a substantial increase in search performance.
* `memory` - A codec that loads and stores terms and postinglists in memory using a FST. Acts like a cached postingslist.
* `bloom_default` - Maintains a bloom filter for the indexed terms, which is stored to disk and builds on top of the `default` postings format. This postings format is useful for low document frequency terms and offers a fail fast for seeks to terms that don't exist.
* `bloom_pulsing` - Similar to the `bloom_default` postings format, but builds on top of the `pulsing` postings format.
* `default` - The default postings format. The default if none is specified.

On all fields it possible to configure a `postings_format` attribute. Example mapping:
```
{
  "person" : {
     "properties" : {
         "second_person_id" : {"type" : "string", "postings_format" : "pulsing"}
     }
  }
}
```

## Configuring a custom postingsformat
It is possible the instantiate custom postingsformats. This can be specified via the index settings.
```
{
   "codec" : {
      "postings_format" : {
         "my_format" : {
            "type" : "pulsing40"
            "freq_cut_off" : "5"
         }
      }
   }
}
```
In the above example the `freq_cut_off` is set the 5 (defaults to 1). This tells the pulsing postings format to inline the postinglist of terms with a document frequency lower or equal to 5 in the term dictionary.

Closes #2411
2012-11-14 23:54:29 +01:00
Igor Motov 120560bd0a Using non-mapped fields in prefix queries shouldn't cause NullPointerException
Fixes #2408
2012-11-14 18:34:54 +01:00
Igor Motov f47d62cc30 Date fields shouldn't be returned as longs by Get API 2012-11-13 21:36:28 +01:00
Igor Motov d1281d283b Add `index.routing.allocation.require....` and `cluster.routing.allocation.require....` settings
Fixes #2404
2012-11-13 19:29:20 +01:00
Igor Motov ea2732a967 lucene 4: field visitors shouldn't return fields that were not present in the visited document 2012-11-13 07:35:54 -05:00
Shay Banon 258244ef37 Deriving the REST status code from a failure can, very rarely, cause an infinite loop
fixes #2402
2012-11-12 17:09:34 +01:00
Nicholas Tung 46e1886975 Allow both .yml and .yaml as valid YAML configuration file extensions 2012-11-12 14:06:04 +01:00
Igor Motov 3ff54c0b5c Add logging for environment paths on startup 2012-11-12 14:04:13 +01:00
Shay Banon 6ede42bc37 merge lucene_4 2012-11-12 13:50:14 +01:00
Martijn van Groningen 978c95649e lucene 4: Fixed SimpleQueryTests 2012-11-12 13:44:42 +01:00
Martijn van Groningen 05746adeb2 lucene 4: Set number of replicas to 0. Makes the test run faster. 2012-11-12 13:44:42 +01:00
Martijn van Groningen e2c33ed659 lucene 4: Fixed BitsetExecutionChildQuerySearchTests class. 2012-11-12 13:44:42 +01:00
Shay Banon 9a79fb40bf lucene 4: sort values on hit are Text, not BytesRef 2012-11-12 13:44:42 +01:00
Igor Motov c46228254d lucene 4: fix TTL 2012-11-12 13:44:42 +01:00
Igor Motov c2f3eab7d3 lucene 4: fix sorting 2012-11-12 13:44:42 +01:00
Shay Banon 2b58c2dfff lucene 4: optimize read/write BytesRef handling 2012-11-12 13:44:42 +01:00
Igor Motov c8cf72d657 lucene 4: fix handling of deleted docs in TermFilter 2012-11-12 13:44:42 +01:00