honeymoose
/
OpenSearch
mirror of https://github.com/honeymoose/OpenSearch.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Scripting: Replace advanced and native scripts with ScriptEngine docs (#24603) 

This commit documents how to write a `ScriptEngine` in order to use
expert internal apis, such as using Lucene directly to find index term
statistics. These documents prepare the way to remove both native
scripts and IndexLookup.

The example java code is actually compiled and tested under a new gradle
subproject for example plugins. This change does not yet breakup
jvm-example into the new examples dir, which should be done separately.

relates #19359
relates #19966
This commit is contained in:
Ryan Ernst 2017-05-11 12:15:16 -07:00 committed by GitHub
parent 17d01550c2
commit c1f1f66509
14 changed files with 394 additions and 284 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
core/src/main/java/org/elasticsearch/script/LeafSearchScript.java
View File

@ -19,18 +19,30 @@
package org.elasticsearch.script; package org.elasticsearch.script;
import org.apache.lucene.search.Scorer;
import org.elasticsearch.common.lucene.ScorerAware; import org.elasticsearch.common.lucene.ScorerAware;
import java.util.Map; import java.util.Map;
/** /**
* A per-segment {@link SearchScript}. * A per-segment {@link SearchScript}.
*
* This is effectively a functional interface, requiring at least implementing {@link #runAsDouble()}.
*/ */
public interface LeafSearchScript extends ScorerAware, ExecutableScript { public interface LeafSearchScript extends ScorerAware, ExecutableScript {
void setDocument(int doc); /**
* Set the document this script will process next.
*/
default void setDocument(int doc) {}
void setSource(Map<String, Object> source); @Override
default void setScorer(Scorer scorer) {}
/**
* Set the source for the current document.
*/
default void setSource(Map<String, Object> source) {}
/** /**
* Sets per-document aggregation {@code _value}. * Sets per-document aggregation {@code _value}.
@ -44,8 +56,23 @@ public interface LeafSearchScript extends ScorerAware, ExecutableScript {
setNextVar("_value", value); setNextVar("_value", value);
} }
long runAsLong(); @Override
default void setNextVar(String field, Object value) {}
/**
* Return the result as a long. This is used by aggregation scripts over long fields.
*/
default long runAsLong() {
throw new UnsupportedOperationException("runAsLong is not implemented");
}
@Override
default Object run() {
return runAsDouble();
}
/**
* Return the result as a double. This is the main use case of search script, used for document scoring.
*/
double runAsDouble(); double runAsDouble();
} }

4
core/src/main/java/org/elasticsearch/script/ScriptEngine.java
View File

@ -38,7 +38,9 @@ public interface ScriptEngine extends Closeable {
/** /**
* The extension for file scripts in this language. * The extension for file scripts in this language.
*/ */
String getExtension(); default String getExtension() {
return getType();
}
/** /**
* Compiles a script. * Compiles a script.

6
docs/reference/modules/scripting.asciidoc
View File

@ -51,7 +51,7 @@ certain tasks.
|built-in |built-in
|templates |templates
|<<modules-scripting-native, `java`>> |<<modules-scripting-engine, `java`>>
|n/a |n/a
|you write it! |you write it!
|expert API |expert API
@ -83,6 +83,4 @@ include::scripting/painless-debugging.asciidoc[]
include::scripting/expression.asciidoc[] include::scripting/expression.asciidoc[]
include::scripting/native.asciidoc[] include::scripting/engine.asciidoc[]
include::scripting/advanced-scripting.asciidoc[]

189
docs/reference/modules/scripting/advanced-scripting.asciidoc
View File

@ -1,189 +0,0 @@
[[modules-advanced-scripting]]
=== Advanced text scoring in scripts
experimental[The functionality described on this page is considered experimental and may be changed or removed in a future release]
Text features, such as term or document frequency for a specific term can be
accessed in scripts with the `_index` variable. This can be useful if, for
example, you want to implement your own scoring model using for example a
script inside a <<query-dsl-function-score-query,function score query>>.
Statistics over the document collection are computed *per shard*, not per
index.
It should be noted that the `_index` variable is not supported in the painless language, but `_index` is defined when using the groovy language.
[float]
=== Nomenclature:
[horizontal]
`df`::
document frequency. The number of documents a term appears in. Computed
per field.
`tf`::
term frequency. The number times a term appears in a field in one specific
document.
`ttf`::
total term frequency. The number of times this term appears in all
documents, that is, the sum of `tf` over all documents. Computed per
field.
`df` and `ttf` are computed per shard and therefore these numbers can vary
depending on the shard the current document resides in.
[float]
=== Shard statistics:
`_index.numDocs()`::
Number of documents in shard.
`_index.maxDoc()`::
Maximal document number in shard.
`_index.numDeletedDocs()`::
Number of deleted documents in shard.
[float]
=== Field statistics:
Field statistics can be accessed with a subscript operator like this:
`_index['FIELD']`.
`_index['FIELD'].docCount()`::
Number of documents containing the field `FIELD`. Does not take deleted documents into account.
`_index['FIELD'].sumttf()`::
Sum of `ttf` over all terms that appear in field `FIELD` in all documents.
`_index['FIELD'].sumdf()`::
The sum of `df` s over all terms that appear in field `FIELD` in all
documents.
Field statistics are computed per shard and therefore these numbers can vary
depending on the shard the current document resides in.
The number of terms in a field cannot be accessed using the `_index` variable. See <<token-count>> for how to do that.
[float]
=== Term statistics:
Term statistics for a field can be accessed with a subscript operator like
this: `_index['FIELD']['TERM']`. This will never return null, even if term or field does not exist.
If you do not need the term frequency, call `_index['FIELD'].get('TERM', 0)`
to avoid unnecessary initialization of the frequencies. The flag will have only
affect is your set the <<index-options,`index_options`>> to `docs`.
`_index['FIELD']['TERM'].df()`::
`df` of term `TERM` in field `FIELD`. Will be returned, even if the term
is not present in the current document.
`_index['FIELD']['TERM'].ttf()`::
The sum of term frequencies of term `TERM` in field `FIELD` over all
documents. Will be returned, even if the term is not present in the
current document.
`_index['FIELD']['TERM'].tf()`::
`tf` of term `TERM` in field `FIELD`. Will be 0 if the term is not present
in the current document.
[float]
=== Term positions, offsets and payloads:
If you need information on the positions of terms in a field, call
`_index['FIELD'].get('TERM', flag)` where flag can be
[horizontal]
`_POSITIONS`:: if you need the positions of the term
`_OFFSETS`:: if you need the offsets of the term
`_PAYLOADS`:: if you need the payloads of the term
`_CACHE`:: if you need to iterate over all positions several times
The iterator uses the underlying lucene classes to iterate over positions. For efficiency reasons, you can only iterate over positions once. If you need to iterate over the positions several times, set the `_CACHE` flag.
You can combine the operators with a `|` if you need more than one info. For
example, the following will return an object holding the positions and payloads,
as well as all statistics:
`_index['FIELD'].get('TERM', _POSITIONS | _PAYLOADS)`
Positions can be accessed with an iterator that returns an object
(`POS_OBJECT`) holding position, offsets and payload for each term position.
`POS_OBJECT.position`::
The position of the term.
`POS_OBJECT.startOffset`::
The start offset of the term.
`POS_OBJECT.endOffset`::
The end offset of the term.
`POS_OBJECT.payload`::
The payload of the term.
`POS_OBJECT.payloadAsInt(missingValue)`::
The payload of the term converted to integer. If the current position has
no payload, the `missingValue` will be returned. Call this only if you
know that your payloads are integers.
`POS_OBJECT.payloadAsFloat(missingValue)`::
The payload of the term converted to float. If the current position has no
payload, the `missingValue` will be returned. Call this only if you know
that your payloads are floats.
`POS_OBJECT.payloadAsString()`::
The payload of the term converted to string. If the current position has
no payload, `null` will be returned. Call this only if you know that your
payloads are strings.
Example: sums up all payloads for the term `foo`.
[source,groovy]
---------------------------------------------------------
termInfo = _index['my_field'].get('foo',_PAYLOADS);
score = 0;
for (pos in termInfo) {
score = score + pos.payloadAsInt(0);
}
return score;
---------------------------------------------------------
[float]
=== Term vectors:
The `_index` variable can only be used to gather statistics for single terms. If you want to use information on all terms in a field, you must store the term vectors (see <<term-vector>>). To access them, call
`_index.termVectors()` to get a
https://lucene.apache.org/core/4_0_0/core/org/apache/lucene/index/Fields.html[Fields]
instance. This object can then be used as described in https://lucene.apache.org/core/4_0_0/core/org/apache/lucene/index/Fields.html[lucene doc] to iterate over fields and then for each field iterate over each term in the field.
The method will return null if the term vectors were not stored.

57
docs/reference/modules/scripting/engine.asciidoc Normal file
View File

@ -0,0 +1,57 @@
[[modules-scripting-engine]]
=== Advanced scripts using script engines
A `ScriptEngine` is a backend for implementing a scripting language. It may also
be used to write scripts that need to use advanced internals of scripting. For example,
a script that wants to use term frequencies while scoring.
The plugin {plugins}/plugin-authors.html[documentation] has more information on
how to write a plugin so that Elasticsearch will properly load it. To register
the `ScriptEngine`, your plugin should implement the `ScriptPlugin` interface
and override the `getScriptEngine(Settings settings)` method.
The following is an example of a custom `ScriptEngine` which uses the language
name `expert_scripts`. It implements a single script called `pure_df` which
may be used as a search script to override each document's score as
the document frequency of a provided term.
["source","java",subs="attributes,callouts,macros"]
--------------------------------------------------
include-tagged::{docdir}/../../plugins/examples/script-expert-scoring/src/main/java/org/elasticsearch/example/expertscript/ExpertScriptPlugin.java[expert_engine]
--------------------------------------------------
You can execute the script by specifying its `lang` as `expert_scripts`, and the name
of the script as the the script source:
[source,js]
--------------------------------------------------
POST /_search
{
"query": {
"function_score": {
"query": {
"match": {
"body": "foo"
}
},
"functions": [
{
"script_score": {
"script": {
"inline": "pure_df",
"lang" : "expert_scripts",
"params": {
"field": "body",
"term": "foo"
}
}
}
}
]
}
}
}
--------------------------------------------------
// CONSOLE
// TEST[skip:we don't have an expert script plugin installed to test this]

86
docs/reference/modules/scripting/native.asciidoc
View File

@ -1,86 +0,0 @@
[[modules-scripting-native]]
=== Native (Java) Scripts
Sometimes `painless` and <<modules-scripting-expression, expression>> aren't enough. For those times you can
implement a native script.
The best way to implement a native script is to write a plugin and install it.
The plugin {plugins}/plugin-authors.html[documentation] has more information on
how to write a plugin so that Elasticsearch will properly load it.
To register the actual script you'll need to implement `NativeScriptFactory`
to construct the script. The actual script will extend either
`AbstractExecutableScript` or `AbstractSearchScript`. The second one is likely
the most useful and has several helpful subclasses you can extend like
`AbstractLongSearchScript` and `AbstractDoubleSearchScript`.
Finally, your plugin should register the native script by implementing the
`ScriptPlugin` interface.
If you squashed the whole thing into one class it'd look like:
[source,java]
--------------------------------------------------
public class MyNativeScriptPlugin extends Plugin implements ScriptPlugin {
@Override
public List<NativeScriptFactory> getNativeScripts() {
return Collections.singletonList(new MyNativeScriptFactory());
}
public static class MyNativeScriptFactory implements NativeScriptFactory {
@Override
public ExecutableScript newScript(@Nullable Map<String, Object> params) {
return new MyNativeScript();
}
@Override
public boolean needsScores() {
return false;
}
@Override
public String getName() {
return "my_script";
}
}
public static class MyNativeScript extends AbstractDoubleSearchScript {
@Override
public double runAsDouble() {
double a = (double) source().get("a");
double b = (double) source().get("b");
return a * b;
}
}
}
--------------------------------------------------
You can execute the script by specifying its `lang` as `native`, and the name
of the script as the `id`:
[source,js]
--------------------------------------------------
POST /_search
{
"query": {
"function_score": {
"query": {
"match": {
"body": "foo"
}
},
"functions": [
{
"script_score": {
"script": {
"inline": "my_script",
"lang" : "native"
}
}
}
]
}
}
}
--------------------------------------------------
// CONSOLE
// TEST[skip:we don't have a native plugin installed to test this]

10
docs/reference/redirects.asciidoc
View File

@ -463,3 +463,13 @@ index that make warmers not necessary anymore.
=== Index time boosting === Index time boosting
The index time boost mapping has been replaced with query time boost (see <<mapping-boost>>). The index time boost mapping has been replaced with query time boost (see <<mapping-boost>>).
[role="exclude",id="modules-scripting-native"]
=== Native scripting
Native scripts have been replaced with writing custom `ScriptEngine` backends (see <<modules-scripting-engine>>).
[role="exclude",id="modules-advanced-scripting"]
=== Advanced scripting
Using `_index` in scripts has been replaced with writing `ScriptEngine` backends (see <<modules-scripting-engine>>).

0
plugins/examples/build.gradle Normal file
View File

28
plugins/examples/script-expert-scoring/build.gradle Normal file
View File

@ -0,0 +1,28 @@
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
apply plugin: 'elasticsearch.esplugin'
esplugin {
name 'script-expert-scoring'
description 'An example script engine to use low level Lucene internals for expert scoring'
classname 'org.elasticsearch.example.expertscript.ExpertScriptPlugin'
}
test.enabled = false

145
plugins/examples/script-expert-scoring/src/main/java/org/elasticsearch/example/expertscript/ExpertScriptPlugin.java Normal file
View File

@ -0,0 +1,145 @@
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.elasticsearch.example.expertscript;
import java.io.IOException;
import java.io.UncheckedIOException;
import java.util.Map;
import java.util.Objects;
import java.util.function.Function;
import org.apache.lucene.index.LeafReaderContext;
import org.apache.lucene.index.PostingsEnum;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.Scorer;
import org.elasticsearch.common.Nullable;
import org.elasticsearch.common.settings.Settings;
import org.elasticsearch.plugins.Plugin;
import org.elasticsearch.plugins.ScriptPlugin;
import org.elasticsearch.script.CompiledScript;
import org.elasticsearch.script.ExecutableScript;
import org.elasticsearch.script.LeafSearchScript;
import org.elasticsearch.script.ScriptEngine;
import org.elasticsearch.script.SearchScript;
import org.elasticsearch.search.lookup.SearchLookup;
/**
* An example script plugin that adds a {@link ScriptEngine} implementing expert scoring.
*/
public class ExpertScriptPlugin extends Plugin implements ScriptPlugin {
@Override
public ScriptEngine getScriptEngine(Settings settings) {
return new MyExpertScriptEngine();
}
/** An example {@link ScriptEngine} that uses Lucene segment details to implement pure document frequency scoring. */
// tag::expert_engine
private static class MyExpertScriptEngine implements ScriptEngine {
@Override
public String getType() {
return "expert_scripts";
}
@Override
public Function<Map<String,Object>,SearchScript> compile(String scriptName, String scriptSource, Map<String, String> params) {
// we use the script "source" as the script identifier
if ("pure_df".equals(scriptSource)) {
return p -> new SearchScript() {
final String field;
final String term;
{
if (p.containsKey("field") == false) {
throw new IllegalArgumentException("Missing parameter [field]");
}
if (p.containsKey("term") == false) {
throw new IllegalArgumentException("Missing parameter [term]");
}
field = p.get("field").toString();
term = p.get("term").toString();
}
@Override
public LeafSearchScript getLeafSearchScript(LeafReaderContext context) throws IOException {
PostingsEnum postings = context.reader().postings(new Term(field, term));
if (postings == null) {
// the field and/or term don't exist in this segment, so always return 0
return () -> 0.0d;
}
return new LeafSearchScript() {
int currentDocid = -1;
@Override
public void setDocument(int docid) {
// advance has undefined behavior calling with a docid <= its current docid
if (postings.docID() < docid) {
try {
postings.advance(docid);
} catch (IOException e) {
throw new UncheckedIOException(e);
}
}
currentDocid = docid;
}
@Override
public double runAsDouble() {
if (postings.docID() != currentDocid) {
// advance moved past the current doc, so this doc has no occurrences of the term
return 0.0d;
}
try {
return postings.freq();
} catch (IOException e) {
throw new UncheckedIOException(e);
}
}
};
}
@Override
public boolean needsScores() {
return false;
}
};
}
throw new IllegalArgumentException("Unknown script name " + scriptSource);
}
@Override
@SuppressWarnings("unchecked")
public SearchScript search(CompiledScript compiledScript, SearchLookup lookup, @Nullable Map<String, Object> params) {
Function<Map<String,Object>,SearchScript> scriptFactory = (Function<Map<String,Object>,SearchScript>) compiledScript.compiled();
return scriptFactory.apply(params);
}
@Override
public ExecutableScript executable(CompiledScript compiledScript, @Nullable Map<String, Object> params) {
throw new UnsupportedOperationException();
}
@Override
public boolean isInlineScriptEnabled() {
return true;
}
@Override
public void close() {}
}
// end::expert_engine
}

38
plugins/examples/script-expert-scoring/src/test/java/org/elasticsearch/example/expertscript/ExpertScriptClientYamlTestSuiteIT.java Normal file
View File

@ -0,0 +1,38 @@
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.elasticsearch.example.expertscript;
import com.carrotsearch.randomizedtesting.annotations.Name;
import com.carrotsearch.randomizedtesting.annotations.ParametersFactory;
import org.elasticsearch.test.rest.yaml.ClientYamlTestCandidate;
import org.elasticsearch.test.rest.yaml.ESClientYamlSuiteTestCase;
public class ExpertScriptClientYamlTestSuiteIT extends ESClientYamlSuiteTestCase {
public ExpertScriptClientYamlTestSuiteIT(@Name("yaml") ClientYamlTestCandidate testCandidate) {
super(testCandidate);
}
@ParametersFactory
public static Iterable<Object[]> parameters() throws Exception {
return ESClientYamlSuiteTestCase.createParameters();
}
}

13
plugins/examples/script-expert-scoring/src/test/resources/rest-api-spec/test/script_expert_scoring/10_basic.yaml Normal file
View File

@ -0,0 +1,13 @@
# Integration tests for the expert scoring script example plugin
#
"Plugin loaded":
- do:
cluster.state: {}
# Get master node id
- set: { master_node: master }
- do:
nodes.info: {}
- match: { nodes.$master.plugins.0.name: script-expert-scoring }

53
plugins/examples/script-expert-scoring/src/test/resources/rest-api-spec/test/script_expert_scoring/20_score.yaml Normal file
View File

@ -0,0 +1,53 @@
# Integration tests for the expert scoring script example plugin
#
---
setup:
- do:
indices.create:
index: test
- do:
index:
index: test
type: test
id: 1
body: { "important_field": "foo" }
- do:
index:
index: test
type: test
id: 2
body: { "important_field": "foo foo foo" }
- do:
index:
index: test
type: test
id: 3
body: { "important_field": "foo foo" }
- do:
indices.refresh: {}
---
"document scoring":
- do:
search:
index: test
body:
query:
function_score:
query:
match:
important_field: "foo"
functions:
- script_score:
script:
inline: "pure_df"
lang: "expert_scripts"
params:
field: "important_field"
term: "foo"
- length: { hits.hits: 3 }
- match: {hits.hits.0._id: "2" }
- match: {hits.hits.1._id: "3" }
- match: {hits.hits.2._id: "1" }

14
settings.gradle
View File

@ -78,6 +78,15 @@ List projects = [
'qa:wildfly' 'qa:wildfly'
] ]
File examplePluginsDir = new File(rootProject.projectDir, 'plugins/examples')
List<String> examplePlugins = []
for (File example : examplePluginsDir.listFiles()) {
if (example.isDirectory() == false) continue;
if (example.name.startsWith('build') || example.name.startsWith('.')) continue;
projects.add("example-plugins:${example.name}".toString())
examplePlugins.add(example.name)
}
boolean isEclipse = System.getProperty("eclipse.launcher") != null || gradle.startParameter.taskNames.contains('eclipse') || gradle.startParameter.taskNames.contains('cleanEclipse') boolean isEclipse = System.getProperty("eclipse.launcher") != null || gradle.startParameter.taskNames.contains('eclipse') || gradle.startParameter.taskNames.contains('cleanEclipse')
if (isEclipse) { if (isEclipse) {
// eclipse cannot handle an intermediate dependency between main and test, so we must create separate projects // eclipse cannot handle an intermediate dependency between main and test, so we must create separate projects
@ -88,6 +97,11 @@ if (isEclipse) {
include projects.toArray(new String[0]) include projects.toArray(new String[0])
project(':build-tools').projectDir = new File(rootProject.projectDir, 'buildSrc') project(':build-tools').projectDir = new File(rootProject.projectDir, 'buildSrc')
project(':example-plugins').projectDir = new File(rootProject.projectDir, 'plugins/examples')
for (String example : examplePlugins) {
project(":example-plugins:${example}").projectDir = new File(rootProject.projectDir, "plugins/examples/${example}")
}
if (isEclipse) { if (isEclipse) {
project(":core").projectDir = new File(rootProject.projectDir, 'core/src/main') project(":core").projectDir = new File(rootProject.projectDir, 'core/src/main')