custom-components.html - incubator-nlpcraft-website - Git at Google

 ---
 active_crumb: Docs
 layout: documentation
 id: custom-components
 ---

 <!--
  Licensed to the Apache Software Foundation (ASF) under one or more
  contributor license agreements.  See the NOTICE file distributed with
  this work for additional information regarding copyright ownership.
  The ASF 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.
 -->

 <div class="col-md-8 second-column">
     <section id="overview">
         <h2 class="section-title">Custom components <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>

         <p>
             NLPCraft provides a numeric of useful built-in components which allow to solve a wide range of tasks
             without coding.
             But you can need to extend provided functionality and develop your own components.
             Let's look how to do it and when it can be useful for all kind of components step by step.
         </p>
     </section>
     <section id="token-parser">
         <h2 class="section-title">Token parser <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenParser.html">NCTokenParser</a> trait.
         </p>
         <p>
             It's not often situation when you need to prepare your own language tokenizer.
             Mostly it can be necessary if you want to work with some new language.
             You have to prepare new implementation once and can use it for all projects on this language.
             Usually you just should find open source solution and wrap it for
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenParser.html">NCTokenParser</a> trait.
         </p>
         <pre class="brush: scala, highlight: [2, 6]">
             import org.apache.nlpcraft.*
             import org.languagetool.tokenizers.fr.FrenchWordTokenizer
             import scala.jdk.CollectionConverters.*

             class NCFrTokenParser extends NCTokenParser:
                 private val tokenizer = new FrenchWordTokenizer

                 override def tokenize(text: String): List[NCToken] =
                     val toks = collection.mutable.ArrayBuffer.empty[NCToken]
                     var sumLen = 0

                     for ((word, idx) <- tokenizer.tokenize(text).asScala.zipWithIndex)
                         val start = sumLen
                         val end = sumLen + word.length

                         if word.strip.nonEmpty then
                             toks += new NCPropertyMapAdapter with NCToken:
                                 override def getText: String = word
                                 override def getIndex: Int = idx
                                 override def getStartCharIndex: Int = start
                                 override def getEndCharIndex: Int = end

                         sumLen = end

                     toks.toList
         </pre>
         <ul>
             <li>
                 <code>NCFrTokenParser</code> is a simple wrapper which implements <code>NCTokenParser</code> based on
                 open source <a href="https://languagetool.org">Language Tool</a> library.
             </li>
         </ul>
     </section>

     <section id="token-enricher">
         <h2 class="section-title">Token enricher <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenEnricher.html">NCTokenEnricher</a> trait.
         </p>
         <p>
             Tokens enricher is component which allows to add additional properties to prepared tokens.
             These tokens properties are used later when entities detection.
         </p>
         <pre class="brush: scala, highlight: [25, 26]">
             import org.apache.nlpcraft.*
             import org.languagetool.AnalyzedToken
             import org.languagetool.tagging.ru.RussianTagger
             import scala.jdk.CollectionConverters.*

             class NCRuLemmaPosTokenEnricher extends NCTokenEnricher:
                 private def nvl(v: String, dflt : => String): String = if v != null then v else dflt

                 override def enrich(req: NCRequest, cfg: NCModelConfig, toks: List[NCToken]): Unit =
                     val tags = RussianTagger.INSTANCE.tag(toks.map(_.getText).asJava).asScala

                     require(toks.size == tags.size)

                     toks.zip(tags).foreach { case (tok, tag) =>
                         val readings = tag.getReadings.asScala

                         val (lemma, pos) = readings.size match
                             // No data. Lemma is word as is, POS is undefined.
                             case 0 => (tok.getText, "")
                             // Takes first. Other variants ignored.
                             case _ =>
                                 val aTok: AnalyzedToken = readings.head
                                 (nvl(aTok.getLemma, tok.getText), nvl(aTok.getPOSTag, ""))

                         tok.put("pos", pos)
                         tok.put("lemma", lemma)

                         () // Otherwise NPE.
                     }
         </pre>
         <ul>
             <li>
                 <code>Lines 25 and 26</code> enriches <a href="apis/latest/org/apache/nlpcraft/NCToken.html">NCToken</a>
                 by two new properties which can be used for <a href="intent-matching.html">Intent matching</a> later.
             </li>
         </ul>
     </section>

     <section id="token-validator">
         <h2 class="section-title">Token validator <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenValidator.html">NCTokenValidator</a> trait.
         </p>

         <p>
             There are tokens are inspected and exception can be thrown from user code to break user input processing.
         </p>

         <pre class="brush: scala, highlight: [3]">
             new NCTokenValidator:
                 override def validate(req: NCRequest, cfg: NCModelConfig, toks: List[NCToken]): Unit =
                     if toks.exists(_.contains("restrictionFlag"))
                         then throw new NCException("Sentence cannot be processed.")
         </pre>

         <ul>
             <li>
                 There is anonymous instance of <a href="apis/latest/org/apache/nlpcraft/NCTokenValidator.html">NCTokenValidator</a>
                 created.
             </li>
             <li>
                 <code>Lines 3</code> defines the rule when exception should be thrown and sentence processing should be stopped.
             </li>
         </ul>
     </section>

     <section id="entity-parser">
         <h2 class="section-title">Entity parser <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityParser.html">NCEntityParser</a> trait.
         </p>

         <p>
             Most important component which finds user specific data.
             These defined entities are input for <a href="intent-matching.html">Intent matching</a> conditions.
             If built-in <a href="apis/latest/org/apache/nlpcraft/nlp/parsers/NCSemanticEntityParser.html">NCSemanticEntityParser</a>
             is not enough, you can implement your own NER searching here.
             There is point for potential integrations with neural networks or any other solutions which
             help you find and mark your domain specific named entities.
         </p>

         <pre class="brush: scala, highlight: [5]">
             import org.apache.nlpcraft.*

             class CommentsEntityParser extends NCEntityParser :
                 def parse(req: NCRequest, cfg: NCModelConfig, toks: List[NCToken]): List[NCEntity] =
                     if req.getText.trim.startsWith("--") then
                         List(
                             new NCPropertyMapAdapter with NCEntity :
                                 override def getTokens: List[NCToken] = toks
                                 override def getRequestId: String = req.getRequestId
                                 override def getId: String = "comment"
                         )
                     else
                         List.empty
         </pre>
         <ul>
             <li>
                 In given example whole input sentence is marked as single element <code>comment</code> if
                 condition defined on <code>line 5</code> is <code>true</code>.
             </li>
         </ul>
     </section>

     <section id="entity-enricher">
         <h2 class="section-title">Entity enricher <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityEnricher.html">NCEntityEnricher</a> trait.
         </p>
         <p>
             Entity enricher is component which allows to add additional properties to prepared entities.
             Can be useful for extending existing entity enrichers functionality.
         </p>

         <pre class="brush: scala, highlight: [4, 10, 11]">
             import org.apache.nlpcraft.*

             object CityPopulationEntityEnricher:
                 val citiesPopulation: Map[String, Int] = someExternalService.getCitiesPopulation()

             import CityPopulationEntityEnricher.*

             class CityPopulationEntityEnricher extends NCEntityEnricher :
                 def enrich(req: NCRequest, cfg: NCModelConfig, ents: List[NCEntity]): Unit =
                     ents.
                         filter(_.getId == "city").
                         foreach(e => e.put("city:population", citiesPopulation(e("city:name"))))
         </pre>

         <ul>
             <li>
                 <code>Line 4</code> defines getting cities population data from some external service.
             </li>
             <li>
                 <code>Line 10</code> filters entities by <code>ID</code>.
             </li>
             <li>
                 <code>Line 11</code> enriches entities by new <code>city:population</code> property.
             </li>
         </ul>
     </section>

     <section id="entity-mapper">
         <h2 class="section-title">Entity mapper<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityMapper.html">NCEntityMapper</a> trait.
         </p>

         <p>
             Entity mapper is component which allows to map one set of entities into another after the entities
             were parsed and enriched. Can be useful for building complex parsers based on existing.
         </p>

         <pre class="brush: scala, highlight: [4, 10, 12, 13, 14]">
             import org.apache.nlpcraft.*

             object CityPopulationEntityMapper:
                 val citiesPopulation: Map[String, Int] = externalService.getCitiesPopulation()

             import CityPopulationEntityMapper.*

             class CityPopulationEntityMapper extends NCEntityMapper :
                 def map(req: NCRequest, cfg: NCModelConfig, ents: List[NCEntity]): List[NCEntity] =
                     val cities = ents.filter(_.getId == "city")

                     ents.filterNot(_.getId == "city") ++
                     cities ++
                     cities.filter(city => citiesPopulation(city("city:name")) > 1000000).
                         map(city =>
                             new NCPropertyMapAdapter with NCEntity :
                                 override def getTokens: List[NCToken] = city.getTokens
                                 override def getRequestId: String = req.getRequestId
                                 override def getId: String = "big-city"
                         )
         </pre>
         <ul>
             <li>
                 <code>Line 4</code> defines getting cities population data from some external service.
             </li>
             <li>
                 <code>Line 10</code> filters entities by <code>ID</code>.
             </li>
             <li>
                 <code>Line 12, 13 and 14</code> define component result entities set.
                 It contains previously defined <code>city</code> elements, new elements <code>big-city</code> and
                 another not city elements.
             </li>
         </ul>
     </section>

     <section id="entity-validator">
         <h2 class="section-title">Entity validator<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityValidator.html">NCEntityValidator</a> trait.
         </p>
         <p>
             Entities validator is user defined component, where prepared entities are inspected and exceptions
             can be thrown from user code to break user input processing.
         </p>

         <pre class="brush: scala, highlight: [3]">
             new NCEntityValidator :
                 override def validate(req: NCRequest, cfg: NCModelConfig, ents: List[NCEntity]): Unit =
                     if ents.exists(_.getId == "restrictedID")
                         then throw new NCException("Sentence cannot be processed.")
         </pre>

         <ul>
             <li>
                 There is anonymous instance of <a href="apis/latest/org/apache/nlpcraft/NCEntityValidator.html">NCEntityValidator</a>
                 created.
             </li>
             <li>
                 <code>Lines 3</code> defines the rule when exception should be thrown and sentence processing should be stopped.
             </li>
         </ul>
     </section>

     <section id="variant-filter">
         <h2 class="section-title">Variant filter<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
         <p>
             You have to implement <a href="apis/latest/org/apache/nlpcraft/NCVariantFilter.html">NCVariantFilter</a> trait.
         </p>

         <p>
             Component which allows filtering detected variants, rejecting undesirable.
         </p>

         <pre class="brush: scala, highlight: [3]">
             new NCVariantFilter :
                 def filter(req: NCRequest, cfg: NCModelConfig, vars: List[NCVariant]): List[NCVariant] =
                     vars.filter(_.getEntities.exists(_.getId == "requiredID"))
         </pre>

         <ul>
             <li>
                 There is anonymous instance of <a href="apis/latest/org/apache/nlpcraft/NCVariantFilter.html">NCVariantFilter</a>
                 created.
             </li>
             <li>
                 <code>Lines 3</code> defines variant's filter,
                 it passed only variants which contain  <code>requiredID</code> elements.
             </li>
         </ul>
     </section>

 </div>
 <div class="col-md-2 third-column">
     <ul class="side-nav">
         <li class="side-nav-title">On This Page</li>
         <li><a href="#overview">Overview</a></li>
         <li><a href="#token-parser">Token parser</a></li>
         <li><a href="#token-enricher">Token enricher</a></li>
         <li><a href="#token-validator">Token validator</a></li>
         <li><a href="#entity-parser">Entity parser</a></li>
         <li><a href="#entity-enricher">Entity enricher</a></li>
         <li><a href="#entity-mapper">Entity mapper</a></li>
         <li><a href="#entity-validator">Entity validator</a></li>
         <li><a href="#variant-filter">Variant filter</a></li>
         {% include quick-links.html %}
     </ul>
 </div>
	---
	active_crumb: Docs
	layout: documentation
	id: custom-components
	---

	<!--
	Licensed to the Apache Software Foundation (ASF) under one or more
	contributor license agreements. See the NOTICE file distributed with
	this work for additional information regarding copyright ownership.
	The ASF 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.
	-->

	<div class="col-md-8 second-column">
	<section id="overview">
	<h2 class="section-title">Custom components <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>

	<p>
	NLPCraft provides a numeric of useful built-in components which allow to solve a wide range of tasks
	without coding.
	But you can need to extend provided functionality and develop your own components.
	Let's look how to do it and when it can be useful for all kind of components step by step.
	</p>
	</section>
	<section id="token-parser">
	<h2 class="section-title">Token parser <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenParser.html">NCTokenParser</a> trait.
	</p>
	<p>
	It's not often situation when you need to prepare your own language tokenizer.
	Mostly it can be necessary if you want to work with some new language.
	You have to prepare new implementation once and can use it for all projects on this language.
	Usually you just should find open source solution and wrap it for
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenParser.html">NCTokenParser</a> trait.
	</p>
	<pre class="brush: scala, highlight: [2, 6]">
	import org.apache.nlpcraft.*
	import org.languagetool.tokenizers.fr.FrenchWordTokenizer
	import scala.jdk.CollectionConverters.*

	class NCFrTokenParser extends NCTokenParser:
	private val tokenizer = new FrenchWordTokenizer

	override def tokenize(text: String): List[NCToken] =
	val toks = collection.mutable.ArrayBuffer.empty[NCToken]
	var sumLen = 0

	for ((word, idx) <- tokenizer.tokenize(text).asScala.zipWithIndex)
	val start = sumLen
	val end = sumLen + word.length

	if word.strip.nonEmpty then
	toks += new NCPropertyMapAdapter with NCToken:
	override def getText: String = word
	override def getIndex: Int = idx
	override def getStartCharIndex: Int = start
	override def getEndCharIndex: Int = end

	sumLen = end

	toks.toList
	</pre>
	<ul>
	<li>
	<code>NCFrTokenParser</code> is a simple wrapper which implements <code>NCTokenParser</code> based on
	open source <a href="https://languagetool.org">Language Tool</a> library.
	</li>
	</ul>
	</section>

	<section id="token-enricher">
	<h2 class="section-title">Token enricher <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenEnricher.html">NCTokenEnricher</a> trait.
	</p>
	<p>
	Tokens enricher is component which allows to add additional properties to prepared tokens.
	These tokens properties are used later when entities detection.
	</p>
	<pre class="brush: scala, highlight: [25, 26]">
	import org.apache.nlpcraft.*
	import org.languagetool.AnalyzedToken
	import org.languagetool.tagging.ru.RussianTagger
	import scala.jdk.CollectionConverters.*

	class NCRuLemmaPosTokenEnricher extends NCTokenEnricher:
	private def nvl(v: String, dflt : => String): String = if v != null then v else dflt

	override def enrich(req: NCRequest, cfg: NCModelConfig, toks: List[NCToken]): Unit =
	val tags = RussianTagger.INSTANCE.tag(toks.map(_.getText).asJava).asScala

	require(toks.size == tags.size)

	toks.zip(tags).foreach { case (tok, tag) =>
	val readings = tag.getReadings.asScala

	val (lemma, pos) = readings.size match
	// No data. Lemma is word as is, POS is undefined.
	case 0 => (tok.getText, "")
	// Takes first. Other variants ignored.
	case _ =>
	val aTok: AnalyzedToken = readings.head
	(nvl(aTok.getLemma, tok.getText), nvl(aTok.getPOSTag, ""))

	tok.put("pos", pos)
	tok.put("lemma", lemma)

	() // Otherwise NPE.
	}
	</pre>
	<ul>
	<li>
	<code>Lines 25 and 26</code> enriches <a href="apis/latest/org/apache/nlpcraft/NCToken.html">NCToken</a>
	by two new properties which can be used for <a href="intent-matching.html">Intent matching</a> later.
	</li>
	</ul>
	</section>

	<section id="token-validator">
	<h2 class="section-title">Token validator <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCTokenValidator.html">NCTokenValidator</a> trait.
	</p>

	<p>
	There are tokens are inspected and exception can be thrown from user code to break user input processing.
	</p>

	<pre class="brush: scala, highlight: [3]">
	new NCTokenValidator:
	override def validate(req: NCRequest, cfg: NCModelConfig, toks: List[NCToken]): Unit =
	if toks.exists(_.contains("restrictionFlag"))
	then throw new NCException("Sentence cannot be processed.")
	</pre>

	<ul>
	<li>
	There is anonymous instance of <a href="apis/latest/org/apache/nlpcraft/NCTokenValidator.html">NCTokenValidator</a>
	created.
	</li>
	<li>
	<code>Lines 3</code> defines the rule when exception should be thrown and sentence processing should be stopped.
	</li>
	</ul>
	</section>

	<section id="entity-parser">
	<h2 class="section-title">Entity parser <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityParser.html">NCEntityParser</a> trait.
	</p>

	<p>
	Most important component which finds user specific data.
	These defined entities are input for <a href="intent-matching.html">Intent matching</a> conditions.
	If built-in <a href="apis/latest/org/apache/nlpcraft/nlp/parsers/NCSemanticEntityParser.html">NCSemanticEntityParser</a>
	is not enough, you can implement your own NER searching here.
	There is point for potential integrations with neural networks or any other solutions which
	help you find and mark your domain specific named entities.
	</p>

	<pre class="brush: scala, highlight: [5]">
	import org.apache.nlpcraft.*

	class CommentsEntityParser extends NCEntityParser :
	def parse(req: NCRequest, cfg: NCModelConfig, toks: List[NCToken]): List[NCEntity] =
	if req.getText.trim.startsWith("--") then
	List(
	new NCPropertyMapAdapter with NCEntity :
	override def getTokens: List[NCToken] = toks
	override def getRequestId: String = req.getRequestId
	override def getId: String = "comment"
	)
	else
	List.empty
	</pre>
	<ul>
	<li>
	In given example whole input sentence is marked as single element <code>comment</code> if
	condition defined on <code>line 5</code> is <code>true</code>.
	</li>
	</ul>
	</section>

	<section id="entity-enricher">
	<h2 class="section-title">Entity enricher <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityEnricher.html">NCEntityEnricher</a> trait.
	</p>
	<p>
	Entity enricher is component which allows to add additional properties to prepared entities.
	Can be useful for extending existing entity enrichers functionality.
	</p>

	<pre class="brush: scala, highlight: [4, 10, 11]">
	import org.apache.nlpcraft.*

	object CityPopulationEntityEnricher:
	val citiesPopulation: Map[String, Int] = someExternalService.getCitiesPopulation()

	import CityPopulationEntityEnricher.*

	class CityPopulationEntityEnricher extends NCEntityEnricher :
	def enrich(req: NCRequest, cfg: NCModelConfig, ents: List[NCEntity]): Unit =
	ents.
	filter(_.getId == "city").
	foreach(e => e.put("city:population", citiesPopulation(e("city:name"))))
	</pre>

	<ul>
	<li>
	<code>Line 4</code> defines getting cities population data from some external service.
	</li>
	<li>
	<code>Line 10</code> filters entities by <code>ID</code>.
	</li>
	<li>
	<code>Line 11</code> enriches entities by new <code>city:population</code> property.
	</li>
	</ul>
	</section>

	<section id="entity-mapper">
	<h2 class="section-title">Entity mapper<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityMapper.html">NCEntityMapper</a> trait.
	</p>

	<p>
	Entity mapper is component which allows to map one set of entities into another after the entities
	were parsed and enriched. Can be useful for building complex parsers based on existing.
	</p>

	<pre class="brush: scala, highlight: [4, 10, 12, 13, 14]">
	import org.apache.nlpcraft.*

	object CityPopulationEntityMapper:
	val citiesPopulation: Map[String, Int] = externalService.getCitiesPopulation()

	import CityPopulationEntityMapper.*

	class CityPopulationEntityMapper extends NCEntityMapper :
	def map(req: NCRequest, cfg: NCModelConfig, ents: List[NCEntity]): List[NCEntity] =
	val cities = ents.filter(_.getId == "city")

	ents.filterNot(_.getId == "city") ++
	cities ++
	cities.filter(city => citiesPopulation(city("city:name")) > 1000000).
	map(city =>
	new NCPropertyMapAdapter with NCEntity :
	override def getTokens: List[NCToken] = city.getTokens
	override def getRequestId: String = req.getRequestId
	override def getId: String = "big-city"
	)
	</pre>
	<ul>
	<li>
	<code>Line 4</code> defines getting cities population data from some external service.
	</li>
	<li>
	<code>Line 10</code> filters entities by <code>ID</code>.
	</li>
	<li>
	<code>Line 12, 13 and 14</code> define component result entities set.
	It contains previously defined <code>city</code> elements, new elements <code>big-city</code> and
	another not city elements.
	</li>
	</ul>
	</section>

	<section id="entity-validator">
	<h2 class="section-title">Entity validator<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCEntityValidator.html">NCEntityValidator</a> trait.
	</p>
	<p>
	Entities validator is user defined component, where prepared entities are inspected and exceptions
	can be thrown from user code to break user input processing.
	</p>

	<pre class="brush: scala, highlight: [3]">
	new NCEntityValidator :
	override def validate(req: NCRequest, cfg: NCModelConfig, ents: List[NCEntity]): Unit =
	if ents.exists(_.getId == "restrictedID")
	then throw new NCException("Sentence cannot be processed.")
	</pre>

	<ul>
	<li>
	There is anonymous instance of <a href="apis/latest/org/apache/nlpcraft/NCEntityValidator.html">NCEntityValidator</a>
	created.
	</li>
	<li>
	<code>Lines 3</code> defines the rule when exception should be thrown and sentence processing should be stopped.
	</li>
	</ul>
	</section>

	<section id="variant-filter">
	<h2 class="section-title">Variant filter<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2>
	<p>
	You have to implement <a href="apis/latest/org/apache/nlpcraft/NCVariantFilter.html">NCVariantFilter</a> trait.
	</p>

	<p>
	Component which allows filtering detected variants, rejecting undesirable.
	</p>

	<pre class="brush: scala, highlight: [3]">
	new NCVariantFilter :
	def filter(req: NCRequest, cfg: NCModelConfig, vars: List[NCVariant]): List[NCVariant] =
	vars.filter(_.getEntities.exists(_.getId == "requiredID"))
	</pre>

	<ul>
	<li>
	There is anonymous instance of <a href="apis/latest/org/apache/nlpcraft/NCVariantFilter.html">NCVariantFilter</a>
	created.
	</li>
	<li>
	<code>Lines 3</code> defines variant's filter,
	it passed only variants which contain <code>requiredID</code> elements.
	</li>
	</ul>
	</section>

	</div>
	<div class="col-md-2 third-column">
	<ul class="side-nav">
	<li class="side-nav-title">On This Page</li>
	<li><a href="#overview">Overview</a></li>
	<li><a href="#token-parser">Token parser</a></li>
	<li><a href="#token-enricher">Token enricher</a></li>
	<li><a href="#token-validator">Token validator</a></li>
	<li><a href="#entity-parser">Entity parser</a></li>
	<li><a href="#entity-enricher">Entity enricher</a></li>
	<li><a href="#entity-mapper">Entity mapper</a></li>
	<li><a href="#entity-validator">Entity validator</a></li>
	<li><a href="#variant-filter">Variant filter</a></li>
	{% include quick-links.html %}
	</ul>
	</div>