Merge pull request #747 from honestica/analyzer-def-using-search-mapping

Analyzer def using search mapping
This commit is contained in:
James Agnew 2017-11-13 13:36:19 -05:00 committed by GitHub
commit 8f44af734a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
15 changed files with 85 additions and 74 deletions

View File

@ -2,6 +2,7 @@ package ca.uhn.fhir.jpa.demo;
import java.util.Properties;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
@ -23,6 +24,7 @@ public class FhirDbConfig {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", "target/lucenefiles");
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -26,20 +26,8 @@ import ca.uhn.fhir.rest.api.Constants;
import ca.uhn.fhir.rest.server.exceptions.UnprocessableEntityException;
import org.apache.commons.lang3.builder.ToStringBuilder;
import org.apache.commons.lang3.builder.ToStringStyle;
import org.apache.lucene.analysis.core.LowerCaseFilterFactory;
import org.apache.lucene.analysis.core.StopFilterFactory;
import org.apache.lucene.analysis.miscellaneous.WordDelimiterFilterFactory;
import org.apache.lucene.analysis.ngram.EdgeNGramFilterFactory;
import org.apache.lucene.analysis.ngram.NGramFilterFactory;
import org.apache.lucene.analysis.pattern.PatternTokenizerFactory;
import org.apache.lucene.analysis.phonetic.PhoneticFilterFactory;
import org.apache.lucene.analysis.snowball.SnowballPorterFilterFactory;
import org.apache.lucene.analysis.standard.StandardFilterFactory;
import org.apache.lucene.analysis.standard.StandardTokenizerFactory;
import org.hibernate.annotations.ColumnDefault;
import org.hibernate.annotations.OptimisticLock;
import org.hibernate.search.annotations.*;
import org.hibernate.search.annotations.Parameter;
import javax.persistence.*;
import javax.persistence.Index;
@ -61,53 +49,6 @@ import static org.apache.commons.lang3.StringUtils.defaultString;
@Index(name = "IDX_RES_TYPE", columnList = "RES_TYPE"),
@Index(name = "IDX_INDEXSTATUS", columnList = "SP_INDEX_STATUS")
})
@AnalyzerDefs({
@AnalyzerDef(name = "autocompleteEdgeAnalyzer",
tokenizer = @TokenizerDef(factory = PatternTokenizerFactory.class, params = {
@Parameter(name = "pattern", value = "(.*)"),
@Parameter(name = "group", value = "1")
}),
filters = {
@TokenFilterDef(factory = LowerCaseFilterFactory.class),
@TokenFilterDef(factory = StopFilterFactory.class),
@TokenFilterDef(factory = EdgeNGramFilterFactory.class, params = {
@Parameter(name = "minGramSize", value = "3"),
@Parameter(name = "maxGramSize", value = "50")
}),
}),
@AnalyzerDef(name = "autocompletePhoneticAnalyzer",
tokenizer = @TokenizerDef(factory = StandardTokenizerFactory.class),
filters = {
@TokenFilterDef(factory = StandardFilterFactory.class),
@TokenFilterDef(factory = StopFilterFactory.class),
@TokenFilterDef(factory = PhoneticFilterFactory.class, params = {
@Parameter(name = "encoder", value = "DoubleMetaphone")
}),
@TokenFilterDef(factory = SnowballPorterFilterFactory.class, params = {
@Parameter(name = "language", value = "English")
})
}),
@AnalyzerDef(name = "autocompleteNGramAnalyzer",
tokenizer = @TokenizerDef(factory = StandardTokenizerFactory.class),
filters = {
@TokenFilterDef(factory = WordDelimiterFilterFactory.class),
@TokenFilterDef(factory = LowerCaseFilterFactory.class),
@TokenFilterDef(factory = NGramFilterFactory.class, params = {
@Parameter(name = "minGramSize", value = "3"),
@Parameter(name = "maxGramSize", value = "20")
}),
}),
@AnalyzerDef(name = "standardAnalyzer",
tokenizer = @TokenizerDef(factory = StandardTokenizerFactory.class),
filters = {
@TokenFilterDef(factory = LowerCaseFilterFactory.class),
}),
@AnalyzerDef(name = "exactAnalyzer",
tokenizer = @TokenizerDef(factory = StandardTokenizerFactory.class),
filters = {
})
}
)
//@formatter:on
public class ResourceTable extends BaseHasResource implements Serializable {
static final int RESTYPE_LEN = 30;

View File

@ -52,16 +52,12 @@ import org.apache.commons.lang3.builder.EqualsBuilder;
import org.apache.commons.lang3.builder.HashCodeBuilder;
import org.apache.commons.lang3.builder.ToStringBuilder;
import org.apache.commons.lang3.builder.ToStringStyle;
import org.apache.lucene.analysis.core.WhitespaceTokenizerFactory;
import org.hibernate.search.annotations.Analyze;
import org.hibernate.search.annotations.Analyzer;
import org.hibernate.search.annotations.AnalyzerDef;
import org.hibernate.search.annotations.AnalyzerDefs;
import org.hibernate.search.annotations.Field;
import org.hibernate.search.annotations.Fields;
import org.hibernate.search.annotations.Indexed;
import org.hibernate.search.annotations.Store;
import org.hibernate.search.annotations.TokenizerDef;
import ca.uhn.fhir.jpa.entity.TermConceptParentChildLink.RelationshipTypeEnum;
import ca.uhn.fhir.jpa.search.DeferConceptIndexingInterceptor;
@ -73,12 +69,6 @@ import ca.uhn.fhir.jpa.search.DeferConceptIndexingInterceptor;
}, indexes= {
@Index(name = "IDX_CONCEPT_INDEXSTATUS", columnList="INDEX_STATUS")
})
@AnalyzerDefs({
@AnalyzerDef(name = "conceptParentPidsAnalyzer",
tokenizer = @TokenizerDef(factory = WhitespaceTokenizerFactory.class),
filters = {
})
})
public class TermConcept implements Serializable {
private static final int MAX_DESC_LENGTH = 400;
private static final org.slf4j.Logger ourLog = org.slf4j.LoggerFactory.getLogger(TermConcept.class);

View File

@ -0,0 +1,53 @@
package ca.uhn.fhir.jpa.search;
import org.apache.lucene.analysis.core.LowerCaseFilterFactory;
import org.apache.lucene.analysis.core.StopFilterFactory;
import org.apache.lucene.analysis.core.WhitespaceTokenizerFactory;
import org.apache.lucene.analysis.miscellaneous.WordDelimiterFilterFactory;
import org.apache.lucene.analysis.ngram.EdgeNGramFilterFactory;
import org.apache.lucene.analysis.ngram.NGramFilterFactory;
import org.apache.lucene.analysis.pattern.PatternTokenizerFactory;
import org.apache.lucene.analysis.phonetic.PhoneticFilterFactory;
import org.apache.lucene.analysis.snowball.SnowballPorterFilterFactory;
import org.apache.lucene.analysis.standard.StandardFilterFactory;
import org.apache.lucene.analysis.standard.StandardTokenizerFactory;
import org.hibernate.search.annotations.Factory;
import org.hibernate.search.cfg.SearchMapping;
/**
* Factory for defining the analysers.
*/
public class LuceneSearchMappingFactory {
@Factory
public SearchMapping getSearchMapping() {
SearchMapping mapping = new SearchMapping();
mapping.analyzerDef("autocompleteEdgeAnalyzer", PatternTokenizerFactory.class)
.tokenizerParam("pattern", "(.*)")
.tokenizerParam("group", "1")
.filter(LowerCaseFilterFactory.class)
.filter(StopFilterFactory.class)
.filter(EdgeNGramFilterFactory.class)
.param("minGramSize", "3")
.param("maxGramSize", "50")
.analyzerDef("autocompletePhoneticAnalyzer", StandardTokenizerFactory.class)
.filter(StandardFilterFactory.class)
.filter(StopFilterFactory.class)
.filter(PhoneticFilterFactory.class)
.param("encoder", "DoubleMetaphone")
.filter(SnowballPorterFilterFactory.class)
.param("language", "English")
.analyzerDef("autocompleteNGramAnalyzer", StandardTokenizerFactory.class)
.filter(WordDelimiterFilterFactory.class)
.filter(LowerCaseFilterFactory.class)
.filter(NGramFilterFactory.class)
.param("minGramSize", "3")
.param("maxGramSize", "20")
.analyzerDef("standardAnalyzer", StandardTokenizerFactory.class)
.filter(LowerCaseFilterFactory.class)
.analyzerDef("exactAnalyzer", StandardTokenizerFactory.class)
.analyzerDef("conceptParentPidsAnalyzer", WhitespaceTokenizerFactory.class);
return mapping;
}
}

View File

@ -1,7 +1,7 @@
package ca.uhn.fhir.jpa.config;
import ca.uhn.fhir.jpa.dao.DaoConfig;
import ca.uhn.fhir.jpa.subscription.email.SubscriptionEmailInterceptor;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import ca.uhn.fhir.rest.server.interceptor.RequestValidatingInterceptor;
import ca.uhn.fhir.validation.ResultSeverityEnum;
import org.apache.commons.dbcp2.BasicDataSource;
@ -60,6 +60,7 @@ public class TestDstu2Config extends BaseJavaConfigDstu2 {
extraProperties.put("hibernate.show_sql", "false");
extraProperties.put("hibernate.hbm2ddl.auto", "update");
extraProperties.put("hibernate.dialect", "org.hibernate.dialect.DerbyTenSevenDialect");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "ram");
extraProperties.put("hibernate.search.lucene_version","LUCENE_CURRENT");
return extraProperties;

View File

@ -1,5 +1,15 @@
package ca.uhn.fhir.jpa.config;
import java.sql.Connection;
import java.sql.SQLException;
import java.util.Properties;
import java.util.concurrent.TimeUnit;
import javax.persistence.EntityManagerFactory;
import javax.sql.DataSource;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import net.ttddyy.dsproxy.listener.logging.SLF4JLogLevel;
import ca.uhn.fhir.jpa.dao.DaoConfig;
import ca.uhn.fhir.jpa.subscription.email.IEmailSender;
import ca.uhn.fhir.jpa.subscription.email.JavaMailEmailSender;
@ -141,6 +151,7 @@ public class TestDstu3Config extends BaseJavaConfigDstu3 {
extraProperties.put("hibernate.show_sql", "false");
extraProperties.put("hibernate.hbm2ddl.auto", "update");
extraProperties.put("hibernate.dialect", "org.hibernate.dialect.DerbyTenSevenDialect");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "ram");
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");
extraProperties.put("hibernate.search.autoregister_listeners", "true");

View File

@ -1,6 +1,7 @@
package ca.uhn.fhir.jpa.config;
import ca.uhn.fhir.jpa.dao.DaoConfig;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import ca.uhn.fhir.rest.server.interceptor.RequestValidatingInterceptor;
import ca.uhn.fhir.validation.ResultSeverityEnum;
import net.ttddyy.dsproxy.listener.ThreadQueryCountHolder;
@ -125,6 +126,7 @@ public class TestR4Config extends BaseJavaConfigR4 {
extraProperties.put("hibernate.show_sql", "false");
extraProperties.put("hibernate.hbm2ddl.auto", "update");
extraProperties.put("hibernate.dialect", "org.hibernate.dialect.DerbyTenSevenDialect");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "ram");
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");
extraProperties.put("hibernate.search.autoregister_listeners", "true");

View File

@ -5,6 +5,7 @@ import java.util.Properties;
import javax.persistence.EntityManagerFactory;
import javax.sql.DataSource;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import org.apache.commons.dbcp2.BasicDataSource;
import org.apache.commons.lang3.time.DateUtils;
import org.hibernate.jpa.HibernatePersistenceProvider;
@ -80,6 +81,7 @@ public class FhirServerConfig extends BaseJavaConfigDstu3 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", "target/lucenefiles");
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -1,10 +1,9 @@
package ca.uhn.fhir.jpa.demo;
import ca.uhn.fhir.jpa.config.BaseJavaConfigDstu2;
import ca.uhn.fhir.jpa.config.BaseJavaConfigDstu3;
import ca.uhn.fhir.jpa.dao.DaoConfig;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import ca.uhn.fhir.jpa.util.SubscriptionsRequireManualActivationInterceptorDstu2;
import ca.uhn.fhir.jpa.util.SubscriptionsRequireManualActivationInterceptorDstu3;
import ca.uhn.fhir.rest.server.interceptor.IServerInterceptor;
import ca.uhn.fhir.rest.server.interceptor.LoggingInterceptor;
import ca.uhn.fhir.rest.server.interceptor.ResponseHighlighterInterceptor;
@ -80,6 +79,7 @@ public class FhirServerConfigDstu2 extends BaseJavaConfigDstu2 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", "target/lucenefiles");
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -84,6 +84,7 @@ public class FhirServerConfig extends BaseJavaConfigDstu3 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", SearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", "target/lucenefiles");
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -2,6 +2,7 @@ package ca.uhn.fhirtest.config;
import ca.uhn.fhir.jpa.config.BaseJavaConfigDstu2;
import ca.uhn.fhir.jpa.dao.DaoConfig;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import ca.uhn.fhir.jpa.util.SubscriptionsRequireManualActivationInterceptorDstu2;
import ca.uhn.fhir.rest.api.RestOperationTypeEnum;
import ca.uhn.fhir.rest.server.interceptor.IServerInterceptor;
@ -111,6 +112,7 @@ public class TdlDstu2Config extends BaseJavaConfigDstu2 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider" ,"filesystem");
extraProperties.put("hibernate.search.default.indexBase", myFhirLuceneLocation);
extraProperties.put("hibernate.search.lucene_version","LUCENE_CURRENT");

View File

@ -5,6 +5,7 @@ import java.util.Properties;
import javax.persistence.EntityManagerFactory;
import javax.sql.DataSource;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import org.apache.commons.dbcp2.BasicDataSource;
import org.apache.commons.lang3.time.DateUtils;
import org.hibernate.dialect.DerbyTenSevenDialect;
@ -13,7 +14,6 @@ import org.springframework.beans.factory.annotation.Autowire;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.context.annotation.DependsOn;
import org.springframework.context.annotation.Import;
import org.springframework.context.annotation.Lazy;
import org.springframework.context.support.PropertySourcesPlaceholderConfigurer;
@ -94,6 +94,7 @@ public class TdlDstu3Config extends BaseJavaConfigDstu3 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", myFhirLuceneLocation);
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -2,6 +2,7 @@ package ca.uhn.fhirtest.config;
import ca.uhn.fhir.jpa.config.BaseJavaConfigDstu2;
import ca.uhn.fhir.jpa.dao.DaoConfig;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import ca.uhn.fhir.rest.server.interceptor.IServerInterceptor;
import ca.uhn.fhir.rest.server.interceptor.RequestValidatingInterceptor;
import ca.uhn.fhir.validation.ResultSeverityEnum;
@ -115,6 +116,7 @@ public class TestDstu2Config extends BaseJavaConfigDstu2 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", myFhirLuceneLocation);
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -5,6 +5,7 @@ import java.util.Properties;
import javax.persistence.EntityManagerFactory;
import javax.sql.DataSource;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import org.apache.commons.dbcp2.BasicDataSource;
import org.apache.commons.lang3.time.DateUtils;
import org.hibernate.dialect.DerbyTenSevenDialect;
@ -112,6 +113,7 @@ public class TestDstu3Config extends BaseJavaConfigDstu3 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", myFhirLuceneLocation);
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");

View File

@ -5,11 +5,11 @@ import java.util.Properties;
import javax.persistence.EntityManagerFactory;
import javax.sql.DataSource;
import ca.uhn.fhir.jpa.search.LuceneSearchMappingFactory;
import org.apache.commons.dbcp2.BasicDataSource;
import org.apache.commons.lang3.time.DateUtils;
import org.hibernate.dialect.DerbyTenSevenDialect;
import org.hibernate.dialect.PostgreSQL94Dialect;
import org.hibernate.dialect.PostgreSQL95Dialect;
import org.hibernate.jpa.HibernatePersistenceProvider;
import org.springframework.beans.factory.annotation.Autowire;
import org.springframework.beans.factory.annotation.Value;
@ -110,6 +110,7 @@ public class TestR4Config extends BaseJavaConfigR4 {
extraProperties.put("hibernate.cache.use_second_level_cache", "false");
extraProperties.put("hibernate.cache.use_structured_entries", "false");
extraProperties.put("hibernate.cache.use_minimal_puts", "false");
extraProperties.put("hibernate.search.model_mapping", LuceneSearchMappingFactory.class.getName());
extraProperties.put("hibernate.search.default.directory_provider", "filesystem");
extraProperties.put("hibernate.search.default.indexBase", myFhirLuceneLocation);
extraProperties.put("hibernate.search.lucene_version", "LUCENE_CURRENT");