Created
October 27, 2015 19:12
-
-
Save mumrah/98a0c0f3863cc6700103 to your computer and use it in GitHub Desktop.
Things that must match up:
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package com.lucidworks.fusion.pipeline.index.config; | |
import com.lucidworks.apollo.pipeline.schema.Annotations.Schema; | |
import com.lucidworks.apollo.pipeline.schema.Annotations.SchemaProperty; | |
import com.lucidworks.apollo.pipeline.schema.validation.RESTValidationError; | |
import com.lucidworks.apollo.pipeline.schema.validation.SchemaValidator; | |
import com.lucidworks.fusion.pipeline.Configuration; | |
import org.codehaus.jackson.annotate.JsonCreator; | |
import org.codehaus.jackson.annotate.JsonProperty; | |
import org.codehaus.jackson.annotate.JsonTypeName; | |
import java.util.Collections; | |
import java.util.List; | |
import java.util.regex.Pattern; | |
@JsonTypeName(RegexFilterConfig.TYPE) | |
@Schema( | |
type = RegexFilterConfig.TYPE, | |
title = "Regular Expression Filter", | |
description = "This stage can filter entities using regular expressions" | |
) | |
public class RegexFilterConfig extends Configuration { | |
public static final String TYPE = "regex-filter"; | |
@SchemaProperty(title = "Filters") | |
private final List<RegexFilterRule> filters; | |
@JsonCreator | |
public RegexFilterConfig( | |
@JsonProperty("id") String id, | |
@JsonProperty("filters") List<RegexFilterRule> rules) { | |
super(id); | |
if (rules != null) { | |
this.filters = Collections.unmodifiableList(rules); | |
} else { | |
this.filters = Collections.emptyList(); | |
} | |
} | |
@JsonProperty("filters") | |
public List<RegexFilterRule> getFilters() { | |
return filters; | |
} | |
public static class RegexFilterRule { | |
@SchemaProperty(name = "sourceField", title = "Source Field", required = true) | |
private final String sourceField; | |
@SchemaProperty(name = "pattern", title = "Filter Pattern", required = true) | |
private final Pattern pattern; | |
@JsonCreator | |
public RegexFilterRule( | |
@JsonProperty("sourceField") String sourceField, | |
@JsonProperty("pattern") String pattern) throws RESTValidationError { | |
this.sourceField = sourceField; | |
this.pattern = SchemaValidator.parseRegex("pattern", pattern); | |
} | |
@JsonProperty("sourceField") | |
public String getSourceField() { | |
return sourceField; | |
} | |
@JsonProperty("pattern") | |
public Pattern getPattern() { | |
return pattern; | |
} | |
} | |
} |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package com.lucidworks.fusion.pipeline.index.stages; | |
import com.google.inject.assistedinject.Assisted; | |
import com.lucidworks.apollo.common.pipeline.PipelineDocument; | |
import com.lucidworks.apollo.common.pipeline.PipelineField; | |
import com.lucidworks.fusion.pipeline.AutoDiscover; | |
import com.lucidworks.fusion.pipeline.Context; | |
import com.lucidworks.fusion.pipeline.StageAssistFactory; | |
import com.lucidworks.fusion.pipeline.StageOutput; | |
import com.lucidworks.fusion.pipeline.index.DocumentProcessor; | |
import com.lucidworks.fusion.pipeline.index.config.RegexFilterConfig; | |
import javax.inject.Inject; | |
import java.util.regex.Matcher; | |
public class RegexFilterStage extends DocumentProcessor<RegexFilterConfig> { | |
@AutoDiscover(value=RegexFilterConfig.TYPE, to=RegexFilterStage.class) | |
public interface RegexFilterStageFactory extends StageAssistFactory<RegexFilterStage> { } | |
@Inject | |
public RegexFilterStage(@Assisted String id) { | |
super(id); | |
} | |
@Override | |
public void init(RegexFilterConfig config) { | |
} | |
@Override | |
public void process(PipelineDocument document, Context context, RegexFilterConfig config, StageOutput<PipelineDocument> output) { | |
PipelineDocument tempDocument = new PipelineDocument(); | |
tempDocument.setId(document.getId()); | |
tempDocument.setMetadata(document.getMetadata()); | |
for(String name : document.getFieldNames()) { | |
for (PipelineField field : document.getFields(name)) { | |
boolean removeField = false; | |
for (RegexFilterConfig.RegexFilterRule rule : config.getFilters()) { | |
if (rule.getSourceField().equalsIgnoreCase(name)) { | |
if (field.getValue() != null) { | |
Matcher matcher = rule.getPattern().matcher(field.getValue().toString()); | |
if (matcher.matches()) { | |
removeField = true; | |
break; | |
} | |
} | |
} | |
} | |
if (!removeField) { | |
tempDocument.addField(field); | |
getMetricRegistry().counter("removed.field").inc(); | |
} | |
} | |
} | |
output.send(tempDocument, context); | |
} | |
@Override | |
public Class<RegexFilterConfig> getConfigurationClass() { | |
return RegexFilterConfig.class; | |
} | |
} | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment