ENH: Add ability to edit built-in plugins (Issue #149) - see TestPlugins.pluginsEdit() for an example.

Author: tsegall

ChangeLog.md

@@ -1,6 +1,9 @@
 
 ## Changes ##
 
+### 17.1.7
+ - ENH: Add ability to edit built-in plugins (Issue #149) - see TestPlugins.pluginsEdit() for an example.
+
 ### 17.1.6
  - ENH: Add support for registering Custom Plugins either higher than or lower than built-in plugins (Issue #148)
  - INT: Bump logback-classic to 1.5.20, SpotBugs to 6.4.4

settings.gradle

@@ -40,7 +40,7 @@ includeBuild 'examples/core/speed'
 dependencyResolutionManagement {
 	versionCatalogs {
 		libs {
-			version('fta', '17.1.6')
+			version('fta', '17.1.7')
 			version('jacoco', '0.8.12')
 
 // https://mvnrepository.com/artifact/de.siegmar/fastcsv

types/src/main/java/com/cobber/fta/PluginDefinition.java

@@ -30,7 +30,6 @@
 import com.cobber.fta.core.FTAType;
 import com.cobber.fta.core.HeaderEntry;
 import com.cobber.fta.core.InternalErrorException;
-
 import com.fasterxml.jackson.core.type.TypeReference;
 import com.fasterxml.jackson.databind.ObjectMapper;
 
@@ -192,6 +191,9 @@ public PluginDefinition(final PluginDefinition other) {
 	/**
 	 * Retrieve the Plugin Definition associated with this Semantic Type name.
 	 *
+	 * Note: Unlike the similar function in TextAnalyzer this one accesses the definition file and any edits
+	 * will not impact the current Analyzer.
+	 *
 	 * @param semanticTypeName The name for this Semantic Type
 	 * @return The Plugin Definition associated with the supplied name.
 	 */

types/src/main/java/com/cobber/fta/TextAnalyzer.java

@@ -53,10 +53,6 @@
 import org.apache.commons.text.similarity.LevenshteinDistance;
 import org.slf4j.LoggerFactory;
 
-import com.cobber.fta.TextAnalyzer.Escalation;
-import com.cobber.fta.TextAnalyzer.Observation;
-import com.cobber.fta.TextAnalyzer.OutlierAnalysis;
-import com.cobber.fta.TextAnalyzer.SignStatus;
 import com.cobber.fta.core.FTAMergeException;
 import com.cobber.fta.core.FTAPluginException;
 import com.cobber.fta.core.FTAType;
@@ -1257,29 +1253,52 @@ public Plugins getPlugins() {
 		return plugins;
 	}
 
+	private void loadPlugins() {
+                synchronized (pluginDefinitions) {
+                        if (pluginDefinitions.isEmpty())
+                                try (BufferedReader reader = new BufferedReader(new InputStreamReader(TextAnalyzer.class.getResourceAsStream("/reference/plugins.json"), StandardCharsets.UTF_8))) {
+                                        pluginDefinitions = mapper.readValue(reader, new TypeReference<List<PluginDefinition>>(){});
+                                } catch (Exception e) {
+                                        throw new IllegalArgumentException("Internal error: Issues with plugins file: " + e.getMessage(), e);
+                                }
+                }
+        }
+
 	/**
 	 * Register the default set of plugins for Semantic Type detection.
 	 *
 	 * @param analysisConfig The Analysis configuration used for this analysis.
 	 * Note: The Locale (on the configuration)  will impact both the set of plugins registered as well as the behavior of the individual plugins
 	 */
 	public void registerDefaultPlugins(final AnalysisConfig analysisConfig) {
-		synchronized (pluginDefinitions) {
-			if (pluginDefinitions.isEmpty()) {
-				try (BufferedReader reader = new BufferedReader(new InputStreamReader(TextAnalyzer.class.getResourceAsStream("/reference/plugins.json"), StandardCharsets.UTF_8))) {
-					pluginDefinitions = mapper.readValue(reader, new TypeReference<List<PluginDefinition>>(){});
-				} catch (Exception e) {
-					throw new IllegalArgumentException("Internal error: Issues with plugins file: " + e.getMessage(), e);
-				}
-			}
-			try {
-				plugins.registerPluginsInternal(pluginDefinitions, context.getStreamName(), analysisConfig);
-			} catch (Exception e) {
-				throw new IllegalArgumentException("Internal error: Issues with plugins file: " + e.getMessage(), e);
-			}
+		loadPlugins();
+
+		try {
+			plugins.registerPluginsInternal(pluginDefinitions, context.getStreamName(), analysisConfig);
+		} catch (Exception e) {
+			throw new IllegalArgumentException("Internal error: Issues with plugins file: " + e.getMessage(), e);
 		}
 	}
 
+	/**
+	 * Retrieve the Plugin Definition associated with this Semantic Type name.
+	 *
+	 * Note: Unlike the similar function in PluginDefinition this one accesses the current instance and any edits
+	 * will impact the current Analyzer.
+	 *
+	 * @param semanticTypeName The name for this Semantic Type
+	 * @return The Plugin Definition associated with the supplied name.
+	 */
+	public PluginDefinition findByName(final String semanticTypeName) {
+		loadPlugins();
+
+		for (final PluginDefinition pluginDefinition : pluginDefinitions)
+			if (pluginDefinition.semanticType.equalsIgnoreCase(semanticTypeName))
+				return pluginDefinition;
+
+		return null;
+	}
+
 	private void initialize() throws FTAPluginException, FTAUnsupportedLocaleException {
 		memoryDebug("initialize.entry");
 

types/src/test/java/com/cobber/fta/TestPlugins.java

@@ -48,6 +48,7 @@
 import com.cobber.fta.core.FTAPluginException;
 import com.cobber.fta.core.FTAType;
 import com.cobber.fta.core.FTAUnsupportedLocaleException;
+import com.cobber.fta.core.HeaderEntry;
 import com.cobber.fta.dates.DateTimeParser.DateResolutionMode;
 import com.cobber.fta.plugins.CheckDigitEAN13;
 import com.cobber.fta.plugins.CheckDigitISBN;
@@ -1962,34 +1963,37 @@ public void pluginsPost() throws FTAException {
 
 	@Test(groups = { TestGroups.ALL, TestGroups.DATETIME })
 	public void pluginsEdit() throws FTAException {
-		PluginDefinition defn = PluginDefinition.findByName("NAME.FIRST");
-		for (PluginLocaleEntry localeEntry : defn.validLocales) {
-			System.err.printf("Found - '%s'\n", localeEntry.localeTag);
-			if ("en".equals(localeEntry.localeTag))
-				System.err.printf("Found - '%s'\n", localeEntry.headerRegExps[0]);
-		}
-		PluginDefinition newDefn = new PluginDefinition(defn);
-		System.err.printf("New - '%s'\n", newDefn.semanticType);
-		for (PluginLocaleEntry localeEntry : newDefn.validLocales) {
-			System.err.printf("Found - '%s'\n", localeEntry.localeTag);
-			if ("en".equals(localeEntry.localeTag)) {
-				System.err.printf("Found - '%s'\n", localeEntry.headerRegExps[0].regExp);
-				localeEntry.headerRegExps[0].regExp = localeEntry.headerRegExps[0].regExp.replace("fname", "fname|fn");
-				System.err.printf("Found - '%s'\n", localeEntry.headerRegExps[0].regExp);
-			}
-		}
-//		for (PluginDefinition defn : pluginDefinitions) {
-//			if ("NAME.FIRST".equals(defn.semanticType)) {
-//				System.err.printf("Found - '%s'\n", defn.semanticType);
-//				for (PluginLocaleEntry localeEntry : defn.validLocales) {
-//					System.err.printf("Found - '%s'\n", localeEntry.localeTag);
-//					if ("en".equals(localeEntry.localeTag))
-//						System.err.printf("Found - '%s'\n", localeEntry.headerRegExps[0]);
-//				}
-//			}
-//		}
-	}
+		final String inputs[] = {
+				"Anna", "Buffalo", "Peter", "David", "Janet", "Mary", "Joe", "Tim", "Jeremy", "Reginald",
+				"Roger", "Lois", "Janet", "Polaris", "Sean", "Shawn", "Shaun", "Regina", "Gail", "Max"
+		};
+		final TextAnalyzer analysisPre = new TextAnalyzer("fn");
+		analysisPre.setLocale(Locale.US);
+		for (final String input : inputs)
+			analysisPre.train(input);
+		final TextAnalysisResult resultPre = analysisPre.getResult();
+		assertNull(resultPre.getSemanticType());
+
+		// Edit the FIRST NAME plugin to add our header "fn"
+		PluginDefinition defn = analysisPre.findByName("NAME.FIRST");
+		for (PluginLocaleEntry localeEntry : defn.validLocales)
+		 if ("en".equals(localeEntry.localeTag)) {
+			 final int entries = localeEntry.headerRegExps.length;
+			 HeaderEntry[] newHeaders = new HeaderEntry[entries + 1];
+			 for (int i = 0; i < entries; i++)
+				 newHeaders[i] = new HeaderEntry(localeEntry.headerRegExps[i]);
+			 newHeaders[entries] = new HeaderEntry("(?i)fn", 100);
+			 localeEntry.headerRegExps = newHeaders;
+		 }
+
+		final TextAnalyzer analysisPost = new TextAnalyzer("fn");
+		analysisPost.setLocale(Locale.US);
+		for (final String input : inputs)
+			analysisPost.train(input);
+		final TextAnalysisResult resultPost = analysisPost.getResult();
+		assertEquals(resultPost.getSemanticType(), "NAME.FIRST");
 
+	}
 
 	@Test(groups = { TestGroups.ALL, TestGroups.PLUGINS })
 	public void basicISIN() throws IOException, FTAException {