Merge pull request #1432 from OpenRefine/issue/1296

add find function. issue 1296

main/src/com/google/refine/expr/functions/strings/Find.java

@@ -0,0 +1,53 @@
+
+package com.google.refine.expr.functions.strings;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Properties;
+import java.util.regex.Matcher;
+import java.util.regex.Pattern;
+
+import org.json.JSONException;
+import org.json.JSONWriter;
+
+import com.google.refine.expr.EvalError;
+import com.google.refine.grel.ControlFunctionRegistry;
+import com.google.refine.grel.Function;
+
+public class Find implements Function {
+
+    @Override
+    public Object call(Properties bindings, Object[] args) {
+        List<String> allMatches = new ArrayList<String>();
+        
+        if (args.length == 2) {
+            Object s = args[0];
+            Object p = args[1];
+            
+            if (s != null && p != null && (p instanceof String || p instanceof Pattern)) {
+                
+                Pattern pattern = (p instanceof String) ? Pattern.compile((String) p) : (Pattern) p;
+
+                Matcher matcher = pattern.matcher(s.toString());
+                
+                while (matcher.find()) {
+                    allMatches.add(matcher.group());
+                } 
+            }
+            
+            return allMatches.toArray(new String[0]);
+        }
+        return new EvalError(ControlFunctionRegistry.getFunctionName(this) + " expects a string or a regexp");
+    }
+    
+    @Override
+    public void write(JSONWriter writer, Properties options)
+        throws JSONException {
+    
+        writer.object();
+        writer.key("description"); writer.value("Returns all the occurances of match given regular expression");
+        writer.key("params"); writer.value("string or regexp");
+        writer.key("returns"); writer.value("array of strings");
+        writer.endObject();
+    }
+}

main/src/com/google/refine/grel/ControlFunctionRegistry.java

@@ -109,6 +109,7 @@ import com.google.refine.expr.functions.strings.IndexOf;
 import com.google.refine.expr.functions.strings.LastIndexOf;
 import com.google.refine.expr.functions.strings.MD5;
 import com.google.refine.expr.functions.strings.Match;
+import com.google.refine.expr.functions.strings.Find;
 import com.google.refine.expr.functions.strings.NGram;
 import com.google.refine.expr.functions.strings.NGramFingerprint;
 import com.google.refine.expr.functions.strings.ParseJson;
@@ -226,6 +227,7 @@ public class ControlFunctionRegistry {
         registerFunction("parseJson", new ParseJson());
         registerFunction("ngram", new NGram());
         registerFunction("match", new Match());
+        registerFunction("find", new Find());
 
         // HTML functions from JSoup
         registerFunction("parseHtml", new ParseHtml());

main/tests/server/src/com/google/refine/tests/expr/functions/FindFunctionTests.java

@@ -0,0 +1,75 @@
+
+package com.google.refine.tests.expr.functions;
+
+import java.util.Properties;
+
+import org.slf4j.LoggerFactory;
+import org.testng.Assert;
+import org.testng.annotations.AfterMethod;
+import org.testng.annotations.BeforeMethod;
+import org.testng.annotations.BeforeTest;
+import org.testng.annotations.Test;
+
+import com.google.refine.RefineServlet;
+import com.google.refine.grel.ControlFunctionRegistry;
+import com.google.refine.grel.Function;
+import com.google.refine.tests.RefineServletStub;
+import com.google.refine.tests.RefineTest;
+
+/**
+ * Test cases for find function.
+ */
+public class FindFunctionTests extends RefineTest {
+    static Properties bindings;
+    
+    @Override
+    @BeforeTest
+    public void init() {
+        logger = LoggerFactory.getLogger(this.getClass());
+    }
+
+    // dependencies
+    RefineServlet servlet;
+
+    @BeforeMethod
+    public void SetUp() {
+        bindings = new Properties();
+        
+        servlet = new RefineServletStub();
+    }
+    
+    @AfterMethod
+    public void TearDown() {
+    }
+
+    
+    @Test
+    public void findFunctionFindAllTest() throws Exception {
+        String[] matches = (String[]) invoke("find", "This is a test string for testing find.", "test");
+        Assert.assertEquals(matches[0], "test");
+        Assert.assertEquals(matches[1], "test");
+    }
+    
+    @Test
+    public void findFunctionFindAllTest2() throws Exception {
+        String[] matches = (String[]) invoke("find", "hello 123456 goodbye.", "\\d{6}|hello");
+        Assert.assertEquals(matches[0], "hello");
+        Assert.assertEquals(matches[1], "123456");
+    }
+    
+    /**
+     * Lookup a control function by name and invoke it with a variable number of args
+     */
+    private static Object invoke(String name,Object... args) {
+        // registry uses static initializer, so no need to set it up
+        Function function = ControlFunctionRegistry.getFunction(name);
+        if (function == null) {
+            throw new IllegalArgumentException("Unknown function "+name);
+        }
+        if (args == null) {
+            return function.call(bindings,new Object[0]);
+        } else {
+            return function.call(bindings,args);
+        }
+    }
+}