From 30e3ca49652235e7cf4eda0a168183c724fa91c5 Mon Sep 17 00:00:00 2001
From: David Huynh <dfhuynh@gmail.com>
Date: Wed, 24 Mar 2010 04:30:31 +0000
Subject: [PATCH] Added splitByLengths function.

git-svn-id: http://google-refine.googlecode.com/svn/trunk@351 7d457c2a-affb-35e4-300a-418c747d4874
---
 .../functions/strings/SplitByLengths.java     | 49 +++++++++++++++++++
 .../gel/ControlFunctionRegistry.java          |  2 +
 2 files changed, 51 insertions(+)
 create mode 100644 src/main/java/com/metaweb/gridworks/expr/functions/strings/SplitByLengths.java

diff --git a/src/main/java/com/metaweb/gridworks/expr/functions/strings/SplitByLengths.java b/src/main/java/com/metaweb/gridworks/expr/functions/strings/SplitByLengths.java
new file mode 100644
index 000000000..6f218694c
--- /dev/null
+++ b/src/main/java/com/metaweb/gridworks/expr/functions/strings/SplitByLengths.java
@@ -0,0 +1,49 @@
+package com.metaweb.gridworks.expr.functions.strings;
+
+import java.util.Properties;
+
+import org.json.JSONException;
+import org.json.JSONWriter;
+
+import com.metaweb.gridworks.expr.EvalError;
+import com.metaweb.gridworks.gel.ControlFunctionRegistry;
+import com.metaweb.gridworks.gel.Function;
+
+public class SplitByLengths implements Function {
+
+    public Object call(Properties bindings, Object[] args) {
+        if (args.length >= 2 && args[0] != null) {
+            Object o = args[0];
+            String s = o instanceof String ? (String) o : o.toString();
+            
+            String[] results = new String[args.length - 1];
+            
+            int lastIndex = 0;
+            
+            for (int i = 1; i < args.length; i++) {
+                int thisIndex = lastIndex;
+                
+                Object o2 = args[i];
+                if (o2 instanceof Number) {
+                    thisIndex = Math.min(s.length(), lastIndex + Math.max(0, ((Number) o2).intValue()));
+                }
+                
+                results[i - 1] = s.substring(lastIndex, thisIndex);
+                lastIndex = thisIndex;
+            }
+            
+            return results;
+        }
+        return new EvalError(ControlFunctionRegistry.getFunctionName(this) + " expects 1 string and 1 or more numbers");
+    }
+    
+    public void write(JSONWriter writer, Properties options)
+        throws JSONException {
+    
+        writer.object();
+        writer.key("description"); writer.value("Returns the array of strings obtained by splitting s into substrings with the given lengths");
+        writer.key("params"); writer.value("string s, number n, ...");
+        writer.key("returns"); writer.value("array");
+        writer.endObject();
+    }
+}
diff --git a/src/main/java/com/metaweb/gridworks/gel/ControlFunctionRegistry.java b/src/main/java/com/metaweb/gridworks/gel/ControlFunctionRegistry.java
index fdf3598d5..bd2b1c7b3 100644
--- a/src/main/java/com/metaweb/gridworks/gel/ControlFunctionRegistry.java
+++ b/src/main/java/com/metaweb/gridworks/gel/ControlFunctionRegistry.java
@@ -32,6 +32,7 @@ import com.metaweb.gridworks.expr.functions.math.Pow;
 import com.metaweb.gridworks.expr.functions.math.Round;
 import com.metaweb.gridworks.expr.functions.math.Sum;
 import com.metaweb.gridworks.expr.functions.strings.Contains;
+import com.metaweb.gridworks.expr.functions.strings.SplitByLengths;
 import com.metaweb.gridworks.expr.functions.strings.Diff;
 import com.metaweb.gridworks.expr.functions.strings.EndsWith;
 import com.metaweb.gridworks.expr.functions.strings.Fingerprint;
@@ -123,6 +124,7 @@ public class ControlFunctionRegistry {
         registerFunction("replaceChars", new ReplaceChars());
         registerFunction("split", new Split());
         registerFunction("splitByCharType", new SplitByCharType());
+        registerFunction("splitByLengths", new SplitByLengths());
         registerFunction("partition", new Partition());
         registerFunction("rpartition", new RPartition());
         registerFunction("trim", new Trim());