From 50a04c077fc4ee76e86df33e3b35aa876f71ed77 Mon Sep 17 00:00:00 2001
From: Noble Paul <noble@apache.org>
Date: Mon, 5 Mar 2018 21:37:10 +1100
Subject: [PATCH] SOLR-11267: Add support for "add-distinct" atomic update
 operation

---
 solr/CHANGES.txt                              |  2 +
 .../processor/AtomicUpdateDocumentMerger.java | 40 ++++++++++++++++
 .../update/processor/AtomicUpdatesTest.java   | 48 +++++++++++++++++++
 .../src/updating-parts-of-documents.adoc      |  6 +++
 4 files changed, 96 insertions(+)

diff --git a/solr/CHANGES.txt b/solr/CHANGES.txt
index c7a7446fb32..d5313117a16 100644
--- a/solr/CHANGES.txt
+++ b/solr/CHANGES.txt
@@ -182,6 +182,8 @@ New Features
 
 * SOLR-11795: Add Solr metrics exporter for Prometheus (Minoru Osuka via koji)
 
+* SOLR-11267: Add support for "add-distinct" atomic update operation (Amrit Sarkar via noble )
+
 Bug Fixes
 ----------------------
 
diff --git a/solr/core/src/java/org/apache/solr/update/processor/AtomicUpdateDocumentMerger.java b/solr/core/src/java/org/apache/solr/update/processor/AtomicUpdateDocumentMerger.java
index 2689f0e80c2..1198bc93466 100644
--- a/solr/core/src/java/org/apache/solr/update/processor/AtomicUpdateDocumentMerger.java
+++ b/solr/core/src/java/org/apache/solr/update/processor/AtomicUpdateDocumentMerger.java
@@ -18,6 +18,7 @@ package org.apache.solr.update.processor;
 
 import java.io.IOException;
 import java.lang.invoke.MethodHandles;
+import java.util.ArrayList;
 import java.util.Collection;
 import java.util.Collections;
 import java.util.HashSet;
@@ -116,6 +117,10 @@ public class AtomicUpdateDocumentMerger {
               updateField = true;
               doInc(toDoc, sif, fieldVal);
               break;
+            case "add-distinct":
+              updateField = true;
+              doAddDistinct(toDoc, sif, fieldVal);
+              break;
             default:
               //Perhaps throw an error here instead?
               log.warn("Unknown operation for the an atomic update, operation ignored: " + key);
@@ -317,6 +322,41 @@ public class AtomicUpdateDocumentMerger {
     toDoc.addField(sif.getName(), sf.getType().toNativeType(fieldVal));
   }
 
+  protected void doAddDistinct(SolrInputDocument toDoc, SolrInputField sif, Object fieldVal) {
+    final String name = sif.getName();
+    SolrInputField existingField = toDoc.get(name);
+
+    SchemaField sf = schema.getField(name);
+
+    if (sf != null) {
+      Collection<Object> original = existingField != null ?
+          existingField.getValues() :
+          new ArrayList<>();
+
+      int initialSize = original.size();
+      if (fieldVal instanceof Collection) {
+        for (Object object : (Collection) fieldVal) {
+          if (!original.contains(object)) {
+            original.add(object);
+          }
+        }
+      } else {
+        Object object = sf.getType().toNativeType(fieldVal);
+        if (!original.contains(object)) {
+          original.add(object);
+        }
+      }
+
+      if (original.size() > initialSize) { // update only if more are added
+        if (original.size() == 1) { // if single value, pass the value instead of List
+          doAdd(toDoc, sif, original.toArray()[0]);
+        } else {
+          toDoc.setField(name, original);
+        }
+      }
+    }
+  }
+
   protected void doInc(SolrInputDocument toDoc, SolrInputField sif, Object fieldVal) {
     SolrInputField numericField = toDoc.get(sif.getName());
     SchemaField sf = schema.getField(sif.getName());
diff --git a/solr/core/src/test/org/apache/solr/update/processor/AtomicUpdatesTest.java b/solr/core/src/test/org/apache/solr/update/processor/AtomicUpdatesTest.java
index fefbe5d3cf6..62df01784a4 100644
--- a/solr/core/src/test/org/apache/solr/update/processor/AtomicUpdatesTest.java
+++ b/solr/core/src/test/org/apache/solr/update/processor/AtomicUpdatesTest.java
@@ -926,6 +926,54 @@ public class AtomicUpdatesTest extends SolrTestCaseJ4 {
     assertQ(req("q", "cat:bbb", "indent", "true"), "//result[@numFound = '1']");
   }
 
+  @Test
+  public void testAddDistinct() throws Exception {
+    SolrInputDocument doc = new SolrInputDocument();
+    doc.setField("id", "3");
+    doc.setField("cat", new String[]{"aaa", "ccc"});
+    assertU(adoc(doc));
+
+    doc = new SolrInputDocument();
+    doc.setField("id", "4");
+    doc.setField("cat", new String[]{"aaa", "ccc"});
+    assertU(adoc(doc));
+
+    assertU(commit());
+
+    assertQ(req("q", "cat:*", "indent", "true"), "//result[@numFound = '2']");
+    assertQ(req("q", "cat:bbb", "indent", "true"), "//result[@numFound = '0']");
+
+
+    doc = new SolrInputDocument();
+    doc.setField("id", "3");
+    doc.setField("cat", ImmutableMap.of("add-distinct", "bbb"));
+    assertU(adoc(doc));
+    assertU(commit());
+
+    assertQ(req("q", "cat:*", "indent", "true"), "//result[@numFound = '2']");
+    assertQ(req("q", "cat:bbb", "indent", "true"), "//result[@numFound = '1']");
+    assertQ(req("q", "cat:bbb", "indent", "true"), "//doc/arr[@name='cat'][count(str)=3]");
+
+    doc = new SolrInputDocument();
+    doc.setField("id", "3");
+    doc.setField("cat", ImmutableMap.of("add-distinct", Arrays.asList(new String[]{"bbb", "bbb"})));
+    assertU(adoc(doc));
+    assertU(commit());
+
+    assertQ(req("q", "cat:*", "indent", "true"), "//result[@numFound = '2']");
+    assertQ(req("q", "cat:bbb", "indent", "true"), "//result[@numFound = '1']");
+    assertQ(req("q", "cat:bbb", "indent", "true"), "//doc/arr[@name='cat'][count(str)=3]"); //'bbb' already present will not be added again
+
+    doc = new SolrInputDocument();
+    doc.setField("id", "5");
+    doc.setField("cat", ImmutableMap.of("add-distinct", "bbb"));
+    assertU(adoc(doc));
+    assertU(commit());
+
+    assertQ(req("q", "cat:*", "indent", "true"), "//result[@numFound = '3']");
+    assertQ(req("q", "cat:bbb", "indent", "true"), "//result[@numFound = '2']"); //'cat' field not present, do 'add' atomic operation
+  }
+
   @Test
   public void testSet() throws Exception {
     SolrInputDocument doc;
diff --git a/solr/solr-ref-guide/src/updating-parts-of-documents.adoc b/solr/solr-ref-guide/src/updating-parts-of-documents.adoc
index 3efc00adc79..5e25d51f775 100644
--- a/solr/solr-ref-guide/src/updating-parts-of-documents.adoc
+++ b/solr/solr-ref-guide/src/updating-parts-of-documents.adoc
@@ -40,6 +40,9 @@ May be specified as a single value, or as a list for multiValued fields.
 `add`::
 Adds the specified values to a multiValued field. May be specified as a single value, or as a list.
 
+`add-distinct`::
+Adds the specified values to a multiValued field, only if not already present. May be specified as a single value, or as a list.
+
 `remove`::
 Removes (all occurrences of) the specified values from a multiValued field. May be specified as a single value, or as a list.
 
@@ -67,6 +70,7 @@ If the following document exists in our collection:
  "price":10,
  "popularity":42,
  "categories":["kids"],
+ "sub_categories":["under_5","under_10"],
  "promo_ids":["a123x"],
  "tags":["free_to_try","buy_now","clearance","on_sale"]
 }
@@ -80,6 +84,7 @@ And we apply the following update command:
  "price":{"set":99},
  "popularity":{"inc":20},
  "categories":{"add":["toys","games"]},
+ "sub_categories":{"add-distinct":"under_10"},
  "promo_ids":{"remove":"a123x"},
  "tags":{"remove":["free_to_try","on_sale"]}
 }
@@ -93,6 +98,7 @@ The resulting document in our collection will be:
  "price":99,
  "popularity":62,
  "categories":["kids","toys","games"],
+ "sub_categories":["under_5","under_10"],
  "tags":["buy_now","clearance"]
 }
 ----