From ddb0b2520a59011b444584ef988f337117906bcf Mon Sep 17 00:00:00 2001
From: Damian Mooyman <damian@silverstripe.com>
Date: Mon, 6 Jul 2015 17:12:52 +1200
Subject: [PATCH] API Additional support for custom copy_fields API Additional
 textSpellHtml type for html-safe spelling database generation

---
 code/solr/SolrIndex.php        | 40 ++++++++++++++++++++++++++++++----
 conf/solr/4/templates/types.ss | 11 ++++++++++
 docs/en/Solr.md                |  2 +-
 3 files changed, 48 insertions(+), 5 deletions(-)

diff --git a/code/solr/SolrIndex.php b/code/solr/SolrIndex.php
index bea3b1d..0a72114 100644
--- a/code/solr/SolrIndex.php
+++ b/code/solr/SolrIndex.php
@@ -46,6 +46,14 @@ abstract class SolrIndex extends SearchIndex {
 	 * @config
 	 */
 	private static $default_field = '_text';
+
+	/**
+	 * List of copy fields all fulltext fields should be copied into.
+	 * This will fallback to default_field if not specified
+	 *
+	 * @var array
+	 */
+	private static $copy_fields = array();
 	
 	/**
 	 * @return String Absolute path to the folder containing
@@ -104,6 +112,22 @@ abstract class SolrIndex extends SearchIndex {
 		return $this->config()->default_field;
 	}
 
+	/**
+	 * Get list of fields each text field should be copied into.
+	 * This will fallback to the default field if omitted.
+	 *
+	 * @return array
+	 */
+	protected function getCopyDestinations() {
+		$copyFields = $this->config()->copy_fields;
+		if($copyFields) {
+			return $copyFields;
+		}
+		// Fallback to default field
+		$df = $this->getDefaultField();
+		return array($df);
+	}
+
 	public function getFieldDefinitions() {
 		$xml = array();
 		$stored = $this->getStoredDefault();
@@ -365,14 +389,22 @@ abstract class SolrIndex extends SearchIndex {
 		);
 	}
 
-	function getCopyFieldDefinitions() {
+	/**
+	 * Generate XML for copy field definitions
+	 *
+	 * @return string
+	 */
+	public function getCopyFieldDefinitions() {
 		$xml = array();
 
-		$df = $this->getDefaultField();
-		foreach ($this->fulltextFields as $name => $field) {
-			$xml[] = "<copyField source='{$name}' dest='{$df}' />";
+		// Default copy fields
+		foreach($this->getCopyDestinations() as $copyTo) {
+			foreach ($this->fulltextFields as $name => $field) {
+				$xml[] = "<copyField source='{$name}' dest='{$copyTo}' />";
+			}
 		}
 
+		// Explicit copy fields
 		foreach ($this->copyFields as $source => $fields) {
 			foreach($fields as $fieldAttrs) {
 				$xml[] = $this->toXmlTag('copyField', $fieldAttrs);
diff --git a/conf/solr/4/templates/types.ss b/conf/solr/4/templates/types.ss
index 135c8ac..c4081b8 100644
--- a/conf/solr/4/templates/types.ss
+++ b/conf/solr/4/templates/types.ss
@@ -203,6 +203,17 @@
         </analyzer>
     </fieldType>
 
+    <!-- Text optimized for spelling corrections, with minimal alterations (e.g. no stemming) but also html filtering -->
+    <fieldType name="textSpellHtml" class="solr.TextField" positionIncrementGap="100">
+        <analyzer>
+            <charFilter class="solr.HTMLStripCharFilterFactory"/>
+            <tokenizer class="solr.StandardTokenizerFactory" />
+            <filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords.txt"/>
+            <filter class="solr.LengthFilterFactory" min="4" max="20" />
+            <filter class="solr.LowerCaseFilterFactory" /> 
+            <filter class="solr.RemoveDuplicatesTokenFilterFactory" /> 
+        </analyzer>
+    </fieldType>
 
     <!-- A general unstemmed text field - good if one does not know the language of the field -->
     <fieldType name="textgen" class="solr.TextField" positionIncrementGap="100">
diff --git a/docs/en/Solr.md b/docs/en/Solr.md
index eeef086..de66329 100644
--- a/docs/en/Solr.md
+++ b/docs/en/Solr.md
@@ -183,7 +183,7 @@ This can be fixed by aggregating spell checking data in a separate
 			$xml = parent::getFieldDefinitions();
 			
 			$xml .= "\n\n\t\t<!-- Additional custom fields for spell checking -->";
-			$xml .= "\n\t\t<field name='spellcheckData' type='textSpell' indexed='true' stored='false' multiValued='true' />";
+			$xml .= "\n\t\t<field name='spellcheckData' type='textSpellHtml' indexed='true' stored='false' multiValued='true' />";
 
 			return $xml;
 		}