[FEATURE] Get encoding from XML source, resolves #2

Author: fsuter

ChangeLog

@@ -1,6 +1,7 @@
 2026-03-13 Francois Suter (Idéative)  <typo3@ideative.ch>
 
     * Add new "requestOptions" parameter, resolves #14
+    * Get encoding from XML source, resolves #2
 
 2026-02-02 Francois Suter (Idéative)  <typo3@ideative.ch>
 

Classes/Service/ConnectorFeed.php

@@ -183,13 +183,20 @@ protected function query(): mixed
             );
         }
         // Check if the current charset is the same as the file encoding
-        // Don't do the check if no encoding was defined
-        if (empty($this->parameters['encoding'])) {
-            $encoding = null;
-            $isSameCharset = true;
-        } else {
+        // Use explicitly defined encoding first, then try getting encoding from XML declaration
+        if (!empty($this->parameters['encoding'])) {
             $encoding = $this->parameters['encoding'];
             $isSameCharset = $this->getCharset() === $encoding;
+        } else {
+            $dom = new \DOMDocument();
+            $dom->loadXML($data, LIBXML_PARSEHUGE);
+            if (!empty($dom->xmlEncoding)) {
+                $encoding = $dom->xmlEncoding;
+                $isSameCharset = $this->getCharset() === $encoding;
+            } else {
+                $encoding = null;
+                $isSameCharset = true;
+            }
         }
         // If the charset is not the same, convert data
         if (!$isSameCharset) {

Documentation/Configuration/Index.rst

@@ -117,11 +117,6 @@ Description
   the encoding values recognized by the PHP libray "mbstring". See
   https://www.php.net/manual/en/mbstring.supported-encodings.php
 
-  .. warning::
-
-     If your are aiming for the array format (i.e. calling
-     :code:`fetchArray()`), you should not define this property. Indeed the
-     :code:`\Cobweb\Svconnector\Utility\ConnectorUtility::convertXmlToArray()`
-     which is used in this case relies on the SimpleXML library, which
-     already takes care of the encoding conversion. To avoid a double
-     encoding just ignore this property.
+  It is not necessary to define this parameter if the XML source contains an
+  encoding in its declaration. It will automatically be retrieved. However,
+  if this parameter is defined, it supersedes the encoding from the XML declaration.

Documentation/Installation/Index.rst

@@ -29,6 +29,8 @@ It makes it possible to use any of the request options supported by Guzzle HTTP.
 The "headers" parameter has been deprecated. Headers should be passed as part of the
 "requestOptions" instead.
 
+Encoding is now guessed from the declaration of the XML source, if defined.
+
 
 .. _installation-update-500:
 

Tests/Functional/ConnectorFeedTest.php

@@ -71,7 +71,7 @@ public static function sourceDataProvider(): array
 EOT
 
             ],
-            'ISO-8859-1 data' => [
+            'ISO-8859-1 data with explicit encoding' => [
                 'parameters' => [
                     'uri' => 'EXT:svconnector_feed/Tests/Functional/Fixtures/data_latin1.xml',
                     'encoding' => 'iso-8859-1',
@@ -83,6 +83,19 @@ public static function sourceDataProvider(): array
 		<name>Porte interdùm lacîna c'est euismod.</name>
 	</item>
 </items>
+EOT
+            ],
+            'ISO-8859-1 data with implicit encoding' => [
+                'parameters' => [
+                    'uri' => 'EXT:svconnector_feed/Tests/Functional/Fixtures/data_latin1.xml',
+                ],
+                'result' => <<<EOT
+<?xml version="1.0" encoding="ISO-8859-1" standalone="no"?>
+<items>
+	<item>
+		<name>Porte interdùm lacîna c'est euismod.</name>
+	</item>
+</items>
 EOT
             ],
         ];