Changeset 61841

Timestamp:

03/05/2026 10:17:34 AM (8 days ago)

Author:

audrasjb

Message:

Media: Import images Alt text IPTC metadata.

This changeset introduce a new wp_get_image_alttext() function that extracts Image Alt text metadata from image IPTC metadata.

Props jhmonroe, rishabhwpn, adamsilverstein, sajjad67, ozgursar, joedolson, audrasjb, huzaifaalmesbah, sabernhardt, valentingrenier, louischan, penelopeadrian, mathiaspeguet.
Fixes #63895.

Location:

trunk

Files:

• src/wp-admin/includes/image.php (modified) (4 diffs)
• src/wp-admin/includes/media.php (modified) (6 diffs)
• src/wp-includes/rest-api/endpoints/class-wp-rest-attachments-controller.php (modified) (3 diffs)
• tests/phpunit/tests/image/meta.php (modified) (4 diffs)

trunk/src/wp-admin/includes/image.php

@@ -814 +814 @@
  *
  * The IPTC metadata that is retrieved is APP13, credit, byline, created date
- * and time, caption, copyright, and title. Also includes FNumber, Model,
+ * and time, caption, copyright, alt, and title. Also includes FNumber, Model,
  * DateTimeDigitized, FocalLength, ISOSpeedRatings, and ExposureTime.
  *
@@ -855 +855 @@
         'orientation'       => 0,
         'keywords'          => array(),
+        'alt'               => '',
     );
 
@@ -926 +927 @@
         }
     }
+
+    $meta['alt'] = wp_get_image_alttext( $file );
 
     $exif = array();
@@ -1073 +1076 @@
      */
     return apply_filters( 'wp_read_image_metadata', $meta, $file, $image_type, $iptc, $exif );
+}
+
+/**
+ * Gets the alt text from image meta data.
+ *
+ * @since 7.0.0
+ *
+ * @param string $file File path to the image.
+ * @return string Embedded alternative text.
+ */
+function wp_get_image_alttext( $file ) {
+    $alt_text     = '';
+    $img_contents = file_get_contents( $file );
+    // Find the start and end positions of the XMP metadata.
+    $xmp_start = strpos( $img_contents, '<x:xmpmeta' );
+    $xmp_end   = strpos( $img_contents, '</x:xmpmeta>' );
+
+    if ( ! $xmp_start || ! $xmp_end ) {
+        // No XMP metadata found.
+        return $alt_text;
+    }
+
+    // Extract the XMP metadata from the JPEG contents
+    $xmp_data = substr( $img_contents, $xmp_start, $xmp_end - $xmp_start + 12 );
+
+    // Parse the XMP metadata using DOMDocument.
+    $doc = new DOMDocument();
+    if ( false === $doc->loadXML( $xmp_data ) ) {
+        // Invalid XML in metadata.
+        return $alt_text;
+    }
+
+    // Instantiate an XPath object, used to extract portions of the XMP.
+    $xpath = new DOMXPath( $doc );
+
+    // Register the relevant XML namespaces.
+    $xpath->registerNamespace( 'x', 'adobe:ns:meta/' );
+    $xpath->registerNamespace( 'rdf', 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' );
+    $xpath->registerNamespace( 'Iptc4xmpCore', 'http://iptc.org/std/Iptc4xmpCore/1.0/xmlns/' );
+
+    $node_list = $xpath->query( '/x:xmpmeta/rdf:RDF/rdf:Description/Iptc4xmpCore:AltTextAccessibility' );
+    if ( $node_list && $node_list->count() ) {
+
+        $node = $node_list->item( 0 );
+
+        // Get the site's locale.
+        $locale = get_locale();
+
+        // Get the alt text accessibility alternative most appropriate for the site language.
+        // There are 3 possibilities:
+        //
+        // 1. there is an rdf:li with an exact match on the site locale.
+        // 2. there is an rdf:li with a partial match on the site locale (e.g., site locale is en_US and rdf:li has @xml:lang="en").
+        // 3. there is an rdf:li with an "x-default" lang.
+        //
+        // Evaluate in that order, stopping when we have a match.
+        $alt_text = $xpath->evaluate( "string( rdf:Alt/rdf:li[ @xml:lang = '{$locale}' ] )", $node );
+        if ( ! $alt_text ) {
+            $alt_text = $xpath->evaluate( 'string( rdf:Alt/rdf:li[ @xml:lang = "' . substr( $locale, 0, 2 ) . '" ] )', $node );
+            if ( ! $alt_text ) {
+                $alt_text = $xpath->evaluate( 'string( rdf:Alt/rdf:li[ @xml:lang = "x-default" ] )', $node );
+            }
+        }
+    }
+
+    return $alt_text;
 }
 

trunk/src/wp-admin/includes/media.php

@@ -320 +320 @@
     $content = '';
     $excerpt = '';
+    $alt     = '';
 
     if ( preg_match( '#^audio#', $type ) ) {
@@ -400 +401 @@
                 $excerpt = $image_meta['caption'];
             }
+
+            if ( trim( $image_meta['alt'] ) ) {
+                $alt = $image_meta['alt'];
+            }
         }
     }
@@ -422 +427 @@
     $attachment_id = wp_insert_attachment( $attachment, $file, $post_id, true );
 
+    if ( trim( $alt ) ) {
+        update_post_meta( $attachment_id, '_wp_attachment_image_alt', sanitize_text_field( $alt ) );
+    }
+
     if ( ! is_wp_error( $attachment_id ) ) {
         /*
@@ -478 +487 @@
     $title   = preg_replace( '/\.[^.]+$/', '', wp_basename( $file ) );
     $content = '';
+    $alt     = '';
 
     // Use image exif/iptc data for title and caption defaults if possible.
@@ -489 +499 @@
         if ( trim( $image_meta['caption'] ) ) {
             $content = $image_meta['caption'];
+        }
+        if ( trim( $image_meta['alt'] ) ) {
+            $alt = $image_meta['alt'];
         }
     }
@@ -513 +526 @@
     // Save the attachment metadata.
     $attachment_id = wp_insert_attachment( $attachment, $file, $post_id, true );
+
+    if ( trim( $alt ) ) {
+        update_post_meta( $attachment_id, '_wp_attachment_image_alt', sanitize_text_field( $alt ) );
+    }
 
     if ( ! is_wp_error( $attachment_id ) ) {

trunk/src/wp-includes/rest-api/endpoints/class-wp-rest-attachments-controller.php

@@ -438 +438 @@
         $type = $file['type'];
         $file = $file['file'];
+        $alt  = '';
 
         // Include image functions to get access to wp_read_image_metadata().
@@ -452 +453 @@
             if ( empty( $request['caption'] ) && trim( $image_meta['caption'] ) ) {
                 $request['caption'] = $image_meta['caption'];
+            }
+
+            if ( empty( $request['alt'] ) && trim( $image_meta['alt'] ) ) {
+                $alt = $image_meta['alt'];
             }
         }
@@ -477 +482 @@
         // $post_parent is inherited from $attachment['post_parent'].
         $id = wp_insert_attachment( wp_slash( (array) $attachment ), $file, 0, true, false );
+
+        if ( trim( $alt ) ) {
+            update_post_meta( $id, '_wp_attachment_image_alt', sanitize_text_field( $alt ) );
+        }
 
         if ( is_wp_error( $id ) ) {

trunk/tests/phpunit/tests/image/meta.php

@@ -128 +128 @@
         $this->assertSame( '0', $out['shutter_speed'], 'Shutter speed value not equivalent' );
         $this->assertSame( '', $out['title'], 'Title value not the same' );
+    }
+
+    /**
+     * @ticket 63895
+     */
+    public function test_iptc_alt() {
+        // Image tests alt text from the IPTC photo metadata standard 2025.1.
+        $out = wp_read_image_metadata( DIR_TESTDATA . '/images/IPTC-PhotometadataRef-Std2025.1.jpg' );
+
+        $this->assertSame( 'This is the Alt Text description to support accessibility in 2025.1', $out['alt'], 'Alt text does not match source.' );
     }
 
@@ -201 +211 @@
                     'orientation'       => '3',
                     'keywords'          => array(),
+                    'alt'               => '',
                 ),
             ),
@@ -218 +229 @@
                     'orientation'       => '0',
                     'keywords'          => array(),
+                    'alt'               => '',
                 ),
             ),
@@ -235 +247 @@
                     'orientation'       => '1',
                     'keywords'          => array( 'beach', 'baywatch', 'LA', 'sunset' ),
+                    'alt'               => '',
                 ),
             ),