Commit 0273daf0 authored by Michael Vanetta's avatar Michael Vanetta Committed by Thomas Seidl
Browse files

Issue #3304560 by recrit, Gertlor, drunken monkey: Fixed problems with... 

Issue #3304560 by recrit, Gertlor, drunken monkey: Fixed problems with "Multiple terms" parse mode and invalid UTF-8 strings.
parent 84f6f6bd

CHANGELOG.txt

+2 −0
Original line number Diff line number Diff line 
Search API 1.x, dev (xxxx-xx-xx):

---------------------------------

- #3304560 by recrit, Gertlor, drunken monkey: Fixed problems with "Multiple

  terms" parse mode and invalid UTF-8 strings.

- #3327383 by unstatu, drunken monkey: Fixed handling of the "view any

  unpublished content" permission in the "Content access" processor.

- #3363208 by drunken monkey: Fixed implicit conversion from float to int in

src/Plugin/search_api/parse_mode/Terms.php

+9 −4
Original line number Diff line number Diff line
@@ -2,6 +2,7 @@ 


namespace Drupal\search_api\Plugin\search_api\parse_mode;



use Drupal\Component\Utility\Unicode;

use Drupal\search_api\ParseMode\ParseModePluginBase;



/**
@@ -19,17 +20,21 @@ class Terms extends ParseModePluginBase { 
   * {@inheritdoc}

   */

  public function parseInput($keys) {

    $ret = [

      '#conjunction' => $this->getConjunction(),

    ];



    if (!Unicode::validateUtf8($keys)) {

      return $ret;

    }

    // Split the keys into tokens. Any whitespace is considered as a delimiter

    // for tokens. This covers ASCII white spaces as well as multi-byte "spaces"

    // which for example are common in Japanese.

    $tokens = preg_split('/\s+/u', $keys);

    $tokens = preg_split('/\s+/u', $keys) ?: [];

    $quoted = FALSE;

    $negated = FALSE;

    $phrase_contents = [];



    $ret = [

      '#conjunction' => $this->getConjunction(),

    ];

    foreach ($tokens as $token) {

      // Ignore empty tokens. (Also helps keep the following code simpler.)

      if ($token === '') {

src/Plugin/views/filter/SearchApiFulltext.php

+7 −1
Original line number Diff line number Diff line
@@ -2,6 +2,7 @@ 


namespace Drupal\search_api\Plugin\views\filter;



use Drupal\Component\Utility\Unicode;

use Drupal\Core\Form\FormStateInterface;

use Drupal\search_api\Entity\Index;

use Drupal\search_api\ParseMode\ParseModePluginManager;
@@ -347,12 +348,17 @@ class SearchApiFulltext extends FilterPluginBase { 
      return;

    }



    if (!Unicode::validateUtf8($input)) {

      $msg = $this->t('Invalid input.');

      $form_state->setErrorByName($identifier, $msg);

    }



    // Only continue if there is a minimum word length set.

    if ($this->options['min_length'] < 2) {

      return;

    }



    $words = preg_split('/\s+/', $input);

    $words = preg_split('/\s+/', $input) ?: [];

    foreach ($words as $i => $word) {

      if (mb_strlen($word) < $this->options['min_length']) {

        unset($words[$i]);

tests/src/Unit/TermsParseModeTest.php

+10 −0
Original line number Diff line number Diff line
@@ -136,4 +136,14 @@ class TermsParseModeTest extends UnitTestCase { 
    ];

  }



  /**

   * Tests that invalid UTF-8 in the input string is handled correctly.

   */

  public function testInvalidInput(): void {

    $parsed = $this->plugin->parseInput("\xc3\x28");

    $this->assertEquals([

      '#conjunction' => 'AND',

    ], $parsed);

  }



}