fix(import): use cocur/slugify library to handle non latin text

2023-11-08 13:51:06 +00:00 · 2023-11-08 13:51:06 +00:00 · 4ca7f9ccae
parent 04b2d8bafa
commit 4ca7f9ccae
4 changed files with 96 additions and 122 deletions
--- a/app/Helpers/misc_helper.php
+++ b/app/Helpers/misc_helper.php
@ -4,6 +4,7 @@ declare(strict_types=1);

 use App\Entities\Person;
 use App\Entities\Podcast;
+use Cocur\Slugify\Slugify;
 use Config\App;
 use Config\Images;
 use Modules\Media\Entities\Image;
@ -42,105 +43,8 @@ if (! function_exists('slugify')) {
            $text = substr($text, 0, strrpos(substr($text, 0, $maxLength), ' '));
        }

-        // replace non letter or digits by -
-        $text = preg_replace('~[^\pL\d]+~u', '-', $text);
-
-        $unwanted = [
-            'Š' => 'S',
-            'š' => 's',
-            'Đ' => 'Dj',
-            'đ' => 'dj',
-            'Ž' => 'Z',
-            'ž' => 'z',
-            'Č' => 'C',
-            'č' => 'c',
-            'Ć' => 'C',
-            'ć' => 'c',
-            'À' => 'A',
-            'Á' => 'A',
-            'Â' => 'A',
-            'Ã' => 'A',
-            'Ä' => 'A',
-            'Å' => 'A',
-            'Æ' => 'AE',
-            'Ç' => 'C',
-            'È' => 'E',
-            'É' => 'E',
-            'Ê' => 'E',
-            'Ë' => 'E',
-            'Ì' => 'I',
-            'Í' => 'I',
-            'Î' => 'I',
-            'Ï' => 'I',
-            'Ñ' => 'N',
-            'Ò' => 'O',
-            'Ó' => 'O',
-            'Ô' => 'O',
-            'Õ' => 'O',
-            'Ö' => 'O',
-            'Ø' => 'O',
-            'Œ' => 'OE',
-            'Ù' => 'U',
-            'Ú' => 'U',
-            'Û' => 'U',
-            'Ü' => 'U',
-            'Ý' => 'Y',
-            'Þ' => 'B',
-            'ß' => 'Ss',
-            'à' => 'a',
-            'á' => 'a',
-            'â' => 'a',
-            'ã' => 'a',
-            'ä' => 'a',
-            'å' => 'a',
-            'æ' => 'ae',
-            'ç' => 'c',
-            'è' => 'e',
-            'é' => 'e',
-            'ê' => 'e',
-            'ë' => 'e',
-            'ì' => 'i',
-            'í' => 'i',
-            'î' => 'i',
-            'ï' => 'i',
-            'ð' => 'o',
-            'ñ' => 'n',
-            'ò' => 'o',
-            'ó' => 'o',
-            'ô' => 'o',
-            'õ' => 'o',
-            'ö' => 'o',
-            'ø' => 'o',
-            'œ' => 'OE',
-            'ù' => 'u',
-            'ú' => 'u',
-            'û' => 'u',
-            'ý' => 'y',
-            'þ' => 'b',
-            'ÿ' => 'y',
-            'Ŕ' => 'R',
-            'ŕ' => 'r',
-            '/' => '-',
-            ' ' => '-',
-        ];
-        $text = strtr($text, $unwanted);
-
-        // transliterate
-        $text = iconv('utf-8', 'us-ascii//TRANSLIT', $text);
-
-        // remove unwanted characters
-        $text = preg_replace('~[^\-\w]+~', '', $text);
-
-        // trim
-        $text = trim($text, '-');
-
-        // remove duplicate -
-        $text = preg_replace('~-+~', '-', $text);
-
-        // lowercase
-        $text = strtolower($text);
-
-        return $text;
+        $slugify = new Slugify();
+        return $slugify->slugify($text);
    }
 }

--- a/composer.json
+++ b/composer.json
@ -7,25 +7,26 @@
  "license": "AGPL-3.0-or-later",
  "require": {
    "php": "^8.1",
-    "codeigniter4/framework": "v4.4.1",
-    "james-heinrich/getid3": "^2.0.0-beta5",
-    "whichbrowser/parser": "^v2.1.7",
-    "geoip2/geoip2": "v2.13.0",
-    "league/commonmark": "^2.4.1",
-    "vlucas/phpdotenv": "v5.5.0",
-    "league/html-to-markdown": "5.1.1",
-    "opawg/user-agents-php": "^v1.0",
    "adaures/ipcat-php": "^v1.0.0",
    "adaures/podcast-persons-taxonomy": "^v1.0.1",
-    "phpseclib/phpseclib": "~2.0.45",
-    "michalsn/codeigniter4-uuid": "v1.0.2",
-    "codeigniter4/settings": "v2.1.2",
-    "chrisjean/php-ico": "^1.0.4",
-    "melbahja/seo": "^v2.1.1",
-    "codeigniter4/shield": "v1.0.0-beta.6",
    "aws/aws-sdk-php": "^3.281.12",
-    "mpratt/embera": "^2.0.34",
+    "chrisjean/php-ico": "^1.0.4",
+    "cocur/slugify": "^4.5",
+    "codeigniter4/framework": "v4.4.1",
+    "codeigniter4/settings": "v2.1.2",
+    "codeigniter4/shield": "v1.0.0-beta.6",
    "codeigniter4/tasks": "dev-develop",
+    "geoip2/geoip2": "v2.13.0",
+    "james-heinrich/getid3": "^2.0.0-beta5",
+    "league/commonmark": "^2.4.1",
+    "league/html-to-markdown": "5.1.1",
+    "melbahja/seo": "^v2.1.1",
+    "michalsn/codeigniter4-uuid": "v1.0.2",
+    "mpratt/embera": "^2.0.34",
+    "opawg/user-agents-php": "^v1.0",
+    "phpseclib/phpseclib": "~2.0.45",
+    "vlucas/phpdotenv": "v5.5.0",
+    "whichbrowser/parser": "^v2.1.7",
    "yassinedoghri/podcast-feed": "dev-main"
  },
  "require-dev": {
--- a/composer.lock
+++ b/composer.lock
@ -4,7 +4,7 @@
    "Read more about it at https://getcomposer.org/doc/01-basic-usage.md#installing-dependencies",
    "This file is @generated automatically"
  ],
-  "content-hash": "2f7475f224d54face5bb527c0edc893a",
+  "content-hash": "b7d9181ef215329ee3c3d2c52e23b2d1",
  "packages": [
    {
      "name": "adaures/ipcat-php",
@ -302,6 +302,75 @@
      },
      "time": "2016-09-27T22:00:56+00:00"
    },
+    {
+      "name": "cocur/slugify",
+      "version": "v4.5.1",
+      "source": {
+        "type": "git",
+        "url": "https://github.com/cocur/slugify.git",
+        "reference": "7c6e088228b9f082050876ae8b0cd287b117b840"
+      },
+      "dist": {
+        "type": "zip",
+        "url": "https://api.github.com/repos/cocur/slugify/zipball/7c6e088228b9f082050876ae8b0cd287b117b840",
+        "reference": "7c6e088228b9f082050876ae8b0cd287b117b840",
+        "shasum": ""
+      },
+      "require": {
+        "ext-mbstring": "*",
+        "php": "~8.0.0 || ~8.1.0 || ~8.2.0 || ~8.3.0"
+      },
+      "conflict": {
+        "symfony/config": "<3.4 || >=4,<4.3",
+        "symfony/dependency-injection": "<3.4 || >=4,<4.3",
+        "symfony/http-kernel": "<3.4 || >=4,<4.3",
+        "twig/twig": "<2.12.1"
+      },
+      "require-dev": {
+        "laravel/framework": "^5.0|^6.0|^7.0|^8.0",
+        "latte/latte": "~2.2",
+        "league/container": "^2.2.0",
+        "mikey179/vfsstream": "~1.6.8",
+        "mockery/mockery": "^1.3",
+        "nette/di": "~2.4",
+        "pimple/pimple": "~1.1",
+        "plumphp/plum": "~0.1",
+        "symfony/config": "^3.4 || ^4.3 || ^5.0 || ^6.0",
+        "symfony/dependency-injection": "^3.4 || ^4.3 || ^5.0 || ^6.0",
+        "symfony/http-kernel": "^3.4 || ^4.3 || ^5.0 || ^6.0",
+        "symfony/phpunit-bridge": "^5.4 || ^6.0",
+        "twig/twig": "^2.12.1 || ~3.0",
+        "zendframework/zend-modulemanager": "~2.2",
+        "zendframework/zend-servicemanager": "~2.2",
+        "zendframework/zend-view": "~2.2"
+      },
+      "type": "library",
+      "autoload": {
+        "psr-4": {
+          "Cocur\\Slugify\\": "src"
+        }
+      },
+      "notification-url": "https://packagist.org/downloads/",
+      "license": ["MIT"],
+      "authors": [
+        {
+          "name": "Florian Eckerstorfer",
+          "email": "florian@eckerstorfer.co",
+          "homepage": "https://florian.ec"
+        },
+        {
+          "name": "Ivo Bathke",
+          "email": "ivo.bathke@gmail.com"
+        }
+      ],
+      "description": "Converts a string into a slug.",
+      "keywords": ["slug", "slugify"],
+      "support": {
+        "issues": "https://github.com/cocur/slugify/issues",
+        "source": "https://github.com/cocur/slugify/tree/v4.5.1"
+      },
+      "time": "2023-09-17T07:26:20+00:00"
+    },
    {
      "name": "codeigniter4/framework",
      "version": "v4.4.1",
@ -6517,9 +6586,9 @@
  "aliases": [],
  "minimum-stability": "stable",
  "stability-flags": {
-    "james-heinrich/getid3": 10,
    "codeigniter4/shield": 10,
    "codeigniter4/tasks": 20,
+    "james-heinrich/getid3": 10,
    "yassinedoghri/podcast-feed": 20
  },
  "prefer-stable": true,
--- a/modules/PodcastImport/Commands/PodcastImport.php
+++ b/modules/PodcastImport/Commands/PodcastImport.php
@ -162,7 +162,7 @@ class PodcastImport extends BaseCommand

            $podcastModel = new PodcastModel();
            if (! $podcastModel->update($this->podcast->id, $this->podcast)) {
-                throw new Exception(print_r($podcastModel->errors()));
+                throw new Exception((string) print_r($podcastModel->errors()));
            }

            CLI::showProgress(false);
@ -250,7 +250,7 @@ class PodcastImport extends BaseCommand
        $podcastModel = new PodcastModel();
        if (! ($podcastId = $podcastModel->insert($podcast, true))) {
            $db->transRollback();
-            throw new Exception(print_r($podcastModel->errors()));
+            throw new Exception((string) print_r($podcastModel->errors()));
        }

        $podcast->id = $podcastId;
@ -347,7 +347,7 @@ class PodcastImport extends BaseCommand
                $personGroupSlug,
                $personRoleSlug
            )) {
-                throw new Exception(print_r($podcastPersonModel->errors()));
+                throw new Exception((string) print_r($podcastPersonModel->errors()));
            }
        }

@ -485,7 +485,7 @@ class PodcastImport extends BaseCommand

            if (! ($episodeId = $episodeModel->insert($episode, true))) {
                $db->transRollback();
-                throw new Exception(print_r($episodeModel->errors()));
+                throw new Exception((string) print_r($episodeModel->errors()));
            }

            $this->importEpisodePersons($episodeId, $item->podcast_persons);
@ -535,7 +535,7 @@ class PodcastImport extends BaseCommand
                ]);

                if (! ($newPersonId = $personModel->insert($newPerson))) {
-                    throw new Exception(print_r($personModel->errors()));
+                    throw new Exception((string) print_r($personModel->errors()));
                }
            }

@ -567,7 +567,7 @@ class PodcastImport extends BaseCommand
                $personGroupSlug,
                $personRoleSlug
            )) {
-                throw new Exception(print_r($episodePersonModel->errors()));
+                throw new Exception((string) print_r($episodePersonModel->errors()));
            }
        }
    }