Laravel学习笔记之Seeder填充数据小技巧

说明:本文主要聊一聊Laravel测试数据填充器Seeder的小技巧,同时介绍下Laravel开发插件三件套,这三个插件挺好用哦。同时,作者会将开发过程中的一些截图和代码黏上去,提高阅读效率。

备注:在设计个人博客软件时,总会碰到有分类Category、博客Post、给博客贴的标签Tag、博客内容的评论Comment。
而且,Category与Post是一对多关系One-Many:一个分类下有很多Post,一个Post只能归属于一个Category;Post与Comment是一对多关系One-Many:一篇博客Post下有很多Comment,一条Comment只能归属于一篇Post;Post与Tag是多对多关系Many-Many:一篇Post有很多Tag,一个Tag下有很多Post。
开发环境:Laravel5.2 + MAMP + PHP7 + MySQL5.5

开发插件三件套

在先聊测试数据填充器seeder之前,先装上开发插件三件套,开发神器。先不管这能干些啥,装上再说。
1、barryvdh/laravel-debugbar

composer require barryvdh/laravel-debugbar --dev

2、barryvdh/laravel-ide-helper

composer require barryvdh/laravel-ide-helper --dev

3、mpociot/laravel-test-factory-helper

composer require mpociot/laravel-test-factory-helper --dev

然后在config/app.php文件中填上:

        /          *Develop Plugin        */        Barryvdh\Debugbar\ServiceProvider::class,        Mpociot\LaravelTestFactoryHelper\TestFactoryHelperServiceProvider::class,        Barryvdh\LaravelIdeHelper\IdeHelperServiceProvider::class,

设计表的字段和关联

设计字段

按照上文提到的Category、Post、Comment和Tag之间的关系创建迁移Migration和模型Model,在项目根目录输入:

php artisan make:model Category -mphp artisan make:model Post -mphp artisan make:model Comment -mphp artisan make:model Tag -m

在各个表的迁移migrations文件中根据表的功能设计字段:

//Category表class CreateCategoriesTable extends Migration{    /      * Run the migrations.     *     * @return void     */    public function up()    {        Schema::create('categories', function (Blueprint $table) {            $table->increments('id');            $table->string('name')->comment('分类名称');            $table->integer('hot')->comment('分类热度');            $table->string('image')->comment('分类图片');            $table->timestamps();        });    }    /      * Reverse the migrations.     *     * @return void     */    public function down()    {        Schema::drop('categories');    }}//Post表class CreatePostsTable extends Migration{    /      * Run the migrations.     *     * @return void     */    public function up()    {        Schema::create('posts', function (Blueprint $table) {            $table->increments('id');            $table->integer('category_id')->unsigned()->comment('外键');            $table->string('title')->comment('标题');            $table->string('slug')->unique()->index()->comment('锚点');            $table->string('summary')->comment('概要');            $table->text('content')->comment('内容');            $table->text('origin')->comment('文章来源');            $table->integer('comment_count')->unsigned()->comment('评论次数');            $table->integer('view_count')->unsigned()->comment('浏览次数');            $table->integer('favorite_count')->unsigned()->comment('点赞次数');            $table->boolean('published')->comment('文章是否发布');            $table->timestamps();            //Post表中category_id字段作为外键,与Category一对多关系            $table->foreign('category_id')                  ->references('id')                  ->on('categories')                  ->onUpdate('cascade')                  ->onDelete('cascade');        });    }    /      * Reverse the migrations.     *     * @return void     */    public function down()    {        //删除表时要删除外键约束,参数为外键名称        Schema::table('posts', function(Blueprint $tabel){            $tabel->dropForeign('posts_category_id_foreign');        });        Schema::drop('posts');    }}//Comment表class CreateCommentsTable extends Migration{    /      * Run the migrations.     *     * @return void     */    public function up()    {        Schema::create('comments', function (Blueprint $table) {            $table->increments('id');            $table->integer('post_id')->unsigned()->comment('外键');            $table->integer('parent_id')->comment('父评论id');            $table->string('parent_name')->comment('父评论标题');            $table->string('username')->comment('评论者用户名');            $table->string('email')->comment('评论者邮箱');            $table->string('blog')->comment('评论者博客地址');            $table->text('content')->comment('评论内容');            $table->timestamps();            //Comment表中post_id字段作为外键,与Post一对多关系            $table->foreign('post_id')                  ->references('id')                  ->on('posts')                  ->onUpdate('cascade')                  ->onDelete('cascade');        });    }    /      * Reverse the migrations.     *     * @return void     */    public function down()    {    //删除表时要删除外键约束,参数为外键名称        Schema::table('comments', function(Blueprint $tabel){            $tabel->dropForeign('comments_post_id_foreign');        });        Schema::drop('comments');    }}//Tag表class CreateTagsTable extends Migration{    /      * Run the migrations.     *     * @return void     */    public function up()    {        Schema::create('tags', function (Blueprint $table) {            $table->increments('id');            $table->string('name')->comment('标签名称');            $table->integer('hot')->unsigned()->comment('标签热度');            $table->timestamps();        });    }    /      * Reverse the migrations.     *     * @return void     */    public function down()    {        Schema::drop('tags');    }}

由于Post表与Tag表是多对多关系,还需要一张存放两者关系的表:

//多对多关系,中间表的命名laravel默认按照两张表字母排序来的,写成tag_post会找不到中间表php artisan make:migration create_post_tag_table --create=post_tag

然后填上中间表的字段:

class CreatePostTagTable extends Migration{    /      * Run the migrations.     *     * @return void     */    public function up()    {        Schema::create('post_tag', function (Blueprint $table) {            $table->increments('id');            $table->integer('post_id')->unsigned();            $table->integer('tag_id')->unsigned();            $table->timestamps();            //post_id字段作为外键            $table->foreign('post_id')                  ->references('id')                  ->on('posts')                  ->onUpdate('cascade')                  ->onDelete('cascade');            //tag_id字段作为外键                  $table->foreign('tag_id')                  ->references('id')                  ->on('tags')                  ->onUpdate('cascade')                  ->onDelete('cascade');        });    }    /      * Reverse the migrations.     *     * @return void     */    public function down()    {        Schema::table('post_tag', function(Blueprint $tabel){            $tabel->dropForeign('post_tag_post_id_foreign');            $tabel->dropForeign('post_tag_tag_id_foreign');        });        Schema::drop('post_tag');    }}

设计关联

写上Migration后,还得在Model里写上关联:

class Category extends Model{    //Category-Post:One-Many    public function posts()    {        return $this->hasMany(Post::class);    }}class Post extends Model{    //Post-Category:Many-One    public function category()    {        return $this->belongsTo(Category::class);    }    //Post-Comment:One-Many    public function comments()    {        return $this->hasMany(Comment::class);    }    //Post-Tag:Many-Many    public function tags()    {        return $this->belongsToMany(Tag::class)->withTimestamps();    }}class Comment extends Model{    //Comment-Post:Many-One    public function post()    {        return $this->belongsTo(Post::class);    }}class Tag extends Model{    //Tag-Post:Many-Many    public function posts()    {        return $this->belongsToMany(Post::class)->withTimestamps();    }}

然后执行迁移:

php artisan migrate

数据库中会生成新建表,表的关系如下:

Seeder填充测试数据

好,在聊到seeder测试数据填充之前,看下开发插件三件套能干些啥,下文中命令可在项目根目录输入php artisan指令列表中查看。
1、barryvdh/laravel-ide-helper
执行php artisan ide-helper:generate指令前:

执行php artisan ide-helper:generate指令后:

不仅Facade模式的Route由之前的反白了变为可以定位到源码了,而且输入Config Facade时还方法自动补全auto complete,这个很方便啊。

输入指令php artisan ide-helper:models后,看看各个Model,如Post这个Model:

belongsTo(Category::class);    }    //Post-Comment:One-Many    public function comments()    {        return $this->hasMany(Comment::class);    }    //Post-Tag:Many-Many    public function tags()    {        return $this->belongsToMany(Tag::class)->withTimestamps();    }}

根据迁移到库里的表生成字段属性和对应的方法提示,在控制器里输入方法时会自动补全auto complete字段属性的方法:

2、mpociot/laravel-test-factory-helper
输入指令php artisan test-factory-helper:generate后,database/factory/ModelFactory.php模型工厂文件会自动生成各个模型对应字段数据。Faker是一个好用的生成假数据的第三方库,而这个开发插件会自动帮你生成这些属性,不用自己写了。

define(App\User::class, function (Faker\Generator $faker) {    return [        'name' => $faker->name,        'email' => $faker->safeEmail,        'password' => bcrypt(str_random(10)),        'remember_token' => str_random(10),    ];});$factory->define(App\Category::class, function (Faker\Generator $faker) {    return [        'name' =>  $faker->name ,        'hot' =>  $faker->randomNumber() ,        'image' =>  $faker->word ,    ];});$factory->define(App\Comment::class, function (Faker\Generator $faker) {    return [        'post_id' =>  function () {             return factory(App\Post::class)->create()->id;        } ,        'parent_id' =>  $faker->randomNumber() ,        'parent_name' =>  $faker->word ,        'username' =>  $faker->userName ,        'email' =>  $faker->safeEmail ,        'blog' =>  $faker->word ,        'content' =>  $faker->text ,    ];});$factory->define(App\Post::class, function (Faker\Generator $faker) {    return [        'category_id' =>  function () {             return factory(App\Category::class)->create()->id;        } ,        'title' =>  $faker->word ,        'slug' =>  $faker->slug ,//修改为slug        'summary' =>  $faker->word ,        'content' =>  $faker->text ,        'origin' =>  $faker->text ,        'comment_count' =>  $faker->randomNumber() ,        'view_count' =>  $faker->randomNumber() ,        'favorite_count' =>  $faker->randomNumber() ,        'published' =>  $faker->boolean ,    ];});$factory->define(App\Tag::class, function (Faker\Generator $faker) {    return [        'name' =>  $faker->name ,        'hot' =>  $faker->randomNumber() ,    ];});

在聊第三个debugbar插件前先聊下seeder小技巧,用debugbar来帮助查看。Laravel官方推荐使用模型工厂自动生成测试数据,推荐这么写的:

//先输入指令生成database/seeds/CategoryTableSeeder.php文件: php artisan make:seeder CategoryTableSeedercreate()->each(function($category){            $category->posts()->save(factory(\App\Post::class)->make());        });    }}//然后php artisan db:seed执行数据填充

但是这种方式效率并不高,因为每一次create()都是一次query,而且每生成一个Category也就对应生成一个Post,当然可以在each()里每一次Category继续foreach()生成几个Post,但每一次foreach也是一次query,效率更差。可以用debugbar小能手看看。先在DatabaseSeeder.php文件中填上这次要填充的Seeder:

    public function run()    {        // $this->call(UsersTableSeeder::class);        $this->call(CategoryTableSeeder::class);    }

在路由文件中写上:

Route::get('/artisan', function () {    $exitCode = Artisan::call('db:seed');    return $exitCode;});

输入路由/artisan后用debugbar查看执行了15次query,耗时7.11ms:

实际上才刚刚输入几个数据呢,Category插入了10个,Post插入了5个。
可以用DB::table()->insert()批量插入,拷贝ModelFactory.php中表的字段定义放入每一个表对应Seeder,当然可以有些字段为便利也适当修改对应假数据。

class CategoryTableSeeder extends Seeder{    /      * Run the database seeds.     *     * @return void     */    public function run()    {//        factory(\App\Category::class, 20)->create()->each(function($category){//            $category->posts()->save(factory(\App\Post::class)->make());//        });        $faker = Faker\Factory::create();        $datas = [];        foreach (range(1, 10) as $key => $value) {            $datas[] = [                'name' =>  'category'.$faker->randomNumber() ,                'hot' =>  $faker->randomNumber() ,                'image' =>  $faker->url ,                'created_at' => \Carbon\Carbon::now()->toDateTimeString(),                'updated_at' => \Carbon\Carbon::now()->toDateTimeString()            ];        }        DB::table('categories')->insert($datas);    }}class PostTableSeeder extends Seeder{    /      * Run the database seeds.     *     * @return void     */    public function run()    {        $faker = Faker\Factory::create();        $category_ids = \App\Category::lists('id')->toArray();        $datas = [];        foreach (range(1, 10) as $key => $value) {            $datas[] = [                'category_id' => $faker->randomElement($category_ids),                'title' =>  $faker->word ,                'slug' =>  $faker->slug ,                'summary' =>  $faker->word ,                'content' =>  $faker->text ,                'origin' =>  $faker->text ,                'comment_count' =>  $faker->randomNumber() ,                'view_count' =>  $faker->randomNumber() ,                'favorite_count' =>  $faker->randomNumber() ,                'published' =>  $faker->boolean ,                'created_at' => \Carbon\Carbon::now()->toDateTimeString(),                'updated_at' => \Carbon\Carbon::now()->toDateTimeString()            ];        }        DB::table('posts')->insert($datas);    }}class CommentTableSeeder extends Seeder{    /      * Run the database seeds.     *     * @return void     */    public function run()    {        $faker = Faker\Factory::create();        $post_ids = \App\Post::lists('id')->toArray();        $datas = [];        foreach (range(1, 50) as $key => $value) {            $datas[] = [                'post_id' => $faker->randomElement($post_ids),                'parent_id' =>  $faker->randomNumber() ,                'parent_name' =>  $faker->word ,                'username' =>  $faker->userName ,                'email' =>  $faker->safeEmail ,                'blog' =>  $faker->word ,                'content' =>  $faker->text ,                'created_at' => \Carbon\Carbon::now()->toDateTimeString(),                'updated_at' => \Carbon\Carbon::now()->toDateTimeString()            ];        }        DB::table('comments')->insert($datas);    }}class TagTableSeeder extends Seeder{    /      * Run the database seeds.     *     * @return void     */    public function run()    {        $faker = Faker\Factory::create();        $datas = [];        foreach (range(1, 10) as $key => $value) {            $datas[] = [                'name' =>  'tag'.$faker->randomNumber() ,                'hot' =>  $faker->randomNumber() ,                'created_at' => \Carbon\Carbon::now()->toDateTimeString(),                'updated_at' => \Carbon\Carbon::now()->toDateTimeString()            ];        }        DB::table('tags')->insert($datas);    }}class PostTagTableSeeder extends Seeder{    /      * Run the database seeds.     *     * @return void     */    public function run()    {        $faker = Faker\Factory::create();        $post_ids = \App\Post::lists('id')->toArray();        $tag_ids = \App\Tag::lists('id')->toArray();        $datas = [];        foreach (range(1, 20) as $key => $value) {            $datas[] = [                'post_id' =>  $faker->randomElement($post_ids) ,                'tag_id' =>  $faker->randomElement($tag_ids) ,                'created_at' => \Carbon\Carbon::now()->toDateTimeString(),                'updated_at' => \Carbon\Carbon::now()->toDateTimeString()            ];        }        DB::table('post_tag')->insert($datas);    }}

在DatabaseSeeder.php中按照顺序依次填上Seeder,顺序不能颠倒,尤其有关联关系的表:

class DatabaseSeeder extends Seeder{    /      * Run the database seeds.     *     * @return void     */    public function run()    {        // $this->call(UsersTableSeeder::class);        $this->call(CategoryTableSeeder::class);        $this->call(PostTableSeeder::class);        $this->call(CommentTableSeeder::class);        $this->call(TagTableSeeder::class);        $this->call(PostTagTableSeeder::class);    }}

输入路由/artisan后,生成了10个Category、10个Post、50个Comments、10个Tag和PostTag表中多对多关系,共有9个Query耗时13.52ms:

It is working!!!

表的迁移Migration和关联Relationship都已设计好,测试数据也已经Seeder好了,就可以根据Repository模式来设计一些数据库逻辑了。准备趁着端午节研究下Repository模式的测试,PHPUnit结合Mockery包来TDD测试也是一种不错的玩法。
M(Model)-V(View)-C(Controller)模式去组织代码,很多时候也未必指导性很强,给Model加一个Repository,给Controller加一个Service,给View加一个Presenter,或许代码结构更清晰。具体可看下面分享的一篇文章。

最近一直在给自己充电,研究MySQL,PHPUnit,Laravel,上班并按时打卡,看博客文章,每天喝红牛。很多不会,有些之前没咋学过,哎,头疼。后悔以前读书太少,书到用时方恨少,人丑还需多读书。

研究生学习机器人的,本打算以后读博搞搞机器人的(研一时真是这么想真是这么准备的,too young too simple)。现在做PHP小码农了,只因当时看到智能机就激动得不行,决定以后做个码农试试吧,搞不好是条生路,哈哈。读书时觉悟太晚,耗费了青春,其实我早该踏入这条路的嘛,呵呵。Follow My Heart!

不扯了,在凌晨两点边听音乐边写博客,就容易瞎感慨吧。。

分享下最近发现的一张好图和一篇极赞的文章:

文章链接:Laravel的中大型專案架構

关键字:php, seed

版权声明

本文来自互联网用户投稿,文章观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处。如若内容有涉嫌抄袭侵权/违法违规/事实不符,请点击 举报 进行投诉反馈!

立即
投稿

微信公众账号

微信扫一扫加关注

返回
顶部