This is a backend service for vue-storefront. Provides data access to product catalog (via ElasticSearch) and allows users to place orders into order queue (by default it's Redis queue supported via kqueue library).

Vue Storefront is a standalone PWA (Progressive Web Application ) storefront for your eCommerce, possible to connect with any eCommerce backend (eg. Magento, Prestashop or Shopware) through the API.

Vue Storefront is and always will be in the open source. Anyone can use and support the project, we want it to be a tool for the improvement of the shopping experience. The project is still in the prove of concept phase. We are looking for Contributors and Designer willing to help us the the solution development.

Vue Storefront was build as a all-in-one front-end for eCommerce. For providing the best performance we decided to use Vue.js as a front-end library, Node.js + Express (and maybe GraphQL support) as a server-API, Elastic Search as a database of products and full PWA/off-line support. Here you can read more about the proof of concept for Vue Storefront connected with Magento2.

Besides a big improvement for the shopping experience, we also want to create a great code base for every developer who needs to work on a front-end application for the eCommerce.

• Docker and Docker Compose

Already included in vue-storefront-api Docker image (required locally, if you do not use containerization):

• Node.js 10.x or higher
• Yarn

Start a containerized environment

The legacy (A) mode - starting just the Elastic and Redis containers: docker-compose up -d

The standard (B) mode - starting Elastic, Redis + Vue Storefront API containers: docker-compose -f docker-compose.yml -f docker-compose.nodejs.yml up -d

As a result, all necessary services will be launched:

• Vue Storefront API runtime environment (Node.js with dependencies from package.json)
• ElasticSearch
• Redis
• Kibana (optional)

Import product catalog

Product catalog is imported using elasticdump, which is installed automatically via project dependency. The default ElasticSearch index name is: vue_storefront_catalog

• (A) yarn restore
• (B) docker exec -it vuestorefrontapi_app_1 yarn restore

Then, to update the structures in the database to the latest version (data migrations), do the following:

• (A) yarn migrate
• (B) docker exec -it vuestorefrontapi_app_1 yarn migrate

By default, the application server is started in development mode. It means that code auto reload is enabled along with ESLint, babel support.

It restores JSON documents stored in ./var/catalog.json. The opposite command - used to generate catalog.json file from running ElasticSearch cluster:

• (A) yarn dump
• (B) docker exec -it vuestorefrontapi_app_1 yarn dump

Access ElasticSearch data with Kibana

A Kibana service is available to explore, search and visualize indexed data at the following url:

http://localhost:5601/

At first access it will ask to specify an index pattern, insert vue_storefront*

Note: Kibana is not provided with the Elastic 7 docker file. Please install it on your own when needed or check the es-head Chrome plugin.

By default, Vue Storefront API docker files and config are based on Elastic 5.6. We plan to change the default Elastic version to 7 with the 1.11 stable release. As for now, the Elastic 7 support is marked as experimental.

How to use it?

In order to use the new Elastic 7 please make sure your vue-storefront-api has been checked out from the PR#342 branch. To start the Elastic 7 docker service please use the docker-compose.elastic7.yml file provided:

docker-compose -f docker-compose.elastic7.yml up

Then, please do change the config/local.json to start using the new Elastic API. Key properties in the elasticsearch section are: indexTypes and apiVersion (to be set to 7.1). If you're using the multistore configuration please make sure you adjusted the storeViews.*.elasticsearch section as well - per each separate store.

  "elasticsearch": {
    "host": "localhost",
    "port": 9200,
    "protocol": "http",
    "min_score": 0.01,
    "indices": [
      "vue_storefront_catalog",
      "vue_storefront_catalog_de",
      "vue_storefront_catalog_it"
    ],
    "indexTypes": [
      "product",
      "category",
      "cms_block",
      "cms_page",
      "attribute",
      "taxrule",
      "review"
    ],
    "apiVersion": "7.1"
  }

Starting from Elasitc 6 and 7 we can have just single document type per single index. Vue Storefront used to have product, category ... types defined in the vue_storefront_catalog.

From now on, we're using the separate indexes per each entity type. The convention is: ${indexName}_${entityType}. If your' logical index name is vue_storefront_catalog then it will be mapped to the physical indexes of: vue_storefront_catalog_product, vue_storefront_catalog_category ...

To take the advantage of this new logical->physical index distinction we've provided new Elastic tools: db7, migrate7, dump7, restore7 tools. They can be used exactly the same way like the old tools were. The only distinction is that they work on separate indexes.

Create new index

Before restoreing or importing data you might want to create a new Elastic index with the proper data types/schema applied. You can run just the yarn db7 new command in order to do that.

Restore the data

After you ran the docker file and have Elastic7 up and running you might want to:

a) restore the demo data by running yarn restore7 and yarn restore7_it; yarn restore7_de for default multistores. The data is restored from var/catalog_product.json, var/catalog_category.json and so on...

b) import the data from Magento to proper physical indexes. To do so, currently you can do this only with mage2vuestorefront.

Note: After 1.11 stable release (around November, 2019) we'll replace the standard tools: db, migrate, dump, restore with the Elastic 7 equivalents and it will become the default.

Catalog API calls are compliant with ElasticSearch (it works like a filtering proxy to ES). More on ES queries: ElasticSearch queries tutorial

Elastic search endpoint: http://localhost:8080/api/catalog/search/<INDEX_NAME>/. You can run the following command to check if everything is up and runing (it assumes vue_storefront_catalog as default index name):

curl -i http://elastic:changeme@localhost:8080/api/catalog/vue_storefront_catalog/_search

This backend is using ElasticSearch data formats popularized by ElasticSuite for Magento2 by Smile.fr.

• Product model
• Category model

Please use data migration mechanism provided to manipulate Redis, ElasticSearch or kue. Details: https://github.com/DivanteLtd/vue-storefront-api/tree/master/doc

When adding custom Extensions to the API you might want to define some dependencies inside them. Thanks to Yarn workspaces dependencies defined inside your custom module will be installed when you execute yarn at project root level, so it's way easier and faster than installing all modules dependencies separately.

To do this, define the package.json with your dependencies in your custom module:

• src/api/extensions/{your-custom-extension}/package.json
• src/platforms/{your-custom-platform}/package.json

Executing docker exec -it vue-storefrontapiapp_1 yarn install will also download your custom modules dependencies.

NOTE: npm users will still have to install the dependencies individually in their modules.

To use review feature you need to install custom module for Magento 2: Divante ReviewApi By default new reviews will be added with status "Pending".

  "review": {
    "defaultReviewStatus": 2 
  },

Vue Storefront API supports output cache for catalog operations. Cache is tagged and can by dynamically invalidated. Please find the details how to configure it in our docs.

You can manually clear the Redis cache for specific tags by running the following command:

npm run cache clear
npm run cache clear -- --tag=product,category
npm run cache clear -- --tag=P198
npm run cache clear -- --tag=*

Magento2 data import is now integrated into vue-storefront-api for simplicity. It's still managed by the mage2vuestorefront - added as a dependency to vue-storefront-api.

After setting the config.magento2.api section using Yours Magento2 oauth credentials:

  "magento2": {
    "imgUrl": "http://localhost:8080/media/catalog/product",
    "assetPath": "/../var/magento2-sample-data/pub/media",
    "api": {
      "url": "http://demo-magento2.vuestorefront.io/rest",
      "consumerKey": "byv3730rhoulpopcq64don8ukb8lf2gq",
      "consumerSecret": "u9q4fcobv7vfx9td80oupa6uhexc27rb",
      "accessToken": "040xx3qy7s0j28o3q0exrfop579cy20m",
      "accessTokenSecret": "7qunl3p505rubmr7u1ijt7odyialnih9"
    }
  },

You can run the following command to execute the full import:

 yarn mage2vs import

... or in multistore setup You can run the same command with specified store-code parameter

 yarn mage2vs import --store-code=de

Import of CMS blocks and pages is also performed by the full import. If the CMS API extension ((SnowdogApps/magento2-cms-api)[https://github.com/SnowdogApps/magento2-cms-api]) was not installed in magento 2, it's recommended to skip the CMS import commands.

 yarn mage2vs import --skip-blocks --skip-pages

You can use the following command to run a delta indexer for a specific storeview:

yarn mage2vs productsdelta --store-code=de

Often in non production environment other services are using self signed certificates for secure connection. You can easily setup the application to trust them by putting them in config/certs directory.

MIT