Skip to content

fauna-labs/serverless-fauna

Repository files navigation

This repository contains unofficial patterns, sample code, and tools to help developers build more effectively with Fauna. All Fauna Labs repositories are provided “as-is” and without support. By using this repository or its contents, you agree that this repository may never be officially supported and moved to the Fauna organization.


Serverless Fauna

This Serverless Framework plugin allows you to manage Fauna databases and resources directly in your serverless.yml file. You can integrate it into your test and CI/CD pipeliness to keep your databases in sync across multiple environments. Visit [this repository][serverless-fauna-example] for a sample application demonstrating how to create, update, and delete collections, indexes, roles, and user-defined functions (UDFs).

Installation

$ npm install @fauna-labs/serverless-fauna --save-dev

or using yarn

$ yarn add @fauna-labs/serverless-fauna

NOTE: This package has not reached a 1.0 release yet. Minor version releases may still contain breaking changes. If you wish, you can restrict your projects to only accepting patch updates by prefacing the version number with a "~" in your package.json file. For example, "~0.2.0" or "~0.1.6"

FQL v10

FQL v10 resources follow a different schema than v4. You must declare your v4 and v10 resources in separate files. Specify a v10 schema by setting version: 10, like so:

plugins:
  - "@fauna-labs/serverless-fauna"

fauna:
  version: 10
  client:
    secret: ${env:FAUNA_SECRET}
    endpoint: ${env:FAUNA_ENDPOINT}

  collections:
    MyCollection:
      indexes:
        MyIndex:
          terms:
            - field: name
          values:
            - field: name
              order: asc

  functions:
    MyFunction:
      body: |
        x => x + 1

  roles:
    MyRole:
      privileges:
        - resource: MyFunction
          actions:
            call: true
        - resource: MyCollection
          actions:
            read: x => x.category == "pets"

You still use sls fauna deploy and remove commands to create, update and destroy this schema.

Supported FQL v10 Resources

Notable Differences

  • You don't declare a separate name property on your config. Instead, the key is used as the name.
  • Create and Update actions during a deploy command are handled in a single transaction. If, for some reason, your schema is large enough to cause an error, you should break it up into separate logical files for deployment.
  • Destruction of resources during a deploy command is handled as a single separate transaction following creates/updates.

Migrating to FQL v10

The v10 and v4 plugins exclusively remove resources associated with their versions. Even so, we recommend you update the deletion_policy to retain during an upgrade. Updating ensures that subsequent deploys won't remove a critical resource if a step is missed.

This makes the upgrade path fairly simple:

E.g.

  1. Start with a v4 schema, serverless.yml.
  2. Update the deletion_policy to retain and deploy it.
fauna:
  deletion_policy: retain
  collections:
    Movies:
      name: Movies
    Series:
      name: Series
  1. Create a v10 schema, serverless-v10.yml
fauna:
  version: 10
  collections:
    Movies: {}
  1. Update to v10 by running sls fauna deploy -c serverless-v10.yml.

    You should see this resource updated. Now, the collection metadata tells the plugin that it's managed by a v10 schema.

  2. Remove Movies from serverless.yml.

    NOTE: If you run sls fauna deploy -c serverless.yml without removing Movies, the metadata will be updated again to tell the plugin it's managed by a v4 schema.

Commands

This plugin listens to hooks from default serverless commands, and runs its own logic.

command description
sls fauna deploy Sync Fauna resources specified in the config. All resources created by the plugin have a property created_by_serverless_plugin
sls fauna remove Remove Fauna resources managed by the plugin. read more about deleting policy

FQL v4 Configuration

plugins:
  - "@fauna-labs/serverless-fauna"

fauna:
  client:
    secret: ${env:FAUNA_ROOT_KEY}
    # domain: db.fauna.com
    # port: 443
    # scheme: https
  collections:
    Movies:
      name: Movies
      data:
        some_data_key: some_data_value

  functions:
    double:
      name: double
      body: ${file(./double.fql)}

  indexes:
    movies_by_type:
      name: movies_by_type
      source: ${self:fauna.collections.Movies.name}
      terms:
        fields:
          - data.type

    movies_by_category:
      name: movies_by_category
      source: ${self:fauna.collections.Movies.name}
      data:
        some_data_key: some_data_value
      terms:
        fields:
          - data.category

    sort_by_year:
      name: sort_by_year
      source: ${self:fauna.collections.Movies.name}
      values:
        fields:
          - data.type
          - ref

FQL v4 Collection configuration

Accepts the same params as Fauna's CreateCollection query

collections:
  Movies:
    name: Movies
    history_days: 30
    ttl_days: 10
    data:
      some_data_key: some_data_value

FQL v4 Function configuration

Accepts the same params as Fauna's CreateFunction query

functions:
  double:
    name: double
    body: ${file(./double.fql)}
    role: admin
    data:
      desc: double number

FQL v4 Index configuration

Accepts the same params as Fauna's CreateIndex query.

In Fauna's indexes, terms, values and source can only be set during index creation. If you try to modify those fields in an existing index, the plugin will throw an error.

search_by_category_and_sort_by_year:
  name: search_by_category_and_sort_by_year
  source:
    collection: ${self:fauna.collections.Movies.name}
    fields:
      is_current_year: ${file(./IsCurrentYear.fql)}
  terms:
    fields: -data.category
  values:
    fields:
      - path: data.type
        reverse: true
      - ref
    bindings:
      - is_current_year

FQL v4 Index source

The index source could be a string, and will be interpreted as collection reference.

source: Movie

Or it could be a source object

source:
  collection: Movie

Or it could be an array of objects:

source:
  - collection: Movies
  - collection: Cartoons
  - collection: Series

FQL v4 Index terms

Index terms describe the fields that should be searchable.

terms:
  fields:
    - data.search
  bindings:
    - binding

FQL v4 Index values

Index values describe the fields returned, and have a similar structure to terms, but with an additional reverse field to define sort order.

values:
  fields:
    - path: data.field
      reverse: true
  bindings:
    - binding

FQL v4 Index binding

Index bindings allow you to compute fields for a source while the document is being indexed.

You can specify multiline FQL:

source:
  collection: Movies
  fields:
    is_current_year: >
      Equals([
        Year(Now()),
        ToInteger(Select(['data', 'release_year'], Var('doc')))
      ])

Or you can create file with the .fql extension, and use the Fauna VSCode plugin to handle your .fql files.

source:
  collection: Movies
  fields:
    is_current_year: ${file(./IsCurrentYear.fql)}

FQL v4 Role configuration

Accepts the same params as Fauna's CreateRole query.

roles:
  movies_reader:
    name: movies_reader
    privileges:
      - collection: ${self:fauna.collections.movies.name}
        actions:
          read: true
      - index: ${self:fauna.indexes.movies_by_type.name}
        actions:
          read: true
      - function: ${self:fauna.functions.double.name}
        actions:
          call: true

FQL v4 Role schema privileges

Read more about the privilege configuration object

For schema privileges, specify a field key without a value:

roles:
  read_collections_and indexes:
    name: read_collections
    privileges:
      - collections:
        actions:
          read: true
      - indexes:
        actions:
          read: true

You can also pass action predicates:

editors:
  name: editors
  membership:
    - ${self:fauna.collections.scriptwriters.name}
    - ${self:fauna.collections.directors.name}
  privileges:
    - collection: ${self:fauna.collections.movies.name}
      actions:
        read: true
        write: ${file(./CanModifyMovie.fql)}

Role membership

A membership configuration object dynamically defines which authenticated resources are members of a given role.

It could be a string:

roles:
  actor:
    name: actor
    membership: actor

Or it could be an array:

roles:
  actor:
    name: participant
    membership:
      - actor
      - directors

Or you could pass the full membership object

roles:
  only_active:
    name: only_active
    membership:
      resource: ${self:fauna.collections.users.name}
      predicate: ${file(./IsActiveUser.fql)}

Or even an array of membership objects:

roles:
  only_active:
    name: only_granted
    membership:
      - resource: ${self:fauna.collections.users.name}
        predicate: ${file(./IsGranted.fql)}
      - resource: ${self:fauna.collections.managers.name}
        predicate: ${file(./IsGranted.fql)}

Deletion policy

This plugin keeps your Fauna database in sync with your serverless configuration file. Therefore, the plugin will remove any resources that currently exist in Fauna, but are not declared in your serverless.com configuration file.

If there are resources that you absolutely do not want deleted, even though they might not be in your serverless.com configuration, you can set deletion_policy to retain (the default being destroy) in the top level fauna configuration. In example below, Fauna resources will not be deleted:

fauna:
  deletion_policy: retain

Please note that if you specify the deletion_policy at both the top level and the resource level, the resource level deletion_policy will override it. For example, in the following configuration, the collection logs would be removed and the rest of the resources would be saved:

fauna:
  deletion_policy: retain
collections:
  Movies:
    name: Movies
  logs:
    name: logs
    data:
      deletion_policy: destroy

Developers

To develop on this repository, clone it and make any changes you would like to issue in a pull request.

You can run the test suite by:

  1. Starting a local fauna container with docker run --rm --name faunadb-sls-test -p 8443:8443 fauna/faunadb
  2. Running the test suite with npm test

Copyright Fauna, Inc. or its affiliates. All rights reserved. SPDX-License-Identifier: MIT-0