ddc6fcceb8
This ensures that node can read the sourcemaps and provide useful stacktraces |
||
---|---|---|
scripts | ||
src | ||
.babelrc | ||
.eslintrc | ||
.gitignore | ||
circle.yml | ||
CONTRIBUTING | ||
package.json | ||
README.md |
hapi-sequelize-crud
Automatically generate a RESTful API for your models and associations
This plugin depends on hapi-sequelize
.
npm install -S hapi-sequelize-crud
Configure
Please note that you should register hapi-sequelize-crud
after defining your
associations.
// First, register hapi-sequelize
await register({
register: require('hapi-sequelize'),
options: { ... }
});
// Then, define your associations
let db = server.plugins['hapi-sequelize'].db;
let models = db.sequelize.models;
associations(models); // pretend this function defines our associations
// Now, register hapi-sequelize-crud
await register({
register: require('hapi-sequelize-crud'),
options: {
prefix: '/v1',
name: 'db', // the same name you used for configuring `hapi-sequelize` (options.name)
defaultConfig: { ... }, // passed as `config` to all routes created
// You can specify which models must have routes defined for using the
// `models` property. If you omit this property, all models will have
// models defined for them. e.g.
models: ['cat', 'dog'] // only the cat and dog models will have routes created
// or
models: [
// possible methods: list, get, scope, create, destroy, destroyAll, destroyScope, update
// the cat model only has get and list methods enabled
{model: 'cat', methods: ['get', 'list']},
// the dog model has all methods enabled
{model: 'dog'},
// the cow model also has all methods enabled
'cow',
// the bat model as a custom config for the list method, but uses the default config for create.
// `config` if provided, overrides the default config
{model: 'bat', methods: ['list'], config: { ... }},
{model: 'bat', methods: ['create']}
{model: 'fly', config: {
// interact with the request before hapi-sequelize-crud does
, ext: {
onPreHandler: (request, reply) => {
if (request.auth.hasAccessToFly) reply.continue()
else reply(Boom.unauthorized())
}
}
// change the response data
response: {
schema: {id: joi.string()},
modify: true
}
}}
]
}
});
Methods
- list: get all rows in a table
- get: get a single row
- scope: reference a sequelize scope
- create: create a new row
- destroy: delete a row
- destroyAll: delete all models in the table
- destroyScope: use a sequelize scope to find rows, then delete them
- update: update a row
where
queries
It's easy to restrict your requests using Sequelize's where
query option. Just pass a query parameter.
// returns only teams that have a `city` property of "windsor"
// GET /team?city=windsor
// results in the Sequelize query:
Team.findOne({ where: { city: 'windsor' }})
You can also do more complex queries by setting the value of a key to JSON.
// returns only teams that have a `address.city` property of "windsor"
// GET /team?city={"address": "windsor"}
// or
// GET /team?city[address]=windsor
// results in the Sequelize query:
Team.findOne({ where: { address: { city: 'windsor' }}})
include
queries
Getting related models is easy, just use a query parameter include
.
// returns all teams with their related City model
// GET /teams?include=City
// results in a Sequelize query:
Team.findAll({include: City})
If you want to get multiple related models, just pass multiple include
parameters.
// returns all teams with their related City and Uniform models
// GET /teams?include=City&include=Uniform
// results in a Sequelize query:
Team.findAll({include: [City, Uniform]})
Authorization and other hooks
You can use Hapi's ext
option to interact with the request both before and after this module does. This is useful if you want to enforce authorization, or modify the request before or after this module does. Hapi has a full list of hooks you can use.
Modify the response format
By default, hapi-sequelize-crud
routes will respond with the full model. You can modify this using the built-in hapi settings.
await register({
register: require('hapi-sequelize-crud'),
options: {
…
{model: 'fly', config: {
response: {
// setting this schema will restrict the response to only the id
schema: { id: joi.string() },
// This tells Hapi to restrict the response to the keys specified in `schema`
modify: true
}
}}
}
})
Full list of methods
Let's say you have a many-to-many
association like this:
Team.belongsToMany(Role, { through: 'TeamRoles' });
Role.belongsToMany(Team, { through: 'TeamRoles' });
You get these:
# get an array of records
GET /team/{id}/roles
GET /role/{id}/teams
# might also append `where` query parameters to search for
GET /role/{id}/teams?members=5
GET /role/{id}/teams?city=healdsburg
# you might also use scopes
GET /teams/{scope}/roles/{scope}
GET /team/{id}/roles/{scope}
GET /roles/{scope}/teams/{scope}
GET /roles/{id}/teams/{scope}
# get a single record
GET /team/{id}/role/{id}
GET /role/{id}/team/{id}
# create
POST /team/{id}/role
POST /role/{id}/team
# update
PUT /team/{id}/role/{id}
PUT /role/{id}/team/{id}
# delete
DELETE /team/{id}/roles #search and destroy
DELETE /role/{id}/teams?members=5
DELETE /team/{id}/role/{id}
DELETE /role/{id}/team/{id}
# include
# include nested associations (you can specify an array if includes)
GET /team/{id}/role/{id}?include=SomeRoleAssociation
# you also get routes to associate objects with each other
GET /associate/role/{id}/employee/{id} # associates role {id} with employee {id}
# you can specify a prefix to change the URLs like this:
GET /v1/team/{id}/roles