strapi-plugin-custom-import-export
v1.0.0
Published
Custom Import/Export Entries Plugin for Strapi
Downloads
7
Maintainers
Readme
Strapi Plugin Import Export Entries
Import/Export data from and to your database in just few clicks.
Features
Import
- Import data directly from the Content Manager
- Import data from JSON file or from typing raw text according to user permissions
- Import contents to collection type/single type (components, dynamic zones and media are supported)
- Control which roles can import data from the admin UI.
Export
- Export data directly from the Content Manager
- Export JSON contents according to user permissions
- Download files or copy exported data to clipboard
- Filter & sort data using Content Manager filters & sorting
- Export plugins content types
- Control which roles can export data from the admin UI.
Known Limitations
At the moment, dynamic zones and media are not unit tested. Tests will be implemented in a near future to improve quality and development speed.
Screenshots
Table Of Content
- Requirements
- Feedback
- Contribute
- Installation
- Rebuild The Admin Panel
- Usage
- Importing Data
- Related Plugins
- Author
- Acknowledgments
Requirements
Strapi v4 is required.
Feedback
Access the publicly available product roadmap and suggest features, report bugs or upvote other people suggestions.
Join the Discord Community to give your feedback 📣 and get some help from the community ⛑️
Contribute
See the repo Strapi Contribute.
Installation
- Download
yarn add strapi-plugin-import-export-entries
or
npm i strapi-plugin-import-export-entries
- Enable the plugin
Add in the file config/plugins.js
:
module.exports = ({ env }) => ({
//...
'import-export-entries': {
enabled: true,
config: {
// See `Config` section.
},
},
//...
});
Rebuild The Admin Panel
New releases can introduce changes to the administration panel that require a rebuild. Rebuild the admin panel with one of the following commands:
yarn build --clean
or
npm run build --clean
Usage
Once the plugin is installed and setup, the functionnalities for a collection are accessible on its content management page.
You can also export the whole database from the home page of the plugin.
Access Control
You can define which roles can import and/or export data from the admin UI.
Go to Settings > Roles (under Administration Panel) > Plugins > Import-export-entries
.
Preferences
For a quick and convenient use, you can set your preferences from the home page of the plugin.
Once set, they will be used each time you import/export data.
Config
Available Options
In config/plugins.js
:
module.exports = ({ env }) => ({
//...
'import-export-entries': {
enabled: true,
config: {
/**
* Public hostname of the server.
*
* If you use the local provider to persist medias,
* `serverPublicHostname` should be set to properly export media urls.
*/
serverPublicHostname: 'https://yoga.com', // default: "".
},
},
//...
});
In any collection schema schema.json
:
{
"collectionName": "my-awesome-collection",
"info": {
"displayName": "My Awesome Collection",
},
"pluginOptions": {
"import-export-entries": {
/**
* Define the `idField` used to find an entry of the collection
* when importing data.
*
* `idField` must match the name of an attribute.
* See section _Specifying `idField` Per Collection_ for more details
*/
"idField": "name"
}
},
"attributes": {
/**
* In this example, `name` will be used to find an entry
* of this collection when importing data.
*/
"name": {
"type": "string",
"unique": true
},
}
}
Specifying idField
Per Collection
Importing data will either create entries if they don't exist, or update them otherwise.
When transfering data from a database to another, relying on the id
of an entry is not reliable. For example, if you are transfering data on hospitals with this schema:
interface Hospital {
id: number;
name: string;
employees: Employee[];
patients: Patient[];
}
You will have something similar in your source and target databases:
// data in source database
{
id: 1,
name: "Pitié Salpêtrière",
employees: [2, 3],
patients: [4, 5],
}
// data in target database
{
id: 11,
name: "Pitié Salpêtrière",
employees: [12, 13],
patients: [14, 15],
}
Different databases, different id
s. 🫠
That's why we need a way to define the field used to find an entry in a collection. This field is called an idField
.
To define the idField
of a collection, add it in the pluginOptions
of the collection, under the property import-export-entries
. Using the example above, this is how we would define the idField
of the collection hospital
:
{
"collectionName": "hospitals",
"info": {
"displayName": "Hospital",
},
"options": {},
/**
* In the property `pluginOptions`, define the `idField` under the property `import-export-entries`.
*
* `idField` must match the name of an attribute.
*/
"pluginOptions": {
"import-export-entries": {
"idField": "name"
}
},
"attributes": {
"name": { // 👈 `name` will be used to find a hospital when importing data.
"type": "string",
"unique": true
},
"employees": {
"type": "relation",
"relation": "oneToMany",
"target": "api::employees.employees"
},
"patients": {
"type": "relation",
"relation": "oneToMany",
"target": "api::patients.patients"
},
}
}
For each collection of your application, you can define a different idField
. For example, you can set the name
attribute as the idField
of the collection hospital
, and for the collection patients
use the attribute ssn
(I really hope you're not storing uncyphered SSNs in your database 😬).
How does the search behave if I don't define explicitly the
idField
of a collection?
By default, the idField
of a collection is the id
attribute. We can imagine in a near future to automatically detect unique scalar fields of a collection and use them by default, but it's not the case at the moment.
How does the search behave when I specify the
idField
from the strapi admin UI?
The idField
specified from the import modal of the admin UI takes precedence over the one defined in the pluginOptions
of the collection.
This default behavior could change in the future if user feedback shows it's cumbersome to set it manually on each import. You tell me.
Importing Large Files
When importing data, imported file size may exceed the file size limit of the server. To lift up the limit, configure the Strapi middleware body
:
// ./config/middlewares.js
module.exports = {
// ...
{
name: 'strapi::body',
config: {
jsonLimit: '10mb',
},
},
// ...
}
Filtering and Sorting
The filtering and sorting mechanism relies on Strapi filtering and sorting feature:
- Connect to the content manager page of the model you want to export, and filter and sort the data as you want it to be exported.
- Open the export modal and check the option Apply filters and sort to exported data.
- Click on Fetch Data.
The exported data is filtered and sorted as expected.
Services
/*****************************
* Service "import".
****************************/
/**
* Get the service.
*/
const service = strapi.plugin("import-export-entries").service("import");
/**
* Method importData.
*/
await service.importData(
/**
* Data to import.
* Expected type depends on the specified format:
* - csv: string
* - jso: object | object[]
* - json: string
*/
dataRaw: object | object[] | string,
options: {
/**
* Slug of the imported model.
* - "media" is a custom slug to specifically import media. See section Importing Data > Media below.
*/
slug: "media" | string;
/**
* Format of the imported data.
* - csv
* - jso: javascript object
* - json: javascript object notation
*/
format: "csv" | "jso" | "json";
/** User importing data. */
user: object;
}
) : Promise<{
failures: {
/** Error raised. */
error: Error;
/** Data for which import failed. */
data: object;
}[]
}>;
/*****************************
* Service "export".
****************************/
/**
* Get the service.
*/
const service = strapi.plugin("import-export-entries").service("export");
/**
* Method exportData.
*/
await service.exportData(
options: {
/**
* Slug of the model to export.
* - "media" is a custom slug to specifically export media.
*/
slug: "media" | string;
/**
* Export format.
* - csv
* - json
* - json-v2: json in the new json file format (see section `Importing Data`)
*/
exportFormat: "csv" | "json" | "json-v2";
/** Search query used to select the entries to export. The package `qs` is used to parse the query. */
search?: string;
/** Whether to apply the search query. */
applySearch?: boolean;
/** Whether to export relations as id instead of plain objects. */
relationsAsId?: boolean;
/** Deepness of the exported data. */
deepness?: number;
}
) : Promise<string>;
Content API
Data can be imported/exported through the content api. Endpoints have to be enabled in Settings > Users & Permissions plugin > Roles.
/*****************************
* Import data
*
* POST /api/import-export-entries/content/import
****************************/
type RouteParams = {
/** Slug of the model to export. */
slug: string;
/**
* Data to import.
* if `format` is "csv", data must be a string.
* if `format` is "json", data must be an object or an array of objects.
* */
data: string | Object | Object[];
/** Format of the passed data to import. */
format: 'csv' | 'json';
/** Name of the field to use as a unique identifier for entries. Default: "id" */
idField?: string;
};
type RouteReturn = {
/** Array of failed imports. */
failures: {
/** Error raised during import. */
error: string;
/** Data for which the import failed. */
data: Object;
}[];
};
/*****************************
* Export data
*
* POST /api/import-export-entries/content/export/contentTypes
****************************/
type RouteParams = {
/** Slug of the model to export. */
slug: string;
/** Format to use to export the data. */
exportFormat: 'csv' | 'json';
/** Search query used to select the entries to export. The package `qs` is used to parse the query. Default: "" */
search?: string;
/** Whether to apply the search query. Default: false */
applySearch?: boolean;
/** Whether to export relations as id instead of plain objects. Default: false */
relationsAsId?: boolean;
/** Deepness of the exported data. Default: 5 */
deepness?: number;
};
type RouteReturn = {
/** Exported data. */
data: string;
};
Webhook
At the moment, the webhook is triggered only for media creation, update and deletion. It is not triggered for other data.
Importing Data
JSON v2
JSON v2 introduces a new supported file structure. Data is flattened and dependencies only relies on id
s (object
s are not supported in this new version). Collection types, single types, media and components are all treated the same for ease of use.
Here is an example:
{
"version": 2, // required for the import to work properly.
"data": {
// Each collection has a dedicated key in the `data` property.
"api::collection-name.collection-name": {
// Sub keys are `id`s of imported entries and values hold the data of the entries to import.
"1": {
"id": 1
//...
},
"2": {
"id": 2
//...
}
},
"api::other-collection-name.other-collection-name": {
"1": {
"id": 1,
// Relations are specified by `id`s.
"collectionAbove": [1]
//...
},
"2": {
"id": 2,
"collectionAbove": [1, 2]
//...
}
},
// Import medias.
"plugin::upload.file": {
"1": {
"id": 1
//...
},
"2": {
"id": 2
//...
}
},
// Import components.
"my.component": {
"1": {
"id": 1
//...
},
"2": {
"id": 2
//...
}
}
}
}
JSON v1 (deprecated)
The expected import data structure:
Relation:
object
the relation is searched in db by id
. If an entry is found, it is updated with the provided data. Otherwise, it is created.
number
The relation is treated as an id.
Media:
object
the media must have an id
, hash
, name
or url
property. First the media is searched by id
, then by hash
, then by name
and finally imported from url
if not found previously.
When imported by url
, the hash
and name
of the file are deduced from the url
(the hash
is also deduced because Strapi exports files with their hash
in the url
instead of the name
). The hash
and name
are used to find the media in db. First the media is searched by hash
, then by name
and used if found. Otherwise, the media is uploaded to the db by downloading the file from the url
.
⚠️ Check the server has access to the
url
.
When imported by url
, extra data can be provided to enhance the created file:
id
(defaults to an auto generatedid
)name
(defaults to thename
deduced from the url)caption
(defaults to""
)alternativeText
(defaults to""
)
Examples
{ id: 1 }
{ hash: "alpaga.jpg" }
{ name: "alpaga.jpg" }
{ url: "https://www.thetimes.co.uk/imageserver/image/alpaga.jpg" }
(Deduced filehash
isalpaga
and deducedname
isimageserver-image-alpaga.jpg
){ url: "http://localhost:1337/alpaga.jpg" }
(Deduced filehash
isalpaga
and deducedname
isalpaga.jpg
){ id: 734, url: "http://localhost:1337/alpaga.jpg", name: "Alpacool", caption: "Alpacool In Da Hood", alternativeText: "Alpacool in da hood" }
string
Same as above, except the media provided is treated as a url
.
"https://www.thetimes.co.uk/imageserver/image/alpaga.jpg"
(Deduced filehash
isalpaga
and deducedname
isimageserver-image-alpaga.jpg
)"http://localhost:1337/alpaga.jpg"
(Deduced filehash
isalpaga
and deducedname
isalpaga.jpg
)
number
The media is treated as an id.
7
Related Plugins
- Strapi Plugin Request Id: Add a unique id to each request made to your server and track your users' activity in the logs
Author
Baboo - @Baboo7
Acknowledgments
This plugin (and especially this README) took strong inspiration from the strapi-plugin-import-export-content from EdisonPeM.