Generate JSON database for JSON server using Filltext.com as random JSON data source.

$ npm install -g json-server-init

• create - Create new JSON database.
• collection - Add new collection to existent database file (todo).

Possible options are:

• --name, -n - Specify name of the database JSON file to create (in case of create command) or use (collection command). Default name if not provided is "db.json".
• --help, -h - Show help.
• --version, -v - Show version number.

For example, to create "dev.json" schema file:

$ json-server-init create -n dev.json

Command produces several prompts.

Prompt for collection name and number of rows renders something like this:

> Collection name and number of rows, 5 if omitted (ex: posts 10):

Valid input would be a new collection name with optional number separated by space indicating how many rows should be generated for this collection. For example, users 10 will generate collection "users" with 10 records in it, sessions will result into collection "sessions" with default 5 records, etc.

After collection name is entered one would need to configure what fields collection should have:

>> What fields should "users" have?
   Comma-separated fieldname:fieldtype pairs (ex: id:index, username:username)

Entry must have specific format: fieldname:fieldtype.

• fieldname - name of the field, only alpha-numeric characters.
• fieldtype - type of the data. Corresponds to types filltext generator uses for fields, refer entire list for possible values. Multiple fields concatenation is possible with + operator.

For example, to generate users collection with four fields: id, username, name and age, one could enter this command:

>> What fields should "users" have?
   Comma-separated fieldname:fieldtype pairs (ex: id:index, username:username)
id:index, username:username, name:firstName+lastName, age:numberRange|18,80

You can add as many collections as necessary: after fields prompt there is a confirmation if more collections need to be created:

> Add another collection? (y/n) n

If "y" is entered flow repeats "Collection prompt" step, otherwise it fetches JSON data and saves it to the file.

TODO...

Here is how typical workflow looks like with create command:

$ json-server-init create
> Collection name and number of rows, 5 if omitted (ex: posts 10): users 2
>> What fields should "users" have?
   Comma-separated fieldname:fieldtype pairs (ex: id:index, username:username)
 id:index, username:username, motto:lorem|5
> Add another collection? (y/n) n
db.json saved.

Above will produce db.json file with content similar to this:

{
    "users": [
        {
            "id": 1,
            "username": "RGershowitz",
            "motto": "curabitur et magna placerat tellus"
        },
        {
            "id": 2,
            "username": "NMuroski",
            "motto": "ante nullam dolor sit placerat"
        }
    ]
}

Now you can start json-server:

$ json-server --watch db.json

MIT License © Aliaksandr Astashenkau