@raresail/prisma-mapper

A CLI that adds @map and @@map based on a json

Installation

NPM

npm install -g @raresail/prisma-mapper

Yarn

yarn global add @raresail/prisma-mapper

Features

Adds @map and @@map
Keeps the @db. attributes for fields
Adds the @updatedAt for fields with the name updated_at or updatedAt
The prisma schema is formatted by the @prisma/internals after generation
The cli does not modify the current schema and generates a new one with the information from the current one
.pmignore Can be used to avoid adding the @map/@@map for specific fields when running the map command

Node Versions

Version	Supported
<= 19.x	❌
20.x	✅
21.x	❌
>= 22.x	✅

Getting Started

If the prisma schema is often pulled you can use the cli like

yarn prisma db pull --force
  && yarn prisma-mapper generate
  && yarn prisma-mapper map
  && yarn prisma generate

or for camelCase by default

yarn prisma db pull --force
  && yarn prisma-mapper map --camel
  && yarn prisma generate

With db pull force you get the latest schema updates and force overwrites the file.

The generate command creates a json called prisma-mapper.json in the root, if the json already exists it adds in it any new fields/models. (NOT FOR --camel option)

yarn prisma-mapper generate

The json looks like the following:

The hasMap is added by generate if the prisma model has already a @@map
The name is to add @@map for a model name
In the fields object is to add @map for each field

{
  "model": {
    "hasMap": false,
    "name": null,
    "fields": {
      "field_name_1": null,
      "field_name_2": null
    }
  }
}

The map command modifies the schema.prisma file and adds the @map and @@map for each field and model, if the field/model-name is in the json file, the field/model will be renamed to the new name.

yarn prisma-mapper map

Options

Both 2 commands have an option --schema "new schema path to use the schema from a different location

For both commands

--schema "new schema path
- is to use the schema from a different location
--camel
- for map it does the mapping without a json file
- for generate it defaults the namings to camel case in the generated json

Map options

-o --output "output schema path"
- after mapping save the schema to another location

Ignore File

Example for the .pmignore file

enum_ignore_s_contents.someValue
enum_ignore_contents.*
enum_ignore_model

model_ignore_s_contents.updated_at
model_ignore_s_contents.created_at
model_ignore_s_contents.enum_three
model_ignore_contents.*
model_ignore

RaresAil/prisma-mapper