Hey <!everyone> ✌️ prisma cool We just released Prisma

3.14.0

prisma rainbow 🦦 CockroachDB connector is now Generally Available! We are proud to announce that the CockroachDB connector is now stable and Generally Available. If you’re upgrading from Prisma version `3.9.0`+ or the PostgreSQL connector, you can now run

npx prisma db pull

and review the changes to your schema. To learn more about CockroachDB-specific native types we support, refer to our docs. To learn more about the connector and how it differs from PostgreSQL, head to our documentation. ⚡*️ PostgreSQL

GIN

,

GiST

,

SP-GiST

, and

BRIN

indexes support (Preview)* This release has expanded index type support with the GIN, GiST, SP-GiST, and BRIN indexes. To make use of an index type, you can update your Prisma schema by providing the

type

argument to the

@@index

attribute:

datasource db {
  provider = "postgresql"
  url      = env("DATABASE_URL")
}

generator client {
  provider        = "prisma-client-js"
  previewFeatures = ["extendedIndexes"]
}

model Post {
  id      Int     @id
  title   String
  content String?
  tags    Json?
  
  @@index([tags], type: Gin)
}

The following SQL will be generated in your migration when you run `prisma migrate dev`:

CREATE TABLE "Post" (
    "id" INTEGER NOT NULL,
    "title" TEXT NOT NULL,
    "content" TEXT,
    "tags" JSONB,
    CONSTRAINT "Post_pkey" PRIMARY KEY ("id")
);

CREATE INDEX "Post_tags_idx" ON "Post" USING GIN ("tags");

To learn more about configuring index types in your schema, refer to our documentation. 🪄 Improved

queryRaw

API In this release, we made improvements to the SQL raw API. Some improvements are breaking and will be available behind the new

improvedQueryRaw

Preview feature flag.

generator client {
  provider        = "prisma-client-js"
  previewFeatures = ["improvedQueryRaw"]
}

The

improvedQueryRaw

Preview feature solves most of the issues faced when working with the raw API. We would encourage you to turn on the Preview feature flag, try out the new API, and let us know how we can make it even better. Here are some of the improvements we made to `queryRaw`: 1. Raw scalar values are deserialized as their correct JavaScript types Prisma Client queries such as

findMany

deserialize database scalar values to their corresponding JavaScript types. For example, a

DateTime

value is deserialized as a JavaScript

Date,

and a

Bytes

would be deserialized as a JavaScript

Buffer

. Raw queries now implement the same behavior when the

improvedQueryRaw

Preview feature flag is enabled. The types of values from the database will be used instead of the types in the Prisma schema. 2. PostgreSQL type-casts We’ve also fixed a lot of PostgreSQL type-casts that were broken by enabling the

improvedQueryRaw

Preview feature flag. A consequence of this fix is that some subtle implicit casts are now handled more strictly and would fail. 3. Query parameters are correctly sent to the database Before this release, query parameters of type

BigInt

,

Bytes

, and

Decimal

were incorrectly sent to the database leading to instances of unexpected inserts. Passing the types as query parameters now works:

await prisma.$executeRaw`INSERT INTO "Table" ("bigint", "bytes", "decimal") VALUES (${BigInt("123")}, ${Buffer.from([1, 2, 3])}, ${Decimal("12.23")});`

This improvement is available without the

improvedQueryRaw

Preview feature flag.

📚 Learn more in the release notes For more info and links to documentation, you can read the release notes. 🌟 Help us spread the word about Prisma 🌟 To help spread the word about Prisma, we’d very much appreciate it if you would star the repo 🌟 And if you’re excited about the features in this week’s release, then help us and share your excitement on Twitter. 📰 Join us on Thursday for the “What’s new in Prisma” livestream This week, @nikolasburk and @Sabin Adams will discuss the latest release and other news from the Prisma ecosystem in a this Thursday at 5 pm Berlin | 8 am San Francisco.