Today, we are excited to share the 2.11.0 stable release 🎉

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release.

Major Improvements

Native database types in the Prisma schema (Preview)

We are super excited to share that this release brings you one of the most requested features since we released the initial version of Prisma: The representation of native database types in the Prisma schema.

Up to this release, the Prisma schema only allowed to represent a limited set of types: String, Int, Float, Boolean, DateTime, Json. Each of these types has a default mapping to an underlying database type that's specified for each connector (see the mappings for PostgreSQL and MySQL).

With this release, you can now add an attribute to a field definition in order to determine which specific database type it should be mapped to. Here is an example for a PostgreSQL database that specifies the underlying database types more specifically. Note that you also need to enable the feature via the the nativeTypes feature flag:

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

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

model User {
  id    Int     @id @default(autoincrement()) 
  name  String? @db.VarChar(255)
}

model Post {
  id        Int      @id @default(autoincrement()) 
  title     String   @db.VarChar(100)
  createdAt DateTime @db.Timestamp(6)
  wordCount Int      @db.SmallInt
}

In the above code snippet, all fields of the two Prisma models are annotated with attributes to diverge from the default mapping and determine the exact type that is used in the underlying database. For example, the id fields are represented as db.BigInt which maps to the BIGINT PostgreSQL type. You can find out more about which Prisma types can be annotated with which native type attributes in the docs for the PostgreSQL and MySQL connectors.

Note: Native types don't work with Prisma Migrate yet. For the time being, you still need to configure the type manually in your database and pull it into your schema via introspection.

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know if this feature is already ready for production! (If encounter any problems, please open a new issue here).

📚 Documentation: Extended native type support

New types in the Prisma schema: BigInt, Bytes and Decimal (Preview)

With this release, we are further introducing three new scalar types in the Prisma schema: BigInt, Bytes and Decimal. Here is an overview for how they map to the different databases that are supported by Prisma and how they're represented when queried with Prisma Client JS:

To make use of these new types, you need to enable the nativeTypes feature flag in your Prisma schema:

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

Once you added the feature flag, be sure to run prisma generate again to update your local version of Prisma Client.

📚 Documentation: Prisma types in the nativeTypes preview

Set foreign keys directly (Preview)

We now support writing foreign keys (also known as relation scalars) directly with the uncheckedScalarInputs preview feature.

Consider this sample Prisma schema:

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

model User {
  id    Int    @id @default(autoincrement())
  posts Post[]
}

model Post {
  id       Int  @id @default(autoincrement())
  authorId Int
  author   User @relation(fields: [authorId], references: [id])
}

Instead of using connect when creating a new Post record to wire up a relation with a User record, you can now directly set the authorId value:

await prisma.post.create({
  data: {
    // You can now set the foreign key directly...
    authorId: 1,
    // ... or connect the relationship...
    // author: {
    //   connect: {
    //     id: 1
    //   }
    // }
    // ... but not both at the same time
  },
});

We'd love for you to give it a try. You can give us feedback in this issue.

The $transaction API is now stable

You can now use $transaction API without the transactionApi feature flag.

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

The connectOrCreate API is now stable

The connectOrCreate API offers a convenient way to link or insert nested records.

You can now use it without the connectOrCreate feature flag.

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

Middleware model parameter is now typed

This is a small quality-of-life improvement for developers using the Prisma middleware. The params.model parameter that's passed into a middleware function is now typed to your model so you can now reliably tie middleware to models:

prisma.$use(async (params, next) => {
  switch (params.model) {
    case 'user':
      // ...
    case 'project:
      // ...
  }
})

If you need the old behavior you can cast params.model back to a string by wrapping it String(params.model).

📚 Documentation: Middleware

Deprecation of multi-provider array

You will now see a warning if you have multiple providers in your datasource. We recently realized that supporting an array of providers in the Prisma schema was premature, adding complexity in the process of improving the product, and causing some confusion about our intent.

You can read more about the decision and what to do about it in this issue.

Validate PrismaClient inputs at runtime

We now validate the parameters we accept in new PrismaClient(parameters). This is unlikely to affect you unless you were passing invalid parameters into PrismaClient before.

Prisma Client Go is now Early Access 🎉

We're happy to announce that Prisma Client Go has graduated from experimental to Early Access.

You can get started here. We'd love to learn more about how you use the Go Client during Early Access. You can schedule a call with us or chat with us on Slack in the #prisma-client-go channel. We're looking forward to hearing from you!

📺 Join us for another "What's new in Prisma" livestream

Learn about the latest release and other news from the Prisma community by joining us for another "What's new in Prisma" livestream.

The stream takes place on Youtube on Thursday, November 12 at 5pm Berlin | 8am San Francisco.

👀 Make your online talk recordings more accessible with subtitles

We'd love to help you make your talk recordings (no matter if Meetup or conference talk) more accessible by adding subtitles to it for free! Reach out to [email protected] if you are interested in getting subtitles for your videos!

Fixes and improvements

prisma

• Can't create a record with only the foreign key to another table
• Foreign key field is not being set when using a nested create for a one-to-one relation
• Prisma Client is unable to fetch DateTime lists
• Middleware params model should be an enum or union of strings.
• vercel/pkg broken on windows
• vercel/pkg errors on windows
• Introspection fails instantly MariaDB
• DATABASE_URL verification still attempted when datasource override provided in PrismaClient constructor
• Introspection should try to preserve custom Enum names on MySQL even without @map
• Mutliple 1-n self relations break migrations
• .transaction([]) fails with Request Timeout Error on large dataset
• JSON value truncate floating numbers
• Postgresql: Large double precision float values cause panic
• Error in prisma results in unhandled promise rejection
• Consolidate Undici error handling
• Internal: update our pnpm version used for development and CI
• Internal: Create new milestone check GitHub Action
• Prisma and Netlify, binary is not downloaded automatically since 2.8.0
• Detect and enforce Node Version in preinstall hook so pnpm fails clearly
• Package name reuse (.prisma/client) when generating multiple prisma clients causes issues with some other libraries (e.g. jest)
• Native Types: @db.BigInt should be BigInt
• Prisma generate or introspect don't work when nativeTypes preview is enabled
• Cannot update a model with composite keys through a referenced model
• RangeError: Invalid count value during create
• Prisma Client doesn't load root .env
• You are trying to load env variables which are already present in your project root .env
• Stabilize connectOrCreate
• Stabilize transactionApi
• Add integration test to trigger env var warning in Prisma Client
• Improve error when using prisma migrate --experimental with native types

prisma-client-js

• Runtime validation for Prisma client constructor parameters
• $transaction spread operator

language-tools

• Add Prisma icon to standard icon theme in VSCode
• Gitpod Support
• prisma.plugin.nextjs.addTypesOnSave broken

studio

• Unable to view / edit DateTime lists
• Case insensitive searching
• npx prisma studio doesn't work on WSL 1
• Cannot type "." in a float column
• Websockets fails when studio is behind a reverse proxy
• [design] Cell overflow

prisma-engines

• Introduce support for scalar type BigInt
• Native Type Precisions
• Remove Scalar Type Duration
• Feature gate scalar types Decimal and Bytes
• Remove TimestampWithTimeZone, TimeWithTimeZone and Intervall

Today, we are issuing the 2.10.2 patch release.

Fixes

Client

• You are trying to load env variables which are already present in your project root .env #4098

Language Tools

• Prisma vscode extension is autoformating typescript react code on save #605
• Npm throws error on build creation while Prisma extension is enabled #606

Today, we are issuing the 2.10.1 patch release.

Fixes

CLI

• @prisma/engines are not used by CLI #4097

Client

• Prisma Client doesn't load root .env #4096

Migrate

• Prisma Migrate Crash on change required foreign key to not-required #626
• v2.10.0 #627
• crashed after the first migration #625
• Error in migration engine. #623
• Migration issue on changing userId field to unique. related to #619 #621
• Migration crashed with code #619
• Crash Report #618

Language Tools

• prisma.plugin.nextjs.addTypesOnSave broken #602

Today, we are excited to share the 2.10.0 stable release 🎉

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release.

Major improvements

This release introduces some major new features in Preview!

Support for Microsoft SQL Server (Preview)

With this release, we're introducing support for a new database: Microsoft SQL Server 🥳

You can start using your MS SQL Server databases today with introspection and query them with Prisma Client. Note that Prisma Migrate does not support MS SQL Server yet.

An MS SQL Server database is specified in your Prisma schema via the new sqlserver connector. Note that while it's in Preview, you also need to add the additional previewFeature flag microsoftSqlServer in the generator for Prisma Client:

datasource db {
  provider = "sqlserver"
}

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

You can find a ready-to-run example based on MS SQL Server here.

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know whether this feature is already ready for production! (If you encounter any problems, please open a new issue here).

📚 Documentation: Microsoft SQL Server connector

Single-command schema changes for prototyping (Preview)

With this release, we are introducing a new command that lets you push the state of your Prisma schema file to the database without using migrations: prisma db push --preview-feature

push is the first command that's introduced under the new prisma db namespace. The prisma db namespace will be used for commands that operate directly against the database without e.g., saving or manipulating a migration history on your file system nor in the migrations table. It's therefore well-suited for prototyping and local development environments. In production environments, you'll want to be able to track and replay the changes made to your database which will be handled by the prisma migrate namespace.

The idea behind prisma db push is that you only care about the end state of your database schema, but not about how this end state is achieved.

Also note that the command runs prisma generate for you, so you're saving an extra step after having updated your Prisma schema!

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know whether this feature is already ready for production! (If you encounter any problems, please open a new issue here).

📚 Documentation: db push command

Atomic number operations are now stable

In 2.6.0 we introduced case atomic number operations in Preview, in today's release we're promoting this feature to stable. This means you don't need to include the atomicNumberOperations feature flag in the Prisma Client generator any more:

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

📚 Documentation: Atomic operations on update

Already existing preview features from previous releases

Just a quick reminder:

• In version 2.1.0 we introduced two preview features, namely connectOrCreate and transactionApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

🎃 Join the first Prisma Meetup online

We are excited to bring you the first Prisma Meetup which is going to happen entirely online, so you can dial in from around the globe!

It will be streamed on Youtube this Thursday: October 29th, 2020 at 6pm (CET, Berlin) | 10am (PDT, San Francisco)

We have amazing speakers on board for our first Meetup:

◭ Tammy Butow (Principal SRE @ Gremlin): Database Horror Stories
◭ Hervé Labas (VP of Product @ Prisma): The future of Prisma and its roadmap
◭ Tom Hutchinson (Head of Mobile @ Rapha): Prisma at Rapha

You can watch the event on YouTube here.

Fixes and improvements

prisma

• Graceful shutdown
• Use .executeRaw() with .transaction([])
• Report the database version in the migrate/introspection error reports
• connectOrCreate fails during upsert
• $connect doesn't actually connect, the first query does
• Log when a connection to the database is opened
• Allow .env in the project root with the Prisma Schema somewhere else
• process.on('SIGINT') callback not being called when I import @prisma/client
• Send sql schema description with the introspection error reports
• Feature request: Allow $queryRaw and $executeRaw in transactions
• Support SQL Server in executeRaw and queryRaw
• Introduce stand-alone integration test package
• Separate test helpers into separate folder structure
• Stabilize atomicNumberOperations
• Introduce @prisma/engines package
• Integration tests with SQL Server
• undefined is printed at the top of @prisma/cli@dev

prisma-client-js

• ISO Dates with an offset is not working
• PANIC: called Result::unwrap() on an Err value: Error("EOF while parsing a value", line: 1, column: 0)

migrate

• [PostgreSQL] migrating column from jsonb[] to jsonb fails
• Migration fails when attempting to change the column type in the absence of a working cast
• Error: Error in migration engine. Reason: [libs\sql-schema-describer\src\postgres.rs:159:20] querying for columns: Error { kind: QueryError(Error { kind: Closed, cause: None }), original_code: None, original_message: Some("connection closed") }

language-tools

• Warn about provider array deprecation in the VSCode plugin
• prisma-language-server cannot be ran as command
• LSP: Can't rename if the server getting position that is not the end of word.
• LSP: Please do best effort when the lsp-client doesn't support context option.
• Json should also show squiggly lines for MSSQL

studio

• Studio enforces optional fields to be filled
• Reordering a column also sorts by that column
• Studio swallows all feature flags inside the generator block

prisma-engines

• Query Engine: Native Types
• Introspection Engine: Native Types
• Native Types: Add support for non integer arguments
• Proper parsing of JDBC connection strings
• Composite key ordering in SQL Server
• Do not use magic partial indices with SQL Server
• Gate SQL Server from Migration Engine
• Validator should disallow native type Text in combination with unique attribute for MySQL
• Remove duplication in Introspection Engine tests
• Move all preview feature flags into the generator block
• Research: Understand compatibility of native types with PSL features
• remove validation for optional 1:1 relations in MSSQL
• Remove ScalarType Xml
• Implement default mapping for new scalar type Decimal
• Implement default mapping for the scalar types Bytes

Today, we are excited to share the 2.9.0 stable release 🎉

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release.

Improvements

In today's release, we have many bug fixes, increased test coverage, and improved error messages. The release also lays the foundations for the upcoming native database types support.

Errors thrown by Prisma Client include the version

With this release, error objects thrown by Prisma Client include the clientVersion field which contains the version of Prisma Client. This is useful for debugging and creating issues.

Improved error message of prisma.$transaction

When you use prisma.$transaction to group multiple calls into a single transaction, you need to pass an array of promises. Passing calls that have been awaited will now trigger a more helpful error:

Error: All elements of the array need to be Prisma Client promises. Hint: Please make sure you are not awaiting the Prisma client calls you intended to pass in the $transaction function.

Already existing preview features from previous releases

Just a quick reminder:

• In version 2.6.0 we introduced one preview feature, namely atomicNumberOperations.
• In version 2.1.0 we introduced two preview features, namely connectOrCreate and transactionApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

Issues closed with the release

prisma

• Switch to stable actions/stale once my PR is merged
• We can't release, as the e2e tests are not passing for the ${tag} npm tag! checks wrong npm tag for passing e2e tests when promoting a patch-dev to latest
• Client seems to ignore Custom Binary Specification
• Tests / GitHub Actions check why cached download test is really slow
• Merge PrismaQueryEngineError into PrismaClientUnknownRequestError
• Incorrect SQL generated for Json field { equals: null } filter
• Any Client error message should include the Client version
• prisma -v should include Client version as well (if installed)
• $executeRaw used in a transaction throws: All elements of the array need to be Prisma Client promises.
• Test forks
• Code coverage
• Consolidate integration tests
• Include possible block names on a wrong keyword
• Optional field is required in @@unique and where clause and doesn't accept null
• prisma introspect overwrites a 1:1 mapping as 1:many
• Replace ncc with esbuild in cli
• cli/introspect tests should exit
• Implement env var FORCE_PANIC_INTROSPECTION_ENGINE
• DMMF Enum value types are invalid in 2.8.0
• findFirst should have return type Model | null
• Unexpected field not found (FieldNotFound) error
• Non unique index should not appear in findOne where clause

prisma-client-js

• PostgreSQL <table>WhereUniqueInput including a column which isn't unique when composite primary key is used
• Improve error message when object is passed to prisma.$transaction instead of a Prisma promise
• Export Sql from @prisma/client
• Prisma and node 12.19.0

language-tools

• Quick fixes for typos in block names
• Include reason of publish CI in every workflow and slack notification
• Autocomplete stops showing after the first character
• Changing the datasource name doesn't break the schema
• Can we check for the presence of the other extension and disable one or warn the user clearly
• extract the possible keywords from prisma-fmt error message
• Remove vsce from renovate ignore list
• VS Code extension not formatting for v2.8.1
• Get preview features from prisma-fmt
• Enum @default() autocomplete putting comments as options

studio

• Navigating from relations table triggered Error: Could not find appropriate InputType for this field type. This should never happen.
• t.toUpperCase is not a function - prisma 2.8.0
• Studio doesn't let me save the newly created record
• Studio sometimes gets stuck on Saving Changes

prisma-engines

• Supporting SQL Server in Introspection Engine
• Datamodel validator capability to prevent a foreign key on pointing to a nullable field
• Remove explicit attached name from generated SQL on SQLite
• Remove explicit database name from generated SQL on MySQL
• Schema Parser should validate supplied feature flags
• default() does not work in conjunction with native types
• Postgres: serial native type and @default(autoincrement()) must not be possible at the same time
• Implement a debugPanic RPC in introspection engine
• Add arguments to Mysql native type TINYINT
• Schema validation rejects fields with native type TinyInt and Scalar Type Boolen

Credits

Huge thanks to @rahul3v for helping!

Interested in providing feedback for the upcoming version of Prisma Migrate?

We are seeking users interested in trying out upcoming versions of Prisma Migrate. This will involve trying it out, validating functionality, and sharing feedback with us.

If you're interested in participating, fill out the following form, it won't take longer than a minute!

🎃 Join the first Prisma Meetup online

We are excited to bring you the first Prisma Meetup which is going to happen entirely online, so you can dial in from around the globe!

The Meetup will be streamed on Youtube this Thursday: October 29th, 2020 at 6pm (CET, Berlin) | 10am (PDT, San Francisco)

We have amazing speakers on board for our first Meetup:

◭ Tammy Butow (Principal SRE @ Gremlin): Database Horror Stories
◭ Hervé Labas (VP of Product @ Prisma): The future of Prisma and its roadmap
◭ Tom Hutchinson (Head of Mobile @ Rapha): Prisma at Rapha

You can watch the event on YouTube here.

Today, we are issuing the 2.8.1 patch release.

Fixes

Prisma Client

• DMMF Enum value types are invalid in 2.8.0 #3804
• findFirstshould have return typeModel | null #3863

Studio

• Studio doesn't let me save the newly created record #560
• t.toUpperCase is not a function - prisma 2.8.0 #556
• Navigating from relations table triggered Error: Could not find appropriate InputType for this field type. This should never happen. #554

Today, we are excited to share the 2.8.0 stable release 🎉

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release.

Major improvements

In today's release, we have two new features coming for you!

findFirst

While the findMany API gives you a powerful API to query your database with different filters, it's not ideal if you just want to query a single item.

On the other hand, the findOne API returns single items, but it only allows for filtering by unique fields.

In version 2.8.0, we introduce findFirst - giving you the full power of findMany filters while only returning the first item that matches the filter criteria.

So instead of this:

const usersCalledAlice = await prisma.user.findMany({
  name: "Alice"	
})
const firstUserCalledAlice = usersCalledAlice[0]

You can now do this:

const firstUserCalledAlice = await prisma.user.findFirst({
  name: "Alice"	
})

All filters available for findMany are also available in findFirst.

Case insensitive filters for PostgreSQL are now stable

In 2.5.0 we introduced case insensitive filters for PostgreSQL, in today's release we're promoting this feature to stable. This means you don't need to include the insensitiveFilters feature flag in the Prisma Client generator any more:

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

The new mode option you can pass to findMany influences the corresponding filter (e.g. contains or startsWith) but doesn't change the return type of the findMany query. mode can have two possible values:

• default: Uses the default filter configured on the database level. If the collation is configured as case insensitive in the database, the default mode will be case insensitive as well. In that case, there's no need to use the insensitive mode.
• insensitive: Uses the case insensitive filter (if possible).

const result = await prisma.user.findMany({
  where: {
    email: {
      equals: '[email protected]',
      mode: 'insensitive',
    },
  },
})

Note that this feature will not work if you're using database collations that do not know how to convert between upper- and lowercase letters (e.g. the C collation).

📚 Documentation: Case sensitivity / Case-sensitive filtering

Already existing preview features from previous releases

Just a quick reminder:

• In version 2.6.0 we introduced one preview feature, namely atomicNumberOperations.
• In version 2.1.0 we introduced two preview features, namely connectOrCreate and transactionApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

Fixes and improvements

prisma

• Output prepared GitHub issue creation link after reporting error (CLI)
• Prisma Format should enforce opinionation for order of attributes
• Cursor Pagination not returning expected results
• Warn about syntax error in datasource url (with env var) earlier
• Allow setting scalar lists without set
• Investigate live reload of Prisma Client with next dev
• Output path of schema.prisma file in CLI commands that read it
• Environment variable to trigger crash/panic in Engines
• Unable to connect when connection string contains $
• Implement a findFirst API

prisma-client-js

• Running into an issue with the connectOrCreate operation. The following query should create an entity for a Winery, and assign a Tag to it.
• Integration Tests, Errors: Writing a signed Int to an unsigned column
• ErrorFormat 'colorless' doesn't work

language-tools

• Add option to set the prisma-fmt binary path in vscode options
• Warn about syntax error in datasource url (with env var) earlier #3509
• Slack notifications for failing workflows
• Quick Fix adding native Types preview Feature when using native types
• Extension panic because of comments
• Move e2e tests for published extension in this repo instead of in vscode-e2e-tests
• Prisma extension formatting failure and freeze after upgrade 2.7.1 and VS v1.49.1
• Add number of CI step in visible name of workflow

studio

• Can't filter by null values
• Show why a table cell is invalid
• Improve error modals

prisma-engines

• Add tests using Cockroach DB
• Add tests using Yugabyte DB
• Add test cases for the DMMF command
• Query Engine: Investigate how we could support union types
• After upload: Download binary builds from S3 again and verify expected checksum
• Incorrect error message on missing native Type previewFeature

❓ Are you using Prisma at work?

We'd love to know if you're using Prisma at work. Answer with a quick yes or no in our poll, it won't take longer than a few seconds!

Credits

Huge thanks to @nohns for some great contributions in this release!

Today, we are issuing the 2.7.1 patch release.

Fixes

Prisma Client JS

• Unable to connect when connection string contains $ #3669

Prisma Studio

• Optional relations show up as invalid when you try to create a new record #546

Today we are excited to share the 2.7.0 stable release.

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release.

Major improvements

Prisma Studio is stable 🎉

We are more than excited to share that Prisma Studio has moved out of its experimental state and is promoted to stable with today's release! Prisma Studio is the perfect companion for developers who work with databases in their daily workflows, to quickly get an overview of the state of their database as well as to view and modify the data in it. An announcement blog post will be coming soon!

To use Prisma Studio in your Prisma project, you can now run the prisma studio command without the previously required --experimental flag:

npx prisma studio

Join the #prisma-studio channel in the Prisma Slack and let us know how you like Prisma Studio! 🙌

Configure Prisma schema location via package.json for more flexibility

Previously, your schema.prisma file needed to be available at a specific location when running Prisma CLI commands or provided via the --schema option when invoking a Prisma CLI command.

With this release, you can configure the location of your Prisma schema via a prisma property in your package.json:

{
  "prisma": {
    "schema": "path/to/schema.prisma"
  }
}

The provided location in package.json will be the default location of the Prisma schema file for any Prisma CLI command you invoke. Note that you can still override this default location by manually specifying the --schema option when invoking a Prisma CLI command.

📚 Documentation: Prisma schema file location

Already existing preview features from previous releases

Just a quick reminder:

• In version 2.6.0 we introduced one preview feature, namely atomicNumberOperations.
• In version 2.5.0 we introduced one preview feature, namely insensitiveFilters.
• In version 2.1.0 we introduced two preview features, namely connectOrCreate and transactionApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

Feedback wanted: How does introspection work for you?

In the last few releases we improved Prisma's database introspection, most notably with "More robust introspection by keeping manual changes in the Prisma schema file" but also many smaller bug fixes. We think it is pretty stable now, and would love to hear from you, our users, how you have been using prisma introspect recently!

If you have used introspection in the past, whether it worked well or not so well with your database, please let us know by sharing your feedback on GitHub!

🌟 Help us spread the word about Prisma

To help spread the word about Prisma, we'd very much appreciate if you would star this repo 🌟 And if you're excited about the features in this week's release, then help us and share it on Twitter.

Fixes and improvements

prisma

• Add ability to configure custom schema location instead of requiring --schema arg
• Running prisma init with unknown parameter just outputs null
• Use dotenv-expand instead of dotenv package to allow variable expressions in the env file
• Update denylist in engines and client
• Engine panics when schema.prisma is empty, we could make it an error.
• Schema parser errors shouldn't be reported to error reporting
• Loss of precision on Float fields with many digits
• 2.6.0 Broke some of my update queries
• Log Prisma client version being generated
• Broken set on nested update on 2.6.0
• Multiple many-many relationships appears to cause an introspection failure.
• Query validation error updating Date field
• Error: [libs/datamodel/core/src/dml/model.rs:166:14] We assume an internally valid datamodel before mutating.
• While using PM2 on a prisma related project popus up prisma client cmd terminal on windows
• Separate command and GH action workflow to run integration tests
• The introspect command in v2.6.x does not remember cuid() defaults for a field and user-defined @relation name property
• Do not run GitHub Action Workflows on fork of example repository
• Do not build binaries for FreeBSD 12 any more
• introspect --force does not work

prisma-client-js

• PANIC: invalid or out-of-range date
• aggregateApi and sum returns null if no record

migrate

• Unable to run prisma migrate with dotenv-cli
• Error when running initial migration with a schema in DATABASE_URL

language-tools

• Add vsix bundle to releases
• Create a local release with .vsix file and publish from that
• E2E tests run on every PR should use local LSP
• Abnormal CPU usage
• Change of client/index.d.ts should not show a notification
• This version is built from commit undefined. This is for Prisma CLI ...
• Reorganize READMEs

studio

• Stuck at "Getting things ready" when Studio is open in multiple places
• Filtering issue (not/equals/contains/startsWith) when "" empty string field Value
• Error when opening certain relations
• Tab dirtyness should be cancelled out when it is returned to its original state
• Cannot edit a relation that is null
• Allow typing in partial dates in table cells
• Allow typing in partial dates in filters
• Very slow to load for large Postgres schemas
• When changing a relation, the scalar that constitutes it does not change
• Unable to change relations & relation lists in an uncommitted record
• If you type in an arbitrary id in a scalar that is part of a relation, Save Changes will always fail
• Oopsies: filter UI is broken
• Deleting a model with unsaved changes open in a tab might result in a fatal error

prisma-engines

• Improve error for nonexistent column in SQLite
• Native types: allow specifying unsigned integer types on MySQL
• Validate that prisma enums have at least one variant/value
• Parser must error if native types are used but feature flag is not enabled
• Trigger binary builds via square brackets in commit message

Credits

Huge thanks to @bre7, @rahul3v, @jasonkuhrt and @Weakky for helping!

Today, we are issuing the 2.6.2 patch release.

Fixes

Prisma Client JS

• 2.6.0 Broke some of my update queries #3497
• Query validation error updating Date field #3512

Today, we are issuing the 2.6.1 patch release.

Fixes

Prisma Client JS

• 2.6.0 Broke some of my update queries #3497
• Broken set on nested update on 2.6.0 #3503

Today we are excited to share the 2.6.0 stable release.

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release.

Major improvements

More robust introspection by keeping manual changes in the Prisma schema file

In prior releases, any manual changes to your Prisma schema would be overridden when invoking prisma introspect (e.g. usage of @map or @@map as well as the renaming of relation fields).

Keeping these manual changes in the Prisma schema file has been available as a preview feature via the --experimental-reintrospection already. After getting tested as a preview feature for a few releases, we are excited to promote this functionality to be the default behavior of prisma introspect into this stable release.

Note that you can opt-out of this behavior by using the new --force flag: prisma introspect --force. This will generate a schema purely based on the introspected database and override any prior manual changes.

📚 Documentation: Prisma Introspect

Preview features

New: Atomic number operations on update

With today's release, we introduce atomic number operations for update queries in Prisma Client. It allows you to update the values of certain fields atomically.

Feature flag

Atomic number operations needs to be enabled with the feature flag atomicNumberOperations like so:

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

Usage

We are introducing five atomic operations for Int and Float fields :

• increment: x: Adds x to the current value
• decrement: x: Subtracts x from the current value
• multiply: x: Multiplies the current value by x
• divide: x: Divides the current value by x
• set: x: Sets the value to x (equivalent to data: { age: 18 })

Note: Only one operation can be done per field at a time.

Here's an example of using the new operations:

const result = await prisma.user.update({
  where: {
    email: '[email protected]',
  },
  data: {
    points: {
      set: 99, // this just sets the value, equivalent to `data: { points: 18 }`
    },
    age: {
      increment: 1, // age = age + 1 - Happy Birthday!
    },
    credits: {
     decrement: 2, // credits = credits - 2
    },
    karma: {
      multiply: 3, // karma = karma * 3
    },
    chocolates: {
      divide: 4, // chocolates = chocolates / 4
    },
  },
})

📚 Documentation: Atomic operations on update

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know if this feature is already ready for production! (If encounter any problems, please open a new issue here).

Already existing preview features from previous releases

Just a quick reminder:

• In version 2.5.0 we introduced one preview feature, namely insensitiveFilters.
• In version 2.1.0 we introduced two preview features, namely connectOrCreate and transactionApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

🤔 How would you describe Prisma?

We'd love to hear from you how you describe Prisma to your developer friends and coworkers. If you have 2 minutes, please answer this question via this online form.

🌟 Help us spread the word about Prisma

To help spread the word about Prisma, we'd very much appreciate if you would star this repo 🌟 And if you're excited about the features in this week's release, then help us and share it on Twitter.

Fixes and improvements

prisma

• Rust Panic when querying a model with @@id
• Invalid prisma/.env leads to confusing error message
• [EPIC] Re-Introspection
• prisma generate stack overflow on a large schema
• [Re-Introspection] Comments in the schema are removed by re-introspection
• Self-referential one-to-many relationship query fails: "subquery has too many columns"
• Error in update with empty disconnect field
• Formatting Prisma 2 schema programmatically in JavaScript (file and string)
• Generating to a custom output path shows the wrong usage instructions
• Duplicate identifier 'Action' in node_modules/.prisma/client/index.d.ts
• Use error code for invalid schema.prisma for reintrospection and migrate
• Stabilize Re-Introspection and integrate it into normal introspect
• [Re-Introspection] Keep @updatedAt
• List Relation Filter issue just in 2.5.0
• Filtering by null seems to be broken in 2.5
• JSON fields filtering stopped working in 2.5.0
• Add check of git commit during dowload tests
• Aggregate ignores @map value of field, generating an incorrect query
• Fix JSON autocompletion
• undici dependency cannot install - " Couldn't find match for "e76f6a37836537f08c2d9b7d8805d6ff21d1e744" git://github.com/mcollina/undici.git" - @prisma/engine-core

prisma-client-js

• Hide top-level client properties (as they are visible via JS code intellisense)
• Error when executing findMany query
• Log event typing breaks when adding datasources to PrismaClient constructor
• node script hanging after disconnect
• New error after update to 2.5 for "not null" query
• Implement step 1 of atomic operations on update

migrate

• Multiple failures for error report, error report is just validation error in the end
• Report id 9877: Error querying the database: near ",": syntax error
• Running save outputs connection strings if included in schema file
• schema validation errors should not trigger an error reporting

language-tools

• Document and Surface "Rename" functionality better
• Remove badges from VSCode Marketplace README
• "No result" when trying to rename model where it is used as type, not in model definition
• Enum default autocompletion support
• Detect VSCode theme and tell people to use a different one for better experience
• @@map gets initialized with [] instead of "" when using autocompletion
• Option to stop extension from creating .vscode folder in workspace

studio

• Opening relations in new tabs
• Relation "accordion" is closed when I click somewhere outside of the browser window
• Relation accordion does not indicate which items are/is connected
• Cmd+Click on a relation cell should open a new tab with appropriate filters applied to show only the connected records
• Opening the relations view is considered making a change
• Briefly displaying "No rows to shows" while loading is a bit confusing
• [feature] UX: click-through relations
• Info: The preview flags middlewares were removed, you can now safely remove them from your schema.prisma.
• Relation table is missing a min-height
• Error when trying to add a new record
• Blank screen when trying to filter a string field
• New Tab button (+) should always create a new tab
• Clicking a relation in relation table scrolls to the row in addition to selecting it
• New record disappears when navigating away (if there are no other records in the table)
• Visual feedback when adding new row if scrolled down
• Show fatal errors instead of White Screen of Death
• Update error message to say prisma not @prisma/cli
• Relation table headers overlap if the model name is too long
• Some (not all) column type icons overlap the next column
• skip can be set to negative
• It isn't possible to delete uncommitted records
• Ugly scrollbar in relation editor
• Save [x] changes count not valid after creating a record
• Relation editor table should have a min-height
• Inconsistent state with no tabs
• Render a separator between uncommitted & committed records

prisma-engines

• Datamodel Parser: Native Types
• Log what error caused a rollback in query engine
• Do not generate unnecessary db queries
• Adding optional relations on existing models on SQLite does not add a foreign key constraint
• Multi-line alter tables on SQLite appear not to be working
• Check that unique constraints can be added/removed from existing tables on MySQL
• Check that column arity is preserved by AlterEnum on MySQL
• Enum names must be validated that they are not using a reserved name

Credits

Huge thanks to @peter50216 for helping!

Today, we are issuing the 2.5.1 patch release.

Fixes

Prisma Client JS

• List Relation Filter issue just in 2.5.0 #3342
• Filtering by null seems to be broken in 2.5 #3352
• New error after update to 2.5 for "not null" query #842

Studio

• Bug fixes

Today we are excited to share the 2.5.0 stable release.

🌟 Help us spread the word about Prisma by starring the repo ☝️, tweeting about the release or sharing your experience with Prisma on Reddit. 🌟

Major improvements

Middlewares, removing duplicates with distinct and aggregations are now stable

After running in preview mode for a number of releases, we are excited to promote the following features into this stable release:

• middlewares
• distinct
• aggregateApi

This means you can omit the respective feature flags in the generator block in your Prisma schema:

generator client {
  provider        = "prisma-client-js"
- previewFeatures = ["middlewares", "distinct", "aggregateApi"]
}

Read on to learn about each feature individually!

Middlewares

Prisma Client's middleware lets you intercept Prisma Client queries to manipulate its parameters and interrogate its result. A middleware is a function that's passed to Prisma Client's $use method.

Middlewares are convenient for a number of use cases, for example:

• Logging the time taken to perform a type of query
• Manipulating or validating query parameters
• Contacting another service upon specific queries

The following example includes two middlewares:

const prisma = new PrismaClient()

prisma.$use(async (params, next) => {
  const result = next(params);
  return result;
}

prisma.$use(async (params, next) => {
  return next(params);
}

📚 Documentation: Middleware

Remove duplicates from query results with distinct

Prisma Client allows you to filter duplicate rows from the response to a findMany query using the distinct option:

const distinctCities = await prisma.user.findMany({
  select: {
    city: true,
    country: true,
  },
  distinct: ["city"],
});

📚 Documentation: Distinct

Aggregations with aggregate

Prisma Client allows you to perform aggregations operations on the number fields (such as Int and Float) of a model - for example, you can get the average age of all users:

const aggregations = await prisma.user.aggregate({
  avg: {
    age: true,
  },
});

console.log("Average age:" + aggregations.avg.age);

Prisma Client supports the following aggregations:

• avg (average)
• sum (sum)
• min (minimum value)
• max (maximum value)

📚 Documentation: Aggregations

Preview features

New: Case insensitive filters (PostgreSQL only)

In 2.5.0 we introduce case insensitive filters for querying capabilities to Prisma Client. It allows you to query for fields in a case insensitive way.

Note: This feature will not work if you're using database collations that do not know how to convert between upper- and lowercase letters (e.g. the C collation).

📚 Documentation: Case sensitivity / Case-sensitive filtering

Feature flag

Insensitive filters querying needs to be enabled with the feature flag insensitiveFilters like so:

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

Usage

The new mode option you can pass to findMany influences the corresponding filter (e.g. contains or startsWith) but doesn't change the return type of the findMany query. mode can have two possible values:

• default: Uses the default filter configured on the database level. If the collation is configured as case insensitive in the database, the default mode will be case insensitive as well. In that case, there's no need to use the insensitive mode.
• insensitive: Uses the case insensitive filter (if possible).

const result = await prisma.user.findMany({
  where: {
    email: {
      equals: '[email protected]',
      mode: 'insensitive',
    },
  },
})

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know if this feature is already ready for production! (If encounter any problems, please open a new issue here).

Feature flags for middlewares, distinct and aggregationApi removed

As mentioned above, we were able to promote three features into this stable release. This was thanks to your help and feedback, so please keep trying the preview features if they're useful to you and help us by sharing feedback.

Already existing preview features from 2.1.0

Just a quick reminder:

• In version 2.1.0 we introduced two experimental features, namely connectOrCreate and transactionApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

🌟 Help us spread the word about Prisma

To help spread the word about Prisma, we'd very much appreciate if you would star this repo 🌟 And if you're excited about the features in this week's release, then help us and share it on Twitter.

Fixes and improvements

prisma

• @default(autoincrement()) not working on non id integer fields
• thread 'main' panicked at 'called Option::unwrap() on a None value', libs/datamodel/core/src/validator/validate.rs:178:29
• Improve update notifier message
• Introspect not creating unique relation names
• EPIC: Provide binaries for FreeBSD 12
• Internal: trigger e2e tests for patch branches
• npx prisma introspect errors with [libs/sql-schema-describer/src/mysql.rs:353:27] column_name on indexes using expressions (MySQL 8.0.13+)
• Switch Client to use Unix Domain Sockets
• Add a few sanity tests for Studio in the CLI pipeline
• Output Studio package version in prisma -v
• Suggested CLI commands are missing npx prefix when being triggered via npx prisma itself
• Internal: Move Studio command into cli package
• Transaction is still a reserved word in 2.4.1
• Fluent API chaining seems to be broken as of 2.3.0
• Error: [libs\datamodel\core\src\dml\datamodel.rs:166:9] Every RelationInfo should have a complementary RelationInfo on the opposite relation field.
• Implement deprecation notice of removed preview feature flags in Client

prisma-client-js

• Support case insensitivity when querying and sorting

migrate

• migrate save and up always recreate the db
• Migration fails after changing ENUM
• On failure _new tables/types are left
• Error: Failure during a migration command: Connector error. (error: Error querying the database: Error querying the database: db error: ERROR: type "MembershipRole" already exists
• Error while using prisma migrate init: The above error occurred in the component:
• Invalid connection string not handled gracefully
• Getting error in prisma migrate save --experimental command

language-tools

• Add lint support for no env vars in provider field
• vscodium marketplace availability
• Publish to open-vsx.org
• 2.4.0 breaks extension
• Renaming prisma.schema model

studio

• [feature] Group action button with other buttons
• Cache error
• Column titles overflow on the next column when too long
• Bring back row selection checkboxes
• Improvements to relation pills
• Field level filters for the relation accordion
• Prisma Client errors are not propagated correctly

prisma-engines

• Create CONTRIBUTING.MD
• Validate that there is only one @default(autoincrement()) per table, on an indexed column on MySQL
• Implement missing unexecutable migration warnings
• Validation: default values on enum fields should be valid for the enum
• Switch GH Actions trigger from repository_dispatch to workflow_dispatch to enable running branches
• Add error code for invalid schema.prisma for reintrospection

Today, we are issuing the 2.4.1 patch release.

When we released the new "Order by multiple fields" feature, you gave us feedback, that the order of properties is not stable in JavaScript. Since ES2015 that is not a problem in the language itself anymore.
However, some teams use the eslint sort-keys plugin, which sorts the keys of objects by alphabet.

Therefore we decided to change the API in a patch release to minimize any problems this might cause to you.

Order by multiple fields in 2.4.0

// order by `age` descending and then by `name` ascending
const users = await prisma.user.findMany({
  orderBy: {
    age: 'desc',
    name: 'asc'
  }
})

Order by multiple fields in 2.4.1

// order by `age` descending and then by `name` ascending
const data = await client.user.findMany({
  orderBy: [
    {
      age: 'desc',
    },
    {
      name: 'asc',
    },
  ],
})

Today, we are issuing the 2.4.0 stable release.

🌟 Help us spread the word about Prisma by starring the repo ☝️ or tweeting about the release 🌟

Major improvements

Order by multiple fields

A long-awaited feature - the ability to order by multiple fields in Prisma Client is finally here!

Until now you could only order by a single scalar field of a model. The API design however was already prepared for ordering multiple fields with the object syntax we have. Instead of just one property, that orderBy object can now have as many fields as you want! The order of the fields hereby determines the order of the returned list (see example below).

Note: As this is an incremental change in the API, we introduce this feature without a feature flag.

You can use it like so:

// order by `age` descending and then by `name` ascending
const users = await prisma.user.findMany({
  orderBy: {
    age: 'desc',
    name: 'asc'
  }
})

As mentioned by the comment, the returned objects in users are ordered first by age (descending), and then by name (ascending).

📚 Documentation: Sort on multiple fields

Top-level Client Methods: $ dollar prefix for top-level Prisma Client methods

In recent versions, we introduced a couple of top-level methods in Prisma Client (e.g. prisma.transaction() and prisma.use()) in preview mode. The immediate feedback was that the denylist for model names grew - which breaks Prisma schemas where a model is called Transaction, transaction, use, or Use. And the list goes on...

In order to have a future-proof API, that allows maximum expressibility in terms of model names, we decided to prefix all non-query methods with a dollar sign $. That means Prisma Client will from now on ship with the following methods:

Pre-existing

• $disconnect
• $connect
• $on
• $queryRaw
• $executeRaw

Still in preview

• $use
• $transaction

The preview methods have already been renamed, the pre-existing methods like connect are still available, you just get a deprecation warning. They will be available for a few more releases with an alias, so no hurry to update them yet.

Updates in Prisma Studio

With this release, we shipped a number of improvements for Prisma Studio:

• Refreshed design
• Moved the Reload button from the right of the databrowser to the left
• Moved the Pending Actions bar from the bottom of the databrowser to the top right
• Removed the sidebar; to open a model, you can now press the New Tab button on the top
• Removed the code editor
• Removed the tree view
• Added ability to view and edit JSON fields

Try out Prisma Studio in the online demo or in your Prisma project by running:

npx prisma studio --experimental

Preview features

Changes to middlewares and transactionApi

As mentioned above, we ar changing the names of some the top-level Prisma Client methods. This affects the two preview features middlewares and transactionApi.

middlewares

You still need to enable the preview feature via the generator block in your Prisma schema to access it:

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

Before

prisma.on()

After

prisma.$on()

transactionApi

You still need to enable the preview feature via the generator block in your Prisma schema to access it:

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

Before

prisma.transaction()

After

prisma.$transaction()

Already existing preview features from 2.1.0, 2.2.0, and 2.3.0

Just a quick reminder:

• In version 2.1.0 we introduced two experimental features, namely connectOrCreate and transactionApi.
• In 2.2.0 we introduced aggregateApi.
• In 2.3.0 we introduced middlewares and the distinctApi.

In case they're useful for you, please give them a try and share your feedback! These features remain in preview in this release.

🌟 Help us spread the word about Prisma

To help spread the word about Prisma, we'd very much appreciate if you would star this repo 🌟 And if you're excited about the features in this week's release, then help us and share it on Twitter.

Fixes and improvements

prisma

• [Introspection] Do not print @default("") for String fields
• Prisma introspect: extra field for self-relations
• Prisma 1 CLI (for versions smaller than 1.34) and Prisma 2.0 CLI share the same command
• Clarify schema comments
• CreateInput and UpdateInput have different typings for optional relations
• Loss of precision on Float fields
• pgbouncer=true flag not working as expected with a relation and enums on both sides of the relations
• Relation + enums + pgbouncer fails with a ConnectorError
• Implicit m-n relations have an extra @relation field after introspection
• Supported feature combination coverage (MySQL + SQLite)
• Support setting a timeout for SQLite
• CLI won't run in AWS Lambda
• Prisma Client is generated twice when output is specified on generator
• [Re-Introspection] @@map and @map with argument name loses the name after re-introspection
• Surface schema validation errors during Re-Introspection and change flow
• Validate uniqueness of names of indexes on databases which doesn't allow two indexes with the same name
• [Re-Introspection] Enum with @map will get option commented if it invalid without map
• optional one-to-many relationship: none/every filters: Incorrect SQL generated
• Problems with middleware for tracing
• Self-referential one-to-one relationship. Incorrect SQL for { where: { non_fk_field: null } }

prisma-client-js

• Ability to sort results by multiple attributes
• Upsert should allow where clause with only undefined
• AWS Lambda: Unknown error in Prisma Client

migrate

• Migration ID timestamp issue

language-tools

• Add support for /* Prisma */ syntax highlighting
• Extension only release automation
• Publish Insider extension on every commit
• Extension only sub-patch release process
• A patch release for CLI should be based on a tag and not master
• Dev instructions for working with language server are outdated.
• Streamline debugging of language server before published to npm
• Implement workaround for hanging types in LSP
• Run e2e tests before publishing on mac, windows as well
• Blank lines removed in model one by one in VS Code
• Syntax highlighting broken when using map attribute inside enum
• Formatting does not add missing relation automatically
• Show previewFeatures in generator block as auto-completion suggestion
• @unique only gets suggested once per model block
• Recognize usage of experimentalFeatures and give warning and quick-fix
• Using F2 on relation field leads to broken schema
• Snapshot tests for change-readme script

studio

• Unifying model navigation
• Removing the tree view
• Removing the code editor & saved tabs

prisma-engines

• Run test suite on postgres 13 beta in CI
• CI should build binaries for FreeBSD 12
• Improve error for nonexistent table
• Log if query engine is running in PgBouncer mode
• chore: Get rid of outdated dependencies

Today, we are issuing the 2.3.0 stable release.

Major improvements

Rename helper tool in VS Code Extension

The Prisma VS Code extension now helps you rename models, fields and enums:

Introspection uses order of columns in table to order fields in model

prisma introspect until recently ordered all fields in models alphabetically. Starting with 2.3.0 it now picks up the order from the columns in your database table and uses that same order for the fields in your models in your Prisma schema file.

Preview features

Renaming "Experimental features" to "Preview features"

With 2.1.0 we introduced feature flags for Prisma Client, called "Experimental Features". The property in the schema has been renamed from experimentalFeatures to previewFeatures to better communicate what they actually are.

generator client {
   provider             = "prisma-client-js"
-  experimentalFeatures = ["..."]
+  previewFeatures      = ["..."]
}

New: Distinct API

In 2.3.0 we introduce distinct querying capabilities to Prisma Client. It allows you to query for distinct (unique) rows of a model. In other words: The fields you provide in the distinct argument will be duplicate free.

📚 Documentation: Distinct

Feature flag

Distinct querying needs to be enabled with the feature flag discintApi like so:

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

Usage

Distinct is only a filter and doesn't change the return type of the findMany query.

const result = await prisma.user.findMany({
  where: {},
  distinct: ['name']
})

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know if this feature is already ready for production! (If encounter any problems, please open a new issue here).

New: Middlewares API

In 2.3.0 we introduce a Middlewares API as a preview feature. It allows you to hook into the control flow of Prisma Client.

📚 Documentation: Middleware

Feature flag

Middlewares need to be enabled with the feature flag middlewares like so:

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

Usage

Here is an example that prints the execution time of a query made by Prisma Client:

const client = new PrismaClient()

client.use(async (params, next) => {
  console.log('params', params)
  const before = Date.now()
  const result = await next(params)
  const after = Date.now()
  console.log(`Query ${params.model}.${params.action} took ${after - before}ms`)
  return result
})

const data = await client.user.findMany({})

This will log the following:

params {
  args: {},
  dataPath: [],
  runInTransaction: false,
  action: 'findMany',
  model: 'User'
}

Query User.findMany took 2ms

Middlewares allow you to both manipulate the params and the result.

They are called in an "onion" fashion. If you have multiple middlewares, this is the order of things being called:

const client = new PrismaClient()

client.use(async (params, next) => {
  console.log('1')
  const result = await next(params)
  console.log('4')
  return result
})

client.use(async (params, next) => {
  console.log('2')
  const result = await next(params)
  console.log('3')
  return result
})

Prints:

1
2
3
4

While Prisma Client's middlewares work a bit different, they're by inspired Koa's middlewares.

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know if this feature is already ready for production! (If encounter any problems, please open a new issue here).

Already existing preview features from 2.1.0 and 2.2.0

Just a quick reminder: In version 2.1.0 we introduced two experimental features, namely connectOrCreate and transactionApi.
In 2.2.0 we introduced aggregateApi.

In case they're useful for you, please give them a try and let us know! They stay in preview in this release.

Fixes and improvements

prisma

• [Re-Introspection] Keep order of models of existing Prisma schema
• Sort fields by the user's defined order not alphanumerically
• Add minimal integration test for studio
• Default value for relations
• Supporting querying distinct rows
• Omit full path when running prisma --version
• (Re)Introspection does not respect variables in dotenv file when schema is located in a custom directory
• Use denylist in Introspection already
• Clarify usage of datasource url when using SQLite - file: is what works and sqlite: sqlite:// should error
• Postinstall hook fails with custom location for schema.prisma
• [Introspection] Pick up order of columns in database into schema
• Aggregate function is missing typing for where
• [2.2] Not loading ./prisma/.env variables when ./.env exists
• Rename the following items: - "model Transaction" after upgrade from 2.1 to 2.2
• Output feature flags in prisma -v
• Add previewFeatures as replacement for experimentalFeatures

prisma-client-js

• Investigate Amazon RDS Proxy
• Improve automatically generated error reports
• The ID's/techinique used to looked up nest relationships differs and results in a PANIC error, depending on what fields are included or not included in a select
• Middlewares
• Enum value 'false' breaks Prisma client create

migrate

• Add linebreak at the end of migrate.lock file to avoid git history mess
• Using env("DATABASE_URL") freezes migrations
• Changing IDs in explicit n-m relationship results in no change
• @prisma/migrate makes react types required
• Abort on unexecutable migrations
• Null constraint violation on the fields: (id)
• @prisma/cli calling migrate save with a prisma schema error hangs indefinitely

language-tools

• "Error parsing attribute "@@relation": The relation field post on Model foo must specify the fields argument in the @relation directive."
• Add Rename Helper Tool
• Try to execute the binary in the "is binary download needed?" check
• Keep root package.json version in sync with subfolders
• LSP should have its own test cases
• Separate vscode text/links from language server
• Check existing LSP version in CI before publishing
• CI fails if master changes during the run
• Validation for values of @default attributes when they are functions
• Put caret into created model or enum after creating it via quick-fix
• Add quick-fix gif to README
• @updatedAt not showing up for auto-completion
• Auto-completion adds too many quotation marks
• Formatter not working: Cannpt execute binary file

studio

• How to navigate to related records?
• Loading state shows column titles later than actual empty table
• Console is full with ag-grid warnings
• Text editing operations shouldn't change based on data type
• Consider removing the delete button
• Can't select a related row in "add row" dialog
• Deleting required relation crashes
• Table view doesn’t show new field
• Adding relation in new record won’t load more than one record in other table
• Int fields are 0 instead of null

prisma-engines

• Check whether the ForeignKeyDefaultDrop destructive change check is still needed
• [Introspection] Comment out Models that do not have a strict unique criteria
• Rust cuid() implementation may output more than 25 characters
• Examine our implementation of CUIDs
• Forbid to update int fields that are @default(autoincrement())
• Parser Refactoring: Split Field type into RelationField and ScalarField

Credits

Huge thanks to @RafaelKr for helping!

Today, we are issuing the 2.2.2 patch release.

With 2.2.1 the wrong binary was released, this new patch fixes it.

The new binary hash from npx @prisma/[email protected] -v is a9e8c3d97ef2a0cf59256e6b26097f2a80f0a6a4

Today, we are issuing the 2.2.1 patch release.

Fixes and improvements

Prisma CLI

• Not loading ./prisma/.env variables when ./.env exists #2971

Prisma Client JS

• Aggregate function is missing typing for where #2959
• Unable to generate Prisma Client with aggregateApi enabled #772

Studio

• Bug fixes

Today, we are issuing the 2.2.0 stable release.

Major improvements

Next to a lot of bug fixes, this version includes a number of new features!

Quick fix suggestions in Prisma VSCode Extension

The Prisma VS Code extension now helps you fix any accidental typos and offers more convenient "Quick Fixes" right inside your editor.

Make datasource provider dynamic via environment variables

The provider field on the datasource defininitions in your Prisma schema now also accepts an array of strings (instead of just a plain string). This can be helpful when you want to switch environments which point to different databases based on an environment variable. Learn more about this feature here.

datasource db {
  provider = ["sqlite", "postgres"]
  url      = env("DATABASE_URL")
}

Experimental features

Already existing experimental features

Just a quick reminder: In v2.1.0 we introduced two experimental features, namely connectOrCreate and transactionApi. In case they're useful for you, please give them a try! They stay experimental in this release.

Re-Introspection

We were able to continue to improve the re-introspection flow based on your feedback. Thanks!

Aggregations API

While Prisma Client already provides a count query for each model, we now introduce further aggregation functionality for number fields.

If your model has a field of type Int or Float, the following aggregations will be available:

• sum: Sum up all values of a field for the selected rows.
• min: Get the minimum value for a field for the selected rows.
• max: Get the maximum value for a field for the selected rows.
• avg: Get the average value for a field for the selected rows.

Example

Note, that the count API is always available, while sum, min, max, and avg are only available for a model if it has a number field.

Feature flag

In order to enable this experimental feature, note that you need to set the feature flag aggregateApi in the Prisma Client generator block in your schema:

generator client {
  provider = "prisma-client-js"
  experimentalFeatures = ["aggregateApi"]
}

Please share your feedback on how this feature works for you. We are interested in both positive and negative feedback, so we know if this feature is already ready for production! (If encounter any problems, please open a new issue here).

Fixes and improvements

prisma

• Make datasource provider dynamic via environment variables
• Introspection doesn't recognize enums in PostgreSQL
• [Introspection] Investigate if our Introspection queries (can) return the correct order of columns
• PrismaClient throw error on missing is env even when overwriting database connection
• Pagination: cursor is ignored when using negative take argument
• Generator error hidden
• prisma generate throws exit code 0 in case of error caused by denied keywords
• .env not loaded when running prisma migrate (save | up | down) --schema <file-path>
• MySQL: Unreachable database host leads to unknown error
• Prisma format ignores newlines at the end of files
• Prisma introspection should add a trailing newline
• Introspection: Use env vars to override introspection binary with --url --print
• @prisma/engine-core's package.json uses a github url for its version of undici
• Unify prisma introspect and prisma introspect --url wrt re-introspection
• Strange @@relation error message when running prisma generate
• Support selections using basic aggregations
• Re-introspection error for renamed relations when there are multiple relations to the same table
• Introspection doesn't add or loses trailing new line of schema file
• UUID Cursor Pagination
• Schema validation: @default(autoincrement()) on a String field should be rejected
• Switch to undici
• Introspect removes experimental features from schema
• Add new line checks to integration tests

prisma-client-js

• Add more integration tests for unhappy paths
• Better error message when generated Client and local Prisma schema are out of sync
• Multiple quick count requests result in Invalid prisma.web_hooks.count invocation
• Wrong jsDocs annotations for create and delete
• Raw API concerns
• Improve sqlite error messages
• JSON Filter does not work
• Bulk Operations Naming Conflict
• Transaction API is not writing the transaction
• Expose debug mode to trigger a panic for reproductions

migrate

• migrate up tells me to read in ./migrations/MIGRATION_ID/README.md
• Create new SQL Database missing first char on name
• migrate save/up hangs while creating database
• README.md of migration folder includes quaint in SQL
• Validate @default(now()) so that it only can be used in DateTime fields

vscode

• Publish lsp server to npm
• Quick Fix for fields with unknown types
• Remove test cases that point to a four digit versioning scheme
• Run integration tests on an extension release
• Add instructions for each feature in README
• Composite keys are not considered valid?
• README on marketplace does not show Prisma Logo
• Lerna bootstrap does not install all dependencies
• Connection to server got closed. Server will restart.
• Investigate *** in GH action output

studio

• Allow for column resizing
• Table scroll area should start underneath table header
• Select range of records when holding shift key
• Table's loading illustration may overlap with the table content
• Tooltips for table header columns
• Type to edit Enums and Booleans with autocomplete
• Copy & paste from a cell does not seem to work
• A better databrowser table
• In the "connect to X" view I would like to connect when clicking on a row
• Loading state shows unexpected data
• Scrolling down in relation accordion does not show entries
• After editing relation only the number of unsaved changes is shown in main table
• Styling of editing Int is different from editing String
• Loading state is not really indicated

prisma-engines

• Mask Datasource URLs in all artifacts generated by the Migration Engine
• Do not try setting database defaults for @default(dbgenerated()) fields.
• Multiple datasources in schema must be forbidden