Atlas Configuration

atlas

The atlas is an optional block for configuring Atlas login credentials.

atlas {
  cloud {
    org = "org-name"
  }
}

atlas blocks

atlas.cloud

atlas.cloud attributes

Name and description	Required	Value
org The org attribute specifies the organization to log in to. If Atlas executes using atlas.hcl without logging in to the specified organization, the command will be aborted.	false	string
token	false	string

atlas.cloud constraints

Constraint	Value
Required	true
Require Name	false
One of required sets	[org, token]

check migrate_apply

check.migrate_apply blocks

check.allow

check.allow attributes

Name	Required	Value
condition	true	bool

check.allow constraints

Constraint	Value
Required	false
Require Name (e.g., check.allow "name" )	true

check.deny

check.deny attributes

Name	Required	Value
condition	true	bool
message	false	string

check.deny constraints

Constraint	Value
Required	false
Require Name (e.g., check.deny "name" )	true

check.migrate_apply constraints

Constraint	Value
Required	false
Repeatable	true

data aws_rds_token

The aws_rds_token data source generates a short-lived token for an AWS RDS database using IAM Authentication.

data "aws_rds_token" "db" {
  region = "us-east-1"
  endpoint = local.endpoint
  username = local.user
}

data.aws_rds_token attributes

Name and description	Required	Value
endpoint	true	string
profile The AWS profile name to use from the shared credentials file. If not set, the default profile will be used.	false	string
region	false	string
role_arn ARN of the role to assume for generating the token.	false	string
username	true	string

data blob_dir

The blob_dir data source reads the state of a migration directory from the blob storage. Currently, it supports S3 (s3://).

data "blob_dir" "migrations" {
  url = "s3://my-bucket/migrations?region=us-east-1&profile=aws-atlas"
}

data.blob_dir attributes

Name	Required	Value
url	true	string

data.blob_dir exposed references

Name	Value
url	string

data cloud_databases

The cloud_databases reads the status of migration targets from Atlas Cloud.

data "cloud_databases" "dev" {
  repo = "my-app"
  env  = "dev"
}

data.cloud_databases attributes

Name and description	Required	Value
env	false	string
repo The repo attribute specifies the migrate-directory name to filter the databases by.	true	string

data.cloud_databases exposed references

Name	Value
targets	any

data composite_schema

The composite_schema data source allows the composition of multiple Atlas schemas into a unified schema graph.

data "composite_schema" "app" {
  schema "inventory" {
    url = data.external_schema.sqlalchemy.url
  }
  schema "graph" {
    url = "ent://ent/schema"
  }
  ...
}

data.composite_schema exposed references

Name	Value
url	string

data.composite_schema blocks

data.schema

data.schema attributes

Name	Required	Value
url	true	Composite_schema url can be one of: Object reference string

data.schema constraints

Constraint	Value
Required	true
Require Name	false
Repeatable	true

data external

The external allows the execution of an external program and uses its output in the project.

data "external" "dot_env" {
  program = [
    "npm",
    "run",
    "load-env.js"
  ]
}

data.external attributes

Name	Required	Value
program	true	List of strings
working_dir	false	string

data external_schema

The external_schema data source allows the import of an SQL schema from an external program into Atlas' desired state.

data "external_schema" "graph" {
  program = [
    "npm",
    "run",
    "generate-schema"
  ]
}

data.external_schema attributes

Name	Required	Value
program	true	List of strings
working_dir	false	string

data.external_schema exposed references

Name	Value
url	string

data gcp_cloudsql_token

The gcp_cloudsql_token data source generates a short-lived token for an GCP CloudSQL database using IAM Authentication.

data "gcp_cloudsql_token" "db" {}

data.gcp_cloudsql_token attributes

Name	Required	Value
region	false	string

data hcl_schema

The hcl_schema data source allows the loading of an Atlas HCL schema from a file or directory, with optional variables.

data "hcl_schema" "app" {
  path = "schema.hcl"
  vars = {
    tenant = var.tenant
  }
}

data.hcl_schema attributes

Name	Required	Value
path	false	string
paths	false	List of strings
vars	false	map

data.hcl_schema exposed references

Name	Value
url	string

data http

The http data source makes an HTTP GET/HEAD/POST request to the given URL and exports information about the response. The given URL may be either an http or https URL.

data.http attributes

Name and description	Required	Value
ca_cert_pem Certificate Authority (CA) in PEM (RFC 1421) format.	false	string
client_cert_pem Client certificate in PEM (RFC 1421) format.	false	string
client_key_pem Client key in PEM (RFC 1421) format.	false	string
insecure Disables verification of the server's certificate chain and hostname. Defaults to false.	false	bool
method The HTTP Method for the request. Allowed methods are a subset of methods defined in RFC7231 namely, GET, HEAD, and POST. POST support is only intended for read-only URLs, such as submitting a search. If omitted, the default method is GET.	false	enum (GET, HEAD, POST)
request_body The request body as a string.	false	string
request_headers A map of request header field names and values.	false	map
request_timeout_ms The request timeout in milliseconds.	false	int
url The URL for the request. Supported schemes are http and https.	true	string

data.http exposed references

Name and description	Value
response_body The response body returned as a string.	string
response_body_base64 The response body encoded as base64 (standard) as defined in RFC 4648.	string
response_headers A map of response header field names and values. Duplicate headers are concatenated according to RFC2616.	map
status_code The HTTP response status code.	int
url The URL used for the request.	string

data.http blocks

data.retry

Retry request configuration. By default there are no retries. Configuring this block will result in retries if an error is returned by the client (e.g., connection errors) or if a 5xx-range (except 501) status code is received.

data.retry attributes

Name and description	Required	Value
attempts The number of times the request is to be retried. For example, if 2 is specified, the request will be tried a maximum of 3 times.	false	int
max_delay_ms The maximum delay between retry requests in milliseconds.	false	int
min_delay_ms The minimum delay between retry requests in milliseconds.	false	int

data remote_dir

The remote_dir data source reads the state of a migration directory from Atlas Cloud.

data "remote_dir" "migrations" {
  name = "myapp"
}

data.remote_dir attributes

Name	Required	Value
name	true	string
tag	false	string

data.remote_dir exposed references

Name	Value
url	string

data remote_schema

data.remote_schema attributes

Name	Required	Value
name	true	string
tag	false	string
version	false	string

data.remote_schema exposed references

Name	Value
url	string

data runtimevar

The runtimevar data source allows fetching runtime variables from a remote source.

data.runtimevar attributes

Name	Required	Value
url	true	string

data sql

The sql data source allows executing SQL queries and using their result in the project.

data "sql" "tenants" {
  url   = var.url
  query = "..."
  args  = [var.pattern]
}

data.sql attributes

Name	Required	Value
args	false	Any value
query	true	string
url	true	string

data.sql exposed references

Name	Value
count	int
value	string
values	strings

data template_dir

The template_dir data source renders a migration directory from a template directory.

data "template_dir" "app" {
  path = var.path
  vars = {
    Key1 = "value1"
    Key2 = "value2"
  }
}

data.template_dir attributes

Name	Required	Value
path	true	string
vars	false	map

data.template_dir exposed references

Name	Value
url	string

deployment

The deployment block defines a multi-tenant or deployment strategy with groups.

deployment "staged" {
  variable "tier" {
    type = string
  }
  group "canary" {
    match    = var.tier == "canary"
    on_error = FAIL
  }
  group "rest" {
    depends_on = [group.canary]
    parallel   = 32
    on_error   = CONTINUE
  }
}

deployment blocks

deployment.group

The group block defines a deployment group with its execution settings.

deployment.group attributes

Name and description	Required	Value
depends_on List of groups that must complete before this group starts.	false	List of object reference to group
match Expression to filter which tenants belong to this group. Uses var.* to reference deployment variables.	false	Any value
on_error Error handling mode: FAIL stops the entire deployment, CONTINUE logs and proceeds.	false	enum (FAIL, CONTINUE)
order_by Expression or list of expressions to sort tenants within this group.	false	Any value
parallel Maximum number of tenants to process in parallel within this group. Defaults to 1.	false	int

deployment.group constraints

Constraint	Value
Required	false
Require Name (e.g., deployment.group "name" )	true
Repeatable	true

deployment.variable

deployment.variable attributes

Name	Required	Value
default	false	Any value
description	false	string
type	true	HCL type (string, number, bool, list(string), etc)

deployment.variable blocks

deployment.variable.validation

deployment.variable.validation attributes

Name	Required	Value
condition	true	bool
error_message	false	string

deployment.variable.validation constraints

Constraint	Value
Required	false
Require Name	false
Repeatable	true

deployment.variable constraints

Constraint	Value
Required	false
Require Name (e.g., deployment.variable "name" )	true

deployment constraints

Constraint	Value
Required	false
Require Name (e.g., deployment "name" )	true

dev clickhouse

dev.clickhouse attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.clickhouse exposed references

Name	Value
url	string

dev.clickhouse blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

dev mariadb

dev.mariadb attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.mariadb exposed references

Name	Value
url	string

dev.mariadb blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

dev mysql

dev.mysql attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.mysql exposed references

Name	Value
url	string

dev.mysql blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

dev postgres

dev.postgres attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.postgres exposed references

Name	Value
url	string

dev.postgres blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

dev redshift

dev.redshift attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.redshift exposed references

Name	Value
url	string

dev.redshift blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

dev snowflake

dev.snowflake attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.snowflake exposed references

Name	Value
url	string

dev.snowflake blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

dev sqlserver

dev.sqlserver attributes

Name	Required	Value
baseline	false	string
url	true	string

dev.sqlserver exposed references

Name	Value
url	string

dev.sqlserver blocks

dev.connection

The connection block defines the connection configuration for the dev-database.

dev "postgres" "postgis" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

dev.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

diff clickhouse

diff.clickhouse blocks

diff.add_table

diff.add_table attributes

Name	Required	Value
if_not_exists	false	bool

diff.cluster

diff.cluster attributes

Name	Required	Value
name	false	string

diff.concurrent_index

diff.concurrent_index attributes

Name	Required	Value
add	false	bool
create	false	bool
drop	false	bool

diff.drop_table

diff.drop_table attributes

Name	Required	Value
cascade	false	bool
if_exists	false	bool

diff.materialized

diff.materialized attributes

Name	Required	Value
with_no_data	false	bool

diff.session_settings

diff.session_settings attributes

Name	Required	Value
match	true	string
values	false	map

diff.session_settings constraints

Constraint	Value
Required	false
Require Name	false
Repeatable	true

diff.skip

diff.skip attributes

Name	Required	Value
add_column	false	bool
add_extension	false	bool
add_foreign_key	false	bool
add_func	false	bool
add_index	false	bool
add_proc	false	bool
add_schema	false	bool
add_table	false	bool
add_trigger	false	bool
add_view	false	bool
drop_column	false	bool
drop_extension	false	bool
drop_foreign_key	false	bool
drop_func	false	bool
drop_index	false	bool
drop_proc	false	bool
drop_role	false	bool
drop_schema	false	bool
drop_table	false	bool
drop_trigger	false	bool
drop_user	false	bool
drop_view	false	bool
modify_column	false	bool
modify_extension	false	bool
modify_foreign_key	false	bool
modify_func	false	bool
modify_index	false	bool
modify_proc	false	bool
modify_schema	false	bool
modify_table	false	bool
modify_trigger	false	bool
modify_view	false	bool
rename_constraint	false	bool
rename_func	false	bool
rename_proc	false	bool
rename_table	false	bool
rename_trigger	false	bool
rename_view	false	bool

docker clickhouse

docker.clickhouse attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.clickhouse exposed references

Name	Value
url	string

docker.clickhouse blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

docker dsql

docker.dsql attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.dsql exposed references

Name	Value
url	string

docker.dsql blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

docker mariadb

docker.mariadb attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.mariadb exposed references

Name	Value
url	string

docker.mariadb blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

docker mysql

docker.mysql attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.mysql exposed references

Name	Value
url	string

docker.mysql blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

docker oracle

docker.oracle attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
database The database attribute specifies the Oracle Pluggable Database (PDB) name to connect to. If not specified, the default database is used. This option is only available for the Oracle Enterprise Image.	false	string
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
mode	false	enum (database, schema)
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.oracle exposed references

Name	Value
url	string

docker.oracle blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

docker postgres

docker.postgres attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
database	false	string
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.postgres exposed references

Name	Value
url	string

docker.postgres blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

docker sqlserver

docker.sqlserver attributes

Name and description	Required	Value
args Additional arguments to pass to the docker run command when starting the container.	false	List of strings
baseline	false	string
collate	false	string
command The command attribute specifies the command and its argument to run in the container overriding the default one.	false	List of strings
database	false	string
env	false	List of strings
image	true	string
init The init attribute specifies the initialization script for the Docker-based database created in Atlas. docker "mysql" "dev" { image = "mysql:8.4" init = <<-SQL SET GLOBAL log_bin_trust_function_creators=true; SET GLOBAL restrict_fk_on_non_standard_key=false; SET GLOBAL sql_mode := REPLACE(@@sql_mode, 'NO_ZERO_DATE', ''); SQL }	false	string
mode	false	enum (database, schema)
platform The platform attribute specifies the platform for the container. It defaults to the host platform if not specified. Examples: linux/amd64, linux/arm64, linux/arm/v7, etc.	false	string
schema	false	string
timeout The timeout attribute specifies the timeout for waiting the container being ready to use. Defaults to 1m if not specified.	false	string

docker.sqlserver exposed references

Name	Value
url	string

docker.sqlserver blocks

docker.build

The build block defines the build configuration for the Docker image. The given image is built using the provided Dockerfile and context.

docker "postgres" "pg-vn" {
  image = "postgres:15-vn"
  build {
    context    = "."
    dockerfile = "locale.Dockerfile"
    args = {
      LOCALE = "vi-VN"
    }
  }
}

Alternatively, you can define the Dockerfile inline:

docker "postgres" "dev" {
  image = "postgres:15-custom"
  build {
    context = "."
    dockerfile_inline = <<-EOF
      FROM postgres:15
      RUN apt-get update && apt-get install -y curl
    EOF
  }
}

docker.build attributes

Name and description	Required	Value
args	false	map
context	true	string
dockerfile	false	string
dockerfile_inline The dockerfile_inline attribute allows defining the Dockerfile content inline. This is useful for simple Dockerfiles that don't require a separate file. Cannot be used together with the dockerfile attribute.	false	string
platform	false	List of strings
target	false	string

docker.build constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[dockerfile, dockerfile_inline]

docker.connection

The connection block defines the connection configuration for the Docker-based database created in Atlas.

docker "postgres" "dev" {
  ...
  connection {
    max_open      = 1
    max_idle	  = 1
    max_lifetime  = "30s"
    max_idle_time = "30s"
  }
}

docker.connection attributes

Name	Required	Value
max_idle	false	int
max_idle_time	false	string
max_lifetime	false	string
max_open	false	int

env

The env block describes an environment block that can be selected by using the --env flag.

env attributes

Name	Required	Value
dev	false	string
exclude	false	List of strings
include	false	List of strings
name	false	string
schemas	false	List of strings
src	false	Schema source can be one of: Object reference string List of strings
url	false	string

env blocks

env.check.migrate_apply

env.check.migrate_apply blocks

env.check.allow

env.check.allow attributes

Name	Required	Value
condition	true	bool

env.check.allow constraints

Constraint	Value
Required	false
Require Name (e.g., env.check.allow "name" )	true

env.check.deny

env.check.deny attributes

Name	Required	Value
condition	true	bool
message	false	string

env.check.deny constraints

Constraint	Value
Required	false
Require Name (e.g., env.check.deny "name" )	true

env.check.migrate_apply constraints

Constraint	Value
Required	false
Repeatable	true

env.data

The data block configures static/lookup data management.

data {
  mode    = UPSERT
  include = ["countries", "currencies"]
  exclude = ["temp_*"]
}

env.data attributes

Name and description	Required	Value
exclude Exclude tables matching patterns. Supports glob patterns.	false	List of strings
include Include only tables matching patterns. Supports glob patterns.	false	List of strings
max_rows Maximum number of rows to manage per table. Value is required when querying a database in SYNC mode	false	int
mode The sync mode: INSERT (new rows only), UPSERT (insert/update), or SYNC (insert/update/delete).	true	enum (INSERT, UPSERT, SYNC)
skip_diff Skip objects from diff comparison. Supports glob patterns (e.g., *.updated_at).	false	List of strings

env.diff.clickhouse

env.diff.clickhouse blocks

env.diff.add_table

env.diff.add_table attributes

Name	Required	Value
if_not_exists	false	bool

env.diff.cluster

env.diff.cluster attributes

Name	Required	Value
name	false	string

env.diff.concurrent_index

env.diff.concurrent_index attributes

Name	Required	Value
add	false	bool
create	false	bool
drop	false	bool

env.diff.drop_table

env.diff.drop_table attributes

Name	Required	Value
cascade	false	bool
if_exists	false	bool

env.diff.materialized

env.diff.materialized attributes

Name	Required	Value
with_no_data	false	bool

env.diff.session_settings

env.diff.session_settings attributes

Name	Required	Value
match	true	string
values	false	map

env.diff.session_settings constraints

Constraint	Value
Required	false
Require Name	false
Repeatable	true

env.diff.skip

env.diff.skip attributes

Name	Required	Value
add_column	false	bool
add_extension	false	bool
add_foreign_key	false	bool
add_func	false	bool
add_index	false	bool
add_proc	false	bool
add_schema	false	bool
add_table	false	bool
add_trigger	false	bool
add_view	false	bool
drop_column	false	bool
drop_extension	false	bool
drop_foreign_key	false	bool
drop_func	false	bool
drop_index	false	bool
drop_proc	false	bool
drop_role	false	bool
drop_schema	false	bool
drop_table	false	bool
drop_trigger	false	bool
drop_user	false	bool
drop_view	false	bool
modify_column	false	bool
modify_extension	false	bool
modify_foreign_key	false	bool
modify_func	false	bool
modify_index	false	bool
modify_proc	false	bool
modify_schema	false	bool
modify_table	false	bool
modify_trigger	false	bool
modify_view	false	bool
rename_constraint	false	bool
rename_func	false	bool
rename_proc	false	bool
rename_table	false	bool
rename_trigger	false	bool
rename_view	false	bool

env.export

The export block configures schema export destinations.

export {
  schema {
    inspect = exporter.sql.dir
    diff    = exporter.http.webhook
  }
}

env.export blocks

env.export.schema

Configure schema export.

env.export.schema attributes

Name and description	Required	Value
diff Exporter reference for 'schema diff' command.	false	Object reference to exporter
inspect Exporter reference for 'schema inspect' command.	false	Object reference to exporter

env.format

The format block defines the output format of the different commands

format {
  migrate {
    diff = "{{ sql . \"  \" }}"
  }
}

env.format blocks

env.format.migrate

env.format.migrate attributes

Name	Required	Value
apply	false	string
diff	false	string
down	false	string
lint	false	string
status	false	string

env.format.schema

env.format.schema attributes

Name	Required	Value
apply	false	string
diff	false	string
inspect	false	string

env.lint

The lint block defines the linting configuration.

lint {
  destructive {
    error = false
  }
  concurrent_index {
    error = true
  }
  naming {
    force = true
  }
}

env.lint attributes

Name	Required	Value
format	false	string
latest	false	int
log	false	string
review	false	enum (ALWAYS, WARNING, ERROR)

env.lint blocks

env.lint.check

The check block defines the configuration of a specific lint check.

check "PG301" {
  error = true
}

check "NM101" {
  skip = true
}

env.lint.check attributes

Name and description	Required	Value
error The error attribute specifies whether lint should error when the check fails.	false	bool
force The force attribute specifies whether this check should be enforced even if the atlas:nolint directive is set	false	bool
skip The skip attribute specifies whether lint should skip the check.	false	bool

env.lint.check constraints

Constraint	Value
Required	false
Require Name (e.g., env.lint.check "name" )	true
Mutually exclusive sets	[error, skip, force]
One of required sets	[error, skip, force]

env.lint.concurrent_index

env.lint.concurrent_index attributes

Name	Required	Value
check_create	false	bool
check_drop	false	bool
check_txmode	false	bool
error	false	bool
force	false	bool

env.lint.condrop

env.lint.condrop attributes

Name	Required	Value
error	false	bool
force	false	bool

env.lint.data_depend

env.lint.data_depend attributes

Name	Required	Value
error	false	bool
force	false	bool

env.lint.destructive

env.lint.destructive attributes

Name	Required	Value
error	false	bool
force	false	bool

env.lint.destructive blocks

env.lint.destructive.allow_column

env.lint.destructive.allow_column attributes

Name	Required	Value
match	false	string

env.lint.destructive.allow_schema

env.lint.destructive.allow_schema attributes

Name	Required	Value
match	false	string

env.lint.destructive.allow_table

env.lint.destructive.allow_table attributes

Name	Required	Value
match	false	string

env.lint.git

env.lint.git attributes

Name	Required	Value
base	false	string
dir	false	string

env.lint.incompatible

env.lint.incompatible attributes

Name	Required	Value
error	false	bool
force	false	bool

env.lint.naming

env.lint.naming attributes

Name	Required	Value
error	false	bool
force	false	bool
match	false	string
message	false	string

env.lint.naming blocks

env.lint.naming.check

env.lint.naming.check attributes

Name	Required	Value
match	false	string
message	false	string

env.lint.naming.column

env.lint.naming.column attributes

Name	Required	Value
match	false	string
message	false	string

env.lint.naming.foreign_key

env.lint.naming.foreign_key attributes

Name	Required	Value
match	false	string
message	false	string

env.lint.naming.index

env.lint.naming.index attributes

Name	Required	Value
match	false	string
message	false	string

env.lint.naming.schema

env.lint.naming.schema attributes

Name	Required	Value
match	false	string
message	false	string

env.lint.naming.table

env.lint.naming.table attributes

Name	Required	Value
match	false	string
message	false	string

env.lint.nestedtx

env.lint.nestedtx attributes

Name	Required	Value
error	false	bool
force	false	bool

env.lint.non_linear

env.lint.non_linear attributes

Name	Required	Value
error	false	bool
on_edit	false	enum (IGNORE, WARN, ERROR)

env.lint.rule.hcl

env.lint.rule.hcl attributes

Name	Required	Value
error	false	bool
src	true	List of strings
vars	false	map

env.lint.statement

The statement block defines the configuration of the statement analyzer.

statement {
  allow "only-alter-create" {
    match = "(ALTER|CREATE) .+"
  }
  allow "delete-with-filter" {
    match = "DELETE .+ WHERE .+"
  }
  deny "reject-delete" {
    match = "DELETE .+"
  }
  deny "reject-truncate" {
    match = "TRUNCATE .+"
  }
}

env.lint.statement attributes

Name	Required	Value
error	false	bool
force	false	bool

env.lint.statement blocks

env.lint.statement.allow

env.lint.statement.allow attributes

Name	Required	Value
match	true	string

env.lint.statement.allow constraints

Constraint	Value
Required	false
Require Name (e.g., env.lint.statement.allow "name" )	true

env.lint.statement.deny

env.lint.statement.deny attributes

Name	Required	Value
match	true	string
message	false	string

env.lint.statement.deny constraints

Constraint	Value
Required	false
Require Name (e.g., env.lint.statement.deny "name" )	true

env.lint.statement constraints

Constraint	Value
Required	false
Require Name	false
One of required sets	[deny, allow]

env.lint constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[latest, git], [log, format]

env.log

Use the format block instead

env.log blocks

env.log.migrate

env.log.migrate attributes

Name	Required	Value
apply	false	string
diff	false	string
down	false	string
lint	false	string
status	false	string

env.log.schema

env.log.schema attributes

Name	Required	Value
apply	false	string
diff	false	string
inspect	false	string

env.migration

The migration block defines the migration configuration

migration {
  dir = "file://migrations"
  repo {
    name = "app"
  }
}

env.migration attributes

Name and description	Required	Value
baseline	false	string
dir	false	string
exclude	false	List of strings
exec_order	false	enum (LINEAR, linear, LINEAR_SKIP, linear-skip, NON_LINEAR, non-linear)
format	false	Migration format can be one of: enum (atlas, golang-migrate, goose, flyway, liquibase, dbmate) string
lock_name	false	string
lock_timeout	false	string
revisions_schema	false	string
skip_lock	false	bool
skip_report Skip reporting migrations to Atlas Cloud. Useful for temporary environments (e.g., PR previews).	false	bool
to_version	false	string
tx_mode	false	Transaction mode can be one of: string enum (none, file, all)

env.migration blocks

env.migration.repo

The repository configuration for the migrations directory in the registry

repo {
  name = "app"
}

env.migration.repo attributes

Name	Required	Value
name	true	string

env.post.migrate_apply

env.post.migrate_apply blocks

env.post.exec

env.post.exec attributes

Name	Required	Value
on_error	false	enum (CONTINUE, BREAK)
sql	false	string
src	false	string

env.post.exec constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[src, sql]
One of required sets	[src, sql]

env.post.skip

env.post.skip attributes

Name	Required	Value
condition	true	bool
message	false	string

env.post.migrate_apply constraints

Constraint	Value
Required	false
Repeatable	true

env.post.schema_apply

env.post.schema_apply blocks

env.post.exec

env.post.exec attributes

Name	Required	Value
on_error	false	enum (CONTINUE, BREAK)
sql	false	string
src	false	string

env.post.exec constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[src, sql]
One of required sets	[src, sql]

env.post.skip

env.post.skip attributes

Name	Required	Value
condition	true	bool
message	false	string

env.post.schema_apply constraints

Constraint	Value
Required	false
Repeatable	true

env.pre.migrate_apply

env.pre.migrate_apply blocks

env.pre.exec

env.pre.exec attributes

Name	Required	Value
on_error	false	enum (CONTINUE, BREAK, FAIL)
sql	false	string
src	false	string

env.pre.exec constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[src, sql]
One of required sets	[src, sql]

env.pre.skip

env.pre.skip attributes

Name	Required	Value
condition	true	bool
message	false	string

env.pre.migrate_apply constraints

Constraint	Value
Required	false
Repeatable	true

env.pre.schema_apply

env.pre.schema_apply blocks

env.pre.exec

env.pre.exec attributes

Name	Required	Value
on_error	false	enum (CONTINUE, BREAK, FAIL)
sql	false	string
src	false	string

env.pre.exec constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[src, sql]
One of required sets	[src, sql]

env.pre.skip

env.pre.skip attributes

Name	Required	Value
condition	true	bool
message	false	string

env.pre.schema_apply constraints

Constraint	Value
Required	false
Repeatable	true

env.rollout

The rollout block configures multi-tenant or multi-environment deployment for this environment.

rollout {
  deployment = deployment.staged
  vars = {
    tier   = each.value.tier
    region = each.value.region
  }
}

env.rollout attributes

Name and description	Required	Value
deployment Reference to a deployment block that defines the rollout strategy.	true	Object reference to deployment
vars Map of variable values to pass to the deployment. Keys must match the deployment's variable declarations.	false	map

env.schema

The configuration for the desired schema

schema {
  src = "file://schema.hcl"
  repo {
    name = "app"
  }
}

env.schema attributes

Name	Required	Value
src	false	Schema source can be one of: Object reference string List of strings

env.schema blocks

env.schema.mode

The inspection mode configuration.

By default, tables, views, funcs, types, objects, and triggers are enabled. Roles and permissions are disabled by default. Sensitive values are denied by default.

// Enable roles and permissions inspection.
mode {
  roles       = true
  permissions = true
}

// Disable triggers and functions inspection.
mode {
  triggers = false
  funcs    = false
}

// Allow sensitive values (passwords) in migration planning.
mode {
  roles     = true
  sensitive = ALLOW
}

env.schema.mode attributes

Name and description	Required	Value
funcs	false	bool
objects	false	bool
permissions	false	bool
roles	false	bool
sensitive Control sensitive values (e.g., passwords) in migration planning. Valid values: ALLOW or DENY (default).	false	enum (ALLOW, DENY)
tables	false	bool
triggers	false	bool
types	false	bool
views	false	bool

env.schema.repo

The repository configuration for the desired schema in the registry

repo {
  name = "app"
}

env.schema.repo attributes

Name	Required	Value
name	true	string

env.test

The test block defines the testing configuration.

test {
  schema {
    src = ["schema.test.hcl"]
    vars = {
      seed_file = "filename.sql"
      variable2 = var.name
      variable3 = data.external.value
    }
  }
}

env.test blocks

env.test.migrate

env.test.migrate attributes

Name	Required	Value
src	true	List of strings
vars	false	map

env.test.schema

env.test.schema attributes

Name	Required	Value
src	true	List of strings
vars	false	map

env constraints

Constraint	Value
Required	false
Require Name	false
Repeatable	true
Allow unknown attributes	true
Mutually exclusive sets	[src, schema]

exporter hcl

Export schema as HCL.

// Single file output.
exporter "hcl" "file" {
  path = "schema.hcl"
}

// Split by object (directory output).
exporter "hcl" "dir" {
  path     = "schema/hcl"
  split_by = object
  naming   = lower
}

exporter.hcl attributes

Name and description	Required	Value
ext File extension for split files (defaults to '.hcl').	false	string
naming Naming convention for split files: same (preserve), lower (lowercase), upper (uppercase).	false	enum (same, lower, upper)
path Output path. If split_by is set, this is the directory; otherwise, it's the file path.	true	string
split_by Split strategy: object (per object), schema (per schema), type (per type).	false	enum (object, schema, type)

exporter http

Export schema via HTTP request.

// Using a template for the body.
exporter "http" "webhook" {
  url           = "https://api.example.com/schemas"
  method        = "POST"
  body_template = "{{ sql . }}"
  headers = {
    "Content-Type"  = "application/json"
    "Authorization" = "Bearer ${var.token}"
  }
}

// Using a non-template body.
exporter "http" "simple" {
  url    = "https://api.example.com/schemas"
  method = "POST"
  body   = "simple request body"
}

exporter.http attributes

Name and description	Required	Value
body The request body as a simple string.	false	string
body_template The request body template. Supports template functions like sql, json, mermaid, etc.	false	string
ca_cert_pem Certificate Authority (CA) in PEM (RFC 1421) format.	false	string
client_cert_pem Client certificate in PEM (RFC 1421) format.	false	string
client_key_pem Client key in PEM (RFC 1421) format.	false	string
headers A map of request header field names and values.	false	map
insecure Disables verification of the server's certificate chain and hostname. Defaults to false.	false	bool
method The HTTP method for the request (e.g., GET, POST, PUT, PATCH).	true	string
request_timeout_ms The request timeout in milliseconds.	false	int
url The URL for the request. Supported schemes are http and https.	true	string

exporter multi

Export schema to multiple exporters.

// Export to multiple HTTP endpoints with fail-fast behavior (default).
exporter "multi" "webhooks" {
  exporters = [
    exporter.http.webhook1,
    exporter.http.webhook2,
  ]
}

// Export to both SQL and HCL formats, continue on errors.
exporter "multi" "all" {
  exporters = [
    exporter.sql.file,
    exporter.hcl.file,
  ]
  on_error = CONTINUE // Defaults to FAIL.
}

exporter.multi attributes

Name and description	Required	Value
exporters List of exporter references to execute in sequence.	true	List of object reference to exporter
on_error Error handling mode. Defaults to FAIL (stop on first error). Options: FAIL (default), CONTINUE (continue but collect errors), IGNORE (silently ignore errors).	false	enum (FAIL, CONTINUE, IGNORE)

exporter sql

Export schema as SQL.

// Single file output.
exporter "sql" "file" {
  path   = "schema.sql"
  indent = "  "
}

// Split by object (directory output).
exporter "sql" "dir" {
  path     = "schema/sql"
  split_by = object
  naming   = lower
}

exporter.sql attributes

Name and description	Required	Value
indent Indentation string for SQL output (e.g., " " for 2 spaces).	false	string
naming Naming convention for split files: same (preserve), lower (lowercase), upper (uppercase).	false	enum (same, lower, upper)
path Output path. If split_by is set, this is the directory; otherwise, it's the file path.	true	string
split_by Split strategy. Only object is supported for SQL.	false	enum (object)

hook sql

The hook "sql" "name" block defines an SQL hook configuration.

hook "sql" "timeout" {
  transaction {
    after_begin = [
      "SET statement_timeout TO '50ms'",
    ]
  }
}

hook.sql attributes

Name	Required	Value
skip_errors	false	bool

hook.sql blocks

hook.transaction

hook.transaction attributes

Name	Required	Value
after_begin	false	List of strings
before_commit	false	List of strings
before_rollback	false	List of strings

lint

The lint block defines the linting configuration.

lint {
  destructive {
    error = false
  }
  concurrent_index {
    error = true
  }
  naming {
    force = true
  }
}

lint attributes

Name	Required	Value
format	false	string
latest	false	int
log	false	string
review	false	enum (ALWAYS, WARNING, ERROR)

lint blocks

lint.check

The check block defines the configuration of a specific lint check.

check "PG301" {
  error = true
}

check "NM101" {
  skip = true
}

lint.check attributes

Name and description	Required	Value
error The error attribute specifies whether lint should error when the check fails.	false	bool
force The force attribute specifies whether this check should be enforced even if the atlas:nolint directive is set	false	bool
skip The skip attribute specifies whether lint should skip the check.	false	bool

lint.check constraints

Constraint	Value
Required	false
Require Name (e.g., lint.check "name" )	true
Mutually exclusive sets	[error, skip, force]
One of required sets	[error, skip, force]

lint.concurrent_index

lint.concurrent_index attributes

Name	Required	Value
check_create	false	bool
check_drop	false	bool
check_txmode	false	bool
error	false	bool
force	false	bool

lint.condrop

lint.condrop attributes

Name	Required	Value
error	false	bool
force	false	bool

lint.data_depend

lint.data_depend attributes

Name	Required	Value
error	false	bool
force	false	bool

lint.destructive

lint.destructive attributes

Name	Required	Value
error	false	bool
force	false	bool

lint.destructive blocks

lint.destructive.allow_column

lint.destructive.allow_column attributes

Name	Required	Value
match	false	string

lint.destructive.allow_schema

lint.destructive.allow_schema attributes

Name	Required	Value
match	false	string

lint.destructive.allow_table

lint.destructive.allow_table attributes

Name	Required	Value
match	false	string

lint.git

lint.git attributes

Name	Required	Value
base	false	string
dir	false	string

lint.incompatible

lint.incompatible attributes

Name	Required	Value
error	false	bool
force	false	bool

lint.naming

lint.naming attributes

Name	Required	Value
error	false	bool
force	false	bool
match	false	string
message	false	string

lint.naming blocks

lint.naming.check

lint.naming.check attributes

Name	Required	Value
match	false	string
message	false	string

lint.naming.column

lint.naming.column attributes

Name	Required	Value
match	false	string
message	false	string

lint.naming.foreign_key

lint.naming.foreign_key attributes

Name	Required	Value
match	false	string
message	false	string

lint.naming.index

lint.naming.index attributes

Name	Required	Value
match	false	string
message	false	string

lint.naming.schema

lint.naming.schema attributes

Name	Required	Value
match	false	string
message	false	string

lint.naming.table

lint.naming.table attributes

Name	Required	Value
match	false	string
message	false	string

lint.nestedtx

lint.nestedtx attributes

Name	Required	Value
error	false	bool
force	false	bool

lint.non_linear

lint.non_linear attributes

Name	Required	Value
error	false	bool
on_edit	false	enum (IGNORE, WARN, ERROR)

lint.rule.hcl

lint.rule.hcl attributes

Name	Required	Value
error	false	bool
src	true	List of strings
vars	false	map

lint.statement

The statement block defines the configuration of the statement analyzer.

statement {
  allow "only-alter-create" {
    match = "(ALTER|CREATE) .+"
  }
  allow "delete-with-filter" {
    match = "DELETE .+ WHERE .+"
  }
  deny "reject-delete" {
    match = "DELETE .+"
  }
  deny "reject-truncate" {
    match = "TRUNCATE .+"
  }
}

lint.statement attributes

Name	Required	Value
error	false	bool
force	false	bool

lint.statement blocks

lint.statement.allow

lint.statement.allow attributes

Name	Required	Value
match	true	string

lint.statement.allow constraints

Constraint	Value
Required	false
Require Name (e.g., lint.statement.allow "name" )	true

lint.statement.deny

lint.statement.deny attributes

Name	Required	Value
match	true	string
message	false	string

lint.statement.deny constraints

Constraint	Value
Required	false
Require Name (e.g., lint.statement.deny "name" )	true

lint.statement constraints

Constraint	Value
Required	false
Require Name	false
One of required sets	[deny, allow]

lint constraints

Constraint	Value
Required	false
Require Name	false
Mutually exclusive sets	[latest, git], [log, format]

test

The test block defines the testing configuration.

test {
  schema {
    src = ["schema.test.hcl"]
    vars = {
      seed_file = "filename.sql"
      variable2 = var.name
      variable3 = data.external.value
    }
  }
}

test blocks

test.migrate

test.migrate attributes

Name	Required	Value
src	true	List of strings
vars	false	map

test.schema

test.schema attributes

Name	Required	Value
src	true	List of strings
vars	false	map