@sqlanvil/core 0.0.1 → 1.0.1

package/configs.proto

@@ -0,0 +1,1028 @@
+// Copyright 2023 Google LLC
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+syntax = "proto3";
+
+package sqlanvil;
+
+option java_package = "com.sqlanvil.protos";
+option java_outer_classname = "ConfigsMeta";
+option java_multiple_files = true;
+
+option go_package = "github.com/sqlanvil/sqlanvil/protos/sqlanvil";
+
+import "google/protobuf/struct.proto";
+import "extension.proto";
+
+// Workflow Settings defines the contents of the `workflow_settings.yaml`
+// configuration file.
+message WorkflowSettings {
+  // The desired sqlanvil core version to compile against.
+  string sqlanvil_core_version = 1;
+
+  // Required. The default Google Cloud project (database).
+  string default_project = 2;
+
+  // Required. The default dataset (schema).
+  string default_dataset = 3;
+
+  // Required. The default BigQuery location to use. For more information on
+  // BigQuery locations, see https://cloud.google.com/bigquery/docs/locations.
+  string default_location = 4;
+
+  // Required. The default dataset (schema) for assertions.
+  string default_assertion_dataset = 5;
+
+  // Optional. User-defined variables that are made available to project code
+  // during compilation. An object containing a list of "key": value pairs.
+  map<string, string> vars = 6;
+
+  // Optional. The suffix to append to all Google Cloud project references.
+  string project_suffix = 7;
+
+  // Optional. The suffix to append to all dataset references.
+  string dataset_suffix = 8;
+
+  // Optional. The prefix to append to all action names.
+  string name_prefix = 9;
+
+  // Optional. Default runtime options for Notebook actions.
+  NotebookRuntimeOptionsConfig default_notebook_runtime_options = 10;
+
+  // Optional. The prefix to append to built-in assertion names.
+  string builtin_assertion_name_prefix = 11;
+
+  // Optional. Default config options for Iceberg tables.
+  DefaultIcebergConfig default_iceberg_config = 12;
+
+  // Optional. Disables all assertions including built-in assertions
+  // (uniqueKey, nonNull, rowConditions) and manual assertions (type: assertion).
+  // When true, assertions will still be compiled but marked as disabled.
+  bool disable_assertions = 13;
+
+  // Optional. The default BigQuery reservation to use for execution.
+  // If unset, default BigQuery behavior applies.
+  // sqlanvil CLI only (GCP sqlanvil support pending).
+  string default_reservation = 14;
+
+  // Optional. An external package that provides an extension.
+  Extension extension = 15;
+
+  // Optional. If set to true, unit tests will be included in the compiled graph.
+  bool include_tests_in_compiled_graph = 16;
+
+  // Optional. The database warehouse to use, e.g. "bigquery", "postgres", "supabase".
+  string warehouse = 17;
+}
+
+message DefaultIcebergConfig {
+  // Optional. Bucket name used to construct a storage URI when creating an
+  // Iceberg table.
+  string bucket_name = 1;
+
+  // Optional. Table folder root used to construct a storage URI when creating
+  // an Iceberg table.
+  string table_folder_root = 2;
+
+  // Optional. Table folder subpath used to construct a storage URI when
+  // creating an Iceberg table.
+  string table_folder_subpath = 3;
+
+  // Optional. The connection specifying the credentials to be used to read and
+  // write to external storage, such as Cloud Storage.
+  string connection = 4;
+}
+
+// Action configs defines the contents of `actions.yaml` configuration files.
+message ActionConfigs {
+  repeated ActionConfig actions = 1;
+}
+
+// Action config defines the contents of `actions.yaml` configuration files.
+message ActionConfig {
+  // Target represents a unique action identifier.
+  message Target {
+    // The Google Cloud project (database) of the action.
+    string project = 1;
+
+    // The dataset (schema) of the action. For notebooks, this is the location.
+    string dataset = 2;
+
+    // The name of the action.
+    string name = 4;
+
+    // flag for when we want to add assertions of this dependency in
+    // dependency_targets as well.
+    bool include_dependent_assertions = 5;
+  }
+
+  message ColumnDescriptor {
+    // The identifier for the column, using multiple parts for nested records.
+    repeated string path = 1;
+
+    // A text description of the column.
+    string description = 2;
+
+    // A list of BigQuery policy tags that will be applied to the column.
+    repeated string bigquery_policy_tags = 3;
+
+    // A list of tags for this column which will be applied.
+    repeated string tags = 4;
+  }
+
+  // Options for shorthand specifying assertions, useable for some table-based
+  // action types.
+  message TableAssertionsConfig {
+    // Column(s) which constitute the dataset's unique key index.
+    //  If set, the resulting assertion will fail if there is more than one row
+    //  in the dataset with the same values for all of these column(s).
+    repeated string unique_key = 1;
+
+    // Combinations of column(s), each of which should constitute a unique key
+    // index for the dataset. If set, the resulting assertion(s) will fail if
+    // there is more than one row in the dataset with the same values for all of
+    // the column(s) in the unique key(s).
+    message UniqueKey {
+      repeated string unique_key = 1;
+    }
+    repeated UniqueKey unique_keys = 2;
+
+    // Column(s) which may never be `NULL`.
+    // If set, the resulting assertion will fail if any row contains `NULL`
+    // values for these column(s).
+    repeated string non_null = 3;
+
+    // General condition(s) which should hold true for all rows in the dataset.
+    // If set, the resulting assertion will fail if any row violates any of
+    // these condition(s).
+    repeated string row_conditions = 4;
+  }
+
+  message IcebergTableConfig {
+    // Supported file formats for BigQuery tables.
+    enum FileFormat {
+      // Default value.
+      FILE_FORMAT_UNSPECIFIED = 0;
+      // Apache Parquet format.
+      PARQUET = 1;
+    }
+
+    // The file format for the BigQuery table.
+    FileFormat file_format = 1;
+
+    // The connection specifying the credentials to be used to read and write
+    // to external storage, such as Cloud Storage. The connection can have the
+    // form `{project}.{location}.{connection_id}` or
+    // `projects/{project}/locations/{location}/connections/{connection_id}",
+    // or be set to DEFAULT.
+    string connection = 2;
+
+    // The name of the Cloud Storage bucket where table data is stored. This
+    // value is be used to construct the storage URI in the following way:
+    // `gs://{bucket_name}/{table_folder_root}/{table_folder_subpath}``.
+    // If `storage_uri` is provided, this value is ignored.
+    string bucket_name = 3;
+
+    // The name of the first-level folder inside the Cloud Storage bucket where
+    // table data is stored. This value will be used to construct the storage
+    // URI in the following way:
+    // `gs://{bucket_name}/{table_folder_root}/{table_folder_subpath}``.
+    // If `storage_uri` is provided, this value is ignored.
+    string table_folder_root = 4;
+
+    // The path under the first-level folder of the Cloud Storage bucket where
+    // table data is stored. This value will be used to construct the storage URI
+    // in the following way:
+    // `gs://{bucket_name}/{table_folder_root}/{table_folder_subpath}``.
+    // If `storage_uri` is provided, this value is ignored.
+    string table_folder_subpath = 5;
+  }
+
+  message TableConfig {
+    // The name of the table.
+    string name = 1;
+
+    // The dataset (schema) of the table.
+    string dataset = 2;
+
+    // The Google Cloud project (database) of the table.
+    string project = 3;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 4;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 5;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 6;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Queries to run before `query`. This can be useful for granting
+    // permissions.
+    repeated string pre_operations = 8;
+
+    // Queries to run after `query`.
+    repeated string post_operations = 9;
+
+    // Description of the table.
+    string description = 10;
+
+    // Descriptions of columns within the table.
+    repeated ColumnDescriptor columns = 11;
+
+    // The key by which to partition the table. Typically the name of a
+    // timestamp or the date column. See
+    // https://cloud.google.com/dataform/docs/partitions-clusters.
+    string partition_by = 12;
+
+    // The number of days for which BigQuery stores data in each partition.
+    // The setting applies to all partitions in a table, but is calculated
+    // independently for each partition based on the partition time.
+    int32 partition_expiration_days = 13;
+
+    // Declares whether the partitioned table requires a WHERE clause
+    // predicate filter that filters the partitioning column.
+    bool require_partition_filter = 14;
+
+    // The keys by which to cluster partitions by. See
+    // https://cloud.google.com/dataform/docs/partitions-clusters.
+    repeated string cluster_by = 15;
+
+    // Key-value pairs for BigQuery labels.
+    map<string, string> labels = 16;
+
+    // Key-value pairs of additional options to pass to the BigQuery API. Some
+    // options, for example, partitionExpirationDays, have dedicated
+    // type/validity checked fields. For such options, use the dedicated fields.
+    map<string, string> additional_options = 17;
+
+    // When set to true, assertions dependent upon any dependency will
+    // be add as dedpendency to this action
+    bool depend_on_dependency_assertions = 18;
+
+    // Assertions to be run on the dataset.
+    // If configured, relevant assertions will automatically be created and run
+    // as a dependency of this dataset.
+    TableAssertionsConfig assertions = 19;
+
+    // If true, this indicates that the action only depends on data from
+    // explicitly-declared dependencies. Otherwise if false, it indicates that
+    // the  action depends on data from a source which has not been declared as
+    // a dependency.
+    bool hermetic = 20;
+
+    // Configuration options for an Iceberg table.
+    IcebergTableConfig iceberg = 22;
+
+    // Metadata for this table.
+    Metadata metadata = 23;
+
+    // Optional. The BigQuery reservation to use for execution.
+    // If unset, the value from workflow_settings.yaml is used. If neither is set, default BigQuery behavior applies.
+    // sqlanvil CLI only (GCP sqlanvil support pending).
+    string reservation = 24;
+
+    // Postgres and Supabase specific options.
+    PostgresOptions postgres = 25;
+    SupabaseOptions supabase = 26;
+  }
+
+  message Metadata {
+    // A detailed description of the data object.
+    string overview = 1;
+    // Extra properties of the data object.
+    google.protobuf.Struct extra_properties = 2;
+  }
+
+  message ViewConfig {
+    // The name of the view.
+    string name = 1;
+
+    // The dataset (schema) of the view.
+    string dataset = 2;
+
+    // The Google Cloud project (database) of the view.
+    string project = 3;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 4;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 5;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 6;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Queries to run before `query`. This can be useful for granting
+    // permissions.
+    repeated string pre_operations = 8;
+
+    // Queries to run after `query`.
+    repeated string post_operations = 9;
+
+    // Applies the materialized view optimization, see
+    // https://cloud.google.com/bigquery/docs/materialized-views-intro.
+    bool materialized = 10;
+
+    // Optional. Applicable only to materialized view. The key by which to partition the materialized view. Typically the name of a
+    // timestamp or the date column. See https://cloud.google.com/bigquery/docs/materialized-views-create#partitioned_materialized_views.
+    string partition_by = 18;
+
+    // Optional. Applicable only to materialized view. The keys by which to cluster partitions by. See
+    // https://cloud.google.com/bigquery/docs/materialized-views-create#cluster_materialized_views.
+    repeated string cluster_by = 19;
+
+    // Description of the view.
+    string description = 11;
+
+    // Descriptions of columns within the table.
+    repeated ColumnDescriptor columns = 12;
+
+    // Key-value pairs for BigQuery labels.
+    map<string, string> labels = 13;
+
+    // Key-value pairs of additional options to pass to the BigQuery API. Some
+    // options, for example, partitionExpirationDays, have dedicated
+    // type/validity checked fields. For such options, use the dedicated fields.
+    map<string, string> additional_options = 14;
+
+    // When set to true, assertions dependent upon any dependency will
+    // be add as dedpendency to this action
+    bool depend_on_dependency_assertions = 15;
+
+    // If true, this indicates that the action only depends on data from
+    // explicitly-declared dependencies. Otherwise if false, it indicates that
+    // the  action depends on data from a source which has not been declared as
+    // a dependency.
+    bool hermetic = 16;
+
+    // Assertions to be run on the dataset.
+    // If configured, relevant assertions will automatically be created and run
+    // as a dependency of this dataset.
+    TableAssertionsConfig assertions = 17;
+
+    // Metadata for this view.
+    Metadata metadata = 20;
+
+    // Optional. The BigQuery reservation to use for execution.
+    // If unset, the value from workflow_settings.yaml is used. If neither is set, default BigQuery behavior applies.
+    // sqlanvil CLI only (GCP sqlanvil support pending).
+    string reservation = 21;
+
+    // Postgres-native options. For a materialized view (materialized: true),
+    // `no_data` (CREATE ... WITH NO DATA) and `refresh_policy`
+    // ("on_dependency_change" → in-place REFRESH instead of drop+recreate)
+    // apply; indexes also apply to materialized views.
+    PostgresOptions postgres = 22;
+  }
+
+  enum OnSchemaChange {
+    // Ignore any schema changes (default).
+    IGNORE = 0;
+    // Fails if the query would result in a new column(s) being added, deleted,
+    // or renamed.
+    FAIL = 1;
+    // Does not block any new column(s) from being added.
+    EXTEND = 2;
+    // Does not block any new column(s) from being added, deleted or renamed.
+    SYNCHRONIZE = 3;
+  }
+
+  message IncrementalTableConfig {
+    // The name of the incremental table.
+    string name = 1;
+
+    // The dataset (schema) of the incremental table.
+    string dataset = 2;
+
+    // The Google Cloud project (database) of the incremental table.
+    string project = 3;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 4;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 5;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 6;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Queries to run before `query`. This can be useful for granting
+    // permissions.
+    repeated string pre_operations = 8;
+
+    // Queries to run after `query`.
+    repeated string post_operations = 9;
+
+    // If true, prevents the dataset from being rebuilt from scratch.
+    bool protected = 10;
+
+    // If set, unique key represents a set of names of columns that will act as
+    // a the unique key. To enforce this, when updating the incremental
+    // table, sqlanvil merges rows with `uniqueKey` instead of appending them.
+    repeated string unique_key = 11;
+
+    // Description of the incremental table.
+    string description = 12;
+
+    // Descriptions of columns within the table.
+    repeated ColumnDescriptor columns = 13;
+
+    // The key by which to partition the table. Typically the name of a
+    // timestamp or the date column. See
+    // https://cloud.google.com/dataform/docs/partitions-clusters.
+    string partition_by = 14;
+
+    // The number of days for which BigQuery stores data in each partition.
+    // The setting applies to all partitions in a table, but is calculated
+    // independently for each partition based on the partition time.
+    int32 partition_expiration_days = 15;
+
+    // Declares whether the partitioned table requires a WHERE clause
+    // predicate filter that filters the partitioning column.
+    bool require_partition_filter = 16;
+
+    // SQL-based filter for when incremental updates are applied.
+    string update_partition_filter = 17;
+
+    // The keys by which to cluster partitions by. See
+    // https://cloud.google.com/dataform/docs/partitions-clusters.
+    repeated string cluster_by = 18;
+
+    // Key-value pairs for BigQuery labels.
+    map<string, string> labels = 19;
+
+    // Key-value pairs of additional options to pass to the BigQuery API. Some
+    // options, for example, partitionExpirationDays, have dedicated
+    // type/validity checked fields. For such options, use the dedicated fields.
+    map<string, string> additional_options = 20;
+
+    // When set to true, assertions dependent upon any dependency will
+    // be add as dedpendency to this action
+    bool depend_on_dependency_assertions = 21;
+
+    // Assertions to be run on the dataset.
+    // If configured, relevant assertions will automatically be created and run
+    // as a dependency of this dataset.
+    TableAssertionsConfig assertions = 22;
+
+    // If true, this indicates that the action only depends on data from
+    // explicitly-declared dependencies. Otherwise if false, it indicates that
+    // the  action depends on data from a source which has not been declared as
+    // a dependency.
+    bool hermetic = 23;
+
+    // Defines the action behavior if the selected columns in the query
+    // don't the match columns in the target table.
+    OnSchemaChange on_schema_change = 24;
+
+    // Configuration options for an Iceberg table.
+    IcebergTableConfig iceberg = 25;
+
+    // Metadata for this incremental table.
+    Metadata metadata = 26;
+
+    // Optional. The BigQuery reservation to use for execution.
+    // If unset, the value from workflow_settings.yaml is used. If neither is set, default BigQuery behavior applies.
+    // sqlanvil CLI only (GCP sqlanvil support pending).
+    string reservation = 27;
+
+    // Postgres and Supabase specific options.
+    PostgresOptions postgres = 28;
+    SupabaseOptions supabase = 29;
+  }
+
+  message AssertionConfig {
+    // The name of the assertion.
+    string name = 1;
+
+    // The dataset (schema) of the assertion.
+    string dataset = 2;
+
+    // The Google Cloud project (database) of the assertion.
+    string project = 3;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 4;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 5;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 6;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Description of the assertion.
+    string description = 8;
+
+    // If true, this indicates that the action only depends on data from
+    // explicitly-declared dependencies. Otherwise if false, it indicates that
+    // the  action depends on data from a source which has not been declared as
+    // a dependency.
+    bool hermetic = 9;
+
+    // If true, assertions dependent upon any of the dependencies are added as
+    // dependencies as well.
+    bool depend_on_dependency_assertions = 10;
+
+    // Optional. The BigQuery reservation to use for execution.
+    // If unset, the value from workflow_settings.yaml is used. If neither is set, default BigQuery behavior applies.
+    // sqlanvil CLI only (GCP sqlanvil support pending).
+    string reservation = 11;
+  }
+
+  message OperationConfig {
+    // The name of the operation.
+    string name = 1;
+
+    // The dataset (schema) of the operation.
+    string dataset = 2;
+
+    // The Google Cloud project (database) of the operation.
+    string project = 3;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 4;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 5;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 6;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Declares that this action creates a dataset which should be
+    // referenceable as a dependency target, for example by using the `ref`
+    // function.
+    bool has_output = 8;
+
+    // Description of the operation.
+    string description = 9;
+
+    // Descriptions of columns within the operation. Can only be set if
+    // hasOutput is true.
+    repeated ColumnDescriptor columns = 10;
+
+    // When set to true, assertions dependent upon any dependency will
+    // be add as dedpendency to this action
+    bool depend_on_dependency_assertions = 11;
+
+    // If true, this indicates that the action only depends on data from
+    // explicitly-declared dependencies. Otherwise if false, it indicates that
+    // the  action depends on data from a source which has not been declared as
+    // a dependency.
+    bool hermetic = 12;
+
+    // Optional. The BigQuery reservation to use for execution.
+    // If unset, the value from workflow_settings.yaml is used. If neither is set, default BigQuery behavior applies.
+    // sqlanvil CLI only (GCP sqlanvil support pending).
+    string reservation = 13;
+  }
+
+  message DeclarationConfig {
+    // The name of the declaration.
+    string name = 1;
+
+    // The dataset (schema) of the declaration.
+    string dataset = 2;
+
+    // The Google Cloud project (database) of the declaration.
+    string project = 3;
+
+    // Description of the declaration.
+    string description = 4;
+
+    // Descriptions of columns within the declaration.
+    repeated ColumnDescriptor columns = 5;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 6;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 7;
+  }
+
+  message NotebookConfig {
+    // The name of the notebook.
+    string name = 1;
+
+    // The Google Cloud location of the notebook.
+    string location = 2;
+
+    // The Google Cloud project (database) of the notebook.
+    string project = 3;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 4;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 5;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 6;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Description of the notebook.
+    string description = 8;
+
+    // When set to true, assertions dependent upon any dependency will
+    // be add as dedpendency to this action
+    bool depend_on_dependency_assertions = 9;
+
+    // A notebook runtime field definition could be added here, to allow atomic
+    // runtime settings of notebooks.
+  }
+
+  message DataPreparationConfig {
+    // The name of the data preparation.
+    string name = 1;
+
+    // The dataset (schema) of the destination table.
+    string dataset = 13;
+
+    // The Google Cloud project (database) of the destination table.
+    string project = 14;
+
+    // Targets of actions that this action is dependent on.
+    repeated Target dependency_targets = 2;
+
+    // Path to the source file that the contents of the action is loaded from.
+    string filename = 3;
+
+    // A list of user-defined tags with which the action should be labeled.
+    repeated string tags = 4;
+
+    // If set to true, this action will not be executed. However, the action can
+    // still be depended upon. Useful for temporarily turning off broken
+    // actions.
+    bool disabled = 7;
+
+    // Description of the data preparation.
+    string description = 8;
+
+    ErrorTableConfig error_table = 11;
+
+    LoadModeConfig load_mode = 12;
+
+    message ErrorTableConfig {
+      // The name of the error table.
+      string name = 1;
+
+      // The dataset (schema) of the error table.
+      string dataset = 2;
+
+      // The Google Cloud project (database) of the error table.
+      string project = 3;
+
+      int32 retention_days = 4;
+    }
+  }
+
+  oneof action {
+    TableConfig table = 1;
+    ViewConfig view = 2;
+    IncrementalTableConfig incremental_table = 3;
+    AssertionConfig assertion = 4;
+    OperationConfig operation = 5;
+    DeclarationConfig declaration = 6;
+    NotebookConfig notebook = 7;
+    DataPreparationConfig data_preparation = 8;
+    RlsPolicyConfig rls_policy = 9;
+    RealtimePublicationConfig realtime_publication = 10;
+    ForeignWrapperConfig foreign_wrapper = 11;
+    VectorIndexConfig vector_index = 12;
+  }
+
+  message RlsPolicyConfig {
+    string name = 1;
+    string table = 2;
+    string command = 3;
+    repeated string roles = 4;
+    string using = 5;
+    string with_check = 6;
+    string filename = 7;
+    repeated Target dependency_targets = 8;
+  }
+
+  message RealtimePublicationConfig {
+    string name = 1;
+    string table = 2;
+    repeated string events = 3;
+    string filename = 4;
+    repeated Target dependency_targets = 5;
+  }
+
+  message ForeignWrapperConfig {
+    string name = 1;
+    string wrapper = 2;
+    string server = 3;
+    map<string, string> options = 4;
+    string filename = 5;
+    repeated Target dependency_targets = 6;
+  }
+
+  message VectorIndexConfig {
+    string name = 1;
+    string table = 2;
+    string column = 3;
+    uint32 dimensions = 4;
+    string index_type = 5;
+    map<string, string> params = 6;
+    string filename = 7;
+    repeated Target dependency_targets = 8;
+  }
+
+  message LoadModeConfig {
+    LoadMode mode = 1;
+    // Required when mode is MAXIMUM or UNIQUE
+    string incremental_column = 2;
+    // required when mode is MERGE
+    repeated string unique_key = 3;
+  }
+
+  enum LoadMode {
+    // Replace existing table (default).
+    REPLACE_TABLE = 0;
+    // Insert into destination table.
+    APPEND = 1;
+    // Insert only records where the specified column value exceeds the existing
+    // maximum value in the destination table.
+    MAXIMUM = 2;
+    // Insert only records where the specified column value is not already
+    // present in the destination column values.
+    UNIQUE = 3;
+    // Merge records into the destination table, deduplicating using 1+ unique keys
+    MERGE = 4;
+  }
+}
+
+message RepositorySnapshotDestinationConfig {
+  // Storage URI to upload the repository snapshot to.
+  string repository_snapshot_uri = 1;
+}
+
+message NotebookRuntimeOptionsConfig {
+  oneof output_sink {
+    // Storage bucket to output notebooks to after their execution.
+    string output_bucket = 1;
+  }
+
+  // Colab runtime template (https://cloud.google.com/colab/docs/runtimes), from
+  // which a runtime is created for notebook executions.
+  string runtime_template_name = 2;
+
+  // The destination for the snapshot of repository files to be available for
+  // read-only access inside a notebook runtime.
+  oneof repository_snapshot_storage {
+    // Storage URI to upload the snapshot to.
+    // For empty URI it defaults to the provided output_bucket.
+    RepositorySnapshotDestinationConfig repository_snapshot_destination = 3;
+  }
+
+}
+
+// =============================================================================
+// Postgres-first-class adapter — additions per
+// https://github.com/sqlanvil/docs/blob/main/postgres_first_class_design.md. Phase 3c.
+//
+// These messages are declared here for wiring in subsequent phases:
+//   - PostgresOptions / SupabaseOptions      → ActionConfig table-level blocks
+//   - PostgresConnection / SupabaseConnection / BigQueryConnection
+//                                            → WorkflowSettings.warehouse
+//   - WarehouseConfig                        → discriminated union over the
+//                                              connection variants
+//
+// Adding the messages alone does not change behavior — wiring happens in
+// Phase 4 (CLI) and the parallel rewrite of ActionConfig's table/view/
+// incremental_table sub-messages.
+// =============================================================================
+
+// PostgresOptions — Postgres-native table-level options. Mirrors what
+// BigQueryOptions-style fields do in TableConfig but in idiomatic Postgres.
+//
+// Used as a peer of the existing `bigquery: {...}` shape on action configs:
+//   publish("daily_orders", { postgres: { tablespace: "fast_ssd", ... } })
+message PostgresOptions {
+  // Physical storage placement (CREATE TABLE ... TABLESPACE <name>).
+  string tablespace = 1;
+
+  // Storage parameter — fraction of each page to fill on insert (1-100).
+  uint32 fillfactor = 2;
+
+  // CREATE UNLOGGED TABLE — faster writes, lost on crash. For staging/temp
+  // tables where durability isn't required.
+  bool unlogged = 3;
+
+  // Native Postgres declarative partitioning.
+  message Partition {
+    enum Kind {
+      RANGE = 0;
+      LIST = 1;
+      HASH = 2;
+    }
+    Kind kind = 1;
+    repeated string columns = 2;
+
+    // Child partitions. `values` is the raw FOR VALUES clause body matching the
+    // kind, e.g. "FROM ('2024-01-01') TO ('2025-01-01')" (range),
+    // "IN ('us', 'ca')" (list), or "WITH (MODULUS 4, REMAINDER 0)" (hash).
+    message Bound {
+      string name = 1;
+      string values = 2;
+    }
+    repeated Bound partitions = 3;
+
+    // Also create a catch-all DEFAULT partition so rows outside every bound
+    // still insert (recommended for declarative full-refresh loads).
+    bool include_default = 4;
+  }
+  Partition partition = 4;
+
+  // Indexes to create alongside the table.
+  message Index {
+    string name = 1;
+    repeated string columns = 2;
+
+    enum Method {
+      BTREE = 0;
+      HASH = 1;
+      GIN = 2;
+      GIST = 3;
+      BRIN = 4;
+    }
+    Method method = 3;
+
+    // Partial index predicate (WHERE <expr>).
+    string where = 4;
+
+    bool unique = 5;
+
+    // INCLUDE non-key columns for covering indexes.
+    repeated string include = 6;
+
+    // Operator class applied to each indexed column, e.g. "gin_trgm_ops"
+    // (pg_trgm), "jsonb_path_ops", or "vector_l2_ops" (pgvector). Required for
+    // gin/gist indexes on types without a default opclass.
+    string opclass = 7;
+  }
+  repeated Index indexes = 5;
+
+  // Materialized view: create WITH NO DATA (empty until first refresh). Default
+  // (false) is WITH DATA. Named for the non-default so proto3's false default
+  // means the sensible WITH DATA.
+  bool no_data = 6;
+
+  // Materialized view refresh on re-run: "on_dependency_change" refreshes an
+  // existing matview in place (REFRESH MATERIALIZED VIEW) instead of dropping +
+  // recreating. Default (unset) drops + recreates each run (safe — also picks up
+  // definition changes, which REFRESH does not).
+  string refresh_policy = 7;
+}
+
+// SupabaseOptions — Supabase-specific platform features layered on top of
+// standard Postgres. Used as a peer of `postgres: {...}` for projects
+// targeting `warehouse: { kind: supabase }`.
+message SupabaseOptions {
+  // Standard Postgres options apply. Set these via `postgres:` directly or
+  // nest under `supabase.postgres:` — either is accepted.
+  PostgresOptions postgres = 1;
+
+  // ALTER PUBLICATION supabase_realtime ADD TABLE <this>.
+  // Implicitly sets REPLICA IDENTITY appropriately.
+  bool publish_to_realtime = 2;
+
+  // ALTER TABLE <this> ENABLE ROW LEVEL SECURITY.
+  // Note: only enables RLS — policies are declared via the `rlsPolicy`
+  // action type (see Phase 5).
+  bool enable_rls = 3;
+
+  // OWNER TO <role>. Typically "postgres" or "service_role".
+  string owner_role = 4;
+
+  // pgvector convenience config. Equivalent to declaring a
+  // PostgresOptions.Index with method=HNSW or method=GIST + ivfflat ops,
+  // but more ergonomic for RAG pipelines.
+  message VectorConfig {
+    string column = 1;
+    uint32 dimensions = 2;
+
+    enum IndexType {
+      IVFFLAT = 0;
+      HNSW = 1;
+    }
+    IndexType index_type = 3;
+
+    // ivfflat: { lists }, hnsw: { m, ef_construction }.
+    map<string, string> params = 4;
+  }
+  repeated VectorConfig vectors = 5;
+}
+
+// BigQueryConnection — connection params for warehouse.kind = "bigquery".
+// Mirrors the legacy flat fields on WorkflowSettings (default_project,
+// default_location, default_dataset) but namespaced under warehouse.
+message BigQueryConnection {
+  // The Google Cloud project (database).
+  string project = 1;
+
+  // BigQuery location, e.g. "US", "EU", "europe-west4".
+  string location = 2;
+
+  // Default dataset (schema).
+  string default_dataset = 3;
+}
+
+// PostgresConnection — libpq-style connection params for
+// warehouse.kind = "postgres". Standard Postgres host/port/database/user.
+message PostgresConnection {
+  string host = 1;
+  uint32 port = 2;
+  string database = 3;
+  string user = 4;
+  string password = 5;
+
+  // SSL mode: "disable" | "allow" | "prefer" | "require" | "verify-ca"
+  // | "verify-full". See https://www.postgresql.org/docs/current/libpq-ssl.html.
+  string ssl_mode = 6;
+
+  string default_schema = 7;
+}
+
+// SupabaseConnection — connection params for warehouse.kind = "supabase".
+// Supabase projects expose a Postgres connection via project_ref +
+// service_role_key, or a direct connection string for bypassing PostgREST.
+message SupabaseConnection {
+  // From the Supabase dashboard (project URL host before .supabase.co).
+  string project_ref = 1;
+
+  // Project service-role JWT. NEVER commit literally — use ${ENV_VAR}
+  // interpolation in workflow_settings.yaml.
+  string service_role_key = 2;
+
+  string default_schema = 3;
+
+  // Optional override — direct Postgres URL bypassing the PostgREST proxy.
+  // e.g. "postgresql://postgres:${PASSWORD}@db.<project_ref>.supabase.co:5432/postgres".
+  // If set, takes precedence over project_ref + service_role_key for the
+  // direct DB connection. service_role_key is still used for RLS bypass.
+  string connection_string = 4;
+}
+
+// WarehouseConfig — discriminated union over connection variants. The
+// `kind:` YAML tag selects which `oneof` arm is unmarshalled.
+//
+// Example YAML:
+//   warehouse:
+//     kind: postgres
+//     host: db.example.com
+//     port: 5432
+//     database: analytics
+//     user: sqlanvil_writer
+//     password: ${PG_PASSWORD}
+//     ssl_mode: require
+//     default_schema: public
+message WarehouseConfig {
+  oneof connection {
+    BigQueryConnection bigquery = 1;
+    PostgresConnection postgres = 2;
+    SupabaseConnection supabase = 3;
+  }
+}