GCP (dagster-gcp)¶

BigQuery¶

class dagster_gcp.BigQueryError[source]¶

dagster_gcp.bigquery_resource ResourceDefinition[source]¶

Config Schema:

project (dagster.StringSource, optional):: Project ID for the project which the client acts on behalf of. Will be passed when creating a dataset / job. If not passed, falls back to the default inferred from the environment.
location (dagster.StringSource, optional):: (Optional) Default location for jobs / datasets / tables.

dagster_gcp.bq_create_dataset(context)[source]¶

BigQuery Create Dataset.

This op encapsulates creating a BigQuery dataset.

Expects a BQ client to be provisioned in resources as context.resources.bigquery.

dagster_gcp.bq_delete_dataset(context)[source]¶

BigQuery Delete Dataset.

This op encapsulates deleting a BigQuery dataset.

Expects a BQ client to be provisioned in resources as context.resources.bigquery.

dagster_gcp.bq_op_for_queries(sql_queries)[source]¶

Executes BigQuery SQL queries.

Expects a BQ client to be provisioned in resources as context.resources.bigquery.

dagster_gcp.import_df_to_bq(context, df)[source]¶

dagster_gcp.import_file_to_bq(context, path)[source]¶

dagster_gcp.import_gcs_paths_to_bq(context, paths)[source]¶

Dataproc¶

dagster_gcp.dataproc_op = <dagster._core.definitions.op_definition.OpDefinition object>[source]¶

Config Schema:

job_timeout_in_seconds (Int, optional):

Optional. Maximum time in seconds to wait for the job being completed. Default is set to 1200 seconds (20 minutes).

Default Value: 1200

job_config (strict dict):

Config Schema:

job (strict dict, optional):

A Cloud Dataproc job resource.

Config Schema:

status (strict dict, optional):

Cloud Dataproc job status.

placement (strict dict, optional):

Cloud Dataproc job config.

Config Schema:

clusterName (String, optional):: Required. The name of the cluster where the job will be submitted.

scheduling (strict dict, optional):

Job scheduling options.

Config Schema:

maxFailuresPerHour (Int, optional):: Optional. Maximum number of times per hour a driver may be restarted as a result of driver terminating with non-zero code before job is reported failed.A job may be reported as thrashing if driver exits with non-zero code 4 times within 10 minute window.Maximum value is 10.

pigJob (strict dict, optional):

A Cloud Dataproc job for running Apache Pig (https://pig.apache.org/) queries on YARN.

Config Schema:

queryFileUri (String, optional):

The HCFS URI of the script that contains the Pig queries.

queryList (strict dict, optional):

A list of queries to run on a cluster.

Config Schema:

queries (List[String], optional):: Required. The queries to execute. You do not need to terminate a query with a semicolon. Multiple queries can be specified in one string by separating each with a semicolon. Here is an example of an Cloud Dataproc API snippet that uses a QueryList to specify a HiveJob: “hiveJob”: { “queryList”: { “queries”: [ “query1”, “query2”, “query3;query4”, ] } }

jarFileUris (List[String], optional):

Optional. HCFS URIs of jar files to add to the CLASSPATH of the Pig Client and Hadoop MapReduce (MR) tasks. Can contain Pig UDFs.

scriptVariables (permissive dict, optional):

Optional. Mapping of query variable names to values (equivalent to the Pig command: name=[value]).

loggingConfig (strict dict, optional):

The runtime logging config of the job.

Config Schema:

driverLogLevels (permissive dict, optional):: The per-package log levels for the driver. This may include “root” package name to configure rootLogger. Examples: ‘com.google = FATAL’, ‘root = INFO’, ‘org.apache = DEBUG’

properties (permissive dict, optional):

Optional. A mapping of property names to values, used to configure Pig. Properties that conflict with values set by the Cloud Dataproc API may be overwritten. Can include properties set in /etc/hadoop/conf/*-site.xml, /etc/pig/conf/pig.properties, and classes in user code.

continueOnFailure (Bool, optional):

Optional. Whether to continue executing queries if a query fails. The default value is false. Setting to true can be useful when executing independent parallel queries.

hiveJob (strict dict, optional):

A Cloud Dataproc job for running Apache Hive (https://hive.apache.org/) queries on YARN.

Config Schema:

continueOnFailure (Bool, optional):

Optional. Whether to continue executing queries if a query fails. The default value is false. Setting to true can be useful when executing independent parallel queries.

queryFileUri (String, optional):

The HCFS URI of the script that contains Hive queries.

queryList (strict dict, optional):

A list of queries to run on a cluster.

Config Schema:

queries (List[String], optional):: Required. The queries to execute. You do not need to terminate a query with a semicolon. Multiple queries can be specified in one string by separating each with a semicolon. Here is an example of an Cloud Dataproc API snippet that uses a QueryList to specify a HiveJob: “hiveJob”: { “queryList”: { “queries”: [ “query1”, “query2”, “query3;query4”, ] } }

jarFileUris (List[String], optional):

Optional. HCFS URIs of jar files to add to the CLASSPATH of the Hive server and Hadoop MapReduce (MR) tasks. Can contain Hive SerDes and UDFs.

scriptVariables (permissive dict, optional):

Optional. Mapping of query variable names to values (equivalent to the Hive command: SET name=”value”;).

properties (permissive dict, optional):

Optional. A mapping of property names and values, used to configure Hive. Properties that conflict with values set by the Cloud Dataproc API may be overwritten. Can include properties set in /etc/hadoop/conf/*-site.xml, /etc/hive/conf/hive-site.xml, and classes in user code.

labels (permissive dict, optional):

Optional. The labels to associate with this job. Label keys must contain 1 to 63 characters, and must conform to RFC 1035 (https://www.ietf.org/rfc/rfc1035.txt). Label values may be empty, but, if present, must contain 1 to 63 characters, and must conform to RFC 1035 (https://www.ietf.org/rfc/rfc1035.txt). No more than 32 labels can be associated with a job.

sparkJob (strict dict, optional):

A Cloud Dataproc job for running Apache Spark (http://spark.apache.org/) applications on YARN.

Config Schema:

archiveUris (List[String], optional):

Optional. HCFS URIs of archives to be extracted in the working directory of Spark drivers and tasks. Supported file types: .jar, .tar, .tar.gz, .tgz, and .zip.

mainJarFileUri (String, optional):

The HCFS URI of the jar file that contains the main class.

jarFileUris (List[String], optional):

Optional. HCFS URIs of jar files to add to the CLASSPATHs of the Spark driver and tasks.

loggingConfig (strict dict, optional):

The runtime logging config of the job.

Config Schema:

driverLogLevels (permissive dict, optional):: The per-package log levels for the driver. This may include “root” package name to configure rootLogger. Examples: ‘com.google = FATAL’, ‘root = INFO’, ‘org.apache = DEBUG’

properties (permissive dict, optional):

Optional. A mapping of property names to values, used to configure Spark. Properties that conflict with values set by the Cloud Dataproc API may be overwritten. Can include properties set in /etc/spark/conf/spark-defaults.conf and classes in user code.

args (List[String], optional):

Optional. The arguments to pass to the driver. Do not include arguments, such as –conf, that can be set as job properties, since a collision may occur that causes an incorrect job submission.

fileUris (List[String], optional):

Optional. HCFS URIs of files to be copied to the working directory of Spark drivers and distributed tasks. Useful for naively parallel tasks.

mainClass (String, optional):

The name of the driver’s main class. The jar file that contains the class must be in the default CLASSPATH or specified in jar_file_uris.

sparkSqlJob (strict dict, optional):

A Cloud Dataproc job for running Apache Spark SQL (http://spark.apache.org/sql/) queries.

Config Schema:

queryList (strict dict, optional):

A list of queries to run on a cluster.

Config Schema:

queries (List[String], optional):: Required. The queries to execute. You do not need to terminate a query with a semicolon. Multiple queries can be specified in one string by separating each with a semicolon. Here is an example of an Cloud Dataproc API snippet that uses a QueryList to specify a HiveJob: “hiveJob”: { “queryList”: { “queries”: [ “query1”, “query2”, “query3;query4”, ] } }

queryFileUri (String, optional):

The HCFS URI of the script that contains SQL queries.

scriptVariables (permissive dict, optional):

Optional. Mapping of query variable names to values (equivalent to the Spark SQL command: SET name=”value”;).

jarFileUris (List[String], optional):

Optional. HCFS URIs of jar files to be added to the Spark CLASSPATH.

loggingConfig (strict dict, optional):

The runtime logging config of the job.

Config Schema:

driverLogLevels (permissive dict, optional):: The per-package log levels for the driver. This may include “root” package name to configure rootLogger. Examples: ‘com.google = FATAL’, ‘root = INFO’, ‘org.apache = DEBUG’

properties (permissive dict, optional):

Optional. A mapping of property names to values, used to configure Spark SQL’s SparkConf. Properties that conflict with values set by the Cloud Dataproc API may be overwritten.

pysparkJob (strict dict, optional):

A Cloud Dataproc job for running Apache PySpark (https://spark.apache.org/docs/0.9.0/python-programming-guide.html) applications on YARN.

Config Schema:

jarFileUris (List[String], optional):

Optional. HCFS URIs of jar files to add to the CLASSPATHs of the Python driver and tasks.

loggingConfig (strict dict, optional):

The runtime logging config of the job.

Config Schema:

driverLogLevels (permissive dict, optional):: The per-package log levels for the driver. This may include “root” package name to configure rootLogger. Examples: ‘com.google = FATAL’, ‘root = INFO’, ‘org.apache = DEBUG’

properties (permissive dict, optional):

Optional. A mapping of property names to values, used to configure PySpark. Properties that conflict with values set by the Cloud Dataproc API may be overwritten. Can include properties set in /etc/spark/conf/spark-defaults.conf and classes in user code.

args (List[String], optional):

Optional. The arguments to pass to the driver. Do not include arguments, such as –conf, that can be set as job properties, since a collision may occur that causes an incorrect job submission.

fileUris (List[String], optional):

Optional. HCFS URIs of files to be copied to the working directory of Python drivers and distributed tasks. Useful for naively parallel tasks.

pythonFileUris (List[String], optional):

Optional. HCFS file URIs of Python files to pass to the PySpark framework. Supported file types: .py, .egg, and .zip.

mainPythonFileUri (String, optional):

Required. The HCFS URI of the main Python file to use as the driver. Must be a .py file.

archiveUris (List[String], optional):

Optional. HCFS URIs of archives to be extracted in the working directory of .jar, .tar, .tar.gz, .tgz, and .zip.

reference (strict dict, optional):

Encapsulates the full scoping used to reference a job.

Config Schema:

projectId (String, optional):: Required. The ID of the Google Cloud Platform project that the job belongs to.
jobId (String, optional):: Optional. The job ID, which must be unique within the project.The ID must contain only letters (a-z, A-Z), numbers (0-9), underscores (_), or hyphens (-). The maximum length is 100 characters.If not specified by the caller, the job ID will be provided by the server.

hadoopJob (strict dict, optional):

A Cloud Dataproc job for running Apache Hadoop MapReduce (https://hadoop.apache.org/docs/current/hadoop-mapreduce-client/hadoop-mapreduce-client-core/MapReduceTutorial.html) jobs on Apache Hadoop YARN (https://hadoop.apache.org/docs/r2.7.1/hadoop-yarn/hadoop-yarn-site/YARN.html).

Config Schema:

jarFileUris (List[String], optional):

Optional. Jar file URIs to add to the CLASSPATHs of the Hadoop driver and tasks.

loggingConfig (strict dict, optional):

The runtime logging config of the job.

Config Schema:

driverLogLevels (permissive dict, optional):: The per-package log levels for the driver. This may include “root” package name to configure rootLogger. Examples: ‘com.google = FATAL’, ‘root = INFO’, ‘org.apache = DEBUG’

properties (permissive dict, optional):

Optional. A mapping of property names to values, used to configure Hadoop. Properties that conflict with values set by the Cloud Dataproc API may be overwritten. Can include properties set in /etc/hadoop/conf/*-site and classes in user code.

args (List[String], optional):

Optional. The arguments to pass to the driver. Do not include arguments, such as -libjars or -Dfoo=bar, that can be set as job properties, since a collision may occur that causes an incorrect job submission.

fileUris (List[String], optional):

Optional. HCFS (Hadoop Compatible Filesystem) URIs of files to be copied to the working directory of Hadoop drivers and distributed tasks. Useful for naively parallel tasks.

mainClass (String, optional):

The name of the driver’s main class. The jar file containing the class must be in the default CLASSPATH or specified in jar_file_uris.

archiveUris (List[String], optional):

Optional. HCFS URIs of archives to be extracted in the working directory of Hadoop drivers and tasks. Supported file types: .jar, .tar, .tar.gz, .tgz, or .zip.

mainJarFileUri (String, optional):

The HCFS URI of the jar file containing the main class. Examples: ‘gs://foo-bucket/analytics-binaries/extract-useful-metrics-mr.jar’ ‘hdfs:/tmp/test-samples/custom-wordcount.jar’ ‘file:///home/usr/lib/hadoop-mapreduce/hadoop-mapreduce-examples.jar’

projectId (dagster.StringSource):

Required. Project ID for the project which the client acts on behalf of. Will be passed when creating a dataset / job. If not passed, falls back to the default inferred from the environment.

region (dagster.StringSource):

job_scoped_cluster (Bool, optional):

whether to create a cluster or use an existing cluster

Default Value: True

dagster_gcp.dataproc_resource ResourceDefinition[source]¶

Config Schema:

projectId (dagster.StringSource):

Required. Project ID for the project which the client acts on behalf of. Will be passed when creating a dataset / job. If not passed, falls back to the default inferred from the environment.

region (dagster.StringSource):

clusterName (dagster.StringSource):

Required. The cluster name. Cluster names within a project must be unique. Names of deleted clusters can be reused.

cluster_config (strict dict, optional):

The cluster config.

Config Schema:

masterConfig (strict dict, optional):

Optional. The config settings for Compute Engine resources in an instance group, such as a master or worker group.

Config Schema:

accelerators (List[strict dict], optional):

Optional. The Compute Engine accelerator configuration for these instances.Beta Feature: This feature is still under development. It may be changed before final release.

numInstances (Int, optional):

Optional. The number of VM instances in the instance group. For master instance groups, must be set to 1.

diskConfig (strict dict, optional):

Specifies the config of disk options for a group of VM instances.

Config Schema:

numLocalSsds (Int, optional):: Optional. Number of attached SSDs, from 0 to 4 (default is 0). If SSDs are not attached, the boot disk is used to store runtime logs and HDFS (https://hadoop.apache.org/docs/r1.2.1/hdfs_user_guide.html) data. If one or more SSDs are attached, this runtime bulk data is spread across them, and the boot disk contains only basic config and installed binaries.
bootDiskSizeGb (Int, optional):: Optional. Size in GB of the boot disk (default is 500GB).
bootDiskType (String, optional):: Optional. Type of the boot disk (default is “pd-standard”). Valid values: “pd-ssd” (Persistent Disk Solid State Drive) or “pd-standard” (Persistent Disk Hard Disk Drive).

managedGroupConfig (strict dict, optional):

Specifies the resources used to actively manage an instance group.

isPreemptible (Bool, optional):

Optional. Specifies that this instance group contains preemptible instances.

imageUri (String, optional):

Optional. The Compute Engine image resource used for cluster instances. It can be specified or may be inferred from SoftwareConfig.image_version.

machineTypeUri (String, optional):

Optional. The Compute Engine machine type used for cluster instances.A full URL, partial URI, or short name are valid. Examples: https://www.googleapis.com/compute/v1/projects/[project_id]/zones/us-east1-a/machineTypes/n1-standard-2 projects/[project_id]/zones/us-east1-a/machineTypes/n1-standard-2 n1-standard-2Auto Zone Exception: If you are using the Cloud Dataproc Auto Zone Placement feature, you must use the short name of the machine type resource, for example, n1-standard-2.

secondaryWorkerConfig (strict dict, optional):

Optional. The config settings for Compute Engine resources in an instance group, such as a master or worker group.

Config Schema:

accelerators (List[strict dict], optional):

Optional. The Compute Engine accelerator configuration for these instances.Beta Feature: This feature is still under development. It may be changed before final release.

numInstances (Int, optional):

Optional. The number of VM instances in the instance group. For master instance groups, must be set to 1.

diskConfig (strict dict, optional):

Specifies the config of disk options for a group of VM instances.

Config Schema:

numLocalSsds (Int, optional):: Optional. Number of attached SSDs, from 0 to 4 (default is 0). If SSDs are not attached, the boot disk is used to store runtime logs and HDFS (https://hadoop.apache.org/docs/r1.2.1/hdfs_user_guide.html) data. If one or more SSDs are attached, this runtime bulk data is spread across them, and the boot disk contains only basic config and installed binaries.
bootDiskSizeGb (Int, optional):: Optional. Size in GB of the boot disk (default is 500GB).
bootDiskType (String, optional):: Optional. Type of the boot disk (default is “pd-standard”). Valid values: “pd-ssd” (Persistent Disk Solid State Drive) or “pd-standard” (Persistent Disk Hard Disk Drive).

managedGroupConfig (strict dict, optional):

Specifies the resources used to actively manage an instance group.

isPreemptible (Bool, optional):

Optional. Specifies that this instance group contains preemptible instances.

imageUri (String, optional):

Optional. The Compute Engine image resource used for cluster instances. It can be specified or may be inferred from SoftwareConfig.image_version.

machineTypeUri (String, optional):

encryptionConfig (strict dict, optional):

Encryption settings for the cluster.

Config Schema:

gcePdKmsKeyName (String, optional):: Optional. The Cloud KMS key name to use for PD disk encryption for all instances in the cluster.

securityConfig (strict dict, optional):

Security related configuration, including Kerberos.

Config Schema:

kerberosConfig (strict dict, optional):

Specifies Kerberos related configuration.

Config Schema:

truststorePasswordUri (String, optional):: Optional. The Cloud Storage URI of a KMS encrypted file containing the password to the user provided truststore. For the self-signed certificate, this password is generated by Dataproc.
enableKerberos (Bool, optional):: Optional. Flag to indicate whether to Kerberize the cluster.
truststoreUri (String, optional):: Optional. The Cloud Storage URI of the truststore file used for SSL encryption. If not provided, Dataproc will provide a self-signed certificate.
crossRealmTrustRealm (String, optional):: Optional. The remote realm the Dataproc on-cluster KDC will trust, should the user enable cross realm trust.
rootPrincipalPasswordUri (String, optional):: Required. The Cloud Storage URI of a KMS encrypted file containing the root principal password.
kmsKeyUri (String, optional):: Required. The uri of the KMS key used to encrypt various sensitive files.
crossRealmTrustKdc (String, optional):: Optional. The KDC (IP or hostname) for the remote trusted realm in a cross realm trust relationship.
crossRealmTrustSharedPasswordUri (String, optional):: Optional. The Cloud Storage URI of a KMS encrypted file containing the shared password between the on-cluster Kerberos realm and the remote trusted realm, in a cross realm trust relationship.
tgtLifetimeHours (Int, optional):: Optional. The lifetime of the ticket granting ticket, in hours. If not specified, or user specifies 0, then default value 10 will be used.
keystoreUri (String, optional):: Optional. The Cloud Storage URI of the keystore file used for SSL encryption. If not provided, Dataproc will provide a self-signed certificate.
keyPasswordUri (String, optional):: Optional. The Cloud Storage URI of a KMS encrypted file containing the password to the user provided key. For the self-signed certificate, this password is generated by Dataproc.
keystorePasswordUri (String, optional):: Optional. The Cloud Storage URI of a KMS encrypted file containing the password to the user provided keystore. For the self-signed certificate, this password is generated by Dataproc.
crossRealmTrustAdminServer (String, optional):: Optional. The admin server (IP or hostname) for the remote trusted realm in a cross realm trust relationship.
kdcDbKeyUri (String, optional):: Optional. The Cloud Storage URI of a KMS encrypted file containing the master key of the KDC database.

initializationActions (List[strict dict], optional):

Optional. Commands to execute on each node after config is completed. By default, executables are run on master and all worker nodes. You can test a node’s role metadata to run an executable on a master or worker node, as shown below using curl (you can also use wget): ROLE=$(curl -H Metadata-Flavor:Google http://metadata/computeMetadata/v1/instance/attributes/dataproc-role) if [[ “${ROLE}” == ‘Master’ ]]; then … master specific actions … else … worker specific actions … fi

configBucket (String, optional):

Optional. A Google Cloud Storage bucket used to stage job dependencies, config files, and job driver console output. If you do not specify a staging bucket, Cloud Dataproc will determine a Cloud Storage location (US, ASIA, or EU) for your cluster’s staging bucket according to the Google Compute Engine zone where your cluster is deployed, and then create and manage this project-level, per-location bucket (see Cloud Dataproc staging bucket).

workerConfig (strict dict, optional):

Optional. The config settings for Compute Engine resources in an instance group, such as a master or worker group.

Config Schema:

accelerators (List[strict dict], optional):

Optional. The Compute Engine accelerator configuration for these instances.Beta Feature: This feature is still under development. It may be changed before final release.

numInstances (Int, optional):

Optional. The number of VM instances in the instance group. For master instance groups, must be set to 1.

diskConfig (strict dict, optional):

Specifies the config of disk options for a group of VM instances.

Config Schema:

numLocalSsds (Int, optional):: Optional. Number of attached SSDs, from 0 to 4 (default is 0). If SSDs are not attached, the boot disk is used to store runtime logs and HDFS (https://hadoop.apache.org/docs/r1.2.1/hdfs_user_guide.html) data. If one or more SSDs are attached, this runtime bulk data is spread across them, and the boot disk contains only basic config and installed binaries.
bootDiskSizeGb (Int, optional):: Optional. Size in GB of the boot disk (default is 500GB).
bootDiskType (String, optional):: Optional. Type of the boot disk (default is “pd-standard”). Valid values: “pd-ssd” (Persistent Disk Solid State Drive) or “pd-standard” (Persistent Disk Hard Disk Drive).

managedGroupConfig (strict dict, optional):

Specifies the resources used to actively manage an instance group.

isPreemptible (Bool, optional):

Optional. Specifies that this instance group contains preemptible instances.

imageUri (String, optional):

Optional. The Compute Engine image resource used for cluster instances. It can be specified or may be inferred from SoftwareConfig.image_version.

machineTypeUri (String, optional):

gceClusterConfig (strict dict, optional):

Common config settings for resources of Compute Engine cluster instances, applicable to all instances in the cluster.

Config Schema:

networkUri (String, optional):: Optional. The Compute Engine network to be used for machine communications. Cannot be specified with subnetwork_uri. If neither network_uri nor subnetwork_uri is specified, the “default” network of the project is used, if it exists. Cannot be a “Custom Subnet Network” (see Using Subnetworks for more information).A full URL, partial URI, or short name are valid. Examples: https://www.googleapis.com/compute/v1/projects/[project_id]/regions/global/default projects/[project_id]/regions/global/default default
zoneUri (String, optional):: Optional. The zone where the Compute Engine cluster will be located. On a create request, it is required in the “global” region. If omitted in a non-global Cloud Dataproc region, the service will pick a zone in the corresponding Compute Engine region. On a get request, zone will always be present.A full URL, partial URI, or short name are valid. Examples: https://www.googleapis.com/compute/v1/projects/[project_id]/zones/[zone] projects/[project_id]/zones/[zone] us-central1-f
metadata (permissive dict, optional):: The Compute Engine metadata entries to add to all instances (see Project and instance metadata (https://cloud.google.com/compute/docs/storing-retrieving-metadata#project_and_instance_metadata)).
internalIpOnly (Bool, optional):: Optional. If true, all instances in the cluster will only have internal IP addresses. By default, clusters are not restricted to internal IP addresses, and will have ephemeral external IP addresses assigned to each instance. This internal_ip_only restriction can only be enabled for subnetwork enabled networks, and all off-cluster dependencies must be configured to be accessible without external IP addresses.
serviceAccountScopes (List[String], optional):: Optional. The URIs of service account scopes to be included in Compute Engine instances. The following base set of scopes is always included: https://www.googleapis.com/auth/cloud.useraccounts.readonly https://www.googleapis.com/auth/devstorage.read_write https://www.googleapis.com/auth/logging.writeIf no scopes are specified, the following defaults are also provided: https://www.googleapis.com/auth/bigquery https://www.googleapis.com/auth/bigtable.admin.table https://www.googleapis.com/auth/bigtable.data https://www.googleapis.com/auth/devstorage.full_control
tags (List[String], optional):: The Compute Engine tags to add to all instances (see Tagging instances).
serviceAccount (String, optional):: Optional. The service account of the instances. Defaults to the default Compute Engine service account. Custom service accounts need permissions equivalent to the following IAM roles: roles/logging.logWriter roles/storage.objectAdmin(see https://cloud.google.com/compute/docs/access/service-accounts#custom_service_accounts for more information). Example: [account_id]@[project_id].iam.gserviceaccount.com
subnetworkUri (String, optional):: Optional. The Compute Engine subnetwork to be used for machine communications. Cannot be specified with network_uri.A full URL, partial URI, or short name are valid. Examples: https://www.googleapis.com/compute/v1/projects/[project_id]/regions/us-east1/subnetworks/sub0 projects/[project_id]/regions/us-east1/subnetworks/sub0 sub0

softwareConfig (strict dict, optional):

Specifies the selection and config of software inside the cluster.

Config Schema:

properties (permissive dict, optional):: Optional. The properties to set on daemon config files.Property keys are specified in prefix:property format, for example core:hadoop.tmp.dir. The following are supported prefixes and their mappings: capacity-scheduler: capacity-scheduler.xml core: core-site.xml distcp: distcp-default.xml hdfs: hdfs-site.xml hive: hive-site.xml mapred: mapred-site.xml pig: pig.properties spark: spark-defaults.conf yarn: yarn-site.xmlFor more information, see Cluster properties.
optionalComponents (List[Component], optional):: The set of optional components to activate on the cluster.
imageVersion (String, optional):: Optional. The version of software inside the cluster. It must be one of the supported Cloud Dataproc Versions, such as “1.2” (including a subminor version, such as “1.2.29”), or the “preview” version. If unspecified, it defaults to the latest Debian version.

GCS¶

dagster_gcp.gcs_resource ResourceDefinition[source]¶

Config Schema:

project (Union[dagster.StringSource, None], optional):: Project name

dagster_gcp.gcs.gcs_pickle_io_manager IOManagerDefinition[source]¶

Config Schema:

gcs_bucket (dagster.StringSource):
gcs_prefix (dagster.StringSource, optional):: Default Value: ‘dagster’

Persistent IO manager using GCS for storage.

Serializes objects via pickling. Suitable for objects storage for distributed executors, so long as each execution node has network connectivity and credentials for GCS and the backing bucket.

Assigns each op output to a unique filepath containing run ID, step key, and output name. Assigns each asset to a single filesystem path, at “<base_dir>/<asset_key>”. If the asset key has multiple components, the final component is used as the name of the file, and the preceding components as parent directories under the base_dir.

Subsequent materializations of an asset will overwrite previous materializations of that asset. With a base directory of “/my/base/path”, an asset with key AssetKey([“one”, “two”, “three”]) would be stored in a file called “three” in a directory with path “/my/base/path/one/two/”.

Example usage:

Attach this IO manager to a set of assets.

from dagster import asset, repository, with_resources
from dagster_gcp.gcs import gcs_pickle_io_manager, gcs_resource

@asset
def asset1():
    # create df ...
    return df

@asset
def asset2(asset1):
    return df[:5]

@repository
def repo():
    return with_resources(
        [asset1, asset2],
        resource_defs={
            "io_manager": gcs_pickle_io_manager.configured(
                {"gcs_bucket": "my-cool-bucket", "gcs_prefix": "my-cool-prefix"}
            ),
            "gcs": gcs_resource,
        },
    )
)

Attach this IO manager to your job to make it available to your ops.

from dagster import job
from dagster_gcp.gcs import gcs_pickle_io_manager, gcs_resource

@job(
    resource_defs={
        "io_manager": gcs_pickle_io_manager.configured(
            {"gcs_bucket": "my-cool-bucket", "gcs_prefix": "my-cool-prefix"}
        ),
        "gcs": gcs_resource,
    },
)
def my_job():
    ...

File Manager (Experimental)¶

class dagster_gcp.GCSFileHandle(gcs_bucket, gcs_key)[source]¶: A reference to a file on GCS.

dagster_gcp.gcs_file_manager ResourceDefinition[source]¶

FileManager that provides abstract access to GCS.

Implements the FileManager API.