|
| 1 | +# Copyright 2019 Google LLC |
| 2 | +# |
| 3 | +# Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | +# you may not use this file except in compliance with the License. |
| 5 | +# You may obtain a copy of the License at |
| 6 | +# |
| 7 | +# https://www.apache.org/licenses/LICENSE-2.0 |
| 8 | +# |
| 9 | +# Unless required by applicable law or agreed to in writing, software |
| 10 | +# distributed under the License is distributed on an "AS IS" BASIS, |
| 11 | +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 | +# See the License for the specific language governing permissions and |
| 13 | +# limitations under the License. |
| 14 | + |
| 15 | + |
| 16 | +def quickstart(override_values): |
| 17 | + """Creates a tag template and attach a tag to a BigQuery table.""" |
| 18 | + # [START data_catalog_quickstart] |
| 19 | + # Import required modules. |
| 20 | + from google.cloud import datacatalog_v1 |
| 21 | + |
| 22 | + # TODO: Set these values before running the sample. |
| 23 | + # Google Cloud Platform project. |
| 24 | + project_id = "my_project" |
| 25 | + # Set dataset_id to the ID of existing dataset. |
| 26 | + dataset_id = "demo_dataset" |
| 27 | + # Set table_id to the ID of existing table. |
| 28 | + table_id = "trips" |
| 29 | + # Tag template to create. |
| 30 | + tag_template_id = "example_tag_template" |
| 31 | + |
| 32 | + # [END data_catalog_quickstart] |
| 33 | + |
| 34 | + # To facilitate testing, we replace values with alternatives |
| 35 | + # provided by the testing harness. |
| 36 | + project_id = override_values.get("project_id", project_id) |
| 37 | + dataset_id = override_values.get("dataset_id", dataset_id) |
| 38 | + table_id = override_values.get("table_id", table_id) |
| 39 | + tag_template_id = override_values.get("tag_template_id", tag_template_id) |
| 40 | + |
| 41 | + # [START data_catalog_quickstart] |
| 42 | + # For all regions available, see: |
| 43 | + # https://cloud.google.com/data-catalog/docs/concepts/regions |
| 44 | + location = "us-central1" |
| 45 | + |
| 46 | + # Use Application Default Credentials to create a new |
| 47 | + # Data Catalog client. GOOGLE_APPLICATION_CREDENTIALS |
| 48 | + # environment variable must be set with the location |
| 49 | + # of a service account key file. |
| 50 | + datacatalog_client = datacatalog_v1.DataCatalogClient() |
| 51 | + |
| 52 | + # Create a Tag Template. |
| 53 | + tag_template = datacatalog_v1.types.TagTemplate() |
| 54 | + |
| 55 | + tag_template.display_name = "Demo Tag Template" |
| 56 | + |
| 57 | + tag_template.fields["source"] = datacatalog_v1.types.TagTemplateField() |
| 58 | + tag_template.fields["source"].display_name = "Source of data asset" |
| 59 | + tag_template.fields[ |
| 60 | + "source" |
| 61 | + ].type_.primitive_type = datacatalog_v1.types.FieldType.PrimitiveType.STRING |
| 62 | + |
| 63 | + tag_template.fields["num_rows"] = datacatalog_v1.types.TagTemplateField() |
| 64 | + tag_template.fields["num_rows"].display_name = "Number of rows in data asset" |
| 65 | + tag_template.fields[ |
| 66 | + "num_rows" |
| 67 | + ].type_.primitive_type = datacatalog_v1.types.FieldType.PrimitiveType.DOUBLE |
| 68 | + |
| 69 | + tag_template.fields["has_pii"] = datacatalog_v1.types.TagTemplateField() |
| 70 | + tag_template.fields["has_pii"].display_name = "Has PII" |
| 71 | + tag_template.fields[ |
| 72 | + "has_pii" |
| 73 | + ].type_.primitive_type = datacatalog_v1.types.FieldType.PrimitiveType.BOOL |
| 74 | + |
| 75 | + tag_template.fields["pii_type"] = datacatalog_v1.types.TagTemplateField() |
| 76 | + tag_template.fields["pii_type"].display_name = "PII type" |
| 77 | + |
| 78 | + for display_name in ["EMAIL", "SOCIAL SECURITY NUMBER", "NONE"]: |
| 79 | + enum_value = datacatalog_v1.types.FieldType.EnumType.EnumValue( |
| 80 | + display_name=display_name |
| 81 | + ) |
| 82 | + tag_template.fields["pii_type"].type_.enum_type.allowed_values.append( |
| 83 | + enum_value |
| 84 | + ) |
| 85 | + |
| 86 | + expected_template_name = datacatalog_v1.DataCatalogClient.tag_template_path( |
| 87 | + project_id, location, tag_template_id |
| 88 | + ) |
| 89 | + |
| 90 | + # Create the Tag Template. |
| 91 | + try: |
| 92 | + tag_template = datacatalog_client.create_tag_template( |
| 93 | + parent=f"projects/{project_id}/locations/{location}", |
| 94 | + tag_template_id=tag_template_id, |
| 95 | + tag_template=tag_template, |
| 96 | + ) |
| 97 | + print(f"Created template: {tag_template.name}") |
| 98 | + except OSError as e: |
| 99 | + print(f"Cannot create template: {expected_template_name}") |
| 100 | + print(f"{e}") |
| 101 | + |
| 102 | + # Lookup Data Catalog's Entry referring to the table. |
| 103 | + resource_name = ( |
| 104 | + f"//bigquery.googleapis.com/projects/{project_id}" |
| 105 | + f"/datasets/{dataset_id}/tables/{table_id}" |
| 106 | + ) |
| 107 | + table_entry = datacatalog_client.lookup_entry( |
| 108 | + request={"linked_resource": resource_name} |
| 109 | + ) |
| 110 | + |
| 111 | + # Attach a Tag to the table. |
| 112 | + tag = datacatalog_v1.types.Tag() |
| 113 | + |
| 114 | + tag.template = tag_template.name |
| 115 | + tag.name = "my_super_cool_tag" |
| 116 | + |
| 117 | + tag.fields["source"] = datacatalog_v1.types.TagField() |
| 118 | + tag.fields["source"].string_value = "Copied from tlc_yellow_trips_2018" |
| 119 | + |
| 120 | + tag.fields["num_rows"] = datacatalog_v1.types.TagField() |
| 121 | + tag.fields["num_rows"].double_value = 113496874 |
| 122 | + |
| 123 | + tag.fields["has_pii"] = datacatalog_v1.types.TagField() |
| 124 | + tag.fields["has_pii"].bool_value = False |
| 125 | + |
| 126 | + tag.fields["pii_type"] = datacatalog_v1.types.TagField() |
| 127 | + tag.fields["pii_type"].enum_value.display_name = "NONE" |
| 128 | + |
| 129 | + tag = datacatalog_client.create_tag(parent=table_entry.name, tag=tag) |
| 130 | + print(f"Created tag: {tag.name}") |
| 131 | + # [END data_catalog_quickstart] |
0 commit comments