datacontract-cli 0.10.22__tar.gz → 0.10.23__tar.gz

{datacontract_cli-0.10.22/datacontract_cli.egg-info → datacontract_cli-0.10.23}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: datacontract-cli
-Version: 0.10.22
+Version: 0.10.23
 Summary: The datacontract CLI is an open source command-line tool for working with Data Contracts. It uses data contract YAML files to lint the data contract, connect to data sources and execute schema and quality tests, detect breaking changes, and export to different formats. The tool is written in Python. It can be used as a standalone CLI tool, in a CI/CD pipeline, or directly as a Python library.
 Author-email: Jochen Christ <jochen.christ@innoq.com>, Stefan Negele <stefan.negele@innoq.com>, Simon Harrer <simon.harrer@innoq.com>
 Project-URL: Homepage, https://cli.datacontract.com
@@ -16,12 +16,12 @@ Requires-Dist: pydantic<2.11.0,>=2.8.2
 Requires-Dist: pyyaml~=6.0.1
 Requires-Dist: requests<2.33,>=2.31
 Requires-Dist: fastjsonschema<2.22.0,>=2.19.1
-Requires-Dist: fastparquet==2024.11.0
+Requires-Dist: fastparquet<2025.0.0,>=2024.5.0
 Requires-Dist: numpy<2.0.0,>=1.26.4
 Requires-Dist: python-multipart==0.0.20
 Requires-Dist: rich<13.10,>=13.7
 Requires-Dist: sqlglot<27.0.0,>=26.6.0
-Requires-Dist: duckdb==1.1.2
+Requires-Dist: duckdb<2.0.0,>=1.0.0
 Requires-Dist: soda-core-duckdb<3.5.0,>=3.3.20
 Requires-Dist: setuptools>=60
 Requires-Dist: python-dotenv~=1.0.0
@@ -39,7 +39,7 @@ Provides-Extra: databricks
 Requires-Dist: soda-core-spark-df<3.4.0,>=3.3.20; extra == "databricks"
 Requires-Dist: soda-core-spark[databricks]<3.4.0,>=3.3.20; extra == "databricks"
 Requires-Dist: databricks-sql-connector<3.8.0,>=3.7.0; extra == "databricks"
-Requires-Dist: databricks-sdk<0.41.0; extra == "databricks"
+Requires-Dist: databricks-sdk<0.45.0; extra == "databricks"
 Provides-Extra: iceberg
 Requires-Dist: pyiceberg==0.8.1; extra == "iceberg"
 Provides-Extra: kafka
@@ -48,7 +48,7 @@ Requires-Dist: soda-core-spark-df<3.4.0,>=3.3.20; extra == "kafka"
 Provides-Extra: postgres
 Requires-Dist: soda-core-postgres<3.4.0,>=3.3.20; extra == "postgres"
 Provides-Extra: s3
-Requires-Dist: s3fs==2024.12.0; extra == "s3"
+Requires-Dist: s3fs==2025.2.0; extra == "s3"
 Requires-Dist: aiobotocore<2.20.0,>=2.17.0; extra == "s3"
 Provides-Extra: snowflake
 Requires-Dist: snowflake-connector-python[pandas]<3.14,>=3.6; extra == "snowflake"
@@ -66,7 +66,7 @@ Requires-Dist: pyarrow>=18.1.0; extra == "parquet"
 Provides-Extra: rdf
 Requires-Dist: rdflib==7.0.0; extra == "rdf"
 Provides-Extra: api
-Requires-Dist: fastapi==0.115.6; extra == "api"
+Requires-Dist: fastapi==0.115.8; extra == "api"
 Requires-Dist: uvicorn==0.34.0; extra == "api"
 Provides-Extra: all
 Requires-Dist: datacontract-cli[api,bigquery,csv,databricks,dbml,dbt,iceberg,kafka,parquet,postgres,rdf,s3,snowflake,sqlserver,trino]; extra == "all"
@@ -230,6 +230,12 @@ if not run.has_passed():
     # Abort pipeline, alert, or take corrective actions...
 ```
 
+## How to
+
+- [How to integrate Data Contract CLI in your CI/CD pipeline as a GitHub Action](https://github.com/datacontract/datacontract-action/)
+- [How to run the Data Contract CLI API to test data contracts with POST requests](https://cli.datacontract.com/API)
+- [How to run Data Contract CLI in a Databricks pipeline](https://www.datamesh-architecture.com/howto/build-a-dataproduct-with-databricks#test-the-data-product)
+
 
 ## Installation
 
@@ -378,49 +384,41 @@ Commands
 
 ### test
 ```
-                                                                                
- Usage: datacontract test [OPTIONS] [LOCATION]                                  
-                                                                                
- Run schema and quality tests on configured servers.                            
-                                                                                
-╭─ Arguments ──────────────────────────────────────────────────────────────────╮
-│   location      [LOCATION]  The location (url or path) of the data contract  │
-│                             yaml.                                            │
-│                             [default: datacontract.yaml]                     │
-╰──────────────────────────────────────────────────────────────────────────────╯
-╭─ Options ────────────────────────────────────────────────────────────────────╮
-│ --schema                                       TEXT  The location (url or    │
-│                                                      path) of the Data       │
-│                                                      Contract Specification  │
-│                                                      JSON Schema             │
-│                                                      [default:               │
-│                                                      https://datacontract.c… │
-│ --server                                       TEXT  The server              │
-│                                                      configuration to run    │
-│                                                      the schema and quality  │
-│                                                      tests. Use the key of   │
-│                                                      the server object in    │
-│                                                      the data contract yaml  │
-│                                                      file to refer to a      │
-│                                                      server, e.g.,           │
-│                                                      `production`, or `all`  │
-│                                                      for all servers         │
-│                                                      (default).              │
-│                                                      [default: all]          │
-│ --publish                                      TEXT  The url to publish the  │
-│                                                      results after the test  │
-│                                                      [default: None]         │
-│ --logs                --no-logs                      Print logs              │
-│                                                      [default: no-logs]      │
-│ --ssl-verification    --no-ssl-verification          SSL verification when   │
-│                                                      publishing the data     │
-│                                                      contract.               │
-│                                                      [default:               │
-│                                                      ssl-verification]       │
-│ --help                                               Show this message and   │
-│                                                      exit.                   │
-╰──────────────────────────────────────────────────────────────────────────────╯
-
+                                                                                                    
+ Usage: datacontract test [OPTIONS] [LOCATION]                                                      
+                                                                                                    
+ Run schema and quality tests on configured servers.                                                
+                                                                                                    
+╭─ Arguments ──────────────────────────────────────────────────────────────────────────────────────╮
+│   location      [LOCATION]  The location (url or path) of the data contract yaml.                │
+│                             [default: datacontract.yaml]                                         │
+╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
+╭─ Options ────────────────────────────────────────────────────────────────────────────────────────╮
+│ --schema                                       TEXT     The location (url or path) of the Data   │
+│                                                         Contract Specification JSON Schema       │
+│                                                         [default: None]                          │
+│ --server                                       TEXT     The server configuration to run the      │
+│                                                         schema and quality tests. Use the key of │
+│                                                         the server object in the data contract   │
+│                                                         yaml file to refer to a server, e.g.,    │
+│                                                         `production`, or `all` for all servers   │
+│                                                         (default).                               │
+│                                                         [default: all]                           │
+│ --publish                                      TEXT     The url to publish the results after the │
+│                                                         test                                     │
+│                                                         [default: None]                          │
+│ --output                                       PATH     Specify the file path where the test     │
+│                                                         results should be written to (e.g.,      │
+│                                                         './test-results/TEST-datacontract.xml'). │
+│                                                         [default: None]                          │
+│ --output-format                                [junit]  The target format for the test results.  │
+│                                                         [default: None]                          │
+│ --logs                --no-logs                         Print logs [default: no-logs]            │
+│ --ssl-verification    --no-ssl-verification             SSL verification when publishing the     │
+│                                                         data contract.                           │
+│                                                         [default: ssl-verification]              │
+│ --help                                                  Show this message and exit.              │
+╰──────────────────────────────────────────────────────────────────────────────────────────────────╯  
 ```
 
 Data Contract CLI connects to a data source and runs schema and quality tests to verify that the data contract is valid.
@@ -1037,6 +1035,12 @@ The export function converts the logical data types of the datacontract into the
 if a server is selected via the `--server` option (based on the `type` of that server). If no server is selected, the
 logical data types are exported.
 
+#### DBT & DBT-SOURCES
+
+The export funciton converts the datacontract to dbt models in YAML format, with support for SQL dialects.
+If a server is selected via the `--server` option (based on the `type` of that server) then the DBT column `data_types` match the expected data types of the server.
+If no server is selected, then it defaults to `snowflake`.
+
 #### Spark
 
 The export function converts the data contract specification into a StructType Spark schema. The returned value is a Python code picture of the model schemas.
@@ -1902,7 +1906,7 @@ Python base interpreter should be 3.11.x (unless working on 3.12 release candida
 
 ```bash
 # create venv
-python3 -m venv venv
+python3.11 -m venv venv
 source venv/bin/activate
 
 # Install Requirements

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/README.md

@@ -144,6 +144,12 @@ if not run.has_passed():
     # Abort pipeline, alert, or take corrective actions...
 ```
 
+## How to
+
+- [How to integrate Data Contract CLI in your CI/CD pipeline as a GitHub Action](https://github.com/datacontract/datacontract-action/)
+- [How to run the Data Contract CLI API to test data contracts with POST requests](https://cli.datacontract.com/API)
+- [How to run Data Contract CLI in a Databricks pipeline](https://www.datamesh-architecture.com/howto/build-a-dataproduct-with-databricks#test-the-data-product)
+
 
 ## Installation
 
@@ -292,49 +298,41 @@ Commands
 
 ### test
 ```
-                                                                                
- Usage: datacontract test [OPTIONS] [LOCATION]                                  
-                                                                                
- Run schema and quality tests on configured servers.                            
-                                                                                
-╭─ Arguments ──────────────────────────────────────────────────────────────────╮
-│   location      [LOCATION]  The location (url or path) of the data contract  │
-│                             yaml.                                            │
-│                             [default: datacontract.yaml]                     │
-╰──────────────────────────────────────────────────────────────────────────────╯
-╭─ Options ────────────────────────────────────────────────────────────────────╮
-│ --schema                                       TEXT  The location (url or    │
-│                                                      path) of the Data       │
-│                                                      Contract Specification  │
-│                                                      JSON Schema             │
-│                                                      [default:               │
-│                                                      https://datacontract.c… │
-│ --server                                       TEXT  The server              │
-│                                                      configuration to run    │
-│                                                      the schema and quality  │
-│                                                      tests. Use the key of   │
-│                                                      the server object in    │
-│                                                      the data contract yaml  │
-│                                                      file to refer to a      │
-│                                                      server, e.g.,           │
-│                                                      `production`, or `all`  │
-│                                                      for all servers         │
-│                                                      (default).              │
-│                                                      [default: all]          │
-│ --publish                                      TEXT  The url to publish the  │
-│                                                      results after the test  │
-│                                                      [default: None]         │
-│ --logs                --no-logs                      Print logs              │
-│                                                      [default: no-logs]      │
-│ --ssl-verification    --no-ssl-verification          SSL verification when   │
-│                                                      publishing the data     │
-│                                                      contract.               │
-│                                                      [default:               │
-│                                                      ssl-verification]       │
-│ --help                                               Show this message and   │
-│                                                      exit.                   │
-╰──────────────────────────────────────────────────────────────────────────────╯
-
+                                                                                                    
+ Usage: datacontract test [OPTIONS] [LOCATION]                                                      
+                                                                                                    
+ Run schema and quality tests on configured servers.                                                
+                                                                                                    
+╭─ Arguments ──────────────────────────────────────────────────────────────────────────────────────╮
+│   location      [LOCATION]  The location (url or path) of the data contract yaml.                │
+│                             [default: datacontract.yaml]                                         │
+╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
+╭─ Options ────────────────────────────────────────────────────────────────────────────────────────╮
+│ --schema                                       TEXT     The location (url or path) of the Data   │
+│                                                         Contract Specification JSON Schema       │
+│                                                         [default: None]                          │
+│ --server                                       TEXT     The server configuration to run the      │
+│                                                         schema and quality tests. Use the key of │
+│                                                         the server object in the data contract   │
+│                                                         yaml file to refer to a server, e.g.,    │
+│                                                         `production`, or `all` for all servers   │
+│                                                         (default).                               │
+│                                                         [default: all]                           │
+│ --publish                                      TEXT     The url to publish the results after the │
+│                                                         test                                     │
+│                                                         [default: None]                          │
+│ --output                                       PATH     Specify the file path where the test     │
+│                                                         results should be written to (e.g.,      │
+│                                                         './test-results/TEST-datacontract.xml'). │
+│                                                         [default: None]                          │
+│ --output-format                                [junit]  The target format for the test results.  │
+│                                                         [default: None]                          │
+│ --logs                --no-logs                         Print logs [default: no-logs]            │
+│ --ssl-verification    --no-ssl-verification             SSL verification when publishing the     │
+│                                                         data contract.                           │
+│                                                         [default: ssl-verification]              │
+│ --help                                                  Show this message and exit.              │
+╰──────────────────────────────────────────────────────────────────────────────────────────────────╯  
 ```
 
 Data Contract CLI connects to a data source and runs schema and quality tests to verify that the data contract is valid.
@@ -951,6 +949,12 @@ The export function converts the logical data types of the datacontract into the
 if a server is selected via the `--server` option (based on the `type` of that server). If no server is selected, the
 logical data types are exported.
 
+#### DBT & DBT-SOURCES
+
+The export funciton converts the datacontract to dbt models in YAML format, with support for SQL dialects.
+If a server is selected via the `--server` option (based on the `type` of that server) then the DBT column `data_types` match the expected data types of the server.
+If no server is selected, then it defaults to `snowflake`.
+
 #### Spark
 
 The export function converts the data contract specification into a StructType Spark schema. The returned value is a Python code picture of the model schemas.
@@ -1816,7 +1820,7 @@ Python base interpreter should be 3.11.x (unless working on 3.12 release candida
 
 ```bash
 # create venv
-python3 -m venv venv
+python3.11 -m venv venv
 source venv/bin/activate
 
 # Install Requirements

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/datacontract/cli.py

@@ -5,9 +5,7 @@ from typing import Iterable, List, Optional
 
 import typer
 from click import Context
-from rich import box
 from rich.console import Console
-from rich.table import Table
 from typer.core import TyperGroup
 from typing_extensions import Annotated
 
@@ -19,6 +17,8 @@ from datacontract.integration.datamesh_manager import (
     publish_data_contract_to_datamesh_manager,
 )
 from datacontract.lint.resolve import resolve_data_contract_dict
+from datacontract.output.output_format import OutputFormat
+from datacontract.output.test_results_writer import write_test_result
 
 console = Console()
 
@@ -92,12 +92,19 @@ def lint(
         str,
         typer.Option(help="The location (url or path) of the Data Contract Specification JSON Schema"),
     ] = None,
+    output: Annotated[
+        Path,
+        typer.Option(
+            help="Specify the file path where the test results should be written to (e.g., './test-results/TEST-datacontract.xml'). If no path is provided, the output will be printed to stdout."
+        ),
+    ] = None,
+    output_format: Annotated[OutputFormat, typer.Option(help="The target format for the test results.")] = None,
 ):
     """
     Validate that the datacontract.yaml is correctly formatted.
     """
     run = DataContract(data_contract_file=location, schema_location=schema).lint()
-    _handle_result(run)
+    write_test_result(run, console, output_format, output)
 
 
 @app.command()
@@ -120,6 +127,13 @@ def test(
         ),
     ] = "all",
     publish: Annotated[str, typer.Option(help="The url to publish the results after the test")] = None,
+    output: Annotated[
+        Path,
+        typer.Option(
+            help="Specify the file path where the test results should be written to (e.g., './test-results/TEST-datacontract.xml')."
+        ),
+    ] = None,
+    output_format: Annotated[OutputFormat, typer.Option(help="The target format for the test results.")] = None,
     logs: Annotated[bool, typer.Option(help="Print logs")] = False,
     ssl_verification: Annotated[
         bool,
@@ -141,7 +155,7 @@ def test(
     ).test()
     if logs:
         _print_logs(run)
-    _handle_result(run)
+    write_test_result(run, console, output_format, output)
 
 
 @app.command()
@@ -214,7 +228,7 @@ def export(
     if output is None:
         console.print(result, markup=False, soft_wrap=True)
     else:
-        with output.open("w") as f:
+        with output.open(mode="w", encoding="utf-8") as f:
             f.write(result)
         console.print(f"Written result to {output}")
 
@@ -306,7 +320,7 @@ def import_(
     if output is None:
         console.print(result.to_yaml(), markup=False, soft_wrap=True)
     else:
-        with output.open("w") as f:
+        with output.open(mode="w", encoding="utf-8") as f:
             f.write(result.to_yaml())
         console.print(f"Written result to {output}")
 
@@ -467,77 +481,11 @@ def api(
     uvicorn.run(app="datacontract.api:app", port=port, host=host, reload=True, log_config=LOGGING_CONFIG)
 
 
-def _handle_result(run):
-    _print_table(run)
-    if run.result == "passed":
-        console.print(
-            f"🟢 data contract is valid. Run {len(run.checks)} checks. Took {(run.timestampEnd - run.timestampStart).total_seconds()} seconds."
-        )
-    elif run.result == "warning":
-        console.print("🟠 data contract has warnings. Found the following warnings:")
-        i = 1
-        for check in run.checks:
-            if check.result != "passed":
-                field = to_field(run, check)
-                if field:
-                    field = field + " "
-                else:
-                    field = ""
-                console.print(f"{i}) {field}{check.name}: {check.reason}")
-                i += 1
-    else:
-        console.print("🔴 data contract is invalid, found the following errors:")
-        i = 1
-        for check in run.checks:
-            if check.result != "passed":
-                field = to_field(run, check)
-                if field:
-                    field = field + " "
-                else:
-                    field = ""
-                console.print(f"{i}) {field}{check.name}: {check.reason}")
-                i += 1
-        raise typer.Exit(code=1)
-
-
-def _print_table(run):
-    table = Table(box=box.ROUNDED)
-    table.add_column("Result", no_wrap=True)
-    table.add_column("Check", max_width=100)
-    table.add_column("Field", max_width=32)
-    table.add_column("Details", max_width=50)
-    for check in sorted(run.checks, key=lambda c: (c.result or "", c.model or "", c.field or "")):
-        table.add_row(with_markup(check.result), check.name, to_field(run, check), check.reason)
-    console.print(table)
-
-
-def to_field(run, check):
-    models = [c.model for c in run.checks]
-    if len(set(models)) > 1:
-        if check.field is None:
-            return check.model
-        return check.model + "." + check.field
-    else:
-        return check.field
-
-
 def _print_logs(run):
     console.print("\nLogs:")
     for log in run.logs:
         console.print(log.timestamp.strftime("%y-%m-%d %H:%M:%S"), log.level.ljust(5), log.message)
 
 
-def with_markup(result):
-    if result == "passed":
-        return "[green]passed[/green]"
-    if result == "warning":
-        return "[yellow]warning[/yellow]"
-    if result == "failed":
-        return "[red]failed[/red]"
-    if result == "error":
-        return "[red]error[/red]"
-    return result
-
-
 if __name__ == "__main__":
     app()

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/datacontract/engines/data_contract_test.py

@@ -29,12 +29,7 @@ def execute_data_contract_test(
             reason="Models block is missing. Skip executing tests.",
             engine="datacontract",
         )
-    check_that_datacontract_contains_valid_server_configuration(run, data_contract_specification, server_name)
-    if server_name:
-        server = data_contract_specification.servers.get(server_name)
-    else:
-        server_name = list(data_contract_specification.servers.keys())[0]
-        server = data_contract_specification.servers.get(server_name)
+    server = get_server(data_contract_specification, server_name)
     run.log_info(f"Running tests for data contract {data_contract_specification.id} with server {server_name}")
     run.dataContractId = data_contract_specification.id
     run.dataContractVersion = data_contract_specification.info.version
@@ -49,3 +44,24 @@ def execute_data_contract_test(
     if server.format == "json" and server.type != "kafka":
         check_jsonschema(run, data_contract_specification, server)
     check_soda_execute(run, data_contract_specification, server, spark)
+
+
+def get_server(data_contract_specification: DataContractSpecification, server_name: str = None):
+    """Get the server configuration from the data contract specification.
+
+    Args:
+        data_contract_specification: The data contract specification
+        server_name: Optional name of the server to use. If not provided, uses the first server.
+
+    Returns:
+        The selected server configuration
+    """
+
+    check_that_datacontract_contains_valid_server_configuration(data_contract_specification, server_name)
+
+    if server_name:
+        server = data_contract_specification.servers.get(server_name)
+    else:
+        server_name = list(data_contract_specification.servers.keys())[0]
+        server = data_contract_specification.servers.get(server_name)
+    return server

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/datacontract/engines/datacontract/check_that_datacontract_contains_valid_servers_configuration.py

@@ -1,12 +1,11 @@
 from datacontract.model.data_contract_specification import DataContractSpecification
 from datacontract.model.exceptions import DataContractException
-from datacontract.model.run import Run
 
 
 def check_that_datacontract_contains_valid_server_configuration(
-    run: Run, data_contract: DataContractSpecification, server_name: str
+    data_contract: DataContractSpecification, server_name: str | None
 ):
-    if data_contract.servers is None:
+    if data_contract.servers is None or len(data_contract.servers) == 0:
         raise DataContractException(
             type="lint",
             name="Check that data contract contains valid server configuration",

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/datacontract/engines/soda/check_soda_execute.py

@@ -3,7 +3,7 @@ import uuid
 
 from datacontract.engines.soda.connections.bigquery import to_bigquery_soda_configuration
 from datacontract.engines.soda.connections.databricks import to_databricks_soda_configuration
-from datacontract.engines.soda.connections.duckdb import get_duckdb_connection
+from datacontract.engines.soda.connections.duckdb_connection import get_duckdb_connection
 from datacontract.engines.soda.connections.kafka import create_spark_session, read_kafka_topic
 from datacontract.engines.soda.connections.postgres import to_postgres_soda_configuration
 from datacontract.engines.soda.connections.snowflake import to_snowflake_soda_configuration

datacontract_cli-0.10.22/datacontract/engines/soda/connections/duckdb.py → datacontract_cli-0.10.23/datacontract/engines/soda/connections/duckdb_connection.py

@@ -1,4 +1,5 @@
 import os
+from typing import Any
 
 import duckdb
 
@@ -27,13 +28,13 @@ def get_duckdb_connection(data_contract, server, run: Run):
         run.log_info(f"Creating table {model_name} for {model_path}")
 
         if server.format == "json":
-            format = "auto"
+            json_format = "auto"
             if server.delimiter == "new_line":
-                format = "newline_delimited"
+                json_format = "newline_delimited"
             elif server.delimiter == "array":
-                format = "array"
+                json_format = "array"
             con.sql(f"""
-                        CREATE VIEW "{model_name}" AS SELECT * FROM read_json_auto('{model_path}', format='{format}', hive_partitioning=1);
+                        CREATE VIEW "{model_name}" AS SELECT * FROM read_json_auto('{model_path}', format='{json_format}', hive_partitioning=1);
                         """)
         elif server.format == "parquet":
             con.sql(f"""
@@ -56,7 +57,7 @@ def get_duckdb_connection(data_contract, server, run: Run):
     return con
 
 
-def to_csv_types(model) -> dict:
+def to_csv_types(model) -> dict[Any, str | None] | None:
     if model is None:
         return None
     columns = {}

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/datacontract/export/avro_converter.py

@@ -108,8 +108,8 @@ def to_avro_type(field: Field, field_name: str) -> str | dict:
     elif field.type in ["time"]:
         return "long"
     elif field.type in ["object", "record", "struct"]:
-        if field.config is not None and 'namespace' in field.config:
-            return to_avro_record(field_name ,field.fields ,field.description ,field.config['namespace'])
+        if field.config is not None and "namespace" in field.config:
+            return to_avro_record(field_name, field.fields, field.description, field.config["namespace"])
         return to_avro_record(field_name, field.fields, field.description, None)
     elif field.type in ["binary"]:
         return "bytes"

{datacontract_cli-0.10.22 → datacontract_cli-0.10.23}/datacontract/export/dbt_converter.py

@@ -9,7 +9,7 @@ from datacontract.model.data_contract_specification import DataContractSpecifica
 
 class DbtExporter(Exporter):
     def export(self, data_contract, model, server, sql_server_type, export_args) -> dict:
-        return to_dbt_models_yaml(data_contract)
+        return to_dbt_models_yaml(data_contract, server)
 
 
 class DbtSourceExporter(Exporter):
@@ -27,15 +27,16 @@ class DbtStageExporter(Exporter):
         )
 
 
-def to_dbt_models_yaml(data_contract_spec: DataContractSpecification):
+def to_dbt_models_yaml(data_contract_spec: DataContractSpecification, server: str = None):
     dbt = {
         "version": 2,
         "models": [],
     }
+
     for model_key, model_value in data_contract_spec.models.items():
-        dbt_model = _to_dbt_model(model_key, model_value, data_contract_spec)
+        dbt_model = _to_dbt_model(model_key, model_value, data_contract_spec, adapter_type=server)
         dbt["models"].append(dbt_model)
-    return yaml.dump(dbt, indent=2, sort_keys=False, allow_unicode=True)
+    return yaml.safe_dump(dbt, indent=2, sort_keys=False, allow_unicode=True)
 
 
 def to_dbt_staging_sql(data_contract_spec: DataContractSpecification, model_name: str, model_value: Model) -> str:
@@ -60,7 +61,7 @@ def to_dbt_sources_yaml(data_contract_spec: DataContractSpecification, server: s
     if data_contract_spec.info.owner is not None:
         source["meta"] = {"owner": data_contract_spec.info.owner}
     if data_contract_spec.info.description is not None:
-        source["description"] = data_contract_spec.info.description
+        source["description"] = data_contract_spec.info.description.strip().replace("\n", " ")
     found_server = data_contract_spec.servers.get(server)
     adapter_type = None
     if found_server is not None:
@@ -87,14 +88,16 @@ def _to_dbt_source_table(
     }
 
     if model_value.description is not None:
-        dbt_model["description"] = model_value.description
+        dbt_model["description"] = model_value.description.strip().replace("\n", " ")
     columns = _to_columns(data_contract_spec, model_value.fields, False, adapter_type)
     if columns:
         dbt_model["columns"] = columns
     return dbt_model
 
 
-def _to_dbt_model(model_key, model_value: Model, data_contract_spec: DataContractSpecification) -> dict:
+def _to_dbt_model(
+    model_key, model_value: Model, data_contract_spec: DataContractSpecification, adapter_type: Optional[str]
+) -> dict:
     dbt_model = {
         "name": model_key,
     }
@@ -108,8 +111,8 @@ def _to_dbt_model(model_key, model_value: Model, data_contract_spec: DataContrac
     if _supports_constraints(model_type):
         dbt_model["config"]["contract"] = {"enforced": True}
     if model_value.description is not None:
-        dbt_model["description"] = model_value.description
-    columns = _to_columns(data_contract_spec, model_value.fields, _supports_constraints(model_type), None)
+        dbt_model["description"] = model_value.description.strip().replace("\n", " ")
+    columns = _to_columns(data_contract_spec, model_value.fields, _supports_constraints(model_type), adapter_type)
     if columns:
         dbt_model["columns"] = columns
     return dbt_model
@@ -171,7 +174,7 @@ def _to_column(
             {"dbt_expectations.dbt_expectations.expect_column_values_to_be_of_type": {"column_type": dbt_type}}
         )
     if field.description is not None:
-        column["description"] = field.description
+        column["description"] = field.description.strip().replace("\n", " ")
     if field.required:
         if supports_constraints:
             column.setdefault("constraints", []).append({"type": "not_null"})