Environment Setup¶

This chapter addresses how to set up the shift_left CLI tool to manage Flink project. To install the CLI, which is based on Python, use a virtual environment: venv and pip will work, but we have adopted uv for package and virtual environment management.

Additionally, when using the tool to do migration of existing code to Apache Flink SQL, a Large Language Model is needed see dedicated chapter for instruction.

Pre-requisites¶

• On Windows - enable WSL2. The shift_left tool was developed on Mac and tested on Linux. Windows WSL should work. Powershell use will not work as of now (09/2025).
• All Platforms - install git

• Make is used to encapsulate the confluent cli, to make it easier for Data engineers to deploy Flink statement during development: It is not use by shift_left tool, but shift_left creates the Makefile with the shift_left table init command (see the recipe section).

  • install make for windows
  • Mac OS: brew install make
  • Linux: sudo apt-get install build-essential

• All Platforms - install confluent cli, also used for make execution. This is not mandatory to run shift_left tool.

• If using uv as python package manager (developers of the shift left tool), install it using the documentation.

• All Platforms - Install Python 3.12

• Clone this repository (this will not be necessary once the CLI will be available in pypi):

  git clone  https://github.com/jbcodeforce/shift_left_utils.git
  cd shift_left_utils
  

Set up config.yaml file¶

The configuration file config.yaml is used intensively to tune the shift_left per environment and will be referenced by the environment variables: CONFIG_FILE. You should have different config.yaml for the different kafka cluster, schema registry and Flink environment.

• Copy the config_tmpl.yaml template file to keep some important parameters for the CLI.

  cp src/shift_left/src/shift_left/core/templates/config_tmpl.yaml ./config.yaml
  

• Modify the config.yaml with values from your Confluent Cloud settings. See the tabs below for the different sections of this file:

Kafka sectionConfluent cloudFlinkApp

• Get the Kakfa cluster URL using the Confluent Console > Cluster Settings page. The URL has the cluster id and a second id that is used for the RESP API.

kafka:
  bootstrap.servers: lkc-2qxxxx-pyyyy.us-west-2.aws.confluent.cloud:9092
  cluster_id: lkc-2qxxxx
  pkafka_cluster: lkc-2qxxxx-pyyyy
  security.protocol: SASL_SSL
  sasl.mechanisms: PLAIN
  sasl.username: <key name>
  sasl.password: <key secret> 
  session.timeout.ms: 5000
  cluster_type: stage
  src_topic_prefix: clone

* the sasl.username is the api key name, and the sasl.password is the api secrets. Those are created in the Kafka cluster > API Keys. The key can be downloaded locally.

• Get environment ID from the Environment details in the Confluen Console. The cloud provider and region.

confluent_cloud:
  base_api: api.confluent.cloud/org/v2
  environment_id:  env-20xxxx
  region:  us-west-2
  provider:  aws
  organization_id: 5329.....96
  api_key: T3.....HH
  api_secret: secret-key
  page_size: 100
  glb_name: glb
  url_scope: private

* The organization_id is defined under the user account > Organization settings * the url_scope is set to private or public. When using private link it should be private. * When using a global load balancer, the glb_name to be set. It will be use to define the URL of the REST API. * The API key and secrets are defined per user, and visible under the user account > API keys

• Flink settings are per environment. Get the URL endpoint by going to Environments > one_of_the_env > Flink > Endpoints, copy the private or public endpoint

  flink:
      flink_url: flink-d....7.us-west-2.aws.glb.confluent.cloud
      api_key: MVXI.....HY   
      api_secret: cf.......IztA
      compute_pool_id: lfcp-0...
      catalog_name:  dev-flink-us-west-2
      database_name:  dev-flink-us-west-2
      max_cfu: 50       #-- maximum CFU when creating compute pool via the tool
      max_cfu_percent_before_allocation: .8   #-- percent * 100 to consider before creating a new compute pool 
  

• Define the API keys from Environments > one_of_the_env > Flink > API keys.

• The compute pool id is used as default for running Flink query.
• Catalog name is the name of the environment and database name is the name of the Kafka cluster

The app section defines a set of capabilities to tune the cli.

app:
    delta_max_time_in_min: 15  # this is to apply randomly to the event timestamp to create out of order
    default_PK: __db
    timezone: America/Los_Angeles
    src_table_name_prefix: src_
    logging: INFO
    products: ["p1", "p2", "p3"]
    accepted_common_products: ['common', 'seeds']
    cache_ttl: 120 # in seconds
    sql_content_modifier: shift_left.core.utils.table_worker.ReplaceEnvInSqlContent
    translator_to_flink_sql_agent: shift_left.core.utils.translator_to_flink_sql.DbtTranslatorToFlinkSqlAgent
    dml_naming_convention_modifier: shift_left.core.utils.naming_convention.DmlNameModifier
    compute_pool_naming_convention_modifier: shift_left.core.utils.naming_convention.ComputePoolNameModifier
    data_limit_where_condition : rf"where tenant_id in ( SELECT tenant_id FROM tenant_filter_pipeline WHERE product = {product_name})"
    data_limit_replace_from_reg_ex: r"\s*select\s+\*\s+from\s+final\s*;?"
    data_limit_table_type: source
    data_limit_column_name_to_select_from: tenant_id
    post_fix_unit_test: _ut
    post_fix_integration_test: _it

• post_fix_unit_test, post_fix_integration_test are used to append the given string to table name during unit testing and integration test respectively.
• The data_limit_replace_from_reg_ex, data_limit_table_type, data_limit_column_name_to_select_from are used to add data filtering to all the source tables based on one column name to filter. The regex specifies to file the select * from final which is the last string in most Flink statements using CTEs implementation.
• sql_content_modifier specifies the custom class to use to do some SQL content modification depending of the target environment. This is a way to extend the CLI logic to specific usage.

Environment variables¶

This document explains how to use environment variables to securely manage API keys and secrets instead of storing them in config.yaml files.

The Shift Left utility now supports environment variables for sensitive configuration values. Environment variables take precedence over config.yaml values, allowing you to:

• Keep sensitive data out of configuration files
• Use different credentials for different environments
• Securely manage secrets in CI/CD pipelines
• Follow security best practices

Kafka Section¶

Confluent Cloud Section¶

Flink Section¶

Priority Order¶

The setting will use the following order:

1. Environment Variables (highest priority)
2. Config.yaml values (fallback)
3. Default values set in the code

If an environment variable is set, it will override the corresponding value in config.yaml. If a value is set in config.yaml it will be override the default value.

Setting Environment Variables¶

Bash/Zsh¶

export SL_KAFKA_API_KEY="your-kafka-api-key"
export SL_KAFKA_API_SECRET="your-kafka-api-secret"
export SL_FLINK_API_KEY="your-flink-api-key"
export SL_FLINK_API_SECRET="your-flink-api-secret"
export SL_CONFLUENT_CLOUD_API_KEY="your-confluent-cloud-api-key"
export SL_CONFLUENT_CLOUD_API_SECRET="your-confluent-cloud-api-secret"

Using .env File¶

Create a .env file (don't commit this to version control):

SL_KAFKA_API_KEY=your-kafka-api-key
SL_KAFKA_API_SECRET=your-kafka-api-secret
SL_FLINK_API_KEY=your-flink-api-key
SL_FLINK_API_SECRET=your-flink-api-secret
SL_CONFLUENT_CLOUD_API_KEY=your-confluent-cloud-api-key
SL_CONFLUENT_CLOUD_API_SECRET=your-confluent-cloud-api-secret

Then load it before running your application:

set -a && source .env && set +a

Never commit secrets to version control: Use .gitignore to exclude .env files

Common Issues¶

1. "Missing environment variables" error
2. Check that environment variable names are correct (case-sensitive)
3. Verify that variables are exported in your shell

4. Use `env | grep SL to see what's set

5. To see all supported environment variables, you can call the help function in Python:

   from shift_left.core.utils.app_config import print_env_var_help
   print_env_var_help()
   

Validate the CLI¶

shift_left version
shift_left --help
shift_left project list-topics $PIPELINES

Next>> Recipes section¶