Command Line Interface

The analyst command can be used to validate or execute a script. There are two commands with identical parameters:

• analyst validate: Attempts to parse the script and assemble the DAG, returning any errors
• analyst run: Executes the script
• analyst test: Runs the script, validating the assertions in TEST blocks, returning any failures as errors. All destinations will be mocked.

The parameters are as follows:

• script: The script to evaluate/validate (default: .analyst)
• params: Global options for the script as a JSON object, eg. {"OptName": "OptValue"}.
• v: Verbose (INFO-level events)
• vv: Extra verbose (TRACE-level events)

Full example

analyst validate --script 'myscript.aql' --params "{\"MyOpt\": 1}" --v

Logging

There are four log levels: TRACE, INFO, WARNING and ERROR. Any error condition causes the execution to halt and any managed transactions to be rolled back.

Logging to Slack

You can configure the logger to send log messages above a given level to Slack using an incoming webhook.

Messages will be formatted as <NAME>: SOURCE - LEVEL - MESSAGE.

The options, that you can set using either SET <option_name> = '<option_value>' syntax or via command-line flag params, are:

• SLACK_WEBHOOK_URL: The URL of your webhook.
• SLACK_LOG_LEVEL: Minimum level of messages to log. One of 'TRACE', 'INFO', 'WARNING' or 'ERROR' (case-insensitive).
• SLACK_CHANNEL (optional): Name of Slack channel for the messages.
• SLACK_USER (optional): Name of Slack user for the messages.
• SLACK_EMOJI (optional): Emoji for message.
• SLACK_NAME (optional): Prefix of all messages, so that the script that caused the error can be identified ('' above)