Skip to main content

About debug macro

Requires Core CLI

The debug() macro is only available when using dbt Core CLI in a local development environment. It's not available in dbt platform.

Do not deploy code to production that uses the debug macro.

If developing in dbt platform or using Fusion, you can instead use:

  • {{ print() }} - Print messages to both the log file and standard output (stdout).
  • {{ log() }} - Structured logging that prints messages during Jinja rendering.

The {{ debug() }} macro will open an iPython debugger in the context of a compiled dbt macro. The DBT_MACRO_DEBUGGING environment variable must be set to use the debugger.

This function requires:

  • Interactive terminal access with iPython debugger (ipdb) installed. Fusion doesn't provide a iPython (ipdb) debugger since its built on Rust. It instead outputs a non-interactive snapshot of the MiniJinja render context in the compiled code.
  • Local development environment running dbt Core CLI
  • DBT_MACRO_DEBUGGING environment variable set

Usage​

my_macro.sql

{% macro my_macro() %}

  {% set something_complex = my_complicated_macro() %}
  
  {{ debug() }}

{% endmacro %}

When dbt hits the debug() line, you'll see something like:

$ DBT_MACRO_DEBUGGING=write dbt compile
Running with dbt=1.0
> /var/folders/31/mrzqbbtd3rn4hmgbhrtkfyxm0000gn/T/dbt-macro-compiled-cxvhhgu7.py(14)root()
     13         environment.call(context, (undefined(name='debug') if l_0_debug is missing else l_0_debug)),
---> 14         environment.call(context, (undefined(name='source') if l_0_source is missing else l_0_source), 'src', 'seedtable'),
     15     )

ipdb> l 9,12
      9     l_0_debug = resolve('debug')
     10     l_0_source = resolve('source')
     11     pass
     12     yield '%s\nselect * from %s' % (

Was this page helpful?

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

0
Loading