[
https://issues.apache.org/jira/browse/AIRFLOW-1715?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zelko Nikolic updated AIRFLOW-1715:
-----------------------------------
Description:
In our DAG folder there's a few helper python files and tools. Airflow wrongly
recognizes some of them as DAGs and then reports errors about them, trying to
interpret them as DAGs.
This is completely wrong. There should be a certain signature ID at the top of
the file that distinguishes DAG files from other python files.
Alternatively, there can be a workaround signature that says: "THIS IS NOT A
DAG", so that Airflow can skip that file.
For example a comment that looks like this could be an Airflow directive:
# <AIRFLOW.NOT_DAG>
was:
In our DAG folder there's a few helper python files and tools. Airflow wrongly
recognizes some of them as DAGs and then reports errors about them, trying to
interpret them as DAGs.
This is completely wrong. There should be a certain signature ID at the top of
the file that distinguishes DAG files from other python files.
Alternatively, there can be a workaround signature that says: "THIS IS NOT A
DAG", so that Airflow skips that file. Right now there's no mechanism to tell
Airflow to ignore a certain file.
For example a comment that looks like this could be an Airflow directive:
# <AIRFLOW.NOT_DAG>
> Recognizing non-dag files as DAGs, then reporting non-existing errors
> ---------------------------------------------------------------------
>
> Key: AIRFLOW-1715
> URL: https://issues.apache.org/jira/browse/AIRFLOW-1715
> Project: Apache Airflow
> Issue Type: Bug
> Affects Versions: Airflow 1.8
> Environment: Ubuntu 16.04
> Reporter: Zelko Nikolic
>
> In our DAG folder there's a few helper python files and tools. Airflow
> wrongly recognizes some of them as DAGs and then reports errors about them,
> trying to interpret them as DAGs.
> This is completely wrong. There should be a certain signature ID at the top
> of the file that distinguishes DAG files from other python files.
> Alternatively, there can be a workaround signature that says: "THIS IS NOT A
> DAG", so that Airflow can skip that file.
> For example a comment that looks like this could be an Airflow directive:
> # <AIRFLOW.NOT_DAG>
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
