Custom operator name field dont accept special characters

[aladinoss]

Apache Airflow version: 1.10.2 --> 1.10.11

Kubernetes version (if you are using kubernetes) (use kubectl version):

Environment:

• Cloud provider or hardware configuration: Internal Cloud solution (VM hosting)
• OS (e.g. from /etc/os-release): rhel 7.7
• Kernel (e.g. uname -a): any
• Install tools: pypi
• Others:

What happened:

Cant import operator if name field contains special or caps character

What you expected to happen:

Load the operator even name contains special character

How to reproduce it:
just add special or caps characters to the class field name
example :

import logging

from airflow.models import BaseOperator
from airflow.plugins_manager import Airflowplugin
from airflow.utils.decorators import apply_defaults

log = logging.getLogger(__name__)

class DevMultiplyOperator(BaseOperator):

  @apply_defaults
  def __init__(self, my_operator_param, * args, ** kwargs):
    self.operator_param = my_operator_param
    super(DevMultiplyoperator, self).__init__( * args, ** kwargs)
  
  def execute(self, context):
    try:
       log.info('operator_param: %s', self.operator_param)
       if not isinstance(self.operator_param, int): 
            raise TypeError('Use integer as inputs.')
       return (self.operator_param * 5)
    except Exception as error:
       log.error(error) 
       raise

class DevMultiplyPlugin(AirflowPlugin):
  name = "dev_multiply_plugin"
  operators = [DevMultiplyoperator]

Anything else we need to know:
We are using plugins directory to host operators/hooks/sensor to better organize project.
/dags --> contains dags and subdags
/plugins --> contains operators/hooks/sensor
/config --> contains configuration files
/tests --> contains tests files

[boring-cyborg]

Thanks for opening your first issue here! Be sure to follow the issue template!

[potiuk]

Will take a look at it shortly @aladinoss :)

[ephraimbuddy]

It seems that you are not using super correctly. Can you change the way you are using it?
Change this
super(DevMultiplyoperator, self).__init__( * args, ** kwargs)
to
super().__init__( * args, ** kwargs)

Before initializing your parameters.

import logging

from airflow.models import BaseOperator
from airflow.plugins_manager import Airflowplugin
from airflow.utils.decorators import apply_defaults

log = logging.getLogger(__name__)

class DevMultiplyOperator(BaseOperator):

  @apply_defaults
  def __init__(self, my_operator_param, * args, ** kwargs):
    super().__init__( * args, ** kwargs)
    self.operator_param = my_operator_param
  
  def execute(self, context):
    try:
       log.info('operator_param: %s', self.operator_param)
       if not isinstance(self.operator_param, int): 
            raise TypeError('Use integer as inputs.')
       return (self.operator_param * 5)
    except Exception as error:
       log.error(error) 
       raise

class DevMultiplyPlugin(AirflowPlugin):
  name = "dev_multiply_plugin"
  operators = [DevMultiplyoperator]

I didn't try this but that's how most operators in airflow are designed.

[ashb]

It seems that you are not using super correctly. Can you change the way you are using it?
Change this
super(DevMultiplyoperator, self).__init__( * args, ** kwargs)
to
super().__init__( * args, ** kwargs)

@ephraimbuddy That is Python2 vs Python3 syntax. Airflow code base has switched to the later, but the former will still work on Python3 without trouble, so it's not that.

@aladinoss How were you trying to import this? What error did you receive when you did so?

[aladinoss]

Thanks,
i'll check next week, provide you more details.

[ashb]

Closing due to no response. We can reopen this if we get more details.