potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1366846230
Yep. Let's keep it as the kind of issue that we can send anyone to if they
stumble upon similar case :).
We can even start counting.
Number of people experiencing the
potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1366520133
> Unfortunately we would technically break backward compatibility if we do
this, but I don’t think there’s a good way around it. Maybe alternatively we
should introduce a config to
potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1366424092
HTML and CSS are AWESOME pair. They never case to Amase me.
TiL (@uranusjr thanks for finding this gem).
Yes, we could canonicalise it. But that would have to be
potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1362765044
Yes. If we limit to only some parameters, that might make sense.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1362758189
Or add template_newline_sequence simply ? I think opening it up for "any"
parameters of Template is kinda dangerous.
--
This is an automated message from the Apache Git Service.
potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1362549266
> Perhaps a better solution would be to allos _per-field_ specification. One
simple way would be to accept passing a `jinja2.Template` to a templated field
directly, e.g.
>
potiuk commented on issue #28280:
URL: https://github.com/apache/airflow/issues/28280#issuecomment-1359916847
Good catch.
It is because this is what Jinja2 does by default - it replaces any new line
it detects in the template (not in the values) with the value of
