[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-03-07 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1459080297 Merged. I will try to complete my Local Task Job refactor with it - but feel free to use it to implement the other (simpler) parts of AIP-44 with it (happy to help with reviews). -- T

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-03-07 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1458910076 If there are no more comments, I would love to merge this one (need a committer approval) and then I could follow up withe the example - and then similar approach could be used by other

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-03-07 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1458907186 > Thank you! This looks really cool. > > Just to confirm I understand it correctly: After this PR is merged, we need to change methods using TaskInstance parameter to use `TaskIns

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-03-04 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1454803956 Any more comments @mhenc ? Can we merge this one and proceed with some refactors of existing methods? -- This is an automated message from the Apache Git Service. To respond to the mes

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-02-27 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1447131918 > Quick question. If we take this approach, why do we need then to send the whole serialized object and not the primary keys only? Is it to avoid to refactore too many functions and have

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-02-26 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1445444591 And now I also added all the "references" in DatasetModel with back-references, to show that Pydantic also handles it well -- This is an automated message from the Apache Git Service.

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-02-26 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1445406396 While working on the next step of the POC I also added `task_instance` field to the BaseJobPydantic to show how nicely (and automatically Pydantic takes care about relations - when you u

[GitHub] [airflow] potiuk commented on pull request #29776: Add Pydantic-powered ORM models serialization for internal API.

2023-02-26 Thread via GitHub
potiuk commented on PR #29776: URL: https://github.com/apache/airflow/pull/29776#issuecomment-1445380776 cc: @vincbeck -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To u