Re: [VOTE] AIP-52 Automatic setup and teardown tasks

2023-01-12 Thread Jed Cunningham
+1 (binding)

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Ash Berlin-Taylor
The description of option 1 is wrong/misleading. It sounds like you are proposing DAG(name="tést") but what actually happens its that gets magically changed to `test` behind the users back. -1 veto to that. So it's Option2 or some combo of the "break MySQL in some way" options. -ash On Jan 12

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Ash Berlin-Taylor
Possibly contentious idea: We allow unicode dag_ids for Postgres, MSQQL (and sqllite) but for Mysql we enforce it as ASCII only. On Jan 12 2023, at 6:15 pm, Jarek Potiuk wrote: > As I mentioned multiple times in similar discussions We have a huge problem > with unicode in dag_id. Namely MySQL

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Daniel Standish
Well, I did acknowledge that others thought about it more than me :) Have not followed the discussions in a while and didn't know where we got. I suspect there are others in the same boat and we're asked to vote on it so it's worth asking and surfacing why that has been ruled out, assuming it

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Jarek Potiuk
As I mentioned multiple times in similar discussions We have a huge problem with unicode in dag_id. Namely MySQL limit on indexes. We would have to shorten the Id significantly in the database to workaround MySQL limits for index size. We can have a wishful thinking that we can change dag_id to

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Ash Berlin-Taylor
+1 to what Daniel said On 12 January 2023 08:32:29 GMT, Daniel Standish wrote: >1 appears to have potential fix: >https://github.com/apache/airflow/issues/21127#issuecomment-1030673862 >2. seems to fail due to our own ascii enforcement... what if we remove that? >3. does not appear to be

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Daniel Standish
1 appears to have potential fix: https://github.com/apache/airflow/issues/21127#issuecomment-1030673862 2. seems to fail due to our own ascii enforcement... what if we remove that? 3. does not appear to be unicode-related or dag_id-related but a feature request for user-friendly mapped task

Re: [VOTE][ISSUE-22073] Finalising approach for displaying non-ascii characters in DAG display name

2023-01-12 Thread Abdul Hadi Shakir
Directly using non-ascii characters (unicodes included) in *dag_id* breaks couples of functionalities. See issues: - Fail to download task log if there are Chinese characters in dag_id #21127 - Airflow scheduler with statsd enabled