Add python 3.14 as classifier to project metadata

[agoscinski]

After a discussion in #7240 we decided to add a classifier.

EDIT: You are right, I should at least run the our production test suite on py3.14 before merging.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 29.98%. Comparing base (1607da4) to head (1f4d0ee).
⚠️ Report is 227 commits behind head on main.

❗ There is a different number of reports uploaded between BASE (1607da4) and HEAD (1f4d0ee). Click for more details.

HEAD has 3 uploads less than BASE

Flag	BASE (1607da4)	HEAD (1f4d0ee)
	4	1

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #7263       +/-   ##
===========================================
- Coverage   79.75%   29.98%   -49.77%     
===========================================
  Files         566      566               
  Lines       43895    43849       -46     
===========================================
- Hits        35006    13144    -21862     
- Misses       8889    30705    +21816     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[danielhollas]

Would be great to run the production test suite on python 3.14

Btw @GeigerJ2 this one would be good to merge before release

[agoscinski]

Benchmark results show significant slow downs. See 2.8.0rc1-py3.14 and 2.8.0rc1-py3.12. Need to investigate it before we merge it. For now we decided to mention Python 3.14 support only in the CHANGELOG, because there is no Beta marker for python support.

[danielhollas]

Not sure if useful, but was curious and ran the performance_benchmark_base.py and I don't see a major difference between Python versions

Using presto profile with sqlite and without broker

(I omitted some lines in the output)

Python 3.12

Python 3.12.7 (main, Oct  1 2024, 00:00:00) [GCC 13.3.1 20240913 (Red Hat 13.3.1-3)]
AiiDA 2.7.3.post0
Running on profile presto
Running 100 calculations.
Success: All calculations finished successfully.
Elapsed time: 51.72 seconds.

Performance: 0.52 s / process

Python 3.14

Running on profile presto
Running 100 calculations.  [####################################]  100%          
Success: All calculations finished successfully.
Elapsed time: 46.99 seconds.

Performance: 0.47 s / process

Profile with rabbitmq 3.13 and Postgresql-14

Python 3.12.13

Python 3.12.13 (main, Mar  3 2026, 14:59:34) [Clang 21.1.4 ]
AiiDA 2.7.3.post0
Running on profile presto-rmq-psql
Submitting 100 calculations.  [####################################]  100%          
Submission completed in 8.82 seconds.
Waiting for calculations to complete  [####################################]  100%          
Success: All calculations finished successfully.
Elapsed time: 54.81 seconds.

Performance: 0.55 s / process

Python 3.14.3

Python 3.14.3 (main, Feb 12 2026, 00:42:54) [Clang 21.1.4 ]
AiiDA 2.7.3.post0
Running on profile presto-rmq-psql
Success: Created and configured temporary `Computer` benchmark-30fe4a8e for localhost.
Success: Created temporary `Code` bash for localhost.
Submitting 100 calculations.
Submission completed in 6.80 seconds.
Success: All calculations finished successfully.
Elapsed time: 44.21 seconds.

Performance: 0.44 s / process

[agoscinski]

I just merge it. We tested it now for ZMQ good enough. The performance issue needs to be investigated but should not be a blocker for a classifier.