SIRF #1357

[Dimitra-Kyriakopoulou]

Changes in this pull request

-- Fixed #1357 by applying the same direction already discussed upstream, not a new design.

This patch follows the __array_priority__ / reverse-operator approach that later appears in #1373. #1358 is
already merged.

-- Changes:

• add __array_priority__ = 10 to ArrayContainer
• add __radd__, __rsub__, and __rtruediv__ to DataContainer
• add regression coverage for the reported left-hand cases in src/common/Utilities.py

-- Checked the reported cases: ay * x, 1 + x, ax + x, ax - x, ay / x, 2 / x
These now return SIRF containers instead of numpy.ndarray or TypeError.

Testing performed

• reran the explicit #1357 repro on the patched branch
• ran src/xSTIR/pSTIR/tests/tests_data_container_algebra.py -v

Related issues

Checklist before requesting a review

• I have performed a self-review of my code
• I have added docstrings/doxygen in line with the guidance in the developer guide
• I have implemented unit tests that cover any new or modified functionality
• The code builds and runs on my machine
• CHANGES.md has been updated with any functionality change

Contribution Notes

Please read and adhere to the contribution guidelines.

Please tick the following:

• [x ] The content of this Pull Request (the Contribution) is intentionally submitted for inclusion in SIRF (the Work) under the terms and conditions of the Apache-2.0 License.

[KrisThielemans]

Thanks @Dimitra-Kyriakopoulou. can you clarify? This is lifting the _array_priority_ to top-level in the SIRF hierarchy?

[Dimitra-Kyriakopoulou]

Dear Professor Thielemans, THANK YOU WHOLEHEARTEDLY FOR YOUR MESSAGE!! Yes, the intent was to lift __array_priority__ to ArrayContainer, i.e. to the common Python base for SIRF array containers, instead of relying on per-subclass behavior. That makes NumPy mixed expressions prefer SIRF container arithmetic more consistently. So it is a lift within the Python array-container branch (not to the top of the whole SIRF hierarchy). Also i attach the pdf which describes my 1st attempt to project 6: i uploaded in

…

________________________________ Από: Kris Thielemans ***@***.***> Στάλθηκε: Τετάρτη, 11 Μαρτίου 2026 12:40 μμ Προς: SyneRBI/SIRF ***@***.***> Κοιν.: Dimitra-Kyriakopoulou ***@***.***>; Mention ***@***.***> Θέμα: Re: [SyneRBI/SIRF] SIRF #1357 (PR #1378) [https://avatars.githubusercontent.com/u/6362141?s=20&v=4]KrisThielemans left a comment (SyneRBI/SIRF#1378)<#1378 (comment)> Thanks @Dimitra-Kyriakopoulou<https://github.com/Dimitra-Kyriakopoulou>. can you clarify? This is lifting the _array_priority_ to top-level in the SIRF hierarchy? — Reply to this email directly, view it on GitHub<#1378 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACZD3XRJC6M6OA5FKWJK4MT4QE7BZAVCNFSM6AAAAACWN7X7B6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DAMZYGI2DEOJUG4>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[KrisThielemans]

ok. @evgueni-ovtchinnikov will review this. thanks!

[evgueni-ovtchinnikov]

so far as I can see, this PR just duplicates #1373 - I suggest we rather merge #1373, and if my understanding is correct just drop this one

[Dimitra-Kyriakopoulou]

Dear Professor @evgueni-ovtchinnikov and Professor @KrisThielemans,
I just checked this and Professor Ovtchinnikov is absolutely right. I had myself explicitly referenced #1373, because I understood it as the upstream direction for fixing #1357 rather than as an already complete implementation of it. Because it was still open, and under the time pressure, I did not realise that #1373 already contained the same functional changes, and in fact, as I now see, also more follow-up work (extra result-type checks and docs/changelog updates). I am deeply sorry for taking your time on this, and I wholeheartedly thank Professor Ovtchinnikov for pointing it out and saving me from unintentionally duplicating existing work -in fact in a way that could have come close to indirect plagiarism.

I AM DEEPLY SORRY,
Dimitra

[KrisThielemans]

no problem. Closing this.