CS connection refurbishment by JeanLucPons · Pull Request #88 · python-accelerator-middle-layer/pyaml

JeanLucPons · 2025-12-04T10:34:13Z

The aim of this PR is:

Allow multiple control system to be used.
Remove the int_cs() call.
Provides a better symmetry between lattices and control systems.

It shows minimum modifications to handle the above and change the way to handle external magnet model and CS backend. This new branch is not compatible with the present Tango backend.
If this PR is accepted then I'll port the Tango backend.

If we decide to limit pyAML to a single control system per Accelerator then this PR is not mandatry and we can change the controls field to control and fix the name to live.

control:
    type: tango.pyaml.controlsystem
    tango_host: ebs-simu-3:10000

…rface.

…f pySC ResponseData type)

…ack to default configuration when not specified

…e-arrays Update arrays documentation

TeresiaOlsson · 2025-12-10T13:39:34Z

About the part:

"If we decide to limit pyAML to a single control system per Accelerator then this PR is not mandatry and we can change the controls field to control and fix the name to live."

Wouldn't that mean that you no longer can configure the virtual accelerator? For me that would be a second controls with a different host/PV prefix?

JeanLucPons · 2025-12-10T13:58:31Z

Wouldn't that mean that you no longer can configure the virtual accelerator? For me that would be a second controls with a different host/PV prefix?

If we limit to one single CS per accelerator, it will be possible to load a second accelerator. So instead of writing:

sr.live...
sr.shadow...

You will write:

sr.live...
shadow.live...

…rbit-correction-and-orm Orbit correction and Orbit Response Matrix measurement through pySC.

TeresiaOlsson · 2025-12-10T17:01:28Z

Wouldn't that mean that you no longer can configure the virtual accelerator? For me that would be a second controls with a different host/PV prefix?

If we limit to one single CS per accelerator, it will be possible to load a second accelerator. So instead of writing:
sr.live...
sr.shadow...
You will write:
sr.live...
shadow.live...

Hmm... I think that will be confusing and potentially unsafe since then it's no longer given that live always will be the live machine. We have that setup in MML at the moment if we run it against the twin because then MML requires it to be run in online mode. I don't like it because then you always feel nervous about if online really means twin or you by accident loaded the real machine.

Also, what would that mean for the simulators? There would be one set of simulator objects for the live and another one for the shadow? Feels like that also potentially could cause problems.

…init-hook Added post_init() hook

…n-accelerator-middle-layer/pyaml into cs-refurbishment-multiple-csprefix

JeanLucPons · 2025-12-11T15:58:22Z

It is now possible to write and run it without any backend installed:

sr = Accelerator.load("tests/config/EBSOrbit.yaml",ignore_external=True)
print(sr.design.get_magnets("VCorr").strengths.get())

        ignore_external: bool
            Ignore external modules and return None for object that
            cannot be created. pydantic schema that support that an
            object is not created should handle None fields.

@TeresiaOlsson , @gubaidulinvadim is it ok for you ?

Array of None are necessary

TeresiaOlsson · 2025-12-11T16:30:57Z

It's okay for me. I'm also working on something for the factory for issue #28 which might remove the need for this but I'm not sure yet and it will likely take a while until I manage to finish my suggestion...

JeanLucPons · 2025-12-11T17:44:09Z

It would be nice that this PR is rapidly merged as it has a strong probability of merge conflict.
I will modify the Tango and EPICS backend once it is merged.

TeresiaOlsson · 2025-12-11T18:03:47Z

It would be nice that this PR is rapidly merged as it has a strong probability of merge conflict. I will modify the Tango and EPICS backend once it is merged.

I agree since what you are doing in the load in the Accelerator looks similar to what I'm trying to do. I will try to go through the whole thing tomorrow.

gubaidulinvadim

Approved. Strangely, GitHub shows to me in the diff that tuning_tools/dispersion and some others is added in this PR when it should already be on the main branch.

The merge-base changed after approval.

gubaidulinvadim · 2025-12-11T22:12:36Z

We should also add a minor version bump as it will break compatibility with tango/ophyd-async bindings

kparasch and others added 25 commits November 25, 2025 18:01

first draft of pySC interface

9d19a41

draft for example to measure ORM. For now it is just testing the inte…

33e5a46

…rface.

add pySC as submodule

51237f1

example for measurement of ideal response matrix

47b5661

update to new commit of pySC (fix of a logging bug)

a5e0e44

add example for orbit correction

55cd9ef

object for ORM

c539c2f

pySC updates, no need for rich now

dc983fa

save function in orm and a bugfix in input/output names

d15b0b0

update example by using OrbitResponseMatrix object

da78bf8

update ideal orm data

ea5cc95

remove very large file

396490a

Remove element_holder from OrbitResponseMatrix's ConfigModel

b5a2049

OrbitResponseMatrix.latest_measurement is now a dictionary (instead o…

eb85763

…f pySC ResponseData type)

fix of printing a 'Measured response of None' at initialization

42d93a5

save with different types, to discuss later again on saving data

c0881e2

do relative import pyaml.external.pySC

65a9817

Allow to specify correctors for which to measure response, but fall b…

4b473d1

…ack to default configuration when not specified

use parenthesis to split line isntead

a8bc34d

linting after rebasing

f000ad8

fixed formatting

4010cd1

Correction per plane

2f0fede

comment out some prints

a60fcb3

dispersion measurement

e4e009f

Added connect() function, removed init_cs()

b03d720

JeanLucPons marked this pull request as draft December 4, 2025 10:34

JeanLucPons changed the title ~~CS connection refurbishment~~ WIP: CS connection refurbishment Dec 4, 2025

JeanLucPons mentioned this pull request Dec 4, 2025

Add option to disable the control system when loading a PyAML file #26

Closed

Renamed connect() to attach()

24f57ee

gubaidulinvadim linked an issue Dec 4, 2025 that may be closed by this pull request

Add option to disable the control system when loading a PyAML file #26

Closed

JeanLucPons and others added 2 commits December 10, 2025 13:29

Update arrays documentation

59bc88a

Merge pull request #91 from python-accelerator-middle-layer/doc-updat…

af44df7

…e-arrays Update arrays documentation

Merge pull request #82 from python-accelerator-middle-layer/feature/o…

a0c5ca4

…rbit-correction-and-orm Orbit correction and Orbit Response Matrix measurement through pySC.

JeanLucPons added 12 commits December 11, 2025 10:00

Added post_init() hook

eddc12c

Fix formatting

010aa00

Merge pull request #93 from python-accelerator-middle-layer/add-post-…

2313b02

…init-hook Added post_init() hook

Added connect() function, removed init_cs()

8d50e66

Renamed connect() to attach()

c23c274

Management of attached devices per CS instance

e8e09b4

Fix for flake

addbe5d

Fix for flake

585c355

Merge branch 'cs-refurbishment-multiple-csprefix' of github.com:pytho…

e68c287

…n-accelerator-middle-layer/pyaml into cs-refurbishment-multiple-csprefix

Added ignore_external flag to accelerator

f7c87f2

Fix unit test

c9a017b

Fix unit test

ecc3d48

JeanLucPons marked this pull request as ready for review December 11, 2025 16:19

JeanLucPons requested a review from gubaidulinvadim December 11, 2025 16:19

JeanLucPons requested a review from kparasch December 11, 2025 17:42

gubaidulinvadim previously approved these changes Dec 11, 2025

View reviewed changes

JeanLucPons closed this Dec 12, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

CS connection refurbishment#88

CS connection refurbishment#88
JeanLucPons wants to merge 67 commits intomainfrom
cs-refurbishment-multiple-csprefix

JeanLucPons commented Dec 4, 2025 •

edited

Loading

Uh oh!

TeresiaOlsson commented Dec 10, 2025

Uh oh!

JeanLucPons commented Dec 10, 2025

Uh oh!

TeresiaOlsson commented Dec 10, 2025

Uh oh!

JeanLucPons commented Dec 11, 2025 •

edited

Loading

Uh oh!

TeresiaOlsson commented Dec 11, 2025

Uh oh!

JeanLucPons commented Dec 11, 2025 •

edited

Loading

Uh oh!

TeresiaOlsson commented Dec 11, 2025 •

edited

Loading

Uh oh!

gubaidulinvadim left a comment

Uh oh!

gubaidulinvadim commented Dec 11, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

JeanLucPons commented Dec 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TeresiaOlsson commented Dec 10, 2025

Uh oh!

JeanLucPons commented Dec 10, 2025

Uh oh!

TeresiaOlsson commented Dec 10, 2025

Uh oh!

JeanLucPons commented Dec 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TeresiaOlsson commented Dec 11, 2025

Uh oh!

JeanLucPons commented Dec 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TeresiaOlsson commented Dec 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

gubaidulinvadim left a comment

Choose a reason for hiding this comment

Uh oh!

gubaidulinvadim commented Dec 11, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

JeanLucPons commented Dec 4, 2025 •

edited

Loading

JeanLucPons commented Dec 11, 2025 •

edited

Loading

JeanLucPons commented Dec 11, 2025 •

edited

Loading

TeresiaOlsson commented Dec 11, 2025 •

edited

Loading