Files
leonarski_fandClaude Opus 5.5 283adb8d98 Live stills: a collection whose pre-pass does not fit the image buffer is refused at start
As the owner decided: no capping. Where min(stills_prepass_images, images)
is more than half of the image buffer's slots, JFJochReceiverService::Start
throws InputParameterInvalid before the stills receiver is built - and so
before the detector is armed (JFJochServices starts the receiver first). A
synchronous /start answers 400, an asynchronous one reports the error through
/wait_till_done, and the broker goes back to Idle. The message gives the
slots, the slot size, N, the image buffer it needs (2 x N x slot, as
image_buffer_MiB) and the stills_prepass_images that would fit.
CheckStillsPrePassFits (StillsGPUAnalysis.h) is shared with
jfjoch_stills_gpu_replay, which exits with the same message.

The capping in JFJochReceiverRugnuxStills::OnStart and its "the pre-pass
runs on the first K images" note are removed. JFJOCH_BROKER.md and the
stills_prepass_images description say the collection is refused (C++ model,
Python docs and frontend types regenerated).

Tests: JFJochReceiverRugnux_Stills_PrePassDoesNotFit (7 of a 12-slot buffer:
refused, InputParameterInvalid, nothing sent); BufferFullDuringPrePass keeps
200 frames through a 12-slot buffer with a pre-pass of 6, all written;
CancelDuringRun takes a pre-pass of 2 for its 4 slots;
GPUStills_GeometryRefinedOnModules and GPUStills_cancel_during_prepass get an
image buffer of 2 x N slots. Replay: -b 30000 at -P 1000 is refused; with
2020 slots per-frame results as before on both windows checked.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HVeDQrPqZ52Tp6rzzZGxf3
2026-11-19 15:36:33 +01:00

3.8 KiB

RugnuxSettings

Rugnux live mode: rotation collections from a DECTRIS detector are processed by rugnux inside the broker while they arrive, and the merged result is sent to the writer. Serial collections, from a DECTRIS detector or converted by the FPGA (JUNGFRAU), are analysed on the GPU after a pre-pass on their first images that decides the beam centre, distance and mask (stills in /rugnux/status), and merged as they go.

Properties

Name Type Description Notes
enable bool Process DECTRIS rotation collections with rugnux live, and serial collections on the GPU after a pre-pass [optional] [default to False]
reserved_cpus int CPU threads left to the receiver and the image pusher while rugnux runs [optional] [default to 4]
deadline_s float Fixed cutoff: this long after the last image the analysis is cut, and the collection ends with what it has established (as `POST /cancel` after the last image does). 0 means no cutoff. [optional] [default to 60.0]
refine_solvent bool With a model: the refinement also builds the solvent - waters added at difference-map peaks and removed where the maps no longer hold them, waters and ions relabelled (as `rugnux --refine-solvent`). Reset to false by `POST /initialize`. [optional] [default to False]
force_still bool Process DECTRIS rotation collections as stills, on the serial path (pre-pass, GPU analysis, stills merge), as `rugnux --force-still` does. The written file keeps its goniometer. Reset to false by `POST /initialize`. [optional] [default to False]
stills_downsample int Serial collections: analyse on average 1 of every this many images, to keep up with a fast stream. Each image is chosen on its own with probability 1 / stills_downsample from a hash of its number - not a stride, which would be in step with periodic patterns (pump-probe on/off, storage cells). The images the pre-pass looks at (the first `stills_prepass_images`) are all analysed. Images not chosen are written as they came, without their analysis. 1: every image. Reset to 1 by `POST /initialize`. [optional] [default to 1]
stills_prepass_images int Serial collections: the first images the pre-pass looks at. They are held in the image buffer, and the stream waits, while the pre-pass measures the mask, polarization and integration settings on them and, where at least 6 of them are strong indexed frames (a unit cell given), refines the beam centre, distance and cell; with fewer, the distance and cell of the dataset settings are used, at the beam centre (the stated one or the scattered background's) that indexes better. Each image waits in a slot of the image buffer of its own, and the pre-pass may take half of the slots: N images need 2N slots in the buffer. Where they do not fit, `/start` refuses a serial collection (with the image buffer size it needs). Rotation collections are not affected. Reset to 1000 by `POST /initialize`. [optional] [default to 1000]

Example

from jfjoch_client.models.rugnux_settings import RugnuxSettings

# TODO update the JSON string below
json = "{}"
# create an instance of RugnuxSettings from a JSON string
rugnux_settings_instance = RugnuxSettings.from_json(json)
# print the JSON string representation of the object
print(RugnuxSettings.to_json())

# convert the object into a dict
rugnux_settings_dict = rugnux_settings_instance.to_dict()
# create an instance of RugnuxSettings from a dict
rugnux_settings_from_dict = RugnuxSettings.from_dict(rugnux_settings_dict)

[Back to Model list] [Back to API list] [Back to README]