Skip to content

DIALS 3.25.2#836

Closed
DiamondLightSource-build-server wants to merge 24 commits intodials-3.25from
dials-next
Closed

DIALS 3.25.2#836
DiamondLightSource-build-server wants to merge 24 commits intodials-3.25from
dials-next

Conversation

@DiamondLightSource-build-server
Copy link
Collaborator

Features

Essentially no changes, just accept the value for DETTYPE. No attempt
is made yet (for Photon-III and -IV) to do anything differently
depending on whether counting or integrating mode was used.
Added format class for treating data from MANDI as Laue data.
dwpaley and others added 19 commits August 11, 2025 11:32
Co-authored-by: Aaron Brewster <asbrewster@lbl.gov>
* `FormatTRPX` for reading images compressed by the TRPX algorithm
)

* Working towards History as an object that can be shared between experiments

* Make History pickleable

* Add History to Experiment as a shareable object

* Function to get unique set of history objects in the experiment list

* Add append_history_item function that controls the format of the history
string, and includes a UTC timestamp

* bug fix

* Add functions to (de)serialize and consolidate history.

At the moment, consolidation is just sorting by timestamp. We could
start combining history items though for cases including parallel
file writes (dials.stills_process)

* Change constructor to require history lines

* Test for history

* tidying

* Add a type annotation and a docstring

* News

* Rename newsfragments/xxx.feature to newsfragments/816.feature

* Bugfix for experiment lists with zero length history

* Tidy up consolidation of histories

This is better as a method of the ExperimentList class, providing
an easier interface to history management. When you want to append a
new history line to each experiment, just call el.consolidate_histories(),
which gives you the now unique History object attached to each experiment
and append the line to that object.

* Missed line in SConscript

* Fix issue when an ExperimentList is saved in an interactive session

* Fix idiotic error in 7f1f770

* Changes based on @phyy-nx's suggestion to:

- correctly get dispatcher names for libtbx builds
- get script name when run as main

---------

Co-authored-by: DiamondLightSource-build-server <DiamondLightSource-build-server@users.noreply.github.com>
Add missing index param for FormatMANDI.get_detector
#841)

Fix typo in FormatMANDILaue when calling generate_laue_data_for_panel.
Add tof_helpers.InstrumentDefinitionReader to read instrument information directly from xml data, and update FormatMANDI to use this by default.
Fix typo in tof_helpers.InstrumentDefinitionReader.
Modify the panel hierarchy of FormatMANDI, FormatISISSXD, and FormatESSNMX to enable individual panel positions to be optimised in refinement.
Preserve an imageset's reject list through pickling. Fixes dials/dials#2998
Support mini CBF images from Pilatus and Eiger detectors at SSRF beamlines BL17U1, BL18U1, and BL19U1, which have inverted rotation axes.
* Change read_uint32 and read_uint32_bs to actually return unsigned
int data rather than trying to cast it into signed int data of the
same width.
* Replace FormatBrukerPhoton, FormatBrukerED1 and FormatBrukerELA with a single
Format class, FormatBrukerModern. Most code was shared between these classes
anyway.
* Rename module file to follow the class name
* Remove now-redundant pathlib=True from dials-data calls
* Update cases where dials_data was asked explicitly for py.path
    towncrier --name=dxtbx --version='3.26.0'
This behaviour has changed in cctbx, however has not been
backported to the release.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants