Mantid 3.1.1 Release Notes#
Note
These release notes were migrated from the retired Mantid wiki (archive.mantidproject.org). Trac ticket numbers are given for reference alongside the GitHub issue the ticket was migrated to.
Patch Release |
|
|---|---|
Version |
3.1.1 |
Requested by |
Garrett Granroth |
Approved by |
Nick Draper |
Authorised on |
24 Feb 2014 |
Code Reviewed by |
Russell Taylor |
Code Reviewed on |
24-27 Feb 2014 |
Review Recommendation |
Release |
Released on |
28 Feb 2014 |
Key Features#
This is a patch release to Mantid that corrects some significant issues since the last full release Including instrument definition updates to three instruments and updates to the ISIS SANS and reflectometry interfaces together with a few bug fixes.
Citing this release#
Mantid (2014): Manipulation and Analysis Toolkit for Instrument Data.; Mantid Project.
https://doi.org/10.5286/SOFTWARE/MANTID3.1.1
Changes in this version#
Ticket |
Summary |
Impact |
Solution |
|---|---|---|---|
#9835 (Trac 8992) |
LineCuts units |
Line cuts do not plot correctly (the X unit is wrong) if a multi-dimensional (MD) workspace has been saved and reloaded. |
Propagate the original transformation properly. |
#9880 (Trac 9037) |
Logic error in refl gui |
There was a bug in an angle calculation in the ISIS reflectometry code. |
The bug has been fixed. |
#9888 (Trac 9045) |
TimeSeriesLog autocomplete error |
An unnerving, though harmless, error message appears in the log window when creating an empty TimeSeriesProperty in Python. |
Don’t raise a misleading and unnecessary error. |
#9920 (Trac 9077) |
Refl_gui Improvements - Patch Release Bugfixes |
A few bugs in the ISIS reflectometry GUI are present in v3.1 |
The bugs have been fixed. |
#9749 (Trac 8906) |
Improve handling monitors in SANS |
The monitors were missing when loading event-based data into the ISIS SANS interface. |
Make sure the (separate in this case) monitor workspace is loaded and kept track of. |
#9779 (Trac 8936) |
Uncaught exception on Results Table tab of MuonAnalysis |
In certain rare circumstances it is possible for an error to escape from the Muon Analysis interface and produce an ‘uncaught exception’ error dialog in MantidPlot. |
Introduce more thorough checks on workspace validity in the interface code. |
#9921 (Trac 9078) |
Fitting IkedaCarpenter on workspace in units of DeltaE fails |
Inelastic instruments are likely to experience difficulties in using this function. |
Ensure that the fixed energy value is picked up correctly. |
#9842 (Trac 8999) |
SliceViewer crash |
MantidPlot would crash if attempting to overlay peaks in the slice viewer using a workspace from an ‘instrument’ that is unknown to Mantid. |
Silently swallow the reported error on trying to get information on an instrument we don’t know about. |
#9852 (Trac 9009) |
Fix Units on ConvertMDHistoToMatrixWorkspace |
ConvertMDHistoToMatrixWorkspace is not setting the units correctly. |
Make sure the units are sourced and passed through appropriately. |
#9879 (Trac 9036) |
FileValidator write-access determination issue |
Users could not write to directories where they had permission to write to the directory itself, but did not have write permission on the parent directory. A particular example is the proposal shared directories at the SNS. |
This was just a validator/early warning mechanism. The checking was made less thorough to avoid the bug. |
#9883 (Trac 9040) |
Inter Parameter file Update |
The slit positions have been found to be incorrect. |
Update the instrument definition file. |
#9458 (Trac 8614) |
fix abs units reduction for LET when mono vanadium is measured in event mode |
The absolute reduction for LET at ISIS is not working correctly when all input files are event mode. |
Update the reduction chain to take account of differences with event mode data, in particular in how the monitor data is stored. |
#9794 (Trac 8951) |
Stitch1DMany doesn’t work on GroupWorkspaces |
The algorithm, and thus in certain circumstances the ISIS reflectometry reduction chain, fails if given multi-period data as input. |
Add the capability to process multi-period data (group workspaces) to Stitch1D & Stitch1DMany. |
#9796 (Trac 8953) |
Geometry tab not populated correctly in v3.1 |
The information on the sample geometry is not being displayed for event workspaces in the ISIS SANS interface. |
Correct the loading and retrieval of event monitor data. |
#9813 (Trac 8970) |
Do not delete tmp reduced workspaces |
Cleaning up of temporary workspaces in the ISIS SANS reduction workflow was introduced in v3.1, but users have requested that they be kept. |
Retain temporary sample and can workspaces in a group named ‘sample_can_reductions’ |
#9836 (Trac 8993) |
Update CNCS geometry |
CNCS has a new geometry configuration that is not in the 3.1 release. |
Add an updated instrument definition file. |
#9864 (Trac 9021) |
Remove ghost tubes from BASIS IDF |
The instrument definition contains some tubes that don’t exist in reality! |
Remove them from the definition file. Now they don’t show up in the instrument view. |
#9874 (Trac 9031) |
ORNL SANS: UI tweaks |
The run title is not always fully visible in the catalog view. |
Make the column width adjustable, plus other very minor usability changes (tab reordering). |
#9875 (Trac 9032) |
Get rid of warning message from 8970 |
Ticket 8970 potentially introduced a spurious warning message. |
Suppress the warning message. |
#9904 (Trac 9061) |
Bring new LET_Definition.xml and LET_Parameters.xml to Mantid |
The LET instrument has been updated with respect to the definition that ships with v3.1. |
Add new instrument definition and parameter files. |
#9911 (Trac 9068) |
Fix issue with negative but over zero time range in VSI |
In this case, the starting point is not initially being displayed correctly. |
Ensure that the time control is updated on initial display. |
#9915 (Trac 9072) |
Fix category on PerformIndexOperations |
The algorithm has been categorized incorrectly. |
Correct the category designation. |
#9918 (Trac 9075) |
GetEiT0atSNS fails on unexpected spectra order |
The algorithm was assuming that the required monitors were the first two spectra in the workspace, and was failing if that were not the case |
Be sure to fetch the correct spectra based on the monitor ID. |
Code Review#
Review of Specific Changes#
Ticket |
Summary |
Implementation Notes |
Probability of side effects |
Merged |
|---|---|---|---|---|
#9835 (Trac 8992) |
LineCuts units |
A couple of methods were added to MDGeometry and then used in two MD algorithms and the line plotting class. The new methods replaced existing ones that referred back to the original workspace - which is missing in the case of a reloaded workspace. This work could affect wider areas of Mantid than other tickets in the patch, but was done by the most experienced developer in this area of Mantid. |
Medium |
26 Feb |
#9880 (Trac 9037) |
Logic error in refl gui |
Well-contained small changes to a single python file that backs the ISIS reflectometry GUI |
Low |
25 Feb |
#9888 (Trac 9045) |
TimeSeriesLog autocomplete error |
The getDefault() method of TimeSeriesProperty now returns (correctly) an empty string rather than throwing a NotImplemented exception. |
Low |
25 Feb |
#9920 (Trac 9077) |
Refl_gui Improvements - Patch Release Bugfixes |
Limited changes (about 40 lines) to two files that back up the ISIS reflectometry GUI. |
Low |
25 Feb |
#9749 (Trac 8906) |
Improve handling monitors in SANS |
Changes are small in terms of number of lines are are confined to the python code that handles the ISIS SANS reduction. |
Low |
26 Feb |
#9779 (Trac 8936) |
Uncaught exception on Results Table tab of MuonAnalysis |
Changes are confined to code for a single tab in the Muon interface and are basically introducing more stringent checks on the existence, type and provenance of workspaces. |
Low |
26 Feb |
#9921 (Trac 9078) |
Fitting IkedaCarpenter on workspace in units of DeltaE fails |
Changes to a single, though admittedly quite central method. Changes have been carefully scrutinized and tested. |
Medium |
26 Feb |
#9842 (Trac 8999) |
SliceViewer crash |
This is the addition of one try-catch block to swallow the exception if the instrument is unknown. |
Low |
25 Feb |
#9852 (Trac 9009) |
Fix Units on ConvertMDHistoToMatrixWorkspace |
The actual fix involves about 10 lines in the ConvertMDHistoToMatrixWorkspace algorithm code only. In addition, good test coverage was put in place. |
Low |
25 Feb |
#9879 (Trac 9036) |
FileValidator write-access determination issue |
Validation now fails in fewer circumstances. This could in rare circumstances pass a path that cannot be written, meaning a failure will be caught later (e.g. during execution of an algorithm). Less feedback is sometimes given if a location is found not to be writable, but this is a price worth paying to avoid false positives. |
Medium |
25 Feb |
#9883 (Trac 9040) |
Inter Parameter file Update |
Updates to 2 non-detector positions in the INTER definition file. |
Very low |
24 Feb |
#9458 (Trac 8614) |
fix abs units reduction for LET when mono vanadium is measured in event mode |
Updates are confined to ISIS direct geometry reduction scripts and certain instrument definition/parameter files for these instruments. |
Low |
27 Feb |
#9794 (Trac 8951) |
Stitch1DMany doesn’t work on GroupWorkspaces |
Fairly extensive refactoring to two non-core algorithms. Alongside these the unit tests were beefed up and in addition this area is well covered by system tests. |
Low |
25 Feb |
#9796 (Trac 8953) |
Geometry tab not populated correctly in v3.1 |
Changes in one method of the ISIS SANS GUI code in an area that is informational only and doesn’t affect the reduction itself. |
Very Low |
26 Feb |
#9813 (Trac 8970) |
Do not delete tmp reduced workspaces |
Workspaces that were being deleted are now being kept around, corralled in a group workspace. Code changes are confined to one place in the ISIS SANS reduction script. |
Low |
25 Feb |
#9836 (Trac 8993) |
Update CNCS geometry |
Updates to CNCS Definition XML files only. |
Low |
24 Feb |
#9864 (Trac 9021) |
Remove ghost tubes from BASIS IDF |
Previous attempts to remove these tubes had cause problems to the analysis, but these issues have been resolved so they can be safely removed. |
Low |
25 Feb |
#9874 (Trac 9031) |
ORNL SANS: UI tweaks |
Changes are small and confined to ORNL SANS interface code. |
Low |
24 Feb |
#9875 (Trac 9032) |
Get rid of warning message from 8970 |
Don’t add workspaces to the group if they’re already there in the ISIS SANS reduction script. |
Very Low |
25 Feb |
#9904 (Trac 9061) |
Bring new LET_Definition.xml and LET_Parameters.xml to Mantid |
Addition of new LET Definition and Parameter XML files only. |
Low |
27 Feb |
#9911 (Trac 9068) |
Fix issue with negative but over zero time range in VSI |
Three lines of code to emit the signal to set the displayed value correctly. |
Very Low |
26 Feb |
#9915 (Trac 9072) |
Fix category on PerformIndexOperations |
A change in the category method of a single algorithm. |
Extremely Low |
24 Feb |
#9918 (Trac 9075) |
GetEiT0atSNS fails on unexpected spectra order |
This was caught by spurious monitors appearing in SEQUOIA data. The parameters that specify the monitor IDs to use are not looked at instead of making an assumption. Changes are confined to the named python algorithm. |
Low |
25 Feb |
Are any changes included in the release that are not accounted for above?#
In principle the changes to the BASIS IDF from #9811 (Trac 8967) are also included, though these are superseded by those in 9021.
Is the collective effect of the changes acceptably low risk?#
Yes
Code Reviewer Recommendation#
Release once kits have undergone some unscripted testing - Complete.