This project is mirrored from Pull mirroring updated .
  1. 22 Sep, 2021 2 commits
  2. 27 Mar, 2020 1 commit
  3. 20 Mar, 2020 1 commit
  4. 21 Feb, 2020 14 commits
    • Gemma Guest's avatar
      Remove hack to add workspaces to ADS from GUI · 09774742
      Gemma Guest authored
      This is no longer required now the workspaces are output correctly from
      the algorithms.
      Re #27633
    • Gemma Guest's avatar
      Fix problems with output property lists · b9277f06
      Gemma Guest authored
      When assigning to a list of outputs when calling the algorithm in
      python, the outputs must be specified in the order they are declared,
      even if they are optional outputs like IvsLam. This means that if we
      want to specify the transmission workspace output we must first specify
      IvsLam even if debug is false. Some of the algorithms always output
      IvsLam if the name was given whereas others did not. They now always
      output it if the name is given.
      Note that this means you may now get an error if you don't supply enough
      arguments whereas previously it was ok. It is ok if you just specify the
      IvsQ output(s) but beyond that we do not know whether the output should
      be IvsLam or the transmission, so we require all of them, hence the
      error. This is better than it assigning an unexpected output so I think
      is ok, and is the only way I can think to do this.
      Re #27633
    • Gemma Guest's avatar
      Simplify output properties and grouping · 91a32f89
      Gemma Guest authored
      The new code to declare properties on the fly was causing too many
      problems so I've removed it. I've added proper output properties for the
      transmission outputs from the child algorithms. The other output
      properties I've removed. It means these workspaces will not have history
      but this is low priority because they are the input TOF workspaces so
      the history is not critical. It would be nice to have it for the
      summed/sliced workspaces, but this can always be extracted from the main
      I've also removed the grouping of additional TOF workspaces, so
      behaviour is now reverted to the original where only the loaded TOF
      workspaces are grouped.
      Re #27633
    • Gemma Guest's avatar
      Only declare debug outputs when flag is set · 9d4d731a
      Gemma Guest authored
      Previously debugs were always being output if run as a child algorithm.
      Also throw if debug is true but a name was not provided/found for the
      output debug workspaces.
      Update reflectometry tests with change to transmission outputs.
      Re #27633
    • Gemma Guest's avatar
      Minor refactoring and renaming for clarity · 3420ca07
      Gemma Guest authored
      Re #27633
    • Gemma Guest's avatar
      Apply flake8 formatting fixes · cf317405
      Gemma Guest authored
      Re #27633
    • Gemma Guest's avatar
      Output TOF workspaces individually rather than by group · 33183c4d
      Gemma Guest authored
      Outputting the group is fine if the group does not already exist, but if
      the group exists then I had trouble outputting only the new members of
      the group, and it's inconsistent which outputs you get depending on the
      state before the algorithm was run. I therefore think it's simpler to
      output workspaces when they are created. If the group does not already
      exist in the ADS then it needs to be added and I've had to use the ADS
      directly for this because we can't set the group as an output property
      if its members are already output or their history will be duplicated.
      Re #27633
    • Gemma Guest's avatar
      Fix a bug where some workspaces were added to a group twice · e14bf49d
      Gemma Guest authored
      Use a set rather than a list to ensure we don't try to group the same
      workspace twice. This was happening because of the way the transmission
      workspaces are returned in different ways depending on whether stitching
      is done or not.
      Re #27633
    • Gemma Guest's avatar
      Fix duplicate history being added if TOF group exists · cf0a656e
      Gemma Guest authored
      If the TOF group exists we cannot add it as an output property because
      the existing workspaces' history will be duplicated. Therefore, add each
      workspace as an output property individually.
      Re #27633
    • Gemma Guest's avatar
      Create TRANS_TOF workspace group · 973e8092
      Gemma Guest authored
      Put TOF transmission workspaces into a group similar to the TOF group
      and add this as an output property. This is easier than setting separate
      output properties for the individual and summed input transmission
      Re #27633
    • Gemma Guest's avatar
      Create output properties for interim transmission workspaces · 61c1ab2b
      Gemma Guest authored
      This commit adds output properties for the interim lambda workspaces for
      the first/second transmission run in addition to the main output
      transmission workspace which already has a property.
      Re #27633
    • Gemma Guest's avatar
      Fix workspace history for transmission workspace · d71b9b75
      Gemma Guest authored
      This uses the main transmission workspace output from RROA and makes it
      an output of RILAP.
      Re #27633
    • Gemma Guest's avatar
      Fix workspace history for sliced TOF group · 1b378976
      Gemma Guest authored
      Re #27633
    • Gemma Guest's avatar
      Fix workspace history for TOF group · 16466063
      Gemma Guest authored
      Declare an on-the-fly property to output the group so that it's history
      is included. Previously it was being hacked directly into the ADS.
      Re #27633
  5. 18 Feb, 2020 2 commits
  6. 19 Dec, 2019 1 commit
    • Gemma Guest's avatar
      Add option to specify which filter algorithm to use · 8aa2e000
      Gemma Guest authored
      There are differences between the old FilterByTime algorithm and the new
      FilterEvents algorithm. This commit adds an option so the user can
      specify which to use and sets the default to use the old one.
      Also, when combining the monitors and detectors into a single workspace, we
      were merging the logs, which meant the proton charge is different to
      previous behaviour when scientists were doing this step manually. This
      commit changes MergeLogs to be false in line with the way scientists
      previously did this.
      I'm not sure this is really an issue because the final proton charge for
      the workspace is taken from the monitor workspace, which contains the
      unfiltered proton charge. We may want to adjust this in future to be the
      proton charge for the filtered slice.
      Re #25881
  7. 13 Sep, 2019 1 commit
  8. 12 Sep, 2019 1 commit
  9. 09 Sep, 2019 2 commits
  10. 22 Aug, 2019 2 commits
  11. 21 Aug, 2019 2 commits
  12. 23 Jul, 2019 1 commit
  13. 05 Jul, 2019 1 commit
  14. 04 Jul, 2019 1 commit
    • Gemma Guest's avatar
      Use new workflow algorithm for live data monitor · c8ac651d
      Gemma Guest authored
      - Replace direct use of ReflectometryReductionOneAuto with the new wrapper ReflectometryISISLoadAndProcess. This fixes a problem with invalid properties being passed from the GUI, which also uses the latter. It also gives it more comprehensive support of the new GUI features such as being able to sum input transmission runs.
      - Tidy up property lists in the algorithms.
      - Fix an issue where a clone was being done of the input live workspace which was only updated with the output values when the first chunk of data arrives. This was confusing users. The output workspace now only appears when the data arrives.
      - Make grouping of TOF workspaces optional so that we can disable it for live data.
      - Make reloading of input workspaces optional so that we can disable it for live data - if for some reason the workspace is not there or is the wrong type we would get a confusing error if we try to reload it from file because there is no file for live data.
      Re #26119
  15. 01 Jul, 2019 1 commit
  16. 21 Jun, 2019 1 commit
  17. 12 Jun, 2019 1 commit
  18. 22 Feb, 2019 2 commits
    • Gemma Guest's avatar
      Fix batch presenter unit tests · aef035b8
      Gemma Guest authored
      Due to recent changes, the batch presenter will now only perform processing if the job runner returns a non-empty list of algorithms, so make sure that the mock returns something.
      Add a new unit test for the case where the next batch is started when there are still algorithms left to run after a previous batch has completed.
      Re #24804
    • Gemma Guest's avatar
      Consolidate the functions to check for invalid workspaces · 5bc58299
      Gemma Guest authored
      Re #24603
  19. 21 Feb, 2019 3 commits