[us-commits] [ehb54/ultrascan3] e5e717: somo: document the SAS Data Utility (buffer/data) ...

[emre brookes]

  Branch: refs/heads/ehb54-issue-1055
  Home:   https://github.com/ehb54/ultrascan3
  Commit: e5e717fb69191f44c96293079fc94662f3eee7ca
      https://github.com/ehb54/ultrascan3/commit/e5e717fb69191f44c96293079fc94662f3eee7ca
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-09-08 (Tue, 08 Sep 2026)

  Changed paths:
    A us_somo/somo/doc/manual/somo/saxs_buffer_nth.html
    M us_somo/somo/doc/manual/somo/somo_saxs_buffer.html
    A us_somo/somo/doc/manual/somo/somo_saxs_buffer_conc.html
    A us_somo/somo/doc/manual/somo/somo_saxs_buffer_conc_load.html

  Log Message:
  -----------
  somo: document the SAS Data Utility (buffer/data) module

The Data Utility, opened from the Data button on the SAS I(q) panel, had no
usable manual coverage: its Help button opened a "work in progress" stub dating
from November 2013, and the Help buttons of its three companion dialogs pointed
at pages that did not exist.

Write the four pages:

  somo_saxs_buffer.html           the module itself
  somo_saxs_buffer_conc.html      Solution Concentrations
  somo_saxs_buffer_conc_load.html File Concentration Loader
  saxs_buffer_nth.html            Select curves

The main page covers loading (recognized extensions, the Conc:/PSV:/I0se:/Units:
header keywords, and the rule that one non-positive standard deviation discards
a file's whole error column), the selection and display tools, averaging and
concentration-normalized averaging, buffer/blank/solution roles, the subtraction
model and both routes to its alpha, error propagation, the non-positive-result
policies, range scanning and multi-file mode, the interactive join, the cropping
family including the Crop Common rounding tolerance, Produced Data and the
output format, and the naming conventions of produced curves.

Controls that are inert are marked as such rather than described as working:
Best buffer subtraction and the Guinier checkbox are empty functions, and the
join's Linear button is created disabled and never enabled.

No screenshots in this pass; the manual carries no images of this module.

Fixes ehb54/ultrascan-tickets#1055

Co-Authored-By: Claude Opus 5 <noreply at anthropic.com>


  Commit: 3b7810e83a098aa64cf61e511181ed1a924503e5
      https://github.com/ehb54/ultrascan3/commit/3b7810e83a098aa64cf61e511181ed1a924503e5
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-09-08 (Tue, 08 Sep 2026)

  Changed paths:
    M us_somo/develop/src/us_hydrodyn_saxs_buffer.cpp
    M us_somo/somo/doc/manual/somo/somo_saxs_buffer.html

  Log Message:
  -----------
  somo: fix the SAS Data Utility join fit range, default Cut and overlap bound

Three defects in the interactive join of the Data Utility module.

1. join_fit_scaling(), join_fit_linear() and replot_join() selected the points
   to compare with

       q >= join_start && q >= join_end

   Both comparisons were >=. Since End is above Start, the condition collapsed
   to q >= End with no upper bound, so the automatic Fit and the RMSD readout
   used the tail beyond End instead of the Start-End window the two markers
   advertise. The scaling factor was therefore fitted over the wrong q interval,
   and the displayed RMSD did not correspond to the selected region. Corrected
   to <= join_end at all eight sites.

2. The default join Cut was computed as ( join_low_q + join_high_q / 2 ), which
   evaluates as join_low_q + (join_high_q / 2). The marker started off-centre
   and, for a wide overlap, could land outside the overlap altogether. Corrected
   to ( join_low_q + join_high_q ) / 2.

3. join_start() took the overlap lower bound from f_qs[ join_file ][ 1 ] where
   every neighbouring expression, including the ternary's own test, uses [ 0 ].
   The bound was one point too high whenever the join curve started at the lower
   q, shifting the Start marker, the wheel ranges and the default Cut. Corrected
   to [ 0 ].

Drop the corresponding "current limitations" note from somo_saxs_buffer.html and
state the Start-End window as what the Fit and the RMSD actually use.

Fixes ehb54/ultrascan-tickets#1056

Co-Authored-By: Claude Opus 5 <noreply at anthropic.com>


Compare: https://github.com/ehb54/ultrascan3/compare/e5e717fb6919%5E...3b7810e83a09

To unsubscribe from these emails, change your notification settings at https://github.com/ehb54/ultrascan3/settings/notifications