Giovanni Di Siena authored 4 years ago and Gigg, Martyn Anthony committed 4 years ago

In places other substitutions have been made, e.g
Clang does not yet specialize std::shared_ptr for T[]. Vector
has been used instead. The operator[] methods were incorrectly
marked const but returning a non-const reference - this has been fixed.
Refs #25842

Removes all of the from __future__ imports that open and close the
brackets on the same line. This is most of them.

Updates the CMake configuration to set our debugging targets. This means
a user can debug Workbench or System tests, which have a python entry
point, directly from the MSVC IDE.

Also updated class_maker.py

Fixes all flake8 3.7.9 warnings as part of the Python 2->3 transition.

This commit adds a note in the dev docs on c++ coding standards saying to use #pragma once

The instructions were incorrect - there should be no "discard" warning
when re-loading a batch file if no changes were made.

Re #28033

Add a reminder to test it on Workbench as well as Plot

Add ISIS-only note to ensure that mantid is updated on the beamline PCs and IDAaaS when appropriate.

Remove task performed by Jenkins job.
Clarify what to do with open pull requests.

This commit makes changes given from code review,

- Moved Engineering diffraction 2 gui test doc into it's own directory.
- corrected typographical errors.

This commit creates documentation for unscripted tests for the new Engineering Diffraction gui
It also removes the unscripted tests for the previous gui which is no longer supported.

re: #27967

Updates the Sanitizers to handle whitespace anywhere in the selected
sanitizer mode
Fixes an issue where undefined sanitizer would not work (generator
expressions do not handle whitespace)
Updates documentation following some feedback, and drops some old
information (such as min CMake version)

Fix indentation and suppress all Python warnings

Reverts the indentation changes introduced in an earlier commit by auto
formatting tools. Changes the regex for the address suppressions so we
match Python 2 and 3

Updates CXXTest to inject the environment required for sanitizers rather
than forcing the user to do it themselves

Updates the CIFS mounting instructions to specify SMB 3.0, point to the
DFS share rather than a individual server and add some extra flags for
warnings

Adapt the command interface and user file parser to use sans state
objects, instead of the new containers (which duplicated the data
objects).

This is a large PR as we must change all call sites, partially unlink
the reading of run data (the dict converter still takes data info which
needs fixing), and create classes that facade the building process

As the state folder was starting to fill I also took this chance to move
them into their own package, so we have a seperation of data and impl

Clarify Python prerequisite for running system tests job on Windows

Switches to using the JSON library for parsing SANS state objects. This
provides numerous advantages:

- We do not need to maintain a custom (de)serializer
- JSON is a documented standard
- We can switch to type hinting in Python 3, currently a significant
amount of CPU time is spend reverifying typed params
- History becomes less brittle and will work from top level algorithms
- Allows new code to be written in a more Pythonic way (e.g. not forced
to use class level variables for them to be serialized)

To avoid polluting the dev-docs with algorithm-specific info and to
avoid storing/updating many image files the Abins development notes
are moved back into Concepts. The fast broadening method is broken
into a separate document to avoid too much wading through plots.

Move deprecation plans to the top of this document to make them more
visible. The broadening section is a little rambling and will probably
remain so until the theory is laid out more formally.

Images could use a few tweaks (more consistent sizing, remove
unnecessary x values) but I don't want to fill the Git history with
small changes to image files.

Figures are included as image files and the code to generate them is
included as a comment. This doesn't seem ideal as they may be revised
and bloat the repo (or become outdated); is there any workflow to
generate images on the fly when building the docs?

The section on interpolation method is a bit long and should perhaps
be spun out into another document.

Updates the HDF5 URL to the new one. Using browsers or wget would
correctly redirect the request to the new URL, whilst sphinx would get a
HTTP 403 instead.

Updating this URL should fix the sphinx warning, restoring the master
incremental and clean builds to stable (hopefully)

Dan's requested fixes

Co-Authored-By: Harrietbrown <49688535+Harrietbrown@users.noreply.github.com>