4.1. SCR for ARMI 0.6.4

This is a listing of all the Software Change Request (SCR) changes in the ARMI repository, as part of the current release.

Below, this SCR is organized into the individual changes that comprise the net SCR for this release. Each SCR below explicitly lists its impact on ARMI requirements, if any. It is also important to note ARMI and all its requirements are tested entirely by the automated testing that happens during the ARMI build. None of the SCRs below will be allowed to happen if any single test fails, so it can be guaranteed that all SCRs below have fully passed all testing.

4.1.1. SCR Listing

The following lists display all the SCRs in this release of the ARMI framework.

List of SCRs of type: Code Changes, Features

  • PR #2405: Remove ruamel.yaml dependency pin maximum

    • Rationale: The ruamel.yaml project released a new version with updated API and changed dependencies that we need to conform to.

    • Impact on Requirements: NA

    • Author: Arrielle Opotowsky

    • Reviewer: John Stilley

  • PR #2436: Optimizing the HistoryTracker EOL interaction

    • Rationale: The HistoryTracker interface could take several days to run for a large full core model with 100-200 detailed assemblies being tracked.

    • Impact on Requirements: NA

    • Author: Michael Jarrett

    • Reviewer: John Stilley

  • PR #2442: Defaulting the Database to read mode

    • Rationale: Improving defaults to read to streamline the API.

    • Impact on Requirements: R_ARMI_DB_H5 by making the permission default to ready

    • Author: Drew Johnson

    • Reviewer: John Stilley

  • PR #2453: Adding matProps: a material library

    • Rationale: It would be better if materials could be treated more as “data” than as “code” in the ARMI ecosystem.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

List of SCRs of type: Code Changes, Bugs and Fixes

  • PR #2468: Fixing shuffle file reading in reload runs

    • Rationale: The existing implementation will cause ARMI apps to crash during a restart run that loads from the database, because a.p.ringPostHist will be stored as byte strings instead of integers.

    • Impact on Requirements: NA

    • Author: Michael Jarrett

    • Reviewer: John Stilley

  • PR #2470: Removing unused and broken material UThZr

    • Rationale: We are removing this material because it is broken and unused.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

List of SCRs of type: Code Changes, Maintenance, or Trivial

  • PR #2422: Removing unused parameter fluxAdjPeak

    • Rationale: Our stated policy is to remove unused code.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2430: Moving 27 TH parameters out of ARMI

    • Rationale: Ideally, ARMI should not be responsible for maintain features that are not generally useful to a wide audience, and only used by one downstream project.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2451: Moving 7 Neutronics Settings out of ARMI

    • Rationale: Settings that are defined in ARMI, but not used in ARMI can confuse someone looking at the ARMI codebase, and those Settings will be better documented if they live closer to where they are used.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2456: Anonymizing example user names in IPYNBs

    • Rationale: We want to keep our code clear of file paths specific to people’s workstations.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Drew Johnson

  • PR #2457: Doing misc cleanup and temporary pyDOE issue

    • Rationale: For the sake of improving our README file’s clarity.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2458: Removing broken pyDOE dep and LatinHyperCubeSuiteBuilder

    • Rationale: The pyDOE dependency made some breaking changes that cause us time to support, and the only code that uses pyDOE turns out to be unused.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2459: Removing Core.getAssembliesOfType in favor of iterChildrenWithFlags

    • Rationale: The API should be clean and understandable.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer(s): Arrielle Opotowsky, Drew Johnson

  • PR #2460: Stopping intermittent CI failures

    • Rationale: CI should be as stable as we can make it.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer(s): Arrielle Opotowsky, Tony Alberti

  • PR #2461: Renaming _Material_Test to AbstractMaterialTest

    • Rationale: We follow a naming convention because it makes the code easier to read and understand.

    • Impact on Requirements: Trivial implementation changes to the tests: T_ARMI_MAT_FLUID2, T_ARMI_MAT_FRACS0, T_ARMI_MAT_FRACS1, T_ARMI_MAT_FRACS2, T_ARMI_MAT_FRACS3, and T_ARMI_MAT_FRACS4.

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2462: Switching from coveralls.io to codecov.io

    • Rationale: We want ARMI CI to be more stable.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

List of SCRs of type: Documentation-Only Changes

  • PR #2441: Moving the README text up in the PDF ToC

    • Rationale: The PDF version of the docs should look as clean as the HTML version, if at all possible.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky

  • PR #2443: Starting release cycle for ARMI 0.6.4

    • Rationale: Since ARMI 0.6.3 is released, we want to start the 0.6.4 release cycle.

    • Impact on Requirements: NA

    • Author: John Stilley

    • Reviewer: Arrielle Opotowsky