Tools to support the development of rSFSW2

Test Projects

Complete rSFSW2 projects configured to test different functionality. Code is available to execute these test projects and compare against reference databases.

1. Required setup

   • Folder structure of git clones with adequate branches checked out for
     • ./rSFSW2_tools
   • If needed adjust dir.external, i.e., path to external datasets in p1 of test projects

2. Run test projects

   • Non-interactive use

           cd rSFSW2_tools/
           ./Test_projects/Run_all_test_projects.R --help
     

     • Examples
       • Run test project 4 and clean up ./Test_projects/Run_all_test_projects.R -t=4 -D
       • Run test project 5, update reference database and clean up if successful ./Test_projects/Run_all_test_projects.R -t=5 -r -d
       • Clean up ./Test_projects/Run_all_test_projects.R -t=0

   • Interactive use

     • Start R

     •   setwd("rSFSW2_tools/") # Set working directory
         source("Test_projects/Run_all_test_projects.R")
       

     • If the folder 'Test_projects' is not on the current path, then enter on the command prompt

         Enter path to folder of test projects 'Test_projects':
       

     • Provide details on your test request

         Should the test output be used as future reference (y/n):
         Should the test output be force deleted (y/n):
         Should the test output be deleted (y/n):
         Available tests:
             1) Test1_downscaling_overhaul
             2) Test2_LookupWeatherFolders
             3) Test3_OnlyMeanDailyOutput
             4) Test4_AllOverallAggregations_snow
             5) Test5_AllOverallAggregations_mpi
             6) Test6_wgen_downscaling
             ...
         Which of the 6 tests should be run ('all'; a single number; several numbers separated by commas; zero or a negative number to delete any temporary objects):
       

3. Check the outcomes

   • If the output does not compare favorably with the reference database, then a problem report is generated.
   • The results of each test project run are summarized by a logical table with the columns "has_run", "has_problems", "made_new_refs", "deleted_output", and "referenceDB"

Feedback

Please, report issues and offer pull-request