fail tests if manipulating main.gms with default cfg drops/changes switches

[orichters]

Purpose of this PR

• It was noticed that some switches were not working properly in main.gms, if some comments looked like code assignment, see:
  • manipulateConfig: doesn't handle GAMS unquoted comments that include "=" sign properly pik-piam/lucode2#203
  • manipulateConfig: doesn't handle GAMS comments properly pik-piam/lucode2#121
• This PR fixes two such occurrences and adds a test, making sure it will not happen again
• The tests copies main.gms to a tempfile, reads the default cfg and adjusts the tempfile according to this cfg. Then, it reads the cfg from the tempfile. This should be identical to the default cfg again.

Type of change

• Bug fix
• New feature

Checklist:

• My code follows the coding etiquette
• I performed a self-review of my own code
• I explained my changes within the PR, particularly in hard-to-understand areas
• I checked that the in-code documentation is up-to-date
• I adjusted the reporting in remind2 where it was needed
• I adjusted forbiddenColumnNames in readCheckScenarioConfig.R in case the PR leads to deprecated switches
• All automated model tests pass (FAIL 0 in the output of make test)
• The changelog CHANGELOG.md has been updated correctly

[cchrisgong]

Thank you soo much! :D

[orichters]

This is how it looks if you run this new test without the fixes:

Warning (test_01-manipulateConfig.R:20:7): manipulate config with default configuration does not change main.gms
Applying manipulateConfig with the default configuration leads to this diff between main.gms and main9105c4a453e9f.gms:
<   cm_iteration_max          "number of iterations, if optimization is set to negishi or testOneRegi; is overwritten in Nash mode, except for cm_nash_autoconverge = 0"
>   cm_iteration_max          "number of iterations, if optimization is set to negishi or testOneRegi; is overwritten in Nash mode, except for cm_nash_autoconverge = 1"
< $setGlobal cm_fxIndUe        off  !! def = off
< $setGlobal cm_fxIndUeReg     ""       !! def = ""

Failure (test_01-manipulateConfig.R:24:5): manipulate config with default configuration does not change main.gms
length(diffresult) not equal to 0.
1/1 mismatches
[1] 4 - 0 == 4

Warning (test_01-manipulateConfig.R:30:5): manipulate config with default configuration does not change main.gms
These cfg$gms switches can't be found after manipulation of main.gms, see main9105c4a453e9f.gms.
Please file an issue in the gms package and try to adjust the code until the error goes away:
- cm_fxIndUe
- cm_fxIndUeReg

Failure (test_01-manipulateConfig.R:34:3): manipulate config with default configuration does not change main.gms
length(removedgms) not equal to 0.
1/1 mismatches
[1] 2 - 0 == 2

I shortly discussed with Tonn about the fact that diff might not be available on Windows, but we agreed that we can live with that for now.

[cchrisgong]

Thanks, this warning message would be found in log.txt?

[cchrisgong]

is it also applied to other module declarations or only in core?

[orichters]

• This whole problem should only be relevant for main.gms.
• You find the error if you run make test.