Closed gaelforget closed 9 months ago
Thanks @gaelforget . What's the use case for this? I.e., why would another parser be reading Hector's INI files?
Merging #711 (32d85eb) into master (8087ca7) will not change coverage. The diff coverage is
n/a
.:exclamation: Current head 32d85eb differs from pull request most recent head d00c381. Consider uploading reports for the commit d00c381 to get more accurate results
@@ Coverage Diff @@
## master #711 +/- ##
=======================================
Coverage 79.72% 79.72%
=======================================
Files 61 61
Lines 5978 5978
=======================================
Hits 4766 4766
Misses 1212 1212
Thanks @gaelforget . What's the use case for this? I.e., why would another parser be reading Hector's INI files?
Use case is running Hector via a Julia wrapper. I do this in the following notebook that is part of the example set for ClimateModels.jl (the generalized model wrapper).
http://www.gaelforget.net/notebooks/Hector.html
Similar to what's done in pyhector and the R interface I presume, we read in the ini file to display parameters, let user change parameters, save customized parameters to a new ini file, and call Hector from Julia.
@gaelforget, thanks for your interest in contributing to Hector! I guess I am curious as to why this change was only made for the HFC4310_halocarbon
?
Other comments were single line comments or after variable definitions & did not seem to cause any issue -- unlike these few comments placed after section definitions ([HFC4310_halocarbon ]
).
This does not affect our miscellaneous scripts that are not tested as part of the CI (this is a todo #657) good to merge!
Seems safer since fewer parsers will support comments trailing section lines than single line comments.
With at least one parser, IniFile.jl, these comments lead to an incorrect read -- separating out the comments as single lines fixed it.