The checkbox "Preferences > PCGen > Sources > Allow multi-line objects in sources" should enable the behaviour described in https://pcgenorg.atlassian.net/browse/CODE-1543 .
When a line begins with white-space, it considers it
part of the previous line. It appends the line to the previous line, but
prepended by tabs
I converted all data files to multi-line format, which looks like this:
This causes the parser to die with a series of errors:
I converted the source files to multi-line format using a script: https://github.com/LiaungYip/Morrowind-dnd3.5-PCGen/blob/master/util/lst_to_multiline_2.py . It simply replaces all runs of tabs with newline + tab.
I'm betting the double underscore files (the new ones) were never imbued with the multi-line code.
Excluding files with double-underscores - that is, not re-writing them to the multi-line format - does allow the data set to load.
There is still some strangeness, though. Consider this screenshot of the Feats & Abilities tab...
Apparently this character qualifies for every ability category!
The log file is informative:
Log file is attached.
Looks like it's only the double__underscore files and the ability category files that don't work. All other input files appear to work with multi-line objects.