> > for clarity, I think this patch should be split up further. > > For example: > > - moving an enum and adding a new entry should be avoided As far as adding a new entry, that could probably be done in a separate patch. The submodule scoping is correct (mostly so config options that come from submodule blobs have a sane value to set the new scope field of git_config_source), but moving it is unavoidable as I'd either have to move config_scope or git_config_source for this patch to work, and moving git_config_source seemed like a ton more work > - does the changes to '/config.c' fix something? Another thing, that in hindsight could probably be split out. The other changes do fix the fact that previously recursive calls to do_git_config_sequence() would blow awway the current_parsing_scope information for higher level callers. This is not super common, but does occur when --show-scope is used with the --blob option. > - exposing config_scope_name should have been done before PATCH 4/6 already If that's better/more convenient, I don't have a problem breaking that out and moving it there. -- Matthew Rogers
