Daniel Shahaf wrote: > The value "all" [...] is not handled at all by check_pristines_mode(). > [...] > Or perhaps this patch is at too early a stage for any of this to > matter :-)
That's right: all those values in the patch are just pencil sketches at this stage. The same answer applies to Johan's observation yesterday that the naming of those option values is currently confusing. The point of sending this patch was just to illustrate the rough shape of passing an option value through from config files to in-memory WC data structures to the code that decides whether to hydrate and/or dehydrate pristines. I'll be re-working it once I've digested the discussion over the last couple of days about how to configure and where to store the configured mode. - Julian
