Stefan Fuhrmann wrote: > Stefan Fuhrmann wrote: >> Julian Foad wrote: >>> Stefan Fuhrmann wrote: >>>> Julian Foad wrote: >>>>> -- a quick optimization API -- the definitely/maybe question -- like the >>>>> existing implementations but documented properly and named >>>>>appropriately; and >>>>> >>>>> -- (perhaps) a definite content comparison API: "has the content >>>>>changed?" >>>> >>>> I tend to prefer an extra flag parameter in a bumped version [...] >>> >>> Well, this is just a general API design style issue, but I think [...] >> >> Fair point. I was thinking about revving svn_ra_get_file_revs2 >> as well with the same new option (callers can chose whether >> need exact reports or want to process the delta anyway). >> That would be a pass-through value for svn_fs_*_changed. > > I implemented just that in my working copy and it turned out > to be a mess. Every RA layer had to be updated and in the > end, there would only be the blame function to use it - with > some fixed value for the STRICT option. > > So, I decided to go with the separate API functions for the > different behaviors in r1573111.
Looks good. Thanks. - Julian
