2014-10-06 14:45 GMT+02:00 Eric Blake <[email protected]>: > This feels like feature bloat. What's wrong with using 'find' to locate > the subset of files you are interested in, then running diff on those files?
It's a bit messy to work around with find: find a/ b/ -type f -printf "%P\0" | sort -zu | xargs -0 -n1 -i% diff -Nu a/% b/% Daniel's proposed feature makes this easier and more efficient (as diff won't be invoked for each file), so it looks good to me. Thanks, Andreas
