Re: [PATCH v2 8/9] longest_ancestor_length(): resolve symlinks before comparing paths

Michael Haggerty [off-list ref] writes:

I think I would advocate that the prefix has to match the front of
the path exactly (including any trailing slashes) and either

    strlen(prefix) == 0
    or the prefix ended with a '/'
    or the prefix and path are identical
    or the character in path following the matching part is a '/'

This would allow the "is path its own prefix" policy to be decided by
the caller by either including or omitting a trailing slash on the
prefix argument.

I think that is sensible thing to do.

The primary thing I found questionable was that the function, given
"/net/wink/project/frotz" to check against "/pub:/s" (or "/pub/:/s/"
if you like), will report that "/net/wink/project" directory is the
longest ancestor, when "/s" is a symlink that happens to point at
"/net/wink/project".  It is very counter-intuitive when you view its
two input strings as strings.  By making its sole caller expand the
symbolic links, it would be a lot clearer what is going on to
anybody who follow the codepath.  We have one path obtained from
getcwd() and a set of paths all of which are real paths without
symbolic aliasing, and we check if one among the latter cover
an earlier part of the former.