Dieter Baron wrote:
On 06/01/2008, at 0:50, Roland Illig wrote:
> [con] there's a naming scheme to be learnt
And to be defined. Roland: That part is missing from your proposal;
could you please suggest a naming scheme that
1. does not cause problems on handicaped file systems (e.g. case
insensitive or Windows that disallows `:' in file names
2. maps file names uniquely into patch file names (so the same patch
file won't be reused for different files)
3. is easy to remember and apply?
Up to now, I just took the basename of the file, converted it to lower-case and stripped off all characters that are not in [a-z], because for a long time, these were the only valid patch names.
In the meantime, this validity check seems to have changed, so that patch-* is allowed. (I didn't do that. ;)) This means we can use arbitrary characters from POSIX's Portable Filename Character Set[1].
In that light, I suggest to just take the name of the patched file relative to WRKSRC and replace everything but [A-Za-z-] with a hyphen. (Not an underscore, since Alistair doesn't like them. ;))
This naming scheme may still lead to collisions, but I think they are so rare that we don't have to design much to avoid that.
Item 3 (easy to remember) would best be left to some piece of software that generates the patches in exactly the format we want, so we don't have to care about the exact scheme when using it. Nevertheless, it should still be very easy to recognize.
Roland[1] http://www.opengroup.org/onlinepubs/000095399/basedefs/xbd_chap03.html#tag_03_276