Re: 16855 and our secret neighbors

Daniel Berlin <dberlin@dberlin.org> writes:

> 16855 is the change, currently on trunk, and proposed for 1.3.x, which
> stops outputting "END\n" to empty prop files, wasting 4 bytes (which
> ends up wasting a disk block per file on most FSen, and multiplying out
> for gcc, 250 meg of space per checkout)

16855 only helps if the files and directories have no properties, is
gcc going to drop all properties, including svn:eol-style?

> I also believe we should fully document the WC format in 1.4, so people
> reimplementing it aren't hit by our little implementation niggles.
>
> This probably means moving away from hash_read, which has little niggles
> in it (in fact, it's one of those niggles that allows this change to be
> backwards compatible), to hash_read2, which is unforgiving.

I don't object to documentation, but you appear to mean more than
simply documenting it; I think you mean that we should guarantee that
it won't change until we bump the wc format. I don't really like that
idea, aside from the implementation difficulties (what happens when
the code and the documentation don't match?) I don't see what the
Subversion project gains by doing it.

-- 
Philip Martin
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org