Greg Hudson <ghudson_at_MIT.EDU> writes:
> On Mon, 2008-11-17 at 00:46 -0500, Karl Fogel wrote:
>> The depth-first constraint has always been there, and I think it was
>> added mainly by instinct (Jim Blandy's, though many of us felt the same
>> way at the time). It may be that he wanted to be able to make
>> guarantees about the "size" of the edit not exceeding the depth of the
>> tree at any given time or something like that.
>
> There was some discussion about this in September and October of 2000.
> At the time, we were creating the "XML editor", which we knew would
> never survive to see the light of day. The proposed DTD for the editor
> was hierarchical (<open-dir> <open-file> ... <close-dir>) which meant
> that the editor drive had to be depth-first in order for the XML editor
> to be streamy.
>
> If that sounds like a bad reason to restrict the editor API, there were
> people who thought so at the time too! See:
>
> http://svn.haxx.se/dev/archive-2000-10/0038.shtml
>
> I believe Greg Stein was also pushing for a random-access editor
> contract at the time. However, Jim Blandy in particular had other, more
> theoretical reasons for wanting depth-first edits. He articulated them
> in:
>
> http://svn.haxx.se/dev/archive-2000-10/0041.shtml
Thanks for doing the legwork [that I should have done], Greg.
From Jim's mail:
"So I don't think hierarchy really limits our ability to plug
arbitrary consumers and producers together. The case to watch out for
is a consumer that can't easily produce hierarchical changes --- is
mod_dav such a case?"
Well, that question may be answered now :-).
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-11-19 01:07:34 CET