Re: WC-NG: Perhaps it's time to write a new summary of where we stand at the moment?

[Greg Stein <gs...@gmail.com>]

On Wed, May 12, 2010 at 09:33, Daniel Näslund <da...@longitudo.com> wrote:
> Hi!
>
> Greg [1] identified the following items in the beginning of February.
>
> * remove entry_t usage within libsvn_wc/client

This is very close in the client, but there is still a good chunk of
work in libsvn_wc.

> * remove access_t usage withing libsvn_wc/client

This looks to be done. libsvn_client has no more uses/references.
There are still some references in libsvn_wc, but that appears to be
for backwards compatibility only.

Hyrum was doing some work to clarify access_t uses in libsvn_wc. Not
sure where he is with that.

> * move props into wc.db

Should happen within a week.

> * move to single/root wc.db

No plans yet. We probably need to remove entry_t uses before tackling this.

> * switch to new pristine file mgmt

Julian is working on this. I believe it is getting close for new
working copies, but we do not have a migration/upgrade story yet.

> I have the feeling, we've made some progress. Are there any new
> identifiable items for wc-ng completion?

We have work items which perform database work inside *and* perform
file operations. These need to be turned inside-out to perform the db
operation and queue file updates within one sqlite txn.

There are a number of places where we perform several db operations
outside of a sqlite txn, which means failure could leave us in a
corrupted state. See svn_wc_add_repos_file4 for an example (note:
other than that, the function is now at our wc-ng "ideal" design).

> The WC-NG meta issue [2] contains only one dependency, #3586. Perhaps we
> could add some more to make it easier to track the progress?

Meh. I don't use the tracker as a to-do list. The set of work is too
fluid for something as rigid as that.

> Cheers,
> Daniel
>
> (Hope I'm not too forward asking these questions)

It's always fine!

> [1] http://svn.haxx.se/dev/archive-2010-02/0286.shtml
> [2] http://subversion.tigris.org/issues/show_bug.cgi?id=3357

Cheers,
-g

Re: WC-NG: Perhaps it's time to write a new summary of where we stand at the moment?

["Hyrum K. Wright" <hy...@mail.utexas.edu>]

On Wed, May 12, 2010 at 12:14 PM, Greg Stein <gs...@gmail.com> wrote:

> On Wed, May 12, 2010 at 09:33, Daniel Näslund <da...@longitudo.com>
> wrote:
> > Hi!
> >
> > Greg [1] identified the following items in the beginning of February.
> >
> > * remove entry_t usage within libsvn_wc/client
>
> This is very close in the client, but there is still a good chunk of
> work in libsvn_wc.
>
> > * remove access_t usage withing libsvn_wc/client
>
> This looks to be done. libsvn_client has no more uses/references.
> There are still some references in libsvn_wc, but that appears to be
> for backwards compatibility only.
>
> Hyrum was doing some work to clarify access_t uses in libsvn_wc. Not
> sure where he is with that.
>

I started working in it, then spent a week at a conference, and then had my
laptop motherboard get toasted.  I'm hoping to have the hardware back by
week's end, and then dive back in next week.

-Hyrum