Ivan Zhakov wrote on Thu, Jun 30, 2011 at 23:22:28 +0400:
> On Thu, Jun 30, 2011 at 22:22, Mark Phippard <markphip_at_gmail.com> wrote:
> > On Thu, Jun 30, 2011 at 2:21 PM, <kmradke_at_rockwellcollins.com> wrote:
> >> Mark Phippard <markphip_at_gmail.com> wrote on 06/30/2011 01:08:52 PM:
> >>> On Thu, Jun 30, 2011 at 2:06 PM, Daniel Shahaf <d.s_at_daniel.shahaf.name>
> >>> wrote:
> >>> > You can't currently pack revision shards unless you also pack revprop
> >>> > shards. (but that's easy to change)
> >>>
> >>> FWIW, that was my assumption. I do not personally think that is a big
> >>> deal. To me the key is that revprops are not stored in SQLite unless
> >>> you explicitly pack the repository.
> >>
> >> Good info. Can one assume that the SQLite db is only updated during
> >> an svnadmin pack command? If so, then one just needs to ensure that
> >> the snapshot/backup/copy does not occur during that window of time....
> >
> > The reason SQLite was used is that rev props can be edited via svn
> > prop*--revprop
> >
> It should be easy to implement editing revprops without using SQLite:
> in case someone modify revprop non-packed revprop file is created, in
> read operation non-packed revprop file should be considered as more
> up-to-date. In next svnadmin pack operation these non-packed files
> should be merged back to packed one. Daniel, is it makes sense for
> you?
So you want the unpacked shard to be preferred to the packed one when
both exist.
But when you have just finished packing a revprops shard, you have to
increment db/min_unpacked_revprop before deleting any revprop shards ---
and during that window, both the unpacked shard and the packed shard
exist, but you need to prefer the packed one.
>
> --
> Ivan Zhakov
Received on 2011-06-30 22:04:20 CEST