I just tried this in a workspace I don’t need to work in. It looks like the –force option in checkout does exactly that, though it still downloads everything from the server. Copying the .svn workspace from another workspace seems to be a lot faster.
What I had in mind was an option that re-uses the workspace and the existing pristine copies in the .svn folder, so only properties and metadata needs to be downloaded from the server. From: Nathan Hartman <hartman.nat...@gmail.com> Sent: 12 August 2019 13:17 To: Jens Restemeier <j...@playtonicgames.com> Cc: users@subversion.apache.org Subject: Re: sqlite[s11]: database disk image malformed On Mon, Aug 12, 2019 at 7:19 AM Jens Restemeier <j...@playtonicgames.com <mailto:j...@playtonicgames.com> > wrote: - would it be possible for "svn checkout" to only download the metadata from the server, and keep the workspace files unchanged? "svn checkout" can actually check out into a directory that already exists and contains files. When a file already in the directory has the same name as an incoming file, it is said to be "obstructing" and isn't changed, instead marking it as a conflict (I think). I don't remember the exact details. All I know is that I've done it before, it worked, and afterwards I had to do a "svn revert -R ." in the working copy to un-mark everything and then it was fine. I'm away from my computer so I can't copy and paste the exact explanation but the output of "svn help checkout" (and most likely the svn-book) should tell you about this.
