s...@apache.org wrote on Thu, Dec 15, 2011 at 11:03:08 -0000:
> Author: stsp
> Date: Thu Dec 15 11:03:08 2011
> New Revision: 1214697
>
> URL: http://svn.apache.org/viewvc?rev=1214697&view=rev
> Log:
> New and improved implementation of 'hotcopy' for FSFS.
...
> Modified: subversion/trunk/subversion/libsvn_repos/repos.c
> URL:
> http://svn.apache.org/viewvc/subversion/trunk/subversion/libsvn_repos/repos.c?rev=1214697&r1=1214696&r2=1214697&view=diff
> ==============================================================================
> --- subversion/trunk/subversion/libsvn_repos/repos.c (original)
> +++ subversion/trunk/subversion/libsvn_repos/repos.c Thu Dec 15 11:03:08 2011
> @@ -1777,17 +1793,29 @@ lock_db_logs_file(svn_repos_t *repos,
> +svn_repos_hotcopy2(const char *src_path,
> {
> + SVN_ERR(svn_dirent_get_absolute(&src_abspath, src_path, pool));
> + SVN_ERR(svn_dirent_get_absolute(&dst_abspath, dst_path, pool));
> + if (strcmp(src_abspath, dst_abspath) == 0)
Noting from IRC that this should resolve symlinks.
> + return svn_error_create(SVN_ERR_INCORRECT_PARAMS, NULL,
> + _("Hotcopy source and destination are equal"));
> Modified: subversion/trunk/subversion/svnadmin/main.c
> URL:
> http://svn.apache.org/viewvc/subversion/trunk/subversion/svnadmin/main.c?rev=1214697&r1=1214696&r2=1214697&view=diff
> ==============================================================================
> --- subversion/trunk/subversion/svnadmin/main.c (original)
> +++ subversion/svnadmin/main.c Thu Dec 15 11:03:08 2011
> @@ -332,8 +332,10 @@ static const svn_opt_subcommand_desc2_t
>
> {"hotcopy", subcommand_hotcopy, {0}, N_
> ("usage: svnadmin hotcopy REPOS_PATH NEW_REPOS_PATH\n\n"
> - "Makes a hot copy of a repository.\n"),
> - {svnadmin__clean_logs} },
> + "Makes a hot copy of a repository.\n"
> + "If --incremental is passed, data which already exists at the
> destination\n"
> + "is not copied again (for FSFS repositories only).\n"),
I'd like to clarify what happens for 'hotcopy --incremental' for BDB.
This sounds like it's silently ignored...
How about saying "[for FSFS; not implemented for BDB]"?
