[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn commit: r1548214 - in /subversion/trunk/subversion: libsvn_wc/externals.c libsvn_wc/update_editor.c libsvn_wc/wc-queries.sql libsvn_wc/wc_db.c libsvn_wc/wc_db.h libsvn_wc/wc_db_update_move.c tests/libsvn_wc/op-depth-test.c

From: Stefan Sperling <stsp_at_elego.de>
Date: Fri, 6 Dec 2013 15:37:43 +0100

On Fri, Dec 06, 2013 at 03:18:27PM +0100, Bert Huijben wrote:
> It is a check if dst_cp_relpath is the root node or any of its descendants.
> The root of a move can never be shadowed (as the maximum op_depth of a node
> is its own op depth... which is the move)

Then why not use something like svn_relpath_skip_ancestor()?
I think using strcmp() in this context is misleading.

> A string comparision is quite heavy if we really only need this check. (The
> db query already assures that there will only be nodes at or below
> dst_relpath)
>
> Bert
>
> >
> > > @@ -1643,6 +1644,11 @@ replace_moved_layer(const char *src_relp
> > > scratch_pool));
> > > if (!err)
> > > err = svn_sqlite__step_done(stmt2);
> > > +
> > > + if (!err && strlen(dst_cp_relpath) > strlen(dst_relpath))
> > > + err = svn_wc__db_extend_parent_delete(wcroot, dst_cp_relpath,
> > kind,
> > > + dst_op_depth,
> scratch_pool);
> > > +
> > > if (err)
> > > return svn_error_compose_create(err, svn_sqlite__reset(stmt));
> > >
> > >
Received on 2013-12-06 15:38:25 CET

This is an archived mail posted to the Subversion Dev mailing list.