[ see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=426191 ]
The text in the description of the --append option may lead one to
believe that files that are shorter on the receiving side won't be
updated, due to the following text:
...
Only files on the receiving side that are shorter than the
corresponding file on the sending side (as well as new files)
are sent. ...
IMHO this text should be changed to:
Only files on the receiving side that are shorter than the
corresponding file on the sending side are affected by this option.
Other files are transferred in the usual manner.
Paul Slootman
Firstly, thanks for the patches. I've checked in my slightly tweaked versions into CVS. On Wed, Aug 01, 2007 at 04:02:03PM +0200, Paul Slootman wrote:> The text in the description of the --append option may lead one to > believe that files that are shorter on the receiving side won't be > updated, due to the following text:That is true as long as you meant to say sending side instead of receiving side. Files that are shorter on the _sending side_ are indeed skipped. There was a bug where files with an identical size were getting updated, and I've just checked in a fix for that. ..wayne..
On Thu 02 Aug 2007, Matt McCutchen wrote:> On 8/2/07, Paul Slootman <paul@debian.org> wrote: > > I thought that --append simply meant that in cases where the destination > > file is shorter, data is simply appended without first checksumming the > > existing data, but that transfers would otherwise not be affected in any > > way, i.e. be performed in the usual manner as if --append was not > > specified. In other words, that --append was simply an optimization > > hint.> Oddly, rsync does checksum the existing data. At first I thought thisOnly after the transfer is complete, no? Not during, as it does normally. Paul Slootman