When creating a new branch, obviously everyone will need to download it for maintenance and etc. The naive approach is to simply update and download the new branch from the server. However, with a large branch and with many users simultaneously updating, this quickly overloads the bandwidth and may end up takings hours to complete.
With 1.6, we would take advantage of being able to copy all of trunk to the new branch directory on the local filesystem, and then switch to the actual branch URL and update, therefore only transferring files that actually differ. This is pretty much the process touched upon in the svnbook http://svnbook.red-bean.com/en/1.7/svn.branchmerge.switchwc.html In 1.7 this does not seem to be possible, since there are no entries in the database for the local copies. I've tried a few different approaches but all seem to end up downloading everything from the server. We use TortoiseSVN primarily, as well as the usual command line tools. I know using sparse checkouts can help with not having to download what is not needed, but the paths we need are still quite large. I know I can switch the existing working copy to the new URL as well, but I need to be able to work on both. Despite a few hours of searching through Google and the mailing lists, I've come up with nothing, not even a solid answer on whether this is still possible or not. It seems that there must be some way to I have a feeling I am overlooking something simple. Regardless, how should one handle this situation? Am I stuck downloading a few gigabytes over a relatively slow connection? Thanks for any help, -- - Adam D. Walling
