I have a Subversion repository accessed via AnkhSVN Subversion Plug-in with Visual Studio 2005 and 2013 in a Windows development environment that needs to be cleaned up. Reading stackoverflow Best practices for Subversion and Visual Studio projects there is a bit of clutter to remove from the trunk.
I also want to branch the trunk to a side stream that will have few to no updates for the current release so that I can concentrate on the next release in the trunk. Understanding branch use with Visual Studio 2010, AnkhSVN, and SVN 1.7.4 indicates that I need more than the AnkhSVN plug-in for actions other than updating the repository and/or working copy.
It appears that Tortoise SVN GUI client is needed as well as the AnkhSVN plug-in. Is this correct?
What I would like to have after a repository update and pruning exercise is a current version in the repository trunk which will not have unneeded files however if I have to pull an earlier version from say a year ago, those pruned files will be part of what is pulled.
How do I create an ignore list for AnkhSVN with files that are already in the repository so that it will ignore files in the working folder and not update them into the Subversion repository in the future? It appears that if a file is new and has not been put into the Subversion repository, I can choose to ignore that file from the Pending Updates window of Visual Studio however what about existing files.
What about files that are not displayed in Visual Studio IDE such as .dsw (left over solution file from Visual Studio 6.1) or .suo files which are in the repository and should be removed?
What are the repercussions of deleting currently unused files in the Subversion repository trunk? Does a delete delete all versions of the file in the trunk and branches or just the trunk? Does it do an actual delete or is the file just marked as deleted and unused for the future versions?
Is there a way to mark a file as being unused beginning with a particular update so that pulling from the repository using that update or later will not include files marked as unused? However if an earlier version is pulled, perhaps for researching a field report of an earlier version, the file which may need needed for that earlier version be available?
It looks like AnkhSVN has a merge capability to allow me to merge changes from trunk into a branch. It looks like I would pull down a clean copy of the branch and then do a merge into my working copy from the trunk. Then I would clean up any conflicts, test, and then update the branch with the merged working copy. Is that correct?