| ??? 09/12/08 20:35 Read: times Msg Score: +1 +1 Informative |
#158234 - svn binary storage Responding to: ???'s previous message |
Marshall Brown said:
I run a wee NAS (Network Attached Storage) and have set SVN up on this as a file based server, so no other server is required, it is really easy to setup and once you are up and running you just right click in windows file manager, and press commit, add your comments, and you are all done. Do make sure that you access the file-based server (file:// semantics) only using SVN commands and you're fine. HOWEVER.
one of the downfalls that SVN has over CVS is that the files are stored in its own native binary file format, so when you browse to the SVN directory on your file server it is meaningless. This concerns me as if it gets corrupted you're stuffed. Most everything you can think of that is a data base stores the data in a binary format. Subversion is quite reliable, but like anything where your data are important, you DO back up your repository on a regular basis, right? Both SVN and CVS can store binary files, but they store the entire file, not the changes. (but who cares, you can still roll back you just end up storing lots of data that is essentially the same.) Absolutely not true of Subversion. See here: "SVN book" said:
In the most general sense, Subversion handles binary files more gracefully than CVS does. Because CVS uses RCS, it can only store successive full copies of a changing binary file. But internally, Subversion expresses differences between files using a binary-differencing algorithm, regardless of whether they contain textual or binary data. That means that all files are stored differentially (compressed) in the repository, and small differences are always sent over the network. -a |
