[GLLUG] More questions about Subversion

frank.dolinar at comcast.net frank.dolinar at comcast.net
Thu Feb 14 14:37:15 EST 2008


Hi, everyone,
    As we get started using Subversion here at work, we're running in to some problems sorting out how some things work.  No surprise there.

    Here's the situation:

    The developer here who loaded things out to the repository late yesterday simply did a copy of everything (and I do mean EVERYTHING) from the Visual Studio folder in which the current project was being set up on his C:\ drive and imported it to the Subversion repository.  The repository is not set up with the Trunk / Branches / Tags structure that you and the documentation recommended.  The content on the repository includes all the business layer code, the base classes / templates for our data entry / maintenance forms, and any other things he'd already written.   He happened to be out this morning and we were having some trouble getting things to work.  Having said that, when we did a checkout of the content of this folder from the Subversion repository it works, but I'm concerned that all 600+ files got copied to my C:\ drive.  [Yeah, I know that the first time it makes sense for me to get all of this the first time so I'll have it available.  Am I being inappropriatel
 y para
noid?]

    So, I have another round of questions for any one who chooses to answer.

1.  There have been several mentions of what kinds of files to put in the repository and which to exclude.  These suggest that only the developer code should go in the repository and that the generated code, such as an .exe (or in the case of Visual Studio, an .sln [aka "solution"] file).

2.  There are exceptions for including generated code / files in the repository.  I believe Clay said something about certain situations where binaries might reasonably be included.  I don't have that message with me here, so if I could get a rehash of that from someone that would be helpful.

3.  As noted above the structure is not Trunk / Branches / Tags.  The current structure includes folders for the following:  Business Objects;  Developer Code; classes for Master files / templates; web page application code;  web services; and the hidden   .svn  housekeeping folder.    How much problem is this going to cause us?   Should we find a way to reconfigure to Trunk / Branches / Tags structure?  Should each of these current folders actually have their own Trunk / Branches / Tags structure (which I would guess would be appropriate based on what I've seen so far in the documentation and in use) ?

4.  Any other comments and suggestions to help us clear up the confusion caused by a serious case of a group of newbies trying to lead each other (aka "Blonde leading the blonde.")

Thanks for all the help.

-Frank




More information about the linux-user mailing list