Origin of `import-1.1.1` branch generated by `cvs-fast-export`

85 views Asked by At

I'm trying to migrate cleanly to git the CVS repository for osdsh at sourceforge. It has only 54 CVS commits that get turned into a dozen git commits, so nothing too complicated I guess. I'm interested in trying to keep the intent of the original developer. Both for a conservation stand point, but also to make the history useful as I maintain the software.

I used cvs-fast-export which produce a mostly-ok Git repository that I can edit with reposurgeon.

There are, however, a few quirks I don't understand. Here's the commit DAG in git as generated by cvs-fast-export.

$ git log --all --oneline --graph --decorate
* 13e2a20 (HEAD -> master) *** empty log message ***
* d5a2ff4 Added commands showconntime, connection, connected and disconnect Improved tkosdshconfig to reflect those changes
* e48575f *** empty log message ***
* 152be2b More and more bugfixes, added a Tcl/Tk configuration utility. Some cleanups.
* 5246123 *** empty log message ***
* 339e309 Added the correct header variables, as in "fixed"
* f13457d Initial Release, experimental osdsh with "libxosdsh"
* da7d830 (tag: start) Initial release, experimental osdsh with "libxosdsh"
| * 57238ca (import-1.1.1) Initial release, experimental osdsh with "libxosdsh"
|/  
* 90d94ee *** empty log message ***
* 0395cb0 Initial sources, normal osdsh
* c2df914 *** empty log message ***

Where does this import-1.1.1 branch come from? It has obviously been generated by cvs-fast-export, but I'm unable to understand what information was used to make this branch. Branches in CVS remain a mystery to me.

Also, as far as I can tell from the output of cvs log, the symbolic name start, when it exists, always refer to the very first revision of each file. Therefore, why is the corresponding tag not on the earliest commit in the git repository?

I'm probably just not versed well enough in CVS and would appreciate some pointers.

0

There are 0 answers