On Wed, 27 Mar 2002 00:05:56 -0700, James Knowles <[EMAIL PROTECTED]> wrote:
> > The primary reason for wanting CVS repository data in XML format is for > > data interrogation in order to populate a database. > > Ah! This is very useful information IMHO. > > Unfortunately I'm not aware of any pre-built softare that does this. Others > may be able to complete my gap in knowledge. > > If there isn't, then: > > 1) If your needs are narrow, it seems it would be pretty simple to write a > small wrapper around CVS. This could be run locally as a script or as a > daemon listening on a port. (I'm assuming a Un*x environment where such > things are common practice. Windows IMHO is a big pain in this area.) Yes, it's *nix. > 3) A simpler possibility if you need information to be collected upon commit > (for example), one could trigger a program per the various "*info" files in > CVSROOT. That would be useful. But would make it difficult to add a new project. When first starting off, you need to populate the database to get a starting point. For ongoing data collection as the project proceeds, data is collected from the mailing list messages produced upon each commit. > 4) Others smarter than me probably have better ideas. I want to avoid starting another project. If the feature/tool is there, I'll use it and expand my own project. Adding and supporting another project will only dilute my existing resource pool. But thanks. _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs