Loic Dachary wrote: > > It might actually be a lot faster to -d /cvsroot/$val (without > going thru pserver). Or even, if it turns out to be simple enough, to > parse the history file directly. But it may not be worth the trouble if > using /cvsroot/$val turns out to be efficient enough. The machine is > not exactly overloaded anyways.
your probably right. I dont know all that much about cvs. I learned enough to import/commit code into cvs for phpGW and I goofed around with a few things to get the stats that i did, but hadnt ever used cvs locally. So much to learn... so much > > I also plan to figure out if there is a way to find out about how many > > checkouts have been done, and stuff like that. > > I do have the cvs command to find out how many unique lines of code has > > changed during a given period of time, but to do it I need a checkout of > > the project. So I dont think this will be worth the effort. Commit > > counts and checkout counts would be enough to be interesting > > Asking for O type records should do it. Not sure what you mean. Can you give me a sample command? > And graphing the output > with rrdtool || mrtg, and ... Oh well, raw data will be great ;-) Yes, nice graphs will be on the todo list for me.
