[cvsnt] Re: Latest Updates

[Tony Hoyle]

Oliver Giesen wrote:
> Seeing that you can NOT commit to HEAD, and going by the gut feeling that a
> Head is something that's at the end/top of something rather than the
> something itself, I'm still prefering the view that HEAD is a symbolic name

You can commit to HEAD though...  That's what you're doing when there's 
no branch.  'cvs commit -r HEAD foo.txt' works too...

Try:
cvs update -r HEAD a.txt
cvs status a.txt
cvs commit -fm "test" a.txt

..this is clearly a commit to HEAD :)

>>(with the caveat the if you've changed the default branch it'll be the
>>wrong one, but few people ever do that).
> 
> 
> Hmm, don't like that stance...
>  
It's the only one practical given the fact that changing the default 
branch isn't versioned.

TBH I'm not sure myself even how to change the default branch, and I've 
never had a need to do it.  That doesn't mean there aren't people who do 
it, but I doubt it's the norm.

> 
>>I need to generalize the parsing one day, to allow that kind of syntax
>>everywhere (not just update) and put in the HEAD.1, foo.1 kind of syntax.
> 
> 
> I'm all for the BranchName.# syntax to replace the current revision
> numbering scheme. I'd just prefer the trunk not be named HEAD but something
> more "trunky", like TRUNK or MAIN...
>
I've always referred to the 'HEAD branch' and so have a lot of other 
people, so it seems it's the common usage.  It may not be the best name 
for it, but it's the name we're stuck with.

Defining a different name for HEAD is one of those things that's on the 
wishlist (low priority I think though).

Tony