Donald Sharp wrote:
> > 2) Will a symbolic tag be included in the history if it was specified in the tag
> > command? e.g. 'cvs tag -roldtag -b newtag" would list "oldtag" somewhere as well
>as
> > "newtag". I know this seems somewhat redundant, but it gives the user the ability
> > to make quick analyses such as, "Oh, Betty created a maintenance branch from the
>4.0
> > release"... Perhaps a "from rev" (see below) format similar to, "symtag:numrev"
> > when necessary?
>
> What would you like this too look like( as an example )?
>
> T 2000-11-29 15:22 +0000 sharpd bar.c from <tag/rev> on <rev> [tag_b:A]
>
> >
> > 3) I'd like to see a "from rev" and a "to rev" for a move ("M") command. I know
> > the information is available form playing with taginfo.
>
> Would this be ok?
>
> T 2000-11-29 15:22 +0000 sharpd bar.c from <tag/rev> to <rev> [tag_b:M]
How about something like this:
L 2000-11-29 15:22 +0000 sharpd [AMD] <fromtag:fromrev> <totag:torev> bar.c repo/path
for all records? [AMD] represents an opcode (add | move | delete = A | M | D). And
revision specs would actually be <tag:rev> or just <rev> when a tag wasn't available.
So some examples:
L 2000-11-29 15:22 +0000 sharpd A release_1_0:1.3 bar.c proj1/sdir
L 2000-11-29 15:22 +0000 sharpd A release_1_0:1.113 ChangeLog proj1/sdir
L 2000-11-29 15:23 +0000 sharpd M release_1_0:1.3 release_1_0:1.4 bar.c proj1/sdir
L 2000-11-29 15:23 +0000 sharpd M release_1_0:1.113 release_1_0:1.114 ChangeLog
proj1/sdir
L 2000-11-29 15:25 +0000 sharpd A release_1_0:1.4 maint_1_0:1.4.2 bar.c proj1/sdir
L 2000-11-29 15:25 +0000 sharpd A release_1_0:1.114 maint_1_0:1.114.2 ChangeLog
proj1/sdir
L 2000-11-29 15:22 +0000 sharpd A release_1_0_1:1.4 bar.c proj1/sdir
L 2000-11-29 15:22 +0000 sharpd A release_1_0_1:1.114 ChangeLog proj1/sdir
L 2000-11-29 15:22 +0000 sharpd D release_1_0_1:1.4 bar.c proj1/sdir
L 2000-11-29 15:22 +0000 sharpd D release_1_0_1:1.114 ChangeLog proj1/sdir
That's a little messy, but spaces delimit the fields nicely and machine readibility is
one of the philosophies driving the history command, I believe. I think some of the
other record types use fixed width field. I'm not sure that is usable in the long run,
but I'll let you figure that out.
Is this realizable with the current history setup? Reshuffling the information
wouldn't
bother me, as long as it is all still there and parsable.
> This is starting to get complicated. Should we rework the cvs history
> output for the 'T' record?
'L'? But whatever. I was thinking "Local" tag even though it might be applied to rtag
too. See my other email re why a new record type is indicated.
Derek
--
Derek Price CVS Solutions Architect ( http://CVSHome.org )
mailto:[EMAIL PROTECTED] OpenAvenue ( http://OpenAvenue.com )
--
I know of no safe depository of the ultimate powers of the society but the
people themselves, and if we think them not enlightened enough to exercise that
control with a wholesome discretion, the remedy is not to take it from them,
but to inform their discretion.
- Thomas Jefferson, 1820.
_______________________________________________
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
