* Bill Shannon <[EMAIL PROTECTED]> [2006-11-08 10:10]:
> >o hdrchk
> > This is purely a matter of teaching it 'the other' keyword format.
> > If the format remains what was initially chosen. I have part of this
> > done, though it currently accepts either SCCS or new-style keywords.
> >
> > Not particularly practical without having keywords for Mercurial
> > implemented, however.
>
> Is there a decision on, or a proposal for, an SCCS keyword replacement
> for Mercurial that I missed?
There is no decision as of yet; keywords handling is one of the topics
for the discussion. There are a variety of viewpoints, which could
become proposals:
- replace versions in strings with changeset hash (automated, loss of
monotonically increasing version number),
- require explicit updating of versions in strings (manual),
- change API to have explicit module versioning (technical change,
incomplete scope), and
- drop versions in strings altogether (policy change).
It would be good to hear about requirements other than loadable kernel
modules and ident pragmas.
- Stephen
--
Stephen Hahn, PhD Solaris Kernel Development, Sun Microsystems
[EMAIL PROTECTED] http://blogs.sun.com/sch/
_______________________________________________
tools-discuss mailing list
[email protected]
