Re: Can anyone correct the Bazaar reference "revno:111954.1.97"?

[Joshua Judson Rosen <rozzin@geekspace.com>]

"Eric S. Raymond" <esr@thyrsus.com> writes:
>
> Oh, crap.  Now I'm confused about what to replace that revno with.  The
> alternatives are:
>
> 111964.1.6	2013-03-15T16:03:54Z!handa@gnu.org
> 112051   	2013-03-15T16:06:12Z!handa@gnu.org

Eeek--sorry, I really meant that more as an FYI for anyone who was
already confused. "Don't Panic".

> It's only a difference in timestamp.  Do they have the same content?
> If not, serious can of worms.  If so, the qustion of which timestamp
> to consider authoritative becomes more philosophical.

It does look like the merge (112051) is indeed `just a merge'.

Actually going through and comparing the diffs ("bzr diff -c 112051"
vs. "bzr diff -c 11964.1.6"), the diffs off the two revisions are
identical except for slight differences in context, and that's explained
by the fact that 112051 looks like this:

    $ bzr log --line -r 112051 -n0
    112051: K. Handa 2013-03-16 [merge] Optimize ASCII file reading...
      111964.1.7: K. Handa 2013-03-16 [merge] merge trunk
      111964.1.6: K. Handa 2013-03-16 Optimize ASCII file reading...
      111964.1.5: K. Handa 2013-03-11 [merge] merge trunk

i.e.: the only thing that's at all different between the two
_changes_ is a slight difference in context, due to their different
positions in the DAG. Had the "Optimize ASCII file reading..."
change happened _after_ the second "merge trunk" (11964.1.7)
instead of before it, then I do think even the diff contexts
would be the same (the revision that 11964.1.7 was merging into
Handa's "work" branch was trunk 12050).

-- 
"'tis an ill wind that blows no minds."