Ar an t-ochtú lá déag de mí Lúnasa, scríobh Uwe Brauer:
Which gives me a graphic idea about the history of the repository.
(I
am also a fan of named branches, but this is a different topic). What I
find confusing though is to see several «branches» in a repo which do
not correspond really to branches.
To illustrate my point
Please try
hg log -G | grep 9624523604c5
Or
hg log | grep 9624523604c5
And you will see something very odd. I presume this was generated by the
conversion from CVS to mercurial and that is why I wanted to know with
which tools the conversation was done, and where to find the original
CVS.
No, those changes were made directly in Mercurial, and it was intentional,
Ben was doing a lot of work parallel to the trunk, with intermittent merging
of changes back and forth. This was distributed development working well.
The trunk has been in Mercurial since December 2007. The other repositories
moved to CVS more slowly.
There are problems with the CVS import, among them that changeset number
426, id 43177a4f3daf removed all the C files(!), so most of the C files look
like they date from changeset 428 on hg annotate. I have a suspicion that
dated from the underlying CVS tree, though.
I don’t know of a copy of the underlying CVS repository. They were last
hosted on
alioth.debian.org, which is no longer available.
--
‘As I sat looking up at the Guinness ad, I could never figure out /
How your man stayed up on the surfboard after forty pints of stout’
(C. Moore)