On Thu, 27 Aug 2015, Andreas Schwab wrote:
> Ian Lance Taylor <i...@google.com> writes:
>
> > On Wed, Aug 26, 2015 at 2:29 PM, Eric S. Raymond <e...@thyrsus.com> wrote:
> >>
> >>> > rolfh = rolfh <rolfh>
> >>>
> >>> I think this is Arne H. Juul (was <ar...@lise.unit.no>).
> >>
> >> How did rolfh turn into arne or vice-versa? I'd like to have confirmation
> >> on this one.
> >
> > I don't know where the user name comes from, but this one is from
> > multiple ChangeLog entries that matches commits. I see that Joseph
> > got the same result.
>
> Arne Juul's contributions were committed by different people over time,
> so I don't think he had an account on his own. (There are entries in
> ChangeLog.6, ChangeLog.8 and ChangeLog.10.)
Here is my reasoning regarding the rolfh account. There are just two
commits:
r2512 | rolfh | 1992-10-18 17:50:52 -0700 (Sun, 18 Oct 1992) | 2 lines
r2376 | rolfh | 1992-10-08 18:48:26 -0700 (Thu, 08 Oct 1992) | 2 lines
In the gcc2 repository, these correspond to r1.73 and r1.65 of
Makefile.in,v. The relevant ChangeLog.6 entries are:
Sun Oct 18 17:52:05 1992 Arne H. Juul (ar...@lise.unit.no)
* Makefile.in (libobjc.a): Remove libobjc.a in current directory
before attempting to hardlink it or make will fail in later stages.
(sublibobcj.a): Don't use $(MAKEFLAGS) for objc subdir stuff.
Thu Oct 8 18:45:00 1992 Arne H. Juul (ar...@lise.unit.no)
* Makefile.in (install-common-headers): added missing 'then'
So rolfh corresponds to Arne H. Juul as author, whoever the committer
might have been, and so that seems to be the best mapping for that
username.
--
Joseph S. Myers
jos...@codesourcery.com
