Re: Commit messages and the move to git

[Jonathan Wakely <jwakely dot gcc at gmail dot com>]

On Thu, 5 Dec 2019 at 10:36, Richard Earnshaw (lists)
<Richard.Earnshaw@arm.com> wrote:
>
> On 05/12/2019 10:32, Jonathan Wakely wrote:
> > On Thu, 5 Dec 2019 at 10:25, Jonathan Wakely <jwakely.gcc@gmail.com> wrote:
> >>
> >> On Wed, 4 Dec 2019 at 23:52, Richard Earnshaw (lists) wrote:
> >>> I've just pushed a new trial conversion:
> >>>
> >>> https://gitlab.com/rearnsha/gcc-trial2-20191130
> >>>
> >>> The main differences between this and the previous trial are:
> >>> - The author attributions should now be fixed, please let me know if you
> >>> see any anomalies in this respect.
> >>> - the emptycommit-* tags/branches are now gone.
> >>> - the 'tags' used for revert and backport now use more gittish style
> >>> revert: and backport:
> >>> - the log entries for c++ style functions containing :: are now handled
> >>> correctly by my summary generation script.
> >>>
> >>> Other issues are still being worked on.
> >>
> >> There's a bogus libstdc++ directory at the top level. That hasn't been
> >> present for nearly 20 years, so some commit that should have deleted
> >> it seems to be missing or incorrect.
> >
> > It was removed by r39445 i.e. this commit in the current Git mirror:
> >
> > commit 1030d23bf251914e12fc8c9d521e334e1f445afc
> > Author: mmitchel <mmitchel@138bc75d-0d04-0410-961f-82ee72b054a4>
> > Date:   Mon Feb 5 01:38:47 2001 +0000
> >
> >              Remove V2 C++ library.
> >              * configure.in: Remove --enable-libstdcxx_v3 support.
> >
> >              * Makefile.in (site.exp): Always set HAVE_LIBSTDCXX_V3.
> >              * configure.in: Remove --enable-libstdcxx_v3 support.
> >              * configure: Regenerated.
> >
> >
> >      git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@39445
> > 138bc75d-0d04-0410-961f-82ee72b054a4
> >
>
> Ok, this is one to keep an eye on.  There are a number of anomalous
> commmits at present, which Eric is working on with a new approach to
> replaying the SVN data into reposurgeon.  Once that is done we're hoping
> that this sort of problem will go away.  (of course, if it doesn't then
> the chances are we won't be using reposurgeon for the conversion at all).

OK. There are several other deletions missing, as I see these in your
conversion which aren't on trunk (paths relative to the libstdc++-v3
dir because that's all I'm checking):

Only in ./config/abi/post: solaris2.8
Only in ./config/os: irix
Only in ./config/os: osf
Only in ./config/os/solaris: solaris2.8
Only in ./include/ext/pb_ds/detail/cc_hash_table_map_: standard_policies.hpp
Only in ./include/ext/pb_ds/detail/gp_hash_table_map_: standard_policies.hpp
Only in ./src/c++98: compatibility-parallel_list-2.cc
Only in ./src/c++98: compatibility-parallel_list.cc
Only in ./testsuite/20_util: is_explicitly_convertible
Only in ./testsuite/26_numerics: cmath
Only in ./testsuite/28_regex/algorithms: 02_match
Only in ./testsuite/28_regex/basic_regex: regex.cc
Only in ./testsuite/28_regex/headers: 04_header


> This repo conversion is really for checking that the summaries I'm
> generating are not just garbled, and the authors problem from the
> previous trial are now fixed.

OK, apart from the artefacts caused by me not knowing my own email
address (discussed off-list), those look good to me.