convert: keep branch switching merges with ancestors (issue3340)
When running convert with a filemap, merge parents which are ancestors
of other parents are ignored. This is hardly a problem when parents
belong to the same branch, but the result could be confusing when named
branches are involved. With:
-o-a1-a2-a3... <- A
\ \
b1-b2-b3...-m- <- B
If all b* revisions are discarded, it is useful to preserve 'm' even if
it is empty after filtering to record the branch switch.
This patch makes filemap preserve "ancestor parents" if there is no
"non-ancestor parent" on the same branch than the merge revision.
Remarks:
- I am not completely convinced by the reasons given above and those
detailed by Matt in this thread:
http://selenic.com/pipermail/mercurial-devel/2012-May/040627.html
The properties we try to preserve are not clearly defined. That said,
I know this patch already helped someone on IRC and the tests output
look reasonable.
- This is a new version of the original "convert: filemap must preserve
fast-forward merges" patch. It has exactly the same output for 2
parents merges, the additional complexity is here to handle more than
two parents.
Mercurial for Plan 9 from Bell Labs
===================================
This directory contains support for Mercurial on Plan 9 from Bell Labs
platforms. It is assumed that the version of Python running on these
systems supports the ANSI/POSIX Environment (APE). At the time of this
writing, the bichued/python port is the most commonly installed version
of Python on these platforms. If a native port of Python is ever made,
some minor modification will need to be made to support some of the more
esoteric requirements of the platform rather than those currently made
(cf. posix.py).
By default, installations will have the factotum extension enabled; this
extension permits factotum(4) to act as an authentication agent for
HTTP repositories. Additionally, an extdiff command named 9diff is
enabled which generates diff(1) compatible output suitable for use with
the plumber(4).
Commit messages are plumbed using E if no editor is defined; users must
update the plumbed file to continue, otherwise the hg process must be
interrupted.
Some work remains with regard to documentation. Section 5 manual page
references for hgignore and hgrc need to be re-numbered to section 6 (file
formats) and a new man page writer should be written to support the
Plan 9 man macro set. Until these issues can be resolved, manual pages
are elided from the installation.
Basic install:
% mk install # do a system-wide install
% hg debuginstall # sanity-check setup
% hg # see help
A proto(2) file is included in this directory as an example of how a
binary distribution could be packaged, ostensibly with contrib(1).
See http://mercurial.selenic.com/ for detailed installation
instructions, platform-specific notes, and Mercurial user information.